Quantitative reasoning and analysis:Yes

$FL2@(#) IBM SPSS STATISTICS 64-bit MS Windows 21.0.0.0 � � � ê Y@28 Jun 1609:28:15 � � � �� �� PRESTG10% Rs occupational prestige score (2010) � � � ��� ��� SEI10 � R's socioeconomic index (2010) � � � �� �� AGE � AGE OF RESPONDENT €X@ ÀX@� � � �� �� CHILDS � NUMBER OF CHILDREN "@� � � �� �� CITIZEN / ARE YOU A CITIZEN OF AMERICA? @ "@� � � �� �� COLSCINM0 NUMBER OF COLLEGE-LEVEL SCI COURSES R HAVE TAKEN €X@ ÀX@� � � �� �� CONINC ! FAMILY INCOME IN CONSTANT DOLLARS ~„.A |„.A� � � �� �� CONRINC % RESPONDENT INCOME IN CONSTANT DOLLARS ~„.A |„.A� � � �� �� DEGREE � RS HIGHEST DEGREE �@ @ "@� � � �� �� DWELOWN � DOES R OWN OR RENT HOME? @ "@� � � �� �� EDUC HIGHEST YEAR OF SCHOOL COMPLETED @X@ €X@ ÀX@� � � �� �� EMAILHR � EMAIL HOURS PER WEEK ð¿ 0�@ 8�@� � � �� �� GRASS � SHOULD MARIJUANA BE MADE LEGAL @ "@� � � �� �� HRS2 # NUMBER OF HOURS USUALLY WORK A WEEK ð¿ €X@ ÀX@� � � �� �� MARITAL � MARITAL STATUS "@� � � �� �� OWNGUN � HAVE GUN IN HOME @ "@� � � �� �� PARTYID � POLITICAL PARTY AFFILIATION @ "@� � � �� �� RACE � RACE OF RESPONDENT � � � �� �� REALINC � FAMILY INCOME IN CONSTANT $ ~„.A |„.A� � ýÿÿÿ �� �� RINCOME � RESPONDENTS INCOME *@ ÀX@ � � � �� �� SEX � RESPONDENTS SEX � � � �� �� SEXORNT � SEXUAL ORIENTATION @ "@� � � �� �� SIZE � SIZE OF PLACE IN 1000S ð¿� � � �� �� TVHOURS � HOURS PER DAY WATCHING TV ð¿ €X@ ÀX@� � � �� �� UNION DOES R OR SPOUSE BELONG TO UNION @ "@� � � �� �� USCITZN � IS R US CITIZEN @ "@� � � �� �� WKRACISM% R FEELS DISCRIMINATED BECAUSE OF RACE @ "@� � � �� �� WKSEXISM' R FEELS DISCRIMINATED BECAUSE OF GENDER @ "@� � � �� �� WRKSLF R SELF-EMP OR WORKS FOR SOMEBODY @ "@� � � �� �� WRKSTAT � LABOR FORCE STATUS "@� � � �� �� WWWHR � WWW HOURS PER WEEK ð¿ 0�@ 8�@� � IAP,DK,NA � � � � � IAP,DK,NA � � � � � @V@�89 OR OLDER €X@�DK ÀX@�NA � � � � � @ EIGHT OR MORE "@�DK NA � � � � � �IAP ð?�YES @�NO @�DK "@�NA � � � � � �IAP €X@ DONT KNOW ÀX@ NO ANSWER � � � � � �IAP |„.A DONT KNOW ~„.A�NA � � � � � � � �LT HIGH SCHOOL ð?�HIGH SCHOOL @�JUNIOR COLLEGE �@�BACHELOR �@�GRADUATE �@�IAP @�DK "@�NA � � � � �IAP ð?�OWN OR IS BUYING @ PAYS RENT �@�OTHER @�DK "@�NA � � � � @X@�IAP €X@�DK ÀX@�NA � � � � � ð¿�IAP 0�@�DK 8�@�NA � � � � � � �IAP ð?�LEGAL @ NOT LEGAL @�DK "@�NA � � � � ð¿�IAP €X@�DK ÀX@�NA � � � � � � ð?�MARRIED @�WIDOWED �@�DIVORCED �@ SEPARATED �@ NEVER MARRIED "@�NA � � � � � �IAP ð?�YES @�NO �@�REFUSED @�DK "@�NA � � � � �STRONG DEMOCRAT ð?�NOT STR DEMOCRAT @�IND,NEAR DEM �@�INDEPENDENT �@�IND,NEAR REP �@�NOT STR REPUBLICAN �@�STRONG REPUBLICAN �@�OTHER PARTY @�DK "@�NA � � � � � �IAP ð?�WHITE @�BLACK �@�OTHER � � � � � �IAP ð?�LT $1000 @ $1000 TO 2999 �@ $3000 TO 3999 �@ $4000 TO 4999 �@ $5000 TO 5999 �@ $6000 TO 6999 �@ $7000 TO 7999 @ $8000 TO 9999 "@�$10000 - 14999 $@�$15000 - 19999 &@�$20000 - 24999 (@�$25000 OR MORE *@�REFUSED €X@�DK ÀX@�NA � � � � � ð?�MALE @�FEMALE � � � � � �IAP ð?�Gay, lesbian, or homosexual @�Bisexual �@�Heterosexual or straight @ DONT KNOW "@�NA � � � � � ð¿�NOT ASSIGNED � � � � � �IAP ð? R BELONGS @�SPOUSE BELONGS �@�R AND SPOUSE BELONG �@�NEITHER BELONGS @�DK "@�NA � � � � � �IAP ð?�A U.S. Citizen @�Not a U.S. Citizen �@]A U.S. CITIZEN BORN IN PUERTO RICO, THE U.S. VIRGIN ISLANDS, OR THE NORTHERN MARIANAS ISLANDS �@`BORN OUTSIDE OF THE UNITED STATES TO PARENTS WHO WERE U.S CITIZENS AT THAT TIME (IF VOLUNTEERED) @ DONT KNOW "@�NA � � � � � �IAP ð?�YES @�NO @ DONT KNOW "@ NO ANSWER � � � � � � �IAP ð? SELF-EMPLOYED @�SOMEONE ELSE @�DK "@�NA � � � � �IAP ð?�WORKING FULLTIME @�WORKING PARTTIME �@�TEMP NOT WORKING �@�UNEMPL, LAID OFF �@�RETIRED �@�SCHOOL �@ KEEPING HOUSE @�OTHER "@�NA � � � � ¸x document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 04 OCT 05) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 JUL 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 JUL 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 JUL 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-1998 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-1998 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' CARDS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP RACHISCH SOCREL TO SOCSIBS MEMNUM INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 SWITNUM TO SWITAGE2 CHURHSCH TVRELIG CIDEKNEW AIDSKNOW PARTNERS PARTNERS5 PARTOPEN PARTOPN5 SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME WORDA TO WORDSUM NUMKIDS SEXFREQ NUMWOMEN NUMMEN CIDEREG CIDEREG2 CIDEREG3 AIDSREG AIDSREG2 AIDSREG3 SUIREG SUIREG2 SUIREG3 SUIKNEW SEXFREQ2 DK RDK UNDK WEEKSWRK SIZE AGE1 TO AGE5 WRKYEARS JOBYEARS EMPYEARS COLMAJR1 TO HSCLASS2 ALIKE1 TO ALIKE8 VISITORS FSTSPDEG RELATIVES SPBIBS SPFAM16 SIBORDER OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD10 OLD11 OLD12 OLD13 OLD14 YEAR ****************************************************** *** IMPORTANT NOTE *** ****************************************************** THIS VERSION OF THE 1972-96 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THIS PURPOSE. THUS WEIGHTED WILL ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-1998 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 18 APR 01) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 12 JAN 07) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 09 MAR 06) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 09 MAR 06) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 09 MAR 06) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WT2004 OR WT2004NR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 09 MAR 06) document Length of Interview in 2004 (LNGTHINV) This is based on the CAPI time stamps from the start of the main interview (excluding the HEF) to the end of the CDC risk behavior section. Initially, we had also included the interviewer remarks and validation section in the total time, but had to exclude this section when we found that in a high number of cases interviewers had not finished this section and turned off the computer immediately after the interview, but left it on and finished their remarks at a later time. This greatly exaggerated the length of this final section and thus the total supposed interview time. There were two cases with no times recorded and we recoded two more cases to missing. One shows a total time of 10 seconds. Another has a total time of 28786 seconds, but with its length getting shorter as sections were completed. That still left a number of cases in either tail of the distribution that were highly improbable. About 1% of cases were under 30 minutes and 1% over three hours. We examined the section-by-section times for the long cases and inspected the record of calls for a sample of these cases. The section-by-section timings often indicated extra-long timing for a single section. This may indicate an extended interruption occurring them. However, many of these extra long times were so long (1-2 hours) that they may not represent mere interruptions, but other occurrences such are errors in the CAPI timings or having the interview conducted in two sessions. Unfortunately for most cases the record of calls had no useful information to help explain the unusually long times. However, among the extra long cases for which there was some useful information, many were done in two or more sessions on different days and it is likely that the CAPI program did not correctly combine these different times. Also, the record of calls did partially validate one interview of almost 4 hours. The interview noted that it was done in between the respondent waiting on customers. We likewise looked at the extremely short cases. There was little information that we could find to explain or validate these lengths. Some may be valid since they did overrepresent shorter ballots and respondents (e.g. those not in the labor force) who skipped major sections. However, some also involved cases completed over two or more different dates and we believe that they may reflect only partial times from one of the sessions. Overall, we believe that the extremely short and long cases include a high, but not precisely known, number of errant times. In addition, there are undoubtedly some timing errors among cases in the less extreme range. Looking at reported times (excluding the four missing cases) shows a mean length of 87.3 minutes and a median of 83 minutes. If low times are recoded to a minimum of 30 minutes and high times to a maximum 180 minutes, the mean length is 86.9 and the median still 83 minutes. (Entered 19 SEP 07) document Since CAPI 2002 (Computer Assisted Personal Interview) provides various information on the length of interview time, there are three elasped interview time variables newly added to the remp7202p.sys. The original data set of GSS2002 has three interview time variables. First, it provides a cumulative elapsed time for Section A to I. Second, it also gives an elapsed time for SAQ only. Lastly, it contains a total elapsed time variable for the entire interview. In order to utilize all information in the original data, we created three interview time variables as follows: LNGTHEND: TOTAL ELASPED MINUTES OF INTERVIEW for 2002 LNGTHCUM: CUMULATIVE ELASPED MINUTES FOR SECTIONS A-I for 2002 LNGTHALL: SUMMED ELASPED MINUTES OF SECTION A-I AND SAQ for 2002 (Entered 19 SEP 07) DOCUMENT WHOELSE1-WHOELSE6 HAVE BEEN INCLUDED IN GSS SINCE 1994. IN 1994, THEY WERE ONLY ASKED FOR VERSION 7-9 (Entered 19 SEP 07) DOCUMENT THIS SYSTEM FILE CONTAINS DATA FOR THE 1972-2004 GENERAL SOCIAL SURVEYS, CONDUCTED BY THE NATIONAL OPINION RESEARCH CENTER. *** IT IS STRONGLY RECOMMENDED THAT USERS OF THE *** *** FILE CONSULT THE CUMULATIVE G.S.S. 1972-2004 *** *** CODEBOOK. *** THE LABELS IN THIS FILE ARE BELIEVED TO BE ACCURATE, BUT ARE NO SUBSTITUTE FOR THE QUESTION WORDING AND RESPONSE CATEGORIES AS PRESENTED IN THE CODEBOOK. USERS SHOULD PAY PARTICULAR ATTENTION TO THE TREATMENT OF MISSING DATA. THE 'MISSING VALUES' STATETMENTS ARE INCLUDED IN THIS FILE-CREATION PACKAGE SIMPLY AS A CONVENIENCE. FOR THE MOST PART, THE CATEGORIES 'DONT KNOW', 'NO ANSWER', AND 'NOT APPLICABLE' HAVE BEEN DECLARED MISSING. *** USERS ARE THEREFORE ENCOURAGED TO RE-DECLARE MISSING *** VALUES WHEN ANALYZING THE DATA. *** FOR CERTAIN VARIABLES, BLANK AND ZERO ARE BOTH VALID CODES. TO AVOID CONFUSION, BLANKS IN THE RAW DATA HAVE BEEN RECODED TO NEGATIVE ONE (-1) FOR THESE VARIABLES ONLY. THE RECODE IS DONE ONLY WHERE BOTH BLANK AND ZERO ARE VALID CODES. THEREFORE, WE RECOMMEND THAT USERS MARK IN THEIR CODEBOOKS THE VARIABLES FOR WHICH THIS RECODE HAS BEEN MADE: HRS1 HRS2 COMMUTE SPHRS1 SPHRS2 PARBORN GRANBORN UNRELAT VETYEARS VETKIND DOTDATA TO PADOTPRE POLVIEWY RUSSIA TO EGYPT MAATTEND PAATTEND SPATTEND TITHING LIBTEMP TO JEWTEMP MSLMTEMP RACHISCH SOCREL TO SOCSIBS INCNEED MININC CHLDIDEL CHLDSOON TVHOURS RADIOHRS DIVORCE5 DIVREL1 DIVREL4 UNEMP5 UNREL1 UNREL4 HOSDIS5 HOSREL1 HOSREL4 DEATH5 DEATH16 PADEATH MADEATH CHLDDTH TO TRATOT5 DEFSPDR TO CUTSPDFG NUMGIVEN EDUC1 TO EDUC5 MEMNUM SWITNUM SWITAGE1 TO SWITAGE2 CHURHSCH TVRELIG WRKYEARS EMPYEARS JOBYEARS SISNUM BRONUM DAUNUM SONNUM GRPARNUM TO RELNUM FRINUM FRIWORK FRINEIGH LIVECOM LIVEHOME NUMKIDS CIDEKNEW CIDEREG CIDEREG2 CIDEREG3 SUIKNEW SUIREG SUIREG2 SUIREG3 AIDSKNOW AIDSREG AIDSREG2 AIDSREG3 PARTNERS SEXFREQ SEXFREQ2 NUMWOMEN NUMMEN PARTOPEN PARTOPN5 PARTNRS5 SIBORDER WORDA TO WORDJ WORDSUM AGE1 TO AGE5 WRKYEARS OLD1 OLD2 OLD3 OLD4 OLD5 OLD6 OLD7 OLD8 OLD9 OLD10 OLD11 OLD12 OLD13 OLD14 VISITORS COLMAJR1 TO HSCLASS2 KDEDUC1 TO KDEDUC9 SIZE SBPICKED KDPICKED INCOM16 FSTSPEDC FSTSPMED FSTSPPED SPFAM16 SPPAEDUC SPMAEDUC SBEDUC SBDEG ALIKE1 TO ALIKE8 SPSIBS LIVECOM1 NUMDAYS SHAKEBLU TO PROUD HOWANGRY REACTOK TO SELFRESP AMIMP HRSHLTH TO HRSOTH TOTHLTH TO VALOTH GIVECONG TO GIVEOTH SHOPNUM OCCYRS NUMFREND NUMBWFRD RELHRS2 RELHRS1 NUMCONG OTHYRS SITEYRS LOCYRS YRJOBEND WHYJBEND PAYMASON TO GIVROCC USWHT TO USMIXED EMSENT TO EMGETP WWWMIN WWWHR CHATMIN CHATHR EMAILMIN EMAILHR INCMINW INCHRW JOBMINW JOBHRW ECOMMINH TO WWWHRH ECOMMINW TO WWWHRW ECOMMINO TO WWWHRO ARTSMIN ARTSHR HLTHINFO TO ARTGN POLCANGN WAIT1 TO WAIT11 COMWHT TO COMAMIND PAYMASON TO GIVROCC NUMCNTCT WWWMINO WWWMINH WWWMINW WWWHRO WWWHRW WWWHRH MOREDAYS PHYSHLTH TO HLTHDAYS HURTATWK NUMPROBS BYEMPROB STOCKVAL EXTRAVAL YEARVAL SEECOWRK COWRKHRD TALKSUP COMYEAR NUMRELEX DRINKMAX NUMFRAT TO NUMINFRM YRFRAT1 TO YRINFRM5 YRSKWN1 TO YRSKWN5 DEMTODAY DEM10PST DEM10FUT CANINF1 CANINF2 CANINF3 CANINF4 CANINF5 CANINF6 CANINF7 CANINF8 CANINF9 CANINF10 CANINF11 CANINF12 CANINF13 CANINF14 YEARSJOB SIBNUM KIDNUM FININD1 OWNHH1 EDDONE1 FTWORK1 SUPFAM1 HAVCHLD1 GETMAR1 COWRKFRD NEIFRD OTHFRD COMYEAR RHHWORK SPHHWORK JOBINFO JOBPAPER JOBPUB JOBCOWRK JOBCNTCT JOBFRNDS JOBPLACE JOBTVRAD JOBONWWW HRSRELAX AGERELEX AGERBORN ****************************************************** *** IMPORTANT NOTE ON WEIGHTS *** ****************************************************** THIS VERSION OF THE 1972-2004 CUMULATIVE GENERAL SOCIAL SURVEY REQUIRES WEIGHTING TO ADJUST FOR THE PLANNED OVERSAMPLING OF THE BLACK SUBPOPULATION IN 1982 AND 1987 and for the sub-sampling design used in 2004. A SET OF DECIMAL WEIGHTS IS PROVIDED FOR THE BLCAK OVERSAMPLES. THUS WEIGHTED THE FILE WILL BE ADJUST FOR THE OVERSAMPLE OF BLACKS. THERE ARE THREE CASES WHERE THE USE OF THE WEIGHTS IS APPROPRIATE: 1. THE USER WANTS TO POOL ALL SUBFILES FOR SUBSEQUENT ANALYSIS, E.G., RUN SUBFILES ALL 2. THE USER WANTS TO POOL BOTH THE 1982 (1987) SUBFILES, Y82 (Y87) AND Y82B (Y87B), THE TRADITIONAL CROSS-SECTIONAL SUBFILE AND THE BLACK OVERSAMPLE SUBFILE. 3. THE USER WANTS TO ANALYZE ONLY BLACKS FROM THE POOLED 1972-2004 SAMPLE. IN THESE THREE SITUATIONS IT IS NECESSARY TO WEIGHT THE DATA BY THE VARIABLE OVERSAMP. IF THE USER IS ANALYZING ONLY BLACKS FROM THE Y82 (Y87) AND Y82B (Y82B) SUBFILES, THE USE OF THE WEIGHT IS NOT NECESSARY. ALSO, WHEN USING ANY VARIABLES FROM THE 2004 GSS ONE OF THE WEIGHT VARIABLES, WTSS OR WTSSNR, MUST BE USED. THE REASON FOR USING THE WEIGHTS IS EXPLAINED IN APPENDIX A OF THE CODEBOOK. ******************************************************** * * NOTE THAT THIS IS A VERY LARGE FILE. USERS DOING * INTENSIVE ANALYSIS SHOULD CONSIDER SAVING AN ADDITIONAL * FILE CONTAINING ONLY THE CASES AND VARIABLES NEEDED. ******************************************************** (Entered 19 SEP 07) � � � � � Ð� � � � éý � � � � ÿÿÿÿÿÿïÿÿÿÿÿÿÿï�þÿÿÿÿÿïÿ� � � ] � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � µ� PRESTG10=prestg10 SEI10=sei10 AGE=age CHILDS=childs CITIZEN=citizen COLSCINM=colscinm CONINC=coninc CONRINC=conrinc DEGREE=degree DWELOWN=dwelown EDUC=educ EMAILHR=emailhr GRASS=grass HRS2=hrs2 MARITAL=marital OWNGUN=owngun PARTYID=partyid RACE=race REALINC=realinc RINCOME=rincome SEX=sex SEXORNT=SEXORNT SIZE=size TVHOURS=tvhours UNION=union USCITZN=USCITZN WKRACISM=wkracism WKSEXISM=wksexism WRKSLF=wrkslf WRKSTAT=wrkstat WWWHR=wwwhr� � � � � ê � � � m� prestg10:$@Role('0' )/sei10:$@Role('0' )/age:$@Role('0' )/childs:$@Role('0' )/citizen:$@Role('0' )/colscinm:$@Role('0' )/coninc:$@Role('0' )/conrinc:$@Role('0' )/degree:$@Role('0' )/dwelown:$@Role('0' )/educ:$@Role('0' )/emailhr:$@Role('0' )/grass:$@Role('0' )/hrs2:$@Role('0' )/marital:$@Role('0' )/owngun:$@Role('0' )/partyid:$@Role('0' )/race:$@Role('0' )/realinc:$@Role('0' )/rincome:$@Role('0' )/sex:$@Role('0' )/SEXORNT:$@Role('0' )/size:$@Role('0' )/tvhours:$@Role('0' )/union:$@Role('0' )/USCITZN:$@Role('0' )/wkracism:$@Role('0' )/wksexism:$@Role('0' )/wrkslf:$@Role('0' )/wrkstat:$@Role('0' )/wwwhr:$@Role('0' )� � � � UTF-8� � � k� <?xml version="1.0" encoding="UTF-8"?><DATAVIEW_TABLE_FORMAT TMP_DESCRIPTION="Save file for format properties in dataview table" VAR_IDS=":0:1:2:3:4:5:6:7:8:9:10:11:12:13:14:15:16:17:18:19:20:21:22:23:24:25:26:27:28:29:30" VERSION="1"><BK_COLOR><WHOLE_TABLE VALUE=""/></BK_COLOR><FOREGROUND_COLOR><WHOLE_TABLE VALUE=""/></FOREGROUND_COLOR></DATAVIEW_TABLE_FORMAT>ç� ™ý™dedýýÍÌÌÌÌŒM@ ðÄí@ ðÄí@gdtcecgfieýpegrc Ð ã@hdffeec—ý~dedýýg33333óN@�$(¾1Ÿ�A ðÄí@dtcecefieýpfgrchŒJê„�u Adfffec‹ýÍÌÌÌÌ�C@Ÿeddýdee à<à@qcdcgfjeýdegrhdd @ÉÔ@ddfhc™ýœš™™™™yR@fdeýýget�$(¾1Ÿ�A  ØÃ@}fcedieýŒJê„�u Amfgrfhdfffff ý®g33333�S@edddgducecefgeddfgrcgdddfic’˜œeddýýgeuid�$(¾1Ÿ�Aáî¬��^�AcefjeýpfŒJê„�u Amreddddeendd£fddddeepdeÇedjedqedrhhdddmme™ý†fdeý33333SQ@ ðÄí@ýheuffce `t×@dleýped| Ð ã@fhdddfeedd‰hddýd `tÇ@ddncecifgeýdfg|c @�¾@hddddlcŽý‚gddýýfÍÌÌÌÌÌC@ À ò@ à<à@dscecefggýpfg|ch €�ç@dddfecdd�ffdýddd à<À@icfcefggýdfg|chf @É´@dddkc‡„œdedýddfo à<À@cdcifgeý @É´@deg|nddddflc}ý©iÍÌÌÌÌL7@ddýddflc �Ë@dcgfgeýd `RÁ@eg|hddddfic‡ýŒfe33333sE@dýýefoed Ð#Ñ@ Ð#Ñ@cefeeýof àðÅ@d|fddfffef„ý}deefffffæ:@ýýeftlfc  ØÓ@  ØÓ@idgeýpfg ÀgÉ@œdhdddfjq�ýœgedý333333B@ P[è@ýeftdfce Ð#Ñ@djeýoegœ à-ß@jhdffefy”ý—gddýý3333338@ ðÄí@ ðÄí@edqcfcgfegýpfdœc Ð ã@heddffcˆý’dddýýeÍÌÌÌÌÌB@�$(¾1Ÿ�Aáî¬��^�AdqcecifgeýpedœchŒJê„�u Adddeec‡ýfffffæ4@—fdfýýdf  ØÓ@ à«Î@refcedheýoegœfhg ÀgÉ@fffegyr‹fddýýddl €¦µ@ €Ã²@cfcgfegý ·«@jfdœchfddfecž«‚ffdýýefr‚ Ð#Ñ@ Ð#Ñ@dcgfgfýo àðÅ@fgœedffffenddˆgfddddfpdfcidheddfgághfdddkkyýŽgfd333333+@ýdedocfc à<À@hfeeýdfg @É´@áchfddfkcˆýŠdedý €5@ `t×@dedrcecifegýdegá @�Î@cheddfjcœýŠfdfýýÍÌÌÌÌLK@ `t×@ `t×@edscecifdeýpfgác @�Î@heddeecŠ�€dedýdh �w@dvcfcifffýdfgách �m@dffefc‰ýš™™™™™2@‡fedýýef Ð#Ñ@ Ð#Ñ@qedchfgeýoegáfdd àðÅ@fffee©ý�fffffÆV@jdgýýget €¦õ@ À ò@necedieý ·ë@peg¸efdffeei–ý–f33333³L@ddýýedsc�$(¾1Ÿ�A À ò@lceefeýpŒJê„�u Aeg¸cgdddfec—ýšed33333“S@fýýgdtce à<à@ à<à@cgffeýpe @ÉÔ@g¸chdfffec•ýÇded33333sL@ýdgeuddc À ò@efgeýdfg €�ç@¸gddddfki‘ý¡gedýfffff&C@ ðÄí@ýgftkdcg à<à@ffeýpfg¸ Ð ã@gddfffetddšdedýd�$(¾1Ÿ�Aget‚dŒifieýqfg¸fŒJê„�u Addfefgi~ý“fdsddg €<@dtcfcefmedqeg¸chdddeec…ý333333+@™fedýdee  Øã@pdfcedegýdeg¸hge ÀgÙ@fffee”ý‰3333338@ffddýedp à«Î@cecefgfdofluchffffecyýŠf333333+@ddýýdfod P[è@ û¹@dcifggýl à-ß@fguhdeddfec‘ýƒdeffffffC@dýýeerel À ò@ à<à@cgdheýpe €�ç@gudedfffeeƒý©fedš™™™™ÙC@ýdddncfc `t×@ffieýÆfg @�Î@uchdffffcŠ�‡eedý�$(¾1Ÿ�Aýgdtcece À ò@feeýpeguŒJê„�u Achdfffec¦À¡fddddgeuhdceehedqfg�fddfeff€ ý´gedýdf33333SP@  Øã@eridceejeýded�fd ÀgÙ@dddfid|ý33333³4@Ÿddddddmocdcifgeddeg�gddddflc‘ý~333333?@deddýeer `t×@decidfedpeg�ghdfffej”ý–e P@edýýgdtc €¦õ@ €¦õ@echfjeýp ·ë@fg�chdfffec’˜�fedýýedpce à<à@ à<à@cifdfýpf @ÉÔ@g�cedfffecŸý‚ddd33333SQ@ýýfesddc  Øã@  Øã@ifieýpeg ÀgÙ@�gddfffee‡„Ÿhddý À ò@ýeeoddce P[è@feeýpeg� €�ç@edefffec—ý„dfdýý33333“S@ €¦õ@ €¦õ@hes‚dcefeeýpfg�f ·ë@dffffer€ý„fddýýe3333339@ �Û@ `t×@frjdcefegýpefrhd `RÑ@ffffei“ýffffffC@˜geeýýhe�$(¾1Ÿ�A à<à@vnfcedgeýpfgrdheŒJê„�u Afffes¡ýŽ33333³R@eedýýgdt À ò@  Øã@cecefdeý €�ç@pegrchdfffecƒý�dÍÌÌÌÌÌB@ddýýfdrc ðÄí@  Øã@ecefdgýp Ð ã@egrchefffec“ýµgeffffffC@dýdeepdf P[è@cfdgeýdf à-ß@grghdddfkc—ý—edd33333“S@ýýhevldc�$(¾1Ÿ�A ðÄí@efigýpegŒJê„�u Aredfddee´˜ý|dedýfffff&N@ €Ã²@ýgfufdci �—@fdfýffgr �¨@eddffffn“ý£ddfýd33333óG@ à<À@edscfcffieýdffrc @É´@hdddfic‹ý�dedýýhÍÌÌÌÌìP@ ðÄí@ ðÄí@dvcecifeeýpfgoch Ð ã@efffec‘¢œddxýýge À ò@ À ò@v‚ecidjeýpfeofhe €�ç@fffen ý�33333�S@eedýdgdt�$(¾1Ÿ�AcecefdgýŒJê„�u Adfgocheddfkcˆý—dfffffæB@dhýýhdtc à«Î@ à«Î@fcifeeýo  ¡Ã@eeochdffffc�ý„ddÍÌÌÌÌŒK@oýýgft€e Ð#Ñ@ Ð#Ñ@cidfgýof àðÅ@goehdddfe¸¤ý“ded  T@ýýhdxcec�$(¾1Ÿ�A ðÄí@eeegýpfgŒJê„�u Aoceeddflc›ý“ddný33333³N@ À ò@ýhdtcfci ðÄí@ffgýpfgo €�ç@chefffec ý†dddýýš™™™™ùU@�$(¾1Ÿ�A @Bù@hex}dcefggýpegodŒJê„�u Adfddfei£ý©fdhýýh33333ÓS@�$(¾1Ÿ�A @Bù@dtcfcefdeýpegfcfŒJê„�u Adddeec£ý33333ÓS@¥ddiýýgd�$(¾1Ÿ�Aáî¬��^�AtcfcefieýpfgfchdŒJê„�u Addeec�ý�fffffæR@gdnýýhev�$(¾1Ÿ�Aáî¬��^�AsecgdgeýŒJê„�u Apegfehdfffe}ƒý|dÍÌÌÌÌÌB@dgýdgdtc ðÄí@ecifieýd Ð ã@egfchdddfecƒý„ddš™™™™™3@dýýedpce�$(¾1Ÿ�A à«Î@cifjeýoeŒJê„�u Agrchdfffec–ý—fedfffff&Q@ýdffreec  ØÓ@gdfeýdeg ÀgÉ@rghdddfld¡ý¦fedýš™™™™9T@�$(¾1Ÿ�Aýhdxcece P[è@fdeýpfgrŒJê„�u Achdddfic‹ý—fdjýýÍÌÌÌÌìP@ À ò@ ðÄí@getŒlcedgeýpfgrf €�ç@hdddfei›ý�dexddgš™™™™™B@etfecedgedqegrehdfffep¤ý  T@‹edfýýhd�$(¾1Ÿ�A €¦õ@xcecefdeýpfgrcedŒJê„�u Aeefec¤ý�  T@gedýdhdv�$(¾1Ÿ�AcfcefggýŒJê„�u AdfgrcheddfkcŽýšf33333óM@edýýgdtc�$(¾1Ÿ�A @Bù@ecefeeýpŒJê„�u Aegrchdfffec˜ýžedš™™™™™T@hýýgdtce�$(¾1Ÿ�Aáî¬��^�AcgfdeýpfŒJê„�u Agrchdddfec”ý�ged33333sB@ýdedpcfc `t×@efggýdeg @�Î@ýchdddfk ï¿@c‡ý›hedý33333óE@ Ð#Ñ@ddfjcdciffgýdfgý àðÅ@ ï¿@hdeddflc}ý—gedýýÍÌÌÌÌL7@ à<À@ û¹@dfopfcedfgýledýi @É´@ ï¿@ldffferƒýŒeddýddš™™™™ÙC@ �w@dnceciffgýdfgých �m@ ï¿@eddfhcƒýÍÌÌÌÌÌB@‡eedýded 5¤@ocechfffýdfgýchd Þ™@ ï¿@ddfhc…ý�333333+@dddýddgn �w@gdcefggý �m@degýjddd ï¿@dfkj€ýˆfš™™™™™5@ddýdddnc `tÇ@fcifefýd @�¾@fgýchddd ï¿@fkc”ý‚gd3333338@dýddfmfe à<À@ciddgýdf @É´@gýfhdddf ï¿@kýŠ�œded 8�@ýddfncfc 5¤@idefýdeg Þ™@ýghdddfh ï¿@c‹ý�fddýfffff&N@ À ò@ýgdtcfce P[è@fefýpegý €�ç@ ï¿@cgeddfecƒý}fedýýÍÌÌÌÌÌB@  Øã@  Øã@efpúdcifdeýpegýj ÀgÙ@ ï¿@ddeefe´”ý”dedýýf P@  Øã@  Øã@drcfcifgfýpfgých ÀgÙ@ ï¿@dfffec¦ýfffff&R@‘eegýýfe @Bù@ €¦õ@sfecidieýpegýfed À*ð@ ï¿@fffei‡ý¬33333³7@iddýdeep à«Î@cdcefgfý  ¡Ã@dfdýfded ï¿@dfic—ý�d33333“S@deýýgdtc @Bù@  Øã@ecefdfýp À*ð@fgýchddd ï¿@flc”ý|fe3333338@Æýýdglle `t×@ `tÇ@cidgeýnf @�Î@dýhhdfff ï¿@fv�ýŒgddfffffæE@ýdefsglc à«Î@gdegýdfg  ¡Ã@ýgheddfh ï¿@i‡ý~dddýÍÌÌÌÌÌ5@ à<à@ýffrnd–i à<à@fefýpfgý @ÉÔ@ ï¿@fdfffegj‡ý€eddýdÍÌÌÌÌÌ5@  ØÃ@efqkd–hfegýdfgýh Àg¹@ ï¿@ddfffgp‡ý‰fddýýeš™™™™™?@ �Û@ �Ë@dpcfcifffýnfgých `RÑ@ ï¿@efffecddÇeddddhducecmfmfdqfgýchd ï¿@mffec|ý¢33333³4@leddddfocecfddfddedýhhdd ï¿@dfhcŒý�gÍÌÌÌÌŒF@edýýeeqi À ò@ À ò@eciddfýp €�ç@fgýhedff ï¿@fei�ý„ddfffffæR@gýýhevne ðÄí@ P[è@cidfeýpe Ð ã@gýehefff ï¿@e‚”ý¥gddš™™™™�4@ýýedpcec  ØÓ@  ØÓ@ffdfýpfd ÀgÉ@ýceeddfe ï¿@c“ý„ddeýffffffC@ à<à@ýgeteece  ØÓ@dfeýpfgý @ÉÔ@ ï¿@fhdfffel~ý—dmddd €<@hfundceffedqemýf ï¿@dfffeeª•ý�eeiýýg33333sL@ À ò@ à<à@ftxfcedgeýpegýfh €�ç@ ï¿@dffeen“ýffffffC@µfefýdge À ò@tkfcfdjeýdegýghd €�ç@ ï¿@ddfik’˜®gedýdget P[è@eecgdeeý à-ß@dfdmfhdddfieªýŽffffff†U@eÆýýhexÇ�$(¾1Ÿ�A €¦õ@ecedheýpŒJê„�u Aedmefdfffeý’˜‘ge 0�@eddget fcediedqegmfhdfffeh“ý{dedfffffæL@ýdgftndc €¦õ@ifgfýdeg ·ë@�fdefffes}ý~dfdýÍÌÌÌÌL7@ à«Î@ýefpcfci  ØÃ@dggýmeg�  ¡Ã@ghffffecˆý�dedýýš™™™™™8@ À ò@  ØÓ@efpddcefggýpfm�d €�ç@ddffeervý€eedýýeš™™™™™;@ �Û@ à«Î@dpceciffgýoeg�ch `RÑ@dffeec|ý33333³4@ eedýdgf �Ë@tedcefeeýdfgýfde `RÁ@ ï¿@ddfke¤ý‡  T@dddýýhfv  Øã@  Øã@dlcidfeý ÀgÙ@pfgýeedf ï¿@ffee™ýŠg333333O@egýdgfti @Bù@eceddgýd À*ð@fgýehedd ï¿@fki‘ý‚defffff&C@dýýgftjd ðÄí@ �Û@cifdgýpe Ð ã@gýgddfff ï¿@es¤ý�ddd  T@ýýhfxndc @Bù@ ðÄí@efmgýpeg À*ð@ýeddfffe ï¿@n€ýždedýš™™™™™5@  ØÓ@ýgetcdci  ØÓ@ffeýpfeý ÀgÉ@ ï¿@hddfffec—ý�ddhýdš™™™™YR@ û¹@gdtceciffeýdegýc ¡°@ ï¿@hdddfkc¡ý¡dedýdhš™™™™9T@ à<à@dvcecifeeýdfgýce @ÉÔ@ ï¿@dddfic¡ýš™™™™9T@œeedýýgd  ØÓ@ Ð#Ñ@tcfcifdeýofgýced ÀgÉ@ ï¿@ffffc®ý³ U@eevýdhft  Øã@cfcfdeeý ÀgÙ@dfmýehed ï¿@dfic�ý„d €=@edýýedrc P[è@ P[è@ecifeeýp à-ß@egýcedff ï¿@fec«ý¨ddÍÌÌÌÌ,V@dýdhfwgd  ØÓ@cgfgeýdf ÀgÉ@gýddeddf ï¿@ip“ý½fedffffffC@ýdedscec €¦õ@fffeýdfd ·ë@ýchdddfi ï¿@c´ýŠdedý333333W@�$(¾1Ÿ�Aýhdxclceáî¬��^�AfigýpegýŒJê„�u A ï¿@chefffec”ý�dddýýš™™™™�4@  ØÓ@  ØÓ@fftedcifmfýpfmýf ÀgÉ@ ï¿@ddfffe‚šý±gedýdh33333sO@ @Bù@eufdcffdfýdegýfd À*ð@ ï¿@eddfie™ýš™™™™yR@“eedýýed P[è@ Ð#Ñ@qcechfdfýoegýchd à-ß@ ï¿@fffec£µ™hdÆýýget�$(¾1Ÿ�A €¦õ@xfcedffýŒJê„�u Apegýehee ï¿@ffexŒýšdÍÌÌÌÌŒF@ddýýeere À ò@ À ò@dcgfdfýp €�ç@fgýgdddd ï¿@fie—ý¤gdš™™™™ÙO@dýýhdvce ðÄí@ ðÄí@‘gfdfýpf Ð ã@gýcedfff ï¿@gcƒý¢heeš™™™™ÙC@ddgftcfcidfeddfdýÆhdddfl ï¿@c—ý‚dddýš™™™™ÙO@ P[è@ýhexsdci à<à@feeýpfgý à-ß@ ï¿@eddddfe‚´ý¡fddýý333333W@�$(¾1Ÿ�Aáî¬��^�AhexedceejeýpegýgŒJê„�u A ï¿@ddffeeg‡„£fedýdg  ØÃ@ftcdciekeýdegýfd Àg¹@ ï¿@dddfic¡ýš™™™™9T@�eddýýhe @Bù@ ðÄí@xidcefeeýpedýfdd À*ð@ ï¿@ddfeg”ý�33333sB@fddýýget €¦õ@ ðÄí@gdceeggý ·ë@pfdýgdef ï¿@ffeg—ý�d33333³<@eeýýeere `t×@  ØÃ@fcidieým @�Î@fgýeedff ï¿@fee¤ýŒfd  T@pýýhdxcf À ò@  Øã@ceejeýpe €�ç@g€cgdfffec‘ýyddd333333?@ýýefrsec €¦õ@ à«Î@idgeýoeg ·ë@€fedfffesŽý›jedýffffffH@ P[è@ýedpcece  Øã@fgeýpeg€ à-ß@chdffeec„ýµhedddffffff*@edscfcffjeddfd€cheddfic¡ýŠfedýýhš™™™™9T@ @Bù@ P[è@duce‘efeeýpfg€ch À*ð@dfefgc™ýÍÌÌÌÌŒM@‰fedýýed `t×@ û¹@pcechfkeýled€chd @�Î@ffeec˜ý‘33333óP@fdlýdhdv €¦õ@cfcefdeý ·ë@dfg€cfdddfkcƒý‰dÍÌÌÌÌÌB@dxýýgetk  Øã@  Øã@lcidffýp ÀgÙ@fg�hedfffeyŠý›gdÍÌÌÌÌŒM@nddhftelcgdggdqed�demfffel”ý®fde3333338@ýýgdtcec  Øã@  ØÃ@gfgeýmfg ÀgÙ@�chdffffc¡ý¸deddš™™™™9T@dgdxcfcifeeddfg�cedddfic‚ýšiedýýš™™™™™=@  ØÓ@  ØÓ@edrcfcefgeýpeg�c ÀgÉ@hdfffecvý{eeddýeš™™™™™;@ €Ã²@eqddcifgfdjeg|fddfffee}ý333333)@}efdýýdf `tÇ@ û©@jgeceddgýhfg|fhf @�¾@ffffeŽýµ33333óM@fddýdeer `tÇ@eecgddeý @�¾@dfg|ghfddfih�ý”e333333B@edýýferg À ò@ À ò@dceeheýp €�ç@eg|eddfffer„ý’deš™™™™�8@dýdefpfd `tÇ@cifjeýde @�¾@e|fddddflf‰ý—jddš™™™™™2@ýýddjcfc  ØÓ@  ØÓ@efegýpeg ÀgÉ@|chfeffecŸýºdedý33333SQ@ �Ë@dfercfcfddfýdfd| `RÁ@ihdddfic‘ýšfedýý333333?@  Øã@ `t×@ddncecefdgýpeg|c ÀgÙ@hdffeecŒý‘degýýfš™™™™ÙJ@ ðÄí@ ðÄí@eriecidgeýpfdkgh Ð ã@dfffeý¤ý 0�@  T@¯gedýdhd ðÄí@vcecfefeýdfgkcee Ð ã@ddfic–¨{ddgýýefr �Û@ `t×@eecidfeý `RÑ@pegkehdfffeh ý£eÍÌÌÌÌìP@edýýedqc P[è@ P[è@ecgfdeýp à-ß@fgkchdfffec„ý¶idfffffæA@dýdedpcf à<À@cgffeýdf @É´@dkchdddfic­ý–gdd  U@ýdferddc�$(¾1Ÿ�AgegeýdfgŒJê„�u Akeddddfl ¤ý·hdhý  T@ `t×@dhevfecgdieýdeg§ @�Î@fhdddfid©ý§dddddfffffÆV@hewneceddeddeg§fhdddfizƒý¸fddýdeš™™™™ÙC@ P[è@eqclcedheýdfgõfh à-ß@eddfic–ýÍÌÌÌÌL@@Ÿeedýýgd  ØÃ@ �w@tcecgfdfýefg§ced Àg¹@ffffcˆý‘ÍÌÌÌÌLA@dedýýhdx ðÄí@ ðÄí@cecifdeý Ð ã@pfg§chdfffec™ý›dÍÌÌÌÌŒM@ddýýfdrc P[è@ �Û@eceeieýp à-ß@eg§chdffeec‘¢¢eddýdheucd €Ã²@cgfieýdf �¨@g§nddddekcˆý±gddfffffæB@ýdefpcdc à<À@gfgeýdfg @É´@§gddddfic�ý’dddýfffff¦L@ €¦õ@ýgftndci €¦õ@ffeýpeg~ ·ë@fddddff}’˜Ÿeedýd P[è@edscecgfgeýdeg~c à-ß@fdddficƒýxdedýýe333333D@  ØÓ@ 5¤@dpcecifffýgfg~ch ÀgÉ@dfeffc¤ý33333óQ@£kedýýhd `t×@ `t×@vcfchfdeýpeg~chd @�Î@ffefc«ý«ÍÌÌÌÌ,V@dedýýhet  ØÓ@ à<À@edcgfieý ÀgÉ@mfg~fddfffek”ý‰d P@ddýýefrŒ  Øã@  Øã@ecidfgýp ÀgÙ@eg~hedfffeÞdd�gddýddfpcl �Û@cedgeýdf `RÑ@doehedddkcdd~fedýdedpclc  Øã@efgeýdfg ÀgÙ@ochddddkcŒý’eedýfffff&C@ à<à@ýeepddci 5¤@ffeýgego @ÉÔ@kddfffeg~ýœhfdýd €<@ ðÄí@dfgddcifgeýdfgoh Ð ã@dfddfkn’˜‘eedýýe `t×@  ØÓ@drcecgfegýpfgoch @�Î@dffffcuýfffff¦F@¢fedýýee à<à@ à<à@pdecgdeeýpeg¸fee @ÉÔ@fffed€ý|3333339@dddddefodecidgfddeg¸ehdddfff”ý”h P@ddýýgetŒ P[è@ P[è@dcggffýp à-ß@eg¸hddddfe ’ýƒddfffffæM@dýýedrce à<à@ à<à@cifhfýpe @ÉÔ@g¸chdeffec‹ý‰dedÍÌÌÌÌìP@ýýgdtcec P[è@ P[è@ifdfýpeg à-ß@¸chdfffecŠ�‡fedý à«Î@ýefpie„i à«Î@ddfýofg¸  ¡Ã@iedfffgŒ‡ý~fddýd33333óE@ �Û@edpcfcifeeýdfg¸c `RÑ@hfddfkc}ýšeddýýd333333)@ €Ã²@ €Ã²@dnceceffeýjfgœch �¨@effffc“ýffffffC@ˆeefýýef  ØÓ@  ØÓ@qteciddeýpffœdhd ÀgÉ@fffeˆ“ý³ffffffC@hddýdedp à<À@clcefheý @É´@dfgœchdddfic}ý²i333333)@edýddfoc à<À@fcgddfýd @É´@fdœghdddfic€ý�dd3333339@dýýedpce Ð#Ñ@ Ð#Ñ@ciffeýoe àðÅ@dœchdddffcyrxdeddýeerefc €Ã²@idfgdjfgœehdffffg”ý�fdeý  Q@ À ò@ýedpcfce À ò@fieýpegœ €�ç@chffffec‰ý£dddýdÍÌÌÌÌL@@ �Û@edpcfcgfheýdegœc `RÑ@hdddfhc|ý}dddýýe33333³4@  ØÓ@ à<À@dqcecifefýmegých ÀgÉ@ à|@dddfhc}ý333333)@„fddýdef à<À@pdfcedeeýdfdýhhf @É´@ à|@ddfkl€ý†š™™™™™5@fddýddfi û¹@ddcifefý ¡°@dfgýjddd à|@dfkpƒý’fš™™™™ÙI@edýýgetx ðÄí@ P[è@dceffgýp Ð ã@egýeddff à|@fex}ý–hf333333)@dýýdfmef �Û@ à«Î@cedgeýoe `RÑ@gýrhffff à|@ed�ý†gdd ÀA@ýýefogdc �Û@ �Û@hffgýpfg `RÑ@ýfddfffe à|@–€ýzgddýš™™™™™5@ Á¯@dddocecifgeýdfgý S¤@ à|@chdddfhc{ý‡fddýýÍÌÌÌÌÌ2@ `t×@ `t×@dfjcfcedeeýpedýe @�Î@ ¢­@hffffecƒý�gfdýddš™™™™™3@ Ð#Ñ@fjcfcedfeýdfdýdh àðÅ@ ¢­@fddfkc”ýš™™™™�4@™gfdýýdf Ð#Ñ@ û¹@gddcefgeýlfdýgdf àðÅ@ ¢­@ffffcddšhddýddfj �Ë@cdcefieý `RÁ@dfdýfded ¢­@ddkcdd–ddÇddedxcecigmgdqfdýchddd ¢­@mkc}ýŠgd333333)@dýýefqff à<à@ €¦µ@cedieýkf @ÉÔ@gýfhfdde ¢­@fg}ý‡hfd333333)@ýýdfogfc �Û@ à«Î@eddeýofg `RÑ@ýfhfffee ¢­@eddwdddý à<À@dedpcecifgfýdegý @É´@ ¢­@chedddjc�ýœheddd33333³<@ddmcfceffgddedýc ¢­@heddflc•ý’fdiýýe33333³F@�$(¾1Ÿ�Aáî¬��^�AdtceceffgýpegýceŒJê„�u A ¢­@dffeec‡ý33333óE@¨fddýdde  ØÓ@fcfceddeýdfdýghe ÀgÉ@ ¢­@ddfic„ý�fffffæA@dedýýefr  Øã@  Øã@}dchfdeý ÀgÙ@pfgýeddf ¢­@ffe‚€ý�f3333339@fdddddpcfcefgedqegýchfef ¢­@efcŠ�¡deeýýeeree à«Î@  ØÃ@cedigýmf  ¡Ã@gýheefff ¢­@fiŠýœdedš™™™™™C@ýýeftfdc  Øã@ `tÇ@efheýneg ÀgÙ@ýfddfffe ¢­@i„ý…fddýfffffæA@ Ð#Ñ@defpdfcgdieýdfgý àðÅ@ ¢­@dhdddfkx™ý|defýýÍÌÌÌÌ�P@ à<à@ à<à@gftnecidegýpfgýe @ÉÔ@ ¢­@hdfffei—ý©ddeýýe33333“S@�$(¾1Ÿ�A à«Î@frgeceddeýoeeýfhŒJê„�u A ¢­@dffeex§ýffffffR@¡fddýýhe�$(¾1Ÿ�A ðÄí@uedcefdeýpegýeddŒJê„�u A ¢­@ddefe”ý²3333338@dddýdedq û¹@cecffffý ¡°@dfgíchdddfic}ý�eÍÌÌÌÌÌ7@edýýedpc à<À@ à<À@ecffgeým @É´@egíchdffffc…ý˜fdÍÌÌÌÌL5@dýddfocf 5¤@cgdieýde Þ™@gíhhdddfhc‡ýŸded33333ó@@ýdedpcec �—@ifeeýdeg ��@íchdddfhc´ý©dd‚ý333333W@�$(¾1Ÿ�Aýhexfecgáî¬��^�AddeýpeeíŒJê„�u Aghdffefh’ý‰edfýd33333³N@ ðÄí@fdrcfceedeýdegìc Ð ã@hdddefc¤ý�eedýýg33333óQ@ @Bù@ P[è@dtcfcefeeýpfgìcf À*ð@dfffec�ýš™™™™™N@Ÿfedýdge �—@tedcgffeýdfgìfde ��@ddekg™ý{š™™™™yR@dedýýeds �Ë@ �Ë@cfcifkeý `RÁ@nfgìchdddfjcddwdedýdefpd €¦õ@fcidfgýd ·ë@egìhhddddjd™ý‚ddÍÌÌÌÌ�P@dýýgftsd  ØÓ@ `tÇ@ciffeýnf ÀgÉ@gìeddddfe–dd¯jeddddencdcefheddfgìgdddddkc�ý’fdddÍÌÌÌÌL7@ddficdcifeedqfgý u@fdffffec~ýwdedýdš™™™™™/@  Øã@eepedcifgeýdegýi ÀgÙ@ u@ddfffex¤ý—edsýýh33333óQ@�$(¾1Ÿ�Aáî¬��^�AevŒecidegýpfgýfhŒJê„�u A u@dfffex¤ý  T@¡fdhýýhe @Bù@ P[è@wsecedieýpfgýhgd À*ð@ u@ddfex€ý‘š™™™™™5@gedýýeeq P[è@  ØÓ@decedeeý à-ß@pfgýfhef u@ffee‰ýžgš™™™™™2@ddýýddfc  Øã@  Øã@fcefegýp ÀgÙ@egýchfff u@eec~ý³leš™™™™™/@ddddfkcfcedeeddfdýjheddf u@ic„ý¯hedfffffæA@ýdgdtcfc à«Î@efhgýdfg  ¡Ã@‹cheddfic‘¢½hddý �Ë@defpcdcffjeýded‹ `RÁ@gddddfic“ý�fedýý33333SP@ à«Î@ à«Î@efp‡dŒifggýofg‹e  ¡Ã@ddfffg‚›ý„dddddg33333óQ@et‚dcefhgdqem‹fdefffeÈ‘ýfffff&C@žfedýýge À ò@ à<à@tkfcedjgýpfg‹fhe €�ç@fffekŒý¬fffff&C@gedýýget  Øã@  Øã@ddcefheý ÀgÙ@peg‹gddffeee—ý¦f33333“S@edýdferi P[è@dcgfeeýd à-ß@fg‹fddddfifddyddfýdefrie à<À@cidffýdf @É´@gàhhddddjˆ‡„‘eddýdedpcfc�$(¾1Ÿ�AgeheýdfgŒJê„�u Aàcedddfec‘ýždedý333333M@ P[è@ýeeqdece  Øã@dgeýpegà à-ß@hhdfffecŸý–fddýý àR@ P[è@ �—@eepefsedheýffgàe à-ß@fdddfgižý›eddýdhÍÌÌÌÌ,R@ �w@exeecgddeýdfgàeh �m@dddfi‡ƒýš™™™™ÙC@�eedýýff�$(¾1Ÿ�A €¦õ@rjecgdgeýpegàehdŒJê„�u Affee}ƒýƒfffffæ?@eedýdefq `t×@fecedgeý @�Î@dfgàfhdddfkn|y‚eedýdefqh `t×@eciddeýd @�Î@fgýghddd s@fhg‡}Žjddýddfodd  ØÃ@cifffýdf Àg¹@gýfddddf s@hc ý}hed33333³D@ýýddncfc Ð#Ñ@  ØÃ@efggýmfg àðÅ@ýchdfffe s@c}ýŸeedd333333)@ddfofdcgfeeddfgý s@fddddflf�ý´hddýdÍÌÌÌÌL7@ �Û@deocdcffieýdfgýn `RÑ@ s@ddddfic}ýždedýde333333)@ `tÇ@epcfcgdgeýdfgýlh @�¾@ s@dddflcddŒjedýddf 5¤@qeecedgeýdfgýdhd Þ™@ s@dddkj¥ý²ÍÌÌÌÌLU@eedýdhev à<à@cdchgmeý @ÉÔ@dedýiddd s@dficƒý²ffffffæ?@ddýdeepc Á¯@ecfdeeýd S¤@fdýihedd s@fkc¡ý’geš™™™™9T@eýýhevgf�$(¾1Ÿ�A ðÄí@cedjeýpfŒJê„�u Ag{eedfffef}ý�fedÍÌÌÌÌÌ7@ýýfdvclc À ò@  Øã@eeheýpeg €�ç@{cgdfffec¬ýµgddýš™™™™YS@ À ò@dgetedcefjeýdeg{ €�ç@gddddfid¥ý£gddddÍÌÌÌÌ,T@getgdcefmfddfdýg h@ddddfig|ý”fddýýe33333³4@ à<à@  ØÓ@dpcecgfdfýpegýce @ÉÔ@ h@dfffec–ý33333sJ@”fddýýef P[è@ P[è@pndcifegýpfg®gdd à-ß@fffex¡ý�33333³R@gddýdedq ðÄí@cecefdfý Ð ã@deg®chdddfhc¥ý�e33333³M@ddýýefre �Ë@ �Ë@dcifdeýn `RÁ@fg®eddffffh…ý‚ed333333+@dddefrgechdfeddff®hhdddfh}–ý�ddh33333³L@ýýgfteec à<à@ à<à@iddfýpfg @ÉÔ@®dhdddfhfšý–iedý33333sO@ P[è@ýffridcg P[è@fdfýpfg® à-ß@hddfffei�ý¡fddddÍÌÌÌÌL7@dgnkfcgdgeddfd®ehfddfk€ˆý‚dddýýeÍÌÌÌÌÌB@ P[è@  ØÓ@fqiecedgeýpeg©dh à-ß@eeffe‚—ýš™™™™ÙO@„fedýýhf ðÄí@  Øã@wndcefeeýpeg©edd Ð ã@fffe ddvdddýddfo �w@dfcidkeý �m@dfg©ihddddj‡“ý¹hffffffC@ddýdhewn à<à@dcgfdeýd @ÉÔ@fd©gddddfii€ý—dd3333339@dýdegpce û¹@cgdgeýde ¡°@glhhdddficƒý~fedš™™™™™3@ýýefpxdc  ØÓ@  ØÓ@hffeýpff ÀgÉ@lddefffex¥ýŠed}ýÍÌÌÌ̬U@ @Bù@ýgexrlce @Bù@dkeýpegl À*ð@fhdddfen”ý‹gddýý3333338@ ðÄí@ Ð#Ñ@ddncfcefggýofglc Ð ã@hefffec}ýxdedýýeÍÌÌÌÌÌ7@ À ò@ �Ë@eredciffeýneglhd €�ç@hffff|™ýÍÌÌÌÌŒD@‹gddýýee ðÄí@ �—@sddceeieýffg�edd Ð ã@ddffk�r�fedýýeep @Bù@ û©@edceedeý À*ð@hfg�hddffffi‹�»fedddefpcdcffjeddfd�iddddeicŠ�µiddýdddnce à<à@ceeheýdf @ÉÔ@g�chdddeic‘ýƒeedÍÌÌÌÌLG@ýýedncec  Øã@  Øã@ifheýpeg ÀgÙ@�cedfffec‡}¡gddý ðÄí@ýgetefce  ØÓ@djeýpfg� Ð ã@fhdffeff¢™ƒedfýý à<à@ à«Î@edsceceegeýofg�c @ÉÔ@hdddfecvý¡eddýdeffffff=@  Øã@dpceceedgýdeg�ce ÀgÙ@dffffc‘ý333333M@‹fddýdee  Øã@pffcfdfeýdfg�jhd ÀgÙ@ddfhh®ý¸ U@dedýdgdw�$(¾1Ÿ�AcecifheýŒJê„�u Adegýchdd �¥@dfic™ýšdš™™™™yR@edýdget}�$(¾1Ÿ�AdceffeýdŒJê„�u Afgýedddd �¥@fiiˆý~defffff¦A@dýýefpne à<à@ à<à@cidfgýpe @ÉÔ@eýghdfff �¥@e}¥ý¸fdd33333ÓU@ddhetedcefgeddfgýfdeddei �¥@iž« fdeý�$(¾1Ÿ�Aýgetfeceáî¬��^�AddfýpegýŒJê„�u A �¥@ehdffeee™ý�defýýš™™™™yR@ À ò@ ðÄí@get‚lcidgeýpfgýf €�ç@ �¥@hdfffen‹ý¥dedýýgÍÌÌÌÌìP@  Øã@  Øã@ftidcgfdeýpegýhd ÀgÙ@ �¥@dfffep™ý33333SQ@†dedýýhe�$(¾1Ÿ�A @Bù@wxdcgfefýpfgýedeŒJê„�u A �¥@fffeg”ý²š™™™™�4@fddddddlcfcffdeddfgýchdd �¥@dfic™ý¹e33333SQ@ddýdedrc à«Î@ecgfgeýd  ¡Ã@fdýchedd �¥@fic‡ý­fd33333óE@dýdddocf `tÇ@cffeeýdf @�¾@gýcheddf �¥@ic£ý}dedÍÌÌÌÌŒO@ýýefsedc `t×@ `t×@ifeeýpfg @�Î@ýdddddfj �¥@}¡ý‚ddŒýš™™™™ÙT@�$(¾1Ÿ�Aýhdxcece ðÄí@egeýpfgýŒJê„�u A �¥@chdfffec€ý’gddýý3333339@ P[è@ P[è@efpdfcgddfýpfgýf à-ß@ �¥@edddfef›ý€edfýýeÍÌÌÌÌLS@ �Û@  ØÓ@frhecidefýpfgýdh `RÑ@ �¥@dddfhÜ”ý3333338@Œideýýed  ØÓ@  ØÓ@pcfcffefýpfgýced ÀgÉ@ �¥@ffffcvý}ffffff=@dedýýedp €¦µ@ €¦µ@cecifgfý ·«@keeýchde �¥@ffecƒý›fÍÌÌÌÌ�H@edýýgexd à<à@ à<à@dcefdgýp @ÉÔ@egýpdedd �¥@fkdddwddeýdddocf  ØÓ@ciffgýdf ÀgÉ@fýchdddd �¥@jc€ýƒhedš™™™™™5@ýýddncfc à<À@ û¹@ifhgýlfg @É´@ýchgfffe �¥@c‘ý…gfdý333333?@  ØÓ@ýefpefce  ØÓ@dfgýpegý ÀgÉ@ �¥@fhffffeedd�detýd @Bù@geweecideeýdegqe À*ð@hddddj}ž«¢fefýdg €¦õ@etgfcgdjeýdegqjh ·ë@dddeif ý33333�S@ªfedýdhd €¦õ@uceceedeýdfgqchd ·ë@ddfic“ýž33333SP@dehýýget�$(¾1Ÿ�A P[è@dfcedfeýŒJê„�u Apegqhgdfffedƒýšeš™™™™ÙC@edýýferf P[è@ Á¯@ecedggýi à-ß@fgqhheeefei‡}–eddddedpcecefggddfgqcfdddfkc€ýˆfddš™™™™™5@ýýeemgdc  Øã@ à«Î@efdeýofg ÀgÙ@�fdeddfkeyý†fdgýÍÌÌÌÌL@@ à<à@ýedscfcg Ð#Ñ@ffgýofg� @ÉÔ@chfffffc‡ýwdedýýÍÌÌÌÌÌ5@ ðÄí@ �w@eepddciffgýeff�i Ð ã@ddddejx‹ý‹hfdddeÍÌÌÌÌ�C@dpcfcgfggddfg�chfddfkcƒýÍÌÌÌÌÌB@¦hedýýee ðÄí@ P[è@sgdcefegýpeg�dde Ð ã@fffer©ý­fffffÆV@fedýdhdw ðÄí@cecgfdeý Ð ã@dfg�chdddficŸý™d àR@ddýýefp  à<à@ `t×@ecideeýp @ÉÔ@fg�hhdfffeŒ™ý�fd33333SQ@fýýgetxe P[è@ P[è@cedeeýpe à-ß@g�fheddfei™ý‡ddfÍÌÌÌÌŒM@ýýgeuxec �w@ �w@idgfýeeg �m@�dhdddefxŠ�{edfý à<À@dfeqeecidieýdfg‚ @É´@ehdfffelƒý®hddýdš™™™™™3@ à«Î@ddncfchfheýdfg‚c  ¡Ã@hdddficƒýˆgfdýddÍÌÌÌÌÌ(@ à<à@djcecefigýdfg‚ch @ÉÔ@fddfkc‰ýš™™™™™2@{fedýdef Á¯@nddcifefýdeg‚pdd S¤@ddfjd‹ý¤ÍÌÌÌÌìP@dedýýhev ðÄí@ `tÇ@kdcefjeý Ð ã@nfg�gddddfhg”ý‹dš™™™™�:@dsýýedvc 5¤@ 5¤@eciffgýg Þ™@fg�chefffec˜ý¶fe33333óP@dddhdxcecefmeddeg�chdddfic�ý–ge–š™™™™™N@ýýgevxec�$(¾1Ÿ�A à<À@edeeýmegŒJê„�u A�fhdfffe‚•ý®gedý33333sL@ P[è@dgethdcefdeýdeg� à-ß@fddddfie‘¢‘eedýý�$(¾1Ÿ�A P[è@geu}dcefkeýpfg‚fŒJê„�u Addfffe‚ ý“hddýýd33333óT@ À ò@ À ò@flÈecgdieýpeg‚hh €�ç@dfffenˆýš™™™™™8@©fedýdee  ØÓ@pedcffjeýdfg‚fdd ÀgÉ@ddeie’ýˆ33333óI@fedýýeep €¦õ@ P[è@dfcedieý ·ë@peg‚fedfffef›ýŠgÍÌÌÌÌLO@ddýýeesi @Bù@ À ò@ecedieýp À*ð@eg‚ehdffeen®ýŽed U@hýýhdxcf @Bù@  Øã@ efdeýpf À*ð@g‚cfdfefgc˜�ªfdhýýhdvcec @Bù@ à<À@eeieýmeg À*ð@‚chdffefc�ýªfdfý333333J@ €¦õ@dgetkfcedheýdeg‚ ·ë@fhdddfjg›ý™gddýýš™™™™™B@ À ò@ Ð#Ñ@eeqkdcefkeýofg‚e €�ç@ddfffer–ý¥deeýde33333³L@ P[è@epfecedgeýdeg�gh à-ß@dddfif ý33333óT@�dedýýge ðÄí@ P[è@v‚dceffeýpeg�gdd Ð ã@fffenƒý€š™™™™ÙC@dddýýedq �w@ �w@cecifegý �m@eeg�chdffffcˆý£fÍÌÌÌÌÌB@efddeesfecedjedqem�hhdfffefŠý‡fe33333³9@dýýeephl P[è@ Ð#Ñ@cedgeýof à-ß@gjehdfffff‹ý‡hddÍÌÌÌÌ�C@ýýdfohdc Á¯@ Á¯@ifeeýifd S¤@jjddfffe‚šý–ediý33333sO@ À ò@ýhdtcfce à<À@edeýmfgj €�ç@chdddflcŠ��eedýý  Øã@  ØÓ@edpcecffheýpfgjc ÀgÙ@hdffffc‹�Œeedýýe�$(¾1Ÿ�Aáî¬��^�AepedcgedeýpemjddŒJê„�u AdffeeeˆýfffffæB@„ddhýýed P[è@ �Û@ucecefeeýpfe�chd à-ß@fffec“ý‚33333SP@dedýýgdt�$(¾1Ÿ�A P[è@cfcifjeýŒJê„�u Apfg�chdfffec“ý‚dš™™™™YQ@dfýýgft| ðÄí@ à<à@ecidheýp Ð ã@fg�hhdddfeŒ£ý•gdÍÌÌÌÌLS@dýýget‚d�$(¾1Ÿ�Aáî¬��^�AceeieýpeŒJê„�u Ag�eddfffeŒ‹ýžeedÍÌÌÌÌÌH@ýýeesnec ðÄí@  Øã@eddeýpfg Ð ã@hfedfffe›”ý|dedýš™™™™�:@�$(¾1Ÿ�AdgdtclciegeýdfghŒJê„�u Achdfefec™ý•iddýý33333SQ@�$(¾1Ÿ�A @Bù@getndceeieýpfgheŒJê„�u Addfefen“–¦eeddýe à«Î@dpclceggedoedhchdfffec”ý  Q@�fefýýge P[è@  Øã@t‡lŒgdgeýpegrfhd à-ß@fffg  ý§33333óT@fdxýýhdx�$(¾1Ÿ�Aáî¬��^�AcecefdgýŒJê„�u Apegrchefffec¤ý©fÍÌÌÌÌ�U@dÆýdgdtc�$(¾1Ÿ�AfceejeýdŒJê„�u Afgrcfdddfic™ý¥gdÍÌÌÌÌŒM@dýýedsce ðÄí@ ðÄí@cefdeýpe Ð ã@grchdffeecŠ�˜dddýýeepedc�$(¾1Ÿ�A À ò@egjeýpegŒJê„�u Ardddddfe‚Šý£feeý333333J@�$(¾1Ÿ�Aýgeteeceáî¬��^�AdfeýpegrŒJê„�u Ahhdffffg–ý gddýýfffff&Q@�$(¾1Ÿ�Aáî¬��^�AeeqkdceffeýpegreŒJê„�u AddffeesddxdmÇddeeqýfcidh 8�@eddfdýdh hƒ@ddddjý ý 8�@33333�S@•hegýýge�$(¾1Ÿ�A @Bù@t}fcedieýpegýehdŒJê„�u A hƒ@ffeen¤ýˆ  T@eddýýhev @Bù@ P[è@edcefieý À*ð@pfgýeddf hƒ@ffegƒý{dÍÌÌÌÌÌB@ddddefrŒdcifdfdÆeeýedddd hƒ@ff|‘ýwdd333333?@dýýddncf�$(¾1Ÿ�A `t×@ciefeýpeŒJê„�u Agýchdddf hƒ@ec«ý�hedÍÌÌÌÌ,V@ýýhexidc €¦õ@ P[è@efieýpeg ·ë@ýfdeffff hƒ@sdd…mfdý Ð#Ñ@ddfpcfcedggýdfgn àðÅ@hhfdddkcˆý‹efddý €4@ à<à@dejcfcidegdpegnehffffec‘ý…ffdýýe333333?@  Øã@  Øã@epddcefggýpegned ÀgÙ@ffffee”ý3333338@Šedhýýfd `t×@ `t×@scecifefýpfgnchf @�Î@fffec“ý�ffffffC@hddýýefp ðÄí@  ØÓ@decidefý Ð ã@pfgnlhdddfh‚€ý{d3333339@edýýedoc 5¤@ 5¤@ecigeeýg Þ™@efnchdddfhc‡ý’idÍÌÌÌÌÌ5@dýdeepde  ØÓ@ceddfýdf ÀgÉ@gnifdddfhh}ý{ddd333333)@ýdeeqdec �w@idffýdfg �m@nghdddfhk‡„›gedý ðÄí@ýedpcecg ðÄí@fdfýpegn Ð ã@cedfffec”ýzdedýý3333338@ �Û@ �Û@ffrhdcifefýpff«f `RÑ@ddfffes¡ýªjdiýýhÍÌÌÌÌìQ@  ØÓ@ �Ë@fvkecfddfýnfg«gh ÀgÉ@dddffkvýÍÌÌÌÌL>@�ddfýýff ðÄí@ `t×@siecidgfýpeg«dhe Ð ã@ddfeg•ýžš™™™™YE@fffýýgft ðÄí@ ðÄí@hecgdefý Ð ã@pfg«ihdffffg”ý›h3333338@edýdefre  Øã@dchfdfýd ÀgÙ@fg«xddddfhe®ýˆfe U@nýýhexhf ðÄí@ à<à@cgdffýpf Ð ã@g«ehdfffeh‰ý¥dddÍÌÌÌÌŒA@ddeepfdcgggeddfgkeddddfkidd¤dddý @Bù@dget}dcifeeýqfgk À*ð@gddddfen‡ý�gddýýš™™™™™?@ �Û@ �Û@edpclceedeýpfgkc `RÑ@hdddfic™ýžgddýýeÍÌÌÌÌŒM@ À ò@ À ò@dpcfcgfheýpegkch €�ç@dfffec’˜˜hedýýed�$(¾1Ÿ�A ðÄí@rcfceffeýpfgkchdŒJê„�u Afffec¥ý£ÍÌÌÌ̬U@fdnýýhex  Øã@  Øã@kfcgddeý ÀgÙ@pfgýfhdd Ђ@dfh‚®ýªf U@edýýgdtc�$(¾1Ÿ�A €¦õ@fcefdeýpŒJê„�u Aegýchdff Ђ@fec¤ý£ddš™™™™ùV@lýýhexpe�$(¾1Ÿ�Aáî¬��^�AcidieýpeŒJê„�u Agýhhdddf Ђ@e‚“ýƒeed33333SP@ýýgftndc @Bù@ P[è@efkeýpfg À*ð@ýeddfffe Ђ@‘©ý¼geeýfffffÆV@�$(¾1Ÿ�AdhexeeceddeýdegýŒJê„�u A Ђ@fhdddeie®ý edhýý U@�$(¾1Ÿ�A À ò@hexsfcedffýpfgýdŒJê„�u A Ђ@hdfffen¤ý†defýýg  T@�$(¾1Ÿ�A €¦õ@etkfceddeýpfgýfhŒJê„�u A Ђ@dfffek©ýfffffÆV@‡edlýýhd�$(¾1Ÿ�Aáî¬��^�AwceceffeýpegýchfŒJê„�u A Ђ@fffec”ý¬š™™™™�:@gddýýefq `t×@ Ð#Ñ@gdcifdeý @�Î@ofgtiddfefef”ý±j3333338@edýdefpe  ØÓ@dcffjeýd ÀgÉ@fgtgddddfig‹ý•heš™™™™™8@dýýeepef P[è@ à<à@cedieýpe à-ß@gthheffefe’ýŠedffffffæM@ýýgfukec à<à@ à<à@gdefýpfd @ÉÔ@tehdmmfei ýŒddgý33333�S@ €¦õ@ýgftleci €¦õ@deeýpfgt ·ë@gedddfex›ý‰gdfýý33333óQ@ À ò@ à«Î@gdtcfcefefýofgtc €�ç@hdffefc—ý‡dddýýg33333“S@ À ò@  Øã@euŒdcefeeýpfgtfd €�ç@efffeŒ„ýfffffæA@¢eddýýee à<à@ à<à@qedcfedfýpfgtgdd @ÉÔ@fffef”ý„ P@dedýýeeq ðÄí@ ðÄí@–ecidefý Ð ã@pegtfhdfffe €ýªf3333339@dgýýgdtc  Øã@ Á¯@fcefjeýi ÀgÙ@egtchdddefcvý�fdÍÌÌÌÌL>@gýýgfxjf  ØÓ@ �—@cidkgýfe ÀgÉ@gtnhdddfeidd fegddhew–lcedfedqed”ehdmmfe}�±�eedý�$(¾1Ÿ�Aýhex‘dce ðÄí@fegýpfg”ŒJê„�u Afdeeefe‘¬ý›dddýdš™™™™YS@ À ò@hexedcifjeýdeg”f €�ç@ddddfkn‡ý‹dedýdeš™™™™™?@ À ò@esgdcefhfýdegyfd €�ç@dfffeiˆýš™™™™�?@˜kfdýdgd �Û@scfcefffýdegychf `RÑ@efeec€ýzš™™™™™5@dedýdefn û¹@df‚idefý ¡°@dfgyihdfffg€�ýŽfš™™™™�B@ddýýddmc Ð#Ñ@ Ð#Ñ@fchfeeýo àðÅ@egychfddefcƒý}deš™™™™ÙC@dýýgdtcf ðÄí@ €¦µ@cifefýke Ð ã@gychefffecƒýŽdfdš™™™™ÙC@ýýedpcfc `tÇ@ `tÇ@ifgfýnfg @�¾@ychfddfjcdd’gfdý à<à@dddicfcefhfýdfgy @ÉÔ@chfdddkc„ý½gedýdfffffæA@ à<à@edpcfcffeeýdfgyc @ÉÔ@hdddfkcddvdddýdd @Bù@eogdcifffýdfghhd À*ð@ddddjx™ýÍÌÌÌÌŒM@�dddýýef�$(¾1Ÿ�Aáî¬��^�AoxdceeheýpegheddŒJê„�u AfffeeŸýœ àR@ddfýýhdv�$(¾1Ÿ�Aáî¬��^�AceceeieýŒJê„�u Apeehchdfffec‘–‘eddýýedpc @Bù@ à<À@eceeieým À*ð@eghchdddeec‡ý}de33333sE@dýýgdtce �Û@ �Û@cifeeýpf `RÑ@gýchdfff Pƒ@ec•ý|ded33333sL@ýýgdtcec  Øã@ `t×@ifdeýpeg ÀgÙ@ýchdffee Pƒ@cvý}dddýffffff=@ P[è@ýhftkdci �w@ffgýefgý à-ß@ Pƒ@fdeffffŒ¥ý¦ddfýdÍÌÌÌÌ,T@ à<à@getkeciddeýdfeýh @ÉÔ@ Pƒ@hdddfik“ýˆdddýýhÍÌÌÌÌ�K@ ðÄí@ 5¤@fxndcefeeýgegýdd Ð ã@ Pƒ@dddfex‘ýfffffæM@¥dddýýee P[è@  Øã@pedciefeýpegýedd à-ß@ Pƒ@ddfes ý�33333�S@dedýýheu @Bù@ `t×@ldcifheý À*ð@pegtdddfffeŒ‡„²dedýýeeoc à«Î@ û¹@fcfdgeýl  ¡Ã@edtihdffffc£ýŸfe33333ÓS@dýýhdxce�$(¾1Ÿ�A À ò@cefdeýpeŒJê„�u Ag€chdffefc�ý¥fedš™™™™™N@ýýgeuxdc�$(¾1Ÿ�Aáî¬��^�AefeeýpegŒJê„�u A€eddffeed ý�heddš™™™™ùU@dgducecefeeddfg€chdddfkc™ý dedýýš™™™™yR@�$(¾1Ÿ�A @Bù@hex}lceddeýpfg�fŒJê„�u Ahdfffen¦ý–ediýýgÍÌÌÌ̬T@ ðÄí@  Øã@etjecedfgýpeg�eh Ð ã@dfffexªýfffff†U@¦dedýdgf  ØÃ@thdcifdeýdeg�edd Àg¹@ddein£ý¥ffffffT@mdxýdfft  ØÃ@hfcgdfeý Àg¹@degÍhhdddelhŠýƒdÍÌÌÌÌŒM@ddddhfwsdcefffdÆegÍededdfjs«ý„ddÍÌÌÌÌ,V@dýýhexed P[è@ à«Î@cefeeýoe à-ß@gÍgddddfeiªý€ded33333óT@ýýgftndc ðÄí@ P[è@effeýpeg Ð ã@°fddfffep“ý�eedýffffffC@ `tÇ@ýefpjdcg `tÇ@fdfýnfgÍ @�¾@eddddfjnŠ�«eedýd�$(¾1Ÿ�AhdvceceeieýdegmcŒJê„�u Ahdddeic™ý�ddfýýgš™™™™yR@�$(¾1Ÿ�A ðÄí@dtceceegeýpfgmchŒJê„�u AdddfecŠ�ƒeedýded `t×@ncecifgeýdegmchd @�Î@fffec€ý¥3333339@dddýdedq  ØÃ@cecefeeý Àg¹@degmchdddfic™ý¬fš™™™™yR@edýdgduc P[è@fcefieýd à-ß@egmchdddfic¡ý‘fdš™™™™9T@dýýhfvdd À ò@ `tÇ@cefheýnf €�ç@gmfddddfeeddwdddýddflcec €Ã²@idgeýdeg �¨@Ãfhddddjc‡|{deddddfoddciffeddegÃfddffffs|ý›fedýd33333³4@ û©@edncecgfefýdegÃc ¡ @hdddfhc”ý®ifdýýe3333338@ û¹@ û¹@fpcdcffdeýlfgÃgd ¡°@fddfkc†ýš™™™™™5@ dddýýef à<À@ à<À@pddcefdgýmegÃgdf @É´@ddfes‡|Šdddýdddm 5¤@cfcifdeý Þ™@degÃchdddehcŒý¯ffffffæ6@ddýdefmc à<À@lcfdeeýd @É´@egÃihdddficyr›dedýýfdsce à«Î@ à«Î@cgfgeýof  ¡Ã@gÃchdffffcšý€ded33333sO@ýýgetndc  ØÓ@  ØÓ@iffeýpeg ÀgÉ@�hddfffen�ý¯hddý33333³<@  Øã@ddekddcefeeýded� ÀgÙ@hddddfic”ý€fefýd3333338@  ØÓ@efrhecedgfýdfg�g ÀgÉ@hdddfkjˆý|dddýýeÍÌÌÌÌLF@ @Bù@ `t×@eufdciffeýpfg�fd À*ð@dfffen„ýfffffæA@™eedýdff €Ã²@redcgfeeýdfg�idd �¨@ddfhg�ýŒš™™™™YG@eedýýfdr ðÄí@  Øã@cecgfheý Ð ã@pfg�chdfffec¤ý›fÍÌÌÌÌ�U@ddýýgetf ðÄí@ à<à@dcefgeýp Ð ã@fg�fddfffes˜ýªgdfffffæQ@dýdedpce�$(¾1Ÿ�AceegeýdeŒJê„�u Agjchdddfic…ý˜gdqš™™™™ÙO@ýýgetnec €¦õ@ �Û@gdeeýpfg ·ë@jgedfefexŽƒ£hddý À ò@ýeeqddce �Û@eeeýpegj €�ç@iddfffec”ý©fdgdd ÀL@gdtceceehedqfgjchdddeec¬ýŠfedýýgš™™™™YS@�$(¾1Ÿ�Aáî¬��^�AduceceefeýpegjchŒJê„�u Adfffec•ýfffffæI@£ddxýdgd  ØÓ@tcfciggeýdelŒchd ÀgÉ@ddfhc©ý�fffffÆV@dddýdhfw à<À@ndcmfieý @É´@degŒfddddfh‚ŒýƒgfffffæA@ddýýdepd P[è@  Øã@dcifgeýp à-ß@egŒgddddfek¡ý feš™™™™9T@Æýýhetie @Bù@ ðÄí@cedfeýpf À*ð@gŒeedfffei¬ýŸgddš™™™™YS@ýýgetuec @Bù@ @Bù@gdeeýpeg À*ð@Œhhdddfel—ýšfedýš™™™™ÙO@�$(¾1Ÿ�Aýhdvcfce ðÄí@efeýpfgŒŒJê„�u Acgdfffec©ýŠfedýýfffffÆV@�$(¾1Ÿ�Aáî¬��^�Ahdtceceffeýpeg„cŒJê„�u Ahefffec£µ“fdiýýh�$(¾1Ÿ�A À ò@dvceceeeeýpfg„chŒJê„�u Adfffec—ý33333“S@šdddddgdtcecefegdqfg„chffffec“ý}33333SP@dedýýgdt À ò@ P[è@cecifdeý €�ç@peg„chdfffecƒý†fš™™™™ÙC@fdýdefpe À ò@fcidegýd €�ç@fg„ehfddfkf©ýšfdfffffÆV@‘ýdhdvce�$(¾1Ÿ�AcefdeýdfŒJê„�u Ag„chdddfic’˜�ddÆýýhdvcec À ò@  Øã@eeeeýpfg €�ç@„chdfffec†uydefý  ØÃ@ýeernlci  ØÃ@dfeýmfgý Àg¹@ (‰@ehdffffh‘¢¦fedýý ðÄí@ ðÄí@edpcfcifdeýpegýc Ð ã@ (‰@hdffeec‹�©gddýdd ðÄí@dpcecefdeýdegýcf Ð ã@ (‰@eddeic£¼–fedýýhd�$(¾1Ÿ�A ðÄí@ucecefggýpfgýcleŒJê„�u A (‰@fffec¡ý¥ @N@fddýýeeq P[è@ �Û@fdcefggý à-ß@pfgýgdef (‰@ffed¬ý®hfffff�W@edýýhdwc�$(¾1Ÿ�A @Bù@ecefdeýpŒJê„�u Aegmchdddeic¬ý¨feš™™™™YS@dýdgdtce�$(¾1Ÿ�AcefieýdeŒJê„�u AgmchdddeicŽý¥eed33333³U@ýýhdvcfc à<à@ à<à@geheýpeg @ÉÔ@mchdffeec§ýšfddýffffffR@�$(¾1Ÿ�Aýgdtceceáî¬��^�AffeýpegmŒJê„�u Achdffeec©ý£feeýýfffffÆV@�$(¾1Ÿ�A €¦õ@hetnfcfddeýpegmdŒJê„�u AhdfffffŽý¦eeeýýgš™™™™�;@ €¦õ@ €¦õ@evfecgdfeýpegmgh ·ë@dffeed­ý  U@Œfddýdhd ðÄí@tcecefhgýdfgmche Ð ã@ddfkc¬ý«š™™™™YS@gedýýgdt�$(¾1Ÿ�Aáî¬��^�AcfcgehgýŒJê„�u ApegmcheddficŒýˆgfffffæ6@ddýýdfme `t×@ `t×@fcedggýp @�Î@eg�ghffffej‚ýžfdÍÌÌÌÌ�I@ddýeetdd `t×@cgfdgdpfg�ddefffeh”ý‰fdd3333338@ýýedscfc à<à@ à<à@gfgfýpfg @ÉÔ@�chdfffec�ý—feddš™™™™™N@defsedcgfffdqfg�dddfffef—ý idddd33333³<@feuhdcgfdgddfm�ededdfihšý†eedýýe33333sO@ ðÄí@ ðÄí@dqcecifdfýpfg�ch Ð ã@dlffec‘ýfffff&C@†edlýýhf P[è@ `t×@xsfcgddeýpfmýghe à-ß@ px@ddfkx˜�©dedýýgdt  Øã@ à<à@cfciekeý ÀgÙ@pfeýchdf px@feec™ý¨ffffff&Q@edýýeerg  Øã@  ØÃ@dcefdfým ÀgÙ@fgýgdddd px@fif…ý„geš™™™™�G@dýýedqce À ò@ À ò@ceeffýpe €�ç@gýcedfff px@ecŠý®ded33333³9@ýdeepedc �Û@eejeýdfg `RÑ@ýpddddfi px@d�ýƒgddý33333³3@ Ð#Ñ@ýefmhfci Ð#Ñ@dggýoegý àðÅ@ px@fhfeffei™ýœdddýý33333SQ@ À ò@ û©@getffcedhgýhfgýf €�ç@ px@hefffeh™ý‰eddýýfÍÌÌÌÌŒD@ `t×@ `t×@etfdcifefýpfgýfd @�Î@ px@dddfei�ý333333J@–eedýded @Bù@rcecefeeýdeg{chd À*ð@ddfic•ý£33333³F@dedýýedp À ò@ Ð#Ñ@cecefdeý €�ç@oeg{cedddfhc ý‘d33333�S@egýýffri P[è@  Øã@fciddfýp à-ß@fg{fedfffexŽý–fefffff¦A@dýýddnce À ò@  Øã@ceekeýpe €�ç@g{chdfffecdd{dddýdefqifc �Û@idhgýdeg `RÑ@{ghddddjf�ýydddý33333³<@ @Bù@ýedpceci à<à@edgýpeg{ À*ð@cedddfec}ý dedýýÍÌÌÌÌÌ7@  Øã@  Øã@eepcfcidfeýpeg{e ÀgÙ@edfffecªýŒddfýýe33333óT@ P[è@  Øã@dpcfcefdgýpefých à-ß@ ��@ffffec¡ý33333³R@€dddýýgf €¦õ@ €¦õ@ujdcifdeýpegýedd ·ë@ ��@fffe–¤ý�  T@ddhýýhfx  ØÓ@  ØÓ@secidgeý ÀgÉ@pffýdedf ��@effgddzdfdýdgete `t×@ecidggýd @�Î@efýfhfdd ��@djs„ý�ifš™™™™�8@dýýddjcf  ØÓ@ `tÇ@cifgeýnf ÀgÉ@gýchfeff ��@ec ý�dddš™™™™ùU@ýýhfvgdc À ò@ À ò@efhgýpeg €�ç@ýgdfddfe ��@}‘ýzdedýfffff�T@  Øã@ddeoddciffeýdegý ÀgÙ@ ��@iddffffitý…eddýýš™™™™™*@ P[è@ �w@eep‚dcefhgýefgýi à-ß@ ��@ddffffŒ¡ýŽedfýýgš™™™™9T@ ðÄí@ ðÄí@duclcifeeýpfgýce Ð ã@ ��@dfffec‘¢�dedýýge €¦õ@ ðÄí@tmdcieeeýpeeýedd ·ë@ ��@fffee‹ý«ÍÌÌÌÌ�C@fddýdeer À ò@kecedgeý €�ç@dfg•gfdddfig´ý�g333333W@edýýhdxc�$(¾1Ÿ�Aáî¬��^�AfceffeýpŒJê„�u Aeg•chdffeec€ý¡fe3333339@dddedpcfcefeedded•cheddfic‘ý´hed333333M@ýdefqcfc à«Î@fdjeýdeg  ¡Ã@•iedddficˆý‚ddfýÍÌÌÌÌLF@ À ò@ýgdtcece  Øã@fgeýpfg• €�ç@chdfffec„ý…eeddý333333:@ €¦õ@eepdecedgedpeg•ghdfffef—ý©gdgýýgš™™™™YR@ À ò@ à«Î@ftiecgddeýoegofh €�ç@dddeik˜�‰gedýýge À ò@  ØÃ@sedcefgeýmffofde €�ç@ffeff®ýª U@fedýdhet�$(¾1Ÿ�AkdceeeeýŒJê„�u Adegoeddddfirƒý¤gš™™™™™3@deýýgdtc P[è@ û¹@fcgfjeýl à-ß@fgochdddffcŸý°ee33333SQ@dýdhfved à«Î@cgfdeýdf  ¡Ã@mogddddfifŠ�°fddddeerdfcedgedqfgoghdddffd�ý–fdddfffffæR@dgfumdcgeeedqegoeddffeemœýŠgdfýýÍÌÌÌÌLK@ P[è@ `t×@gdtcfcefhgýpfgýc à-ß@ @p@heddekc’ý±dddýdg33333³N@ `t×@etcdcefegýdfgýfd @�Î@ @p@fddeic–ý33333³L@•fdfýýed  Øã@ `t×@pcfcefggýpfgýche ÀgÙ@ @p@ddfec ý�33333óT@ffdýýeen  Øã@  Øã@slcgdieý ÀgÙ@pfgýfhff @p@ffes”ý•dÍÌÌÌÌŒC@ddýýedrc P[è@ �w@fceejeýe à-ß@fgýchddd @p@fkc“ý¤ddffffffC@dddeepfdcefgeddfg‹gddddfkf•ýˆged33333³F@ddedpcfcefmgdqeg‹chdddfkc�ý•fedýÍÌÌÌÌLJ@ P[è@ýgftedce à<à@egeýpfg‹ à-ß@fddfffek ý‘hddýý33333óT@ ðÄí@ ðÄí@fet‚ecedeeýpeg‹e Ð ã@hfddfen¦ý¨fdhýýeÍÌÌÌ̬T@  Øã@ �Ë@dqcfcefheýnfg‹ch ÀgÙ@dddfic†ýš™™™™YA@³geÆýdee `t×@scfcfddfýdeg‹ehd @�Î@ddfic–ý™33333³L@fddýýeep ðÄí@  Øã@edceeieý Ð ã@peg‹gddfffek–ý®ffffff&Q@dfýdeerk à<à@lcgdjeýd @ÉÔ@fgýhhddd ¸’@fih‘–±dedýdeepgl �Û@cfdheýdf `RÑ@gýhheddf ¸’@if™ý‚eed33333SQ@ýýget‡dc�$(¾1Ÿ�A  Øã@eeieýpfgŒJê„�u Aýeddfefe ¸’@xdd�gdiý�$(¾1Ÿ�AdgdtcfcefjeýdfgýŒJê„�u A ¸’@chddddkc¦ýœiddýýÍÌÌÌÌLO@ `tÇ@ `tÇ@efrxdcgekfýnegýj @�¾@ ¸’@ddddff}ƒýƒhddýýdš™™™™™3@ €Ã²@ €Ã²@gqfdchfgfýjffýkd �¨@ ¸’@dffffn”ý3333338@¥hedýdef û¹@redcgfgfýdfdýhdd ¡°@ ¸’@ddfie’ý›ÍÌÌÌÌ�N@lfdýýdej  ØÓ@  ØÓ@efceddgý ÀgÉ@pegýehff ¸’@ffee…ý›f333333+@edýýeepc à<à@ à<à@dcifgfýp @ÉÔ@fgýiddff ¸’@fec†ýšgeÍÌÌÌÌÌH@dýýedrce à<à@ �Ë@cefjgýne @ÉÔ@gýchdfff ¸’@ec„ý¯eedfffffæ:@ýdefodec à<À@iddfýdeg @É´@ýhhdddfk ¸’@kŠý™fedý ÀN@ `t×@dgftddcgfdfýdfgý @�Î@ ¸’@jddddfiý 0�@”ý‘gddýd3333338@ à<à@fesnfceddfýdfmÙf @ÉÔ@hefffevŒý«eddýýe33333³@@ à<à@ à<à@epdlcgdheýpfgÙfh @ÉÔ@dddfef™ý33333SQ@„fedýýee À ò@ �Û@tefcedheýpfgÙgld €�ç@ddfkrœý…ÍÌÌÌÌ,S@edjýdgev @Bù@dfcedfgý À*ð@dfgÙfheffffn‹ý¬gÍÌÌÌÌìP@djýdhdxc ðÄí@eceffeýd Ð ã@fgÙchdddficŠýªee33333³=@dýdedpcf `t×@ciefeýde @�Î@gocedddfic�ýžede33333óC@ýýeepdec �Û@  ØÓ@gddeýpfg `RÑ@oehdffeef™ýšdddýfffff&Q@ À ò@ýget‚dce P[è@eheýpfgo €�ç@fddddfe‚ƒýªgedýýš™™™™ÙC@ �Ë@ �Ë@edqcfcefjeýnfgoc `RÁ@hdfffecŒý“gedýýeš™™™™�E@ @Bù@ @Bù@dpcfcgedfýpegých À*ð@ (‡@dfffec ý33333SP@“fedýýge À ò@ P[è@tfdceeieýpegýedd €�ç@ (‡@fffef”ý… ÀL@iedýdffr ðÄí@gdceefeý Ð ã@degýfddf (‡@feexŒý“ffffffæ6@ddýýedqc Ð#Ñ@ 5¤@ecefffýg àðÅ@egýchddd (‡@fkc‹�Œfedýdgducf P[è@cifeeýdf à-ß@gýchdddf (‡@jc…ýŽhdd333333+@ýýefsddc `tÇ@ `tÇ@gfefýnff @�¾@ýiddfffe (‡@x®ýœeedý U@�$(¾1Ÿ�Aýhexxdce À ò@ejeýpflýŒJê„�u A (‡@gddeffenŠ�¦gddýd P[è@eesfdceejeýdeghg à-ß@ddddfie¤ý›fdeýýg  T@ ðÄí@ à<à@etjfŒedgeýpfghgh Ð ã@dddfgi’˜®jdeýdee @Bù@qeeceddfýdeghlfd À*ð@ddfi}ˆývš™™™™™8@dedýdeep�$(¾1Ÿ�AddcifheýŒJê„�u Adfgheddffffi”ýžf3333338@ddýdedqc Ð#Ñ@eceegeýd àðÅ@fgichdddficƒý°ldÍÌÌÌÌÌB@dýdferid û©@cfegeýdf ¡ @gifddddfigƒý¦fddš™™™™ÙC@ýdgetddc P[è@gedfýdeg à-ß@ijddddfic“ý¢eedýffffffC@  Øã@ýfdscfce  ØÓ@ejeýpfgi ÀgÙ@chdfffec„ýªledýdffffff*@ Ð#Ñ@eepcfcfdffýdfgih àðÅ@hdddfkc‡ý™fddýýeš™™™™™?@ P[è@ `t×@erifceddfýpfgief à-ß@dfffeg§ýffffffR@·fedýdhd P[è@xcecefdeýdfg¡chd à-ß@ddehc‘ý¦33333³3@dedýýgdt P[è@ Ð#Ñ@cecffeeý à-ß@oee¡chdffffc…ý‚e333333+@edýýefpd @Bù@  ØÓ@ecidffýp À*ð@eg¡ehdfffefŠý¡gd33333³9@dddeepxdcefiedqfg¡fdeddffeŠ�µfedddeepdeceddfddfg¡ihdddekc”ý«demý3333338@ Ð#Ñ@deerefcidjeýdfg¡ àðÅ@ghdddfig ý´gedýd33333�S@ @Bù@hfveecfddeýdfg¡f À*ð@hdddfid¡ý®eepýdgÍÌÌÌÌìQ@ À ò@eugfcedjeýdfg¡he €�ç@dddfil—ý33333“S@�gedýýge�$(¾1Ÿ�Aáî¬��^�Atldceeheýpeg�eddŒJê„�u Affee–™ýžš™™™™yR@fedýýeds @Bù@ À ò@clceeheý À*ð@peg�chdfffecŒý“gÍÌÌÌÌL<@ddýýeeqf  Øã@ �Û@ecgdkeýp ÀgÙ@eg�ehdffeexŽý›idfffff&L@dýdeepde €¦õ@cedgeýdf ·ë@g�ghdddfkfžý�dddš™™™™yT@ýýgftxdc À ò@  Øã@efegýpeg €�ç@�idefffes¥ýˆfddýÍÌÌÌ̬U@�$(¾1Ÿ�Aýget‚dce P[è@fdeýpeg�ŒJê„�u Aeddfffes�ý�hdfýýfffffFR@�$(¾1Ÿ�A ðÄí@getgecedjeýpfgngŒJê„�u Ahdddfei‘ý�gedýdgfffff&C@ @Bù@dtcecefjeýdfgnch À*ð@dddfkc‘ýš™™™™™C@~dedýýed À ò@ à<à@pcecifheýpegnced €�ç@fffec ý¤33333�S@gegýýget @Bù@ €¦õ@xecedeeý À*ð@pegnfhdfffeh’˜«fedýdeepf�$(¾1Ÿ�AdceejeýdŒJê„�u Afgngddddfi€ ý§fe33333�S@dýdgducl �w@cefieýqe �m@gnchdddficddƒeddýdgdtcfc P[è@efggýdfg à-ß@lchfdddkcddŒedgý P[è@dgdtcfcefggýdegl à-ß@chfdddkc‡ý“dddýý33333³7@ Ð#Ñ@ Ð#Ñ@eeoeecidgeýofglf àðÅ@hdfefed�ýªgedýýe €=@ à«Î@ à«Î@dpcfcffeeýofglch  ¡Ã@dfffec ýš™™™™ùU@�dddddgftlfcedggdqfmlfhfddfeh‹ý©ÍÌÌÌÌìP@fdeýdeer À ò@clcedgeý €�ç@degllhdddfic’˜¡eddýýferx À ò@  Øã@lcedeeýp €�ç@fglgfdddflg“ý¢feffffffC@dddferddcefiedqfglfddffffd“ýydedffffffC@ýýedrcec À ò@ `tÇ@ifdeýnff €�ç@ýchdddfj Ø—@c“ý�ddeýffffffC@ à<à@ýedqceci à<à@fdeýpegý @ÉÔ@ Ø—@cfdfffec—ý‡ddfýýš™™™™ÙO@  Øã@  Øã@hfv„ecideeýpegýi ÀgÙ@ Ø—@edfffe–”ý§fdfýde3333338@ `tÇ@eqdecfddfýdfgýjh @�¾@ Ø—@dddfic‡ý33333óE@°gedddddncfcffdfddfgýchd Ø—@ddfic|ý‚33333³4@dddýýedq €Ã²@ €Ã²@cfcifjeý �¨@jegýcedf Ø—@fffcŠý›e33333³9@edýdefqe �—@dcgfdfýd ��@fgýjdddd Ø—@fhe}ý£kdÍÌÌÌÌÌ7@dýdeepcf à<À@cfddfýdf @É´@gýjhdddf Ø—@lc™ý~eddÍÌÌÌÌŒD@ýýfdrcec à<À@ à<À@ifffýmfg @É´@ýchdddfh Ø—@c”ý•fedý3333338@ à<À@ýddocech û©@fdfýhfgý @É´@ Ø—@cedddflc…ý„dddýý333333+@ û¹@ �w@efpjecidjfýeegýf ¡°@ Ø—@hdddfhsƒýŽeehýýgš™™™™™3@  Øã@ �w@etgeceddeýefgsgh ÀgÙ@dfffff…ý33333sF@–eddýdee û¹@qedcifeeýdfgshdd ¡°@fffex‡ý”ÍÌÌÌÌÌ5@gddýddel À ò@dfcedgeý €�ç@dfgseheddfkd}ý°gÍÌÌÌÌÌ7@edddeenkdcfejeddfdspddddfip™ý™edš™™™™yR@dýýeerŒd�$(¾1Ÿ�A ðÄí@cefieýpeŒJê„�u Adsgddfffex„ýƒgddfffffæ:@ýdeeqidc �—@efjeýdfg ��@shddddfkn’˜‰dd}ý P[è@ýgduceci P[è@fffýpfgw à-ß@cedfffec”ý’hedýý3333338@  Øã@  Øã@edpcfcefdfýpfgwc ÀgÙ@fdeffec¥ýŒddvýýhÍÌÌÌÌ,T@ �Û@ �Û@dwclciffeýpfgwch `RÑ@dddfec‹ýÍÌÌÌÌÌ4@}eddýýef `t×@  ØÃ@qddcifefýmfgwgdd @�Î@ddfke‡„¡eddýýeep À ò@ P[è@decedigý €�ç@peg¡ehdddfee“ý–d33333SP@dedýeeq€ 5¤@eceddedgfg¡ffdddfes‡„—gedýýedpce À ò@ À ò@cgeggýpe €�ç@g¡cedfffec•ý�gdd33333³F@ýýeesddc�$(¾1Ÿ�A  Øã@eefeýpegŒJê„�u A¡gddddfhn”ý³iedý3333338@ `t×@ýfdrcfcf �Ë@fffýnfgý @�Î@ f @chdddflc„ýƒfeeýýš™™™™�1@ �Ë@ €Ã²@eeriecgdefýjffýf `RÁ@ f @hdffeffŠý|gddýýeš™™™™™C@  ØÃ@  ØÃ@fqndcifgfýmfgýid Àg¹@ f @dddfex’˜¡gddýýee  Øã@ `tÇ@pnecedffýnfgýifd ÀgÙ@ f @ffffk‘ý¢fffff&C@iddýýedp `t×@ `t×@cfcffefý @�Î@pfgýchdf f @ffec‘ý¤efffffæM@edýdgetd Ð#Ñ@dcmegfýd àðÅ@fgýÆdddd f @fic‡ýšed33333³7@dýýedpce à<à@ à<à@cifefýpe @ÉÔ@ggchdfffecŠ�|eddýýefpeec à<à@ à«Î@idgeýofg @ÉÔ@gkhdddfeg�r³dedý  ØÃ@dddocecfegeýdegg Àg¹@chdddeicŒý…fdhýd33333³@@ `t×@efteecidffýdfggj @�Î@hdddfkx…ý�hddýýe333333+@ û¹@ €Ã²@dpcecefefýjfggch ¡°@dfffec—ýfffff&J@ˆeddýýge P[è@ à<à@tedcieheýpegýfdd à-ß@ ˜�@ddeeiƒýwš™™™™ÙC@dedýdefp�$(¾1Ÿ�AÈecidgfýŒJê„�u Adegýdhdf ˜�@ffeÈ�ýŸfffffffL@ddýýedpc P[è@ P[è@ecgfgeýp à-ß@fgýchdff ˜�@fec‹�£hddýdeeqed À ò@ceeheýde €�ç@gýdddddf ˜�@ip�ý�eddš™™™™™N@ýýedpcec Á¯@ Á¯@ifieýiem S¤@ýchdddfe ˜�@c­ýˆddsd33333sU@dgducecifmedqfgý ˜�@chdddfec™ýˆgddýý33333SQ@ À ò@ ðÄí@hfv}dceedfýpfgýg €�ç@ ˜�@ddfffex|ýŒffdýýe33333³4@ �Û@ �Û@fmcdcefgeýpedýfd `RÑ@ f @ffffecŠ�…hddýýdf à«Î@ `tÇ@ndfcedgfýnegýihd  ¡Ã@ f @fffen‡ý„33333³7@gedýdefq €¦µ@ldchfdeý ·«@dfgýgddf f @ffegˆý“hš™™™™™8@fdýdeepe  ØÓ@dcefgeýd ÀgÉ@ffýfdfdd f @fkgdd‹edeýdeep}f ðÄí@cedjeýqe Ð ã@gýehdddf f @e}–ý}ddf33333³L@ýdeftsfc  Øã@idegýdef ÀgÙ@ýhhdddfe f @n“ýxdegdffffffC@deesflcideeddfgý f @ehdddfjlƒý„hedýýÍÌÌÌÌÌB@ À ò@ P[è@edrcfceeegýpfgýc €�ç@ f @hdfffeczý„dddýýefffffæ2@ à<à@ à<à@fpcecidfeýpegýfh @ÉÔ@ f @dfffecdd’dddýdef  ØÓ@pcfcedggýdfgýhhf ÀgÉ@ f @dddkctý�š™™™™�6@dedýýefp  ØÓ@  ØÓ@cdcgfeeý ÀgÉ@pegýfdef f @ffec‡„ efdýýedpc à<à@ à<à@lcefgeýp @ÉÔ@egýchfff f @fec}ýŸdd333333)@dýýdfmcf `t×@ `t×@cidgeýpf @�Î@gýhhfffe f @ec¡ý“gedš™™™™9T@ýýhdxcfc @Bù@ `tÇ@eeieýnfg À*ð@ýchdffef f @c“ý�eedýš™™™™YO@  Øã@ýffrffci  Øã@djeýpfgý ÀgÙ@ f @ghdfffen©ý§fedddfffffÆV@hexfdcgfieddfgýf f @ddddehy˜ý¥gedýýhfffffæQ@�$(¾1Ÿ�Aáî¬��^�AdvcecgfeeýpefýchŒJê„�u A f @dfffec ý33333�S@¡fdeýýgd `t×@ �w@tcecefieýefgýchd @�Î@ f @ffefczýˆš™™™™™-@hddýdeep Ð#Ñ@eecedheý àðÅ@dfgýdhdd f @dfkx ý‰d33333�S@ddýdegpd �w@ecidkeýd �m@fgýfhddd f @fkx‘ý´eeÍÌÌÌÌLG@dýdeerfd ðÄí@ceejeýde Ð ã@gýiddddf f @idŠ�~ddgýýedrcec @Bù@ Ð#Ñ@iegeýofg À*ð@ÂchdfffecŠý…ddfýÍÌÌÌÌŒM@ @Bù@ýgdtceci  Øã@fjeýpfg À*ð@chdddfecªý�ddfýý33333óT@ €¦õ@ ðÄí@getsfcedjeýpegÂe ·ë@hdfffe‚œý†dddýýgfffffæJ@ P[è@ P[è@ftodcifgeýpfgÂhd à-ß@dfffee¤ýÍÌÌÌÌ�U@–eenýdfe  Øã@refcedgeýdfgÂhhd ÀgÙ@ddfknŠýž ÀN@deeýdfet @Bù@kfcedheý À*ð@dfgÂghdddfksŠý‹dÍÌÌÌÌŒM@deýýhdxc P[è@ P[è@ecifdfýp à-ß@eetchdfffec„ýªgdfffffæA@dýdedpcf à<À@cffdfýdf @É´@gtchdddficyr|iddýýefqhec �Ë@ à<À@idefýmfg `RÁ@tfhdddfejƒýŒhedýš™™™™™3@ Ð#Ñ@ýedrceci Ð#Ñ@fefýofgt àðÅ@chdfffecŽý”fedýý33333óM@ P[è@ P[è@ferddcgejeýpeg¤e à-ß@ddfffee�ržeedýýd 5¤@ 5¤@fkcfcfdfeýged¤fh Þ™@dfffec“ýffffffC@�gddýýee à<à@ à<à@pddcfefeýpeg¤fdd @ÉÔ@ddfed‡}°gddýdden �Û@cfcgdjeý `RÑ@deg¤ghdddfic”ý™ffffff&K@ddýýedoc À ò@  Øã@eceejeýp €�ç@eg¤chdddfecyr‚dedýýfesdd à<à@ �—@ceeieýff @ÉÔ@g¤gddffffg}ý¤fdd333333)@ddddfcfceffedÆfd…chfddffc~ýŸhedý €<@ à<à@defnddcefkeýdfg… @ÉÔ@fddffeff‘ý¼gddddfffff&C@edpcfcffieddfg…chdddflc£µ¡eedýýg €¦õ@ ðÄí@etxdcefdfýpeg…gd ·ë@dfffee€ýš™™™™™5@–fdeýýge Ð#Ñ@ û¹@tdfcedfgýleg…ehe àðÅ@ffffk ý°š™™™™ùU@ddxýdhew  ØÓ@jlcidfgý ÀgÉ@deg…eheddfiedd�dfddddfjkdcifgeddfg…gdfdddkd}ý�fe333333)@dýýefppd û¹@ û¹@cifdfýlf ¡°@gý|ddffe @z@fl€ýydddš™™™™™5@ýýeepedc `t×@ à<À@ifdfýmeg @�Î@ýgddfffe @z@p–ý�deeý33333³L@ À ò@ýgetgeci  Øã@ddfýpfgý €�ç@ @z@ihdfffep•ýŸkddýdÍÌÌÌÌ�O@ P[è@dflkdcefkfýdfgýf à-ß@ @z@ddddfkp—ý„fedýýefffff&J@ �Ë@ 5¤@dqcechfdfýgegých `RÁ@ @z@dffeec†ýš™™™™™5@‰fddýýdd  ØÓ@  ØÓ@jcfcefggýpegÑchf ÀgÉ@fffec—ý‰fffff&J@ffdýýffs Ð#Ñ@ Ð#Ñ@kdcieeeý àðÅ@oegÑedffffe‚ ýŒd33333�S@edýdgfvf  ØÓ@dcifdeýd ÀgÉ@egÑfdefffef”ý—de  Q@eýdferne  ØÓ@cidieýdf ÀgÉ@eÑehdfffe‡ƒý~eddš™™™™™3@ýdedpcfc P[è@efeeýdfg à-ß@Ñchfddfkc¨ý«dedýš™™™™yV@�$(¾1Ÿ�Aýgevkdce P[è@eieýpemÑŒJê„�u Agdeddfii‘ý™gdfýýfffff&C@ P[è@  Øã@gdtcfcgedfýpfg§c à-ß@hdfffec‚ý‘fdeýýeÍÌÌÌÌ�I@  Øã@ `t×@friecedgeýpeg§eh ÀgÙ@dddfen–ýfffff&Q@™fedýýgd @Bù@ @Bù@tcecefieýpeg§chd À*ð@fffecƒýzš™™™™™3@ddfdýeer 5¤@elcidefdgfd§ihdddffpŸýžfš™™™™yR@edýýhdxc €¦õ@ €¦õ@fcgfdfýp ·ë@fg§chdfffec™ý–ed33333SQ@hýýgetle�$(¾1Ÿ�A @Bù@cedjeýpeŒJê„�u Ag§ghdfffet–ý˜eddfffff&Q@ýdddncec P[è@geieýdfg à-ß@€chdddficƒý”hddýÍÌÌÌÌÌB@ ðÄí@ýedpcece `t×@fjeýpfg€ Ð ã@chdddfecŠ�¢fedýd à<à@edpcecefgeýqfg€c @ÉÔ@hdfffec”ý~fedýýeš™™™™™<@ P[è@ �Û@epedcifieýpfg€ed à-ß@dfffef‡ýÍÌÌÌÌÌ5@Ÿhdddýeg �—@pdfcgdfedffg€fhdddfhk‹ý¤ÍÌÌÌÌ�C@eedýdeep ðÄí@decedieý Ð ã@dfg€phdddfie¡ýŠgš™™™™9T@edýýgfvn P[è@  Øã@dcefdfýp à-ß@fg€gddfffen‡ý˜heÍÌÌÌÌÌ5@dýýdflcd `tÇ@ û©@cgfdfýhf @�¾@d§lddddekc€ý€dedš™™™™™5@ddefpýdc 0�@ifjfddfg§Æddddfký¡ý­fedý 0�@š™™™™9T@ P[è@ýhdwcfcg P[è@gefýpfm§ à-ß@chdddficŠý�dddýdÍÌÌÌÌŒM@ û¹@edpcfcefgeýdeg§c ¡°@hdffffc’˜„fddýýf `t×@ `t×@erxdchfdfýpfg§fd @�Î@dfffex}ý333333)@³iedýdde �—@ocfcfddfýdfg§fhd ��@ddfic¤ý”ÍÌÌÌÌ�U@fddýýgex À ò@ P[è@ddcegheý €�ç@pfgÀgdefffev‡ý™eš™™™™™?@ddýýedpc �Ë@  ØÃ@eceffeým `RÁ@egÀchfddfec‡ý‘fd33333ó@@dýýdeodd `tÇ@ `tÇ@cefgeýne @�¾@gÀedfffeec…ý�gef333333+@ýdeerefc  ØÃ@hdeeýdfg Àg¹@Àehdffffn¤ý„eedý  T@ ðÄí@ýhdxceŒe à«Î@fjeýofg¹ Ð ã@cfdfffgc†ý}fedýdš™™™™YA@ à<À@edpcecifefýdfgýc @É´@ Àd@hdfffec�ý³eddddd33333³<@dmcfcfeefddedých Àd@dddfic‰ýš™™™™™@@›eddýded à<À@pcecifdfýdegýchd @É´@ Àd@ddflc’˜‹fdgýýees à<à@ �Û@xlciddfý @ÉÔ@pfgýfhdf Àd@ffen‰ý²fš™™™™™@@edddeerddcffdfddegýldddd Àd@fic¢ýšge333333I@dýýeered P[è@ à<à@ceedfýpe à-ß@gýhddfff Àd@ed|ýxddd33333³4@ýýedpclc `tÇ@ 5¤@ifffýgeg @�¾@ýchdddfl Àd@c’ý¬iddý33333³N@ à<À@dddlcfchfjeýdfgý @É´@ ðx@cheddfic”ý–efdýý3333338@ P[è@ Ð#Ñ@ddpcfcefgeýofgýc à-ß@ ðx@hffffecŠý³gdddde333333J@epcdcgededdegýid ðx@eddfic ý33333�S@½eddýded �—@pcfcgfeeýdfdýche ��@ ðx@ddfic|ý¯ffffffA@eddýdedp û¹@cfcffggý ¡°@dfgýched ðx@dfic~ý™fš™™™™™/@fdýýefpg `tÇ@  ØÃ@dchfgeým @�¾@fgýfdfff ðx@fei‡„“fddýýedpce `t×@ `t×@cefeeýpe @�Î@gýchfddf ðx@ecŠý¯fed33333³9@ýýeepedc P[è@  Øã@efdfýpeg à-ß@yfdefffeddd”eddý `t×@deep|dcefeeýdfgy @�Î@hdfdddkn¡ý£gedýýš™™™™9T@ À ò@  ØÃ@getedcffefýmfgyi €�ç@ddffffrŠ��gedýýh P[è@ P[è@dtcfcefheýpegych à-ß@effeec„ýfffffæA@ƒdfdýýhf à<à@ û©@wfeciddeýhegkfhf @ÉÔ@ffffŒ—ýŒ33333óN@edhýýhdx À ò@ À ò@cecefeeý €�ç@pegkcheddfecˆý®dfffffæB@edýýhevx  Øã@ �Û@ecidheýp ÀgÙ@egkihdffffn‡ý”gdÍÌÌÌÌÌ5@fýdgetef ðÄí@cedgeýdf Ð ã@g‰ehdddfkh–ý¨dddfffff&Q@ýdeereec�$(¾1Ÿ�AedfeýdfgŒJê„�u A‰ghdddfig•ý½fedd33333sL@defpddcffdeddfd‰eddddeic”ý|ddddý3333338@ û©@feqrdcifefdhfg‚gdefffer€ýŒefdýýe3333339@ �Ë@ �Ë@fsefcgdegýneg‚fh `RÁ@fffeefyý333333+@€dddýdef P[è@neecedeeýdeg‚fhd à-ß@ddfejddvdddýdefq �w@kdcifefý �m@dfg‚gdddddjyŠ�…dddýýedqc `t×@ `t×@ecieieýp @�Î@eg‚chefffecŽƒŽeddýýedpcf�$(¾1Ÿ�Aáî¬��^�AcgeieýpeŒJê„�u Ag‚chdffeec�r‡gddýýddgcfc à«Î@ à«Î@efggýoeg  ¡Ã@‚chfddfec�ý‚eedýfffffæE@ @Bù@deerfecedgeýdfg¾ À*ð@fhdddfkt‹ýœfdfddÍÌÌÌÌìP@eeqxecedhgdqeg¾ehdffeex¬ýŒgdÆddhÍÌÌÌÌ�W@dwcmceggedqfm¾chdffffc®ý U@ fdkýýhe�$(¾1Ÿ�A à<à@xslcedgeýpfg¾ghdŒJê„�u AfefefddŒffdddedpcfcefeeddfg¾chfdddkc¬ý‡dš™™™™YS@edýýhexh�$(¾1Ÿ�A @Bù@dcifdeýpŒJê„�u Aef¾dddeffe‚ƒý¦idÍÌÌÌÌÌB@dýýefqdd  Øã@ �w@cgfffýee ÀgÙ@gýgddddf P†@ic‡ýžgddš™™™™™?@ýdeersfc  ØÓ@gddfýdfg ÀgÉ@ýeedffff P†@g‘ý•gddýš™™™™™C@  Øã@ýeeqhdch  Øã@eifýpegý ÀgÙ@ P†@iddddfee•’Ÿdedýý à<à@ û©@gducfcigmfýhegýc @ÉÔ@ P†@hdefefcƒý‚gedýýeš™™™™™3@ `t×@ Á¯@fpddcifdfýifgýgd @�Î@ P†@dddfj‚ƒýÍÌÌÌÌÌB@£fedýdef `tÇ@qdechddfýdegýrhd @�¾@ P†@ddfic‹ý­š™™™™™8@dddýddeo €¦µ@ddcifdfý ·«@degýgddd P†@dfid”ýœhš™™™™�:@ddýdddnc €Ã²@ecifefýd �¨@fg£chdddfkcxý~dd333333%@fýýedscf à<À@ Á¯@cififýie @É´@g£chdfffecˆý~ddnfffffæB@ýýedtcec P[è@ �Û@egdfýpfg à-ß@£chdddeec}ýxeddýÍÌÌÌÌÌ7@ €¦µ@defp|dcifgeýdeg£ ·«@jddddfhp�ý—eedýý €=@ `t×@ `t×@edqcecffheýpfg£c @�Î@hdfeffc‡|‹dddýýe Ð#Ñ@ û¹@epdlcidfeýleg£gh àðÅ@dddfkf—ý33333“S@žgedýýee P[è@  Øã@pŽdcefeeýpfg£edd à-ß@fffep‘ýºfffff&C@fedýdedp à<à@cfceejeý @ÉÔ@dfgnchdddfkc�ý�fÍÌÌÌÌLK@edýýeepy ðÄí@ ðÄí@ecgdeeýp Ð ã@fgnhhdffferƒý“feš™™™™ÙC@dýýgetll�$(¾1Ÿ�Aáî¬��^�AceddeýpeŒJê„�u Adnehdfffeh‡}™fddýýhew}dc @Bù@ €¦õ@efjeýpeg À*ð@neddddfej‘¢‰eddý À ò@ýgewxdce À ò@fieýpegn €�ç@fddddfendd}ddnýd �—@gftifcidieýdfgnd ��@hddddjy‹ýšdddýýdš™™™™™8@ �Û@ �Û@flcdciffgýpegêed `RÑ@dfffec|ý33333³4@¤fedýdde  Øã@pheceddeýdegêmed ÀgÙ@ddfiŒ”ý‰š™™™™�:@jddýdfft à«Î@edcifdfý  ¡Ã@dfgêhddddfjª£ý£gffffffT@edýdeepe `t×@dcegjeýd @�Î@egêhddddfieyrzdddýýefped à<À@ �—@cifgeýfe @É´@gêdddfffe¸ƒý‚fehš™™™™ÙC@ýýffrhfc  Øã@ `tÇ@idffýnfd ÀgÙ@êifdffffjŠý©fedýfffff&H@ à<à@dedrcfcegheýdegê @ÉÔ@cedddficŒý}eedýýfffffæ6@  ØÓ@ à«Î@efqdfciddfýoeg}g ÀgÉ@hdfffee�ýœeddýýeÍÌÌÌÌÌ4@ `t×@  ØÃ@fpedcefeeýmfg}hd @�Î@dfefelŒý33333³H@}eedýdef à<à@pedcifffýdeg}edd @ÉÔ@fffe|’˜›dedýýeep �Û@ �Û@gdcifdeý `RÑ@pfg}gddfffer}ý�fÍÌÌÌÌL7@dgýýefre à«Î@ €Ã²@ecidgeýj  ¡Ã@eg}fhdfffej£ý­feffffffT@dýdedpcf P[è@cgeieýde à-ß@d}chdddfic‰ý™gddš™™™™™@@ýdefoedc û¹@hfdeýdfg ¡°@yhddddfkd”ý}ddjý3333338@ ðÄí@dfewefcidieýdfgy Ð ã@gfdddfkfƒý«gddýdš™™™™™3@ �Ë@dfocfcfdjeýdfgyn `RÁ@hdddfic|ý˜gedýýd33333³4@ à<À@ à<À@flddchfheýmfgyfd @É´@dfffec‹ýÍÌÌÌÌìP@ˆeddýýfd  Øã@  Øã@rclciffeýpfgychd ÀgÙ@fefecddzdedýdefp  ØÃ@dfcidgeý Àg¹@dfgylhddddkýŠ�ƒd 0�@ddýýeetd P[è@  ØÓ@dcefgeýp à-ß@fgyfddfffej�ý¢ddš™™™™�9@dýdedpce `t×@cefgeýde @�Î@gˆchdddfic‘ýždedfffff&C@ýýgdtcec à<à@ à<à@efdeýpee @ÉÔ@ˆcedfefec†ý}eedýš™™™™YA@ À ò@dedpcecifgeýdegˆ €�ç@chdffffc€ý{dddýýš™™™™™5@ à«Î@ à«Î@efqidcifdfýofgˆi  ¡Ã@ddffff‚�ýxdfdýdd33333³3@ €¦µ@foedcifggýdegˆfd ·«@ffffff ý33333�S@Çdefddhevxecedfedqegˆehdfefex”ýw3333338@dddýýffr ðÄí@  ØÓ@rdcifkfý Ð ã@pfgýfddf Àh@ffe‡’ý­h33333³N@ddddeerkdceffeddfgýddddd Àh@efn’˜Œfdeýýgetxf�$(¾1Ÿ�A €¦õ@cedeeýpfŒJê„�u Agýeedfff Àh@ejŠý­edh33333³9@ýdgdtcec �—@ifdeýdfg ��@ýchdddfi Àh@c™‚�fedý ðÄí@ýedrcfce  ØÓ@ffeýpfgý Ð ã@ Àh@chdfffec•ýšfedýý33333sL@ ðÄí@ ðÄí@fdtceceffeýpedýc Ð ã@ Àh@hdffeec ý½fddýde33333SP@ ðÄí@fsdfcedheýdegýgh Ð ã@ Àh@dddfik¦À¨hedýýhe�$(¾1Ÿ�Aáî¬��^�AwcdcgekeýpedýhddŒJê„�u A Àh@ffeec�ý§33333sC@fdgýýgdu À ò@ `t×@ce}eeheý €�ç@pfg’chdddfgc’ý¡eš™™™™�N@ddýýhevx €¦õ@ ðÄí@dcefegýp ·ë@fg’eddddfej›ý‹df33333óQ@dýýferdd�$(¾1Ÿ�A à«Î@cefegýofŒJê„�u Ag’hdfffffd“ý‹gddffffffC@ýdgetedc�$(¾1Ÿ�AeedeýdfgŒJê„�u A’fddddfkx–ý‡eedý33333³L@ ðÄí@dhdvcfcefdeýdfg’ Ð ã@cheddfkc‘ýœgeiýýfffff&C@  Øã@  Øã@fesxecgdeeýpfgªe ÀgÙ@edfffex‡|˜dedýýe `t×@ �—@fpddcefheýfegªhd @�Î@dddfld¡ýÍÌÌÌÌìQ@€dedýýff  Øã@ Á¯@sddceffeýifgªhdd ÀgÙ@fffe}‡}–dedýdddj �w@cecifgeý �m@degªchdddflc„ý°ifffffæ:@ddýddfoc `tÇ@lcfdfeýd @�¾@fgªihdddfic¯ý½fe33333SV@dýdgfted Ð#Ñ@cgfeeýde àðÅ@eªeddddfig�ý�dde33333³<@ýýefqdec `t×@ �—@edfeýfeg @�Î@ªehdfffegdd¢ledý û¹@dddjcecgffeýdfgª ¡°@chddddkc›ýœdedddš™™™™™B@efodeciddedqfmªhhdffeed‡ýŸdddýdeš™™™™™?@  ØÃ@dpceciedfýdfgªch Àg¹@dddfic�ýÍÌÌÌÌLK@†edldýff `t×@sgecidegdpfgýfhd �ƒ@ffffg•ý�33333óB@dedýýgfw À ò@ à«Î@ndceffeý €�ç@ofgýdddf �ƒ@ffel’ý†dÍÌÌÌÌ�N@dfýýgfth  Øã@  Øã@ecidgeýp ÀgÙ@ffýfhddd �ƒ@feŒšý dd33333sO@eýýhdwce à<à@ �Û@cifdeýpf @ÉÔ@gýcedfff �ƒ@fc„ý�dedfffffæ:@ýýhdvcec �Û@  ØÓ@ifdeýpfg `RÑ@ýchdffff �ƒ@cƒýƒhdfý33333³A@  ØÓ@ýffrdece `tÇ@deeýnego ÀgÉ@fhdddfhfdd­fddýd  ØÃ@eepcfcfddeýqegoj Àg¹@edddficƒýžeddýýeš™™™™ÙC@ ðÄí@ à<à@ernfcedeeýpegofh Ð ã@dfffeiŠ�Žfddýýef  Øã@ à<à@pdecgdgeýpfgodhd ÀgÙ@ddfhd ý‡33333óT@eekýýgev P[è@  Øã@xlcedheý à-ß@pegoghdfffen…ý—i333333+@dfýdeesd P[è@echddgýd à-ß@fgofheddfff‡ýšgdÍÌÌÌÌÌ5@dýdddrcf �Û@cefggýdf `RÑ@dochdddelc•ý™fed33333³F@ýýeesddc P[è@ �Û@efgeýpeg à-ß@•fddffeen¤ý˜fdhý  T@ @Bù@ýhdxcece P[è@gheýpeg• À*ð@chdfffec�ý�dedýýÍÌÌÌÌLJ@ À ò@ À ò@eepffcidieýpfg•f €�ç@edfffedŠ�Žddfýýg €¦õ@ €¦õ@dtcecgejeýpeg•ch ·ë@dfffec„ýfffffæA@žeddýýee à<à@  ØÓ@pgecedieýpeg•fhd @ÉÔ@ddfeg™ýŠ33333SQ@gddýdddn  ØÓ@cfcefggý ÀgÉ@dfg•chdddfec¡ýŒfš™™™™9T@dhýýgeuh P[è@ �Û@ecedheýp à-ß@fgÚehhfffenœýƒfdÍÌÌÌÌÌJ@dýýedqce à<à@ à<à@cgfkeýpe @ÉÔ@gÚchdddfec–ýœjde @G@ýýedtcfc  Øã@  Øã@eegfýpfm ÀgÙ@Úchdddfec¤ý†edfý  T@ À ò@ýhducfce à<à@fgeýpegÚ €�ç@cgdfffecˆýžgedýýš™™™™™8@ à<à@ `tÇ@ddocecgfgeýnfg„c @ÉÔ@hdefeec™ý{eedýýeÍÌÌÌÌŒD@  ØÓ@ û©@drceciefeýhfg„ch ÀgÉ@dfffec”ý ÀL@—fegýdhe�$(¾1Ÿ�Ax›fcedeeýdfg„fhdŒJê„�u AfffeŒ›ý™33333³N@fedýýeeq `t×@ Ð#Ñ@efcidkeý @�Î@ofg„hhdfffee¢ý¨d333333I@ddýdddkc à<À@ecgfdeýd @É´@eg„chdddfic”ý«jd3333338@dýdddjcf  ØÓ@ceeeeýdf ÀgÉ@g„chdddfic�ýzeejš™™™™™N@ýýffteec  ØÃ@  ØÃ@idefýmfg Àg¹@„ehdffff}€ý«iddýš™™™™™5@ €¦µ@ddflddcgfgeýdfg„ ·«@hddddfid“ý‚dddýýffffffC@ P[è@ P[è@get‡dcifjeýpfgýd à-ß@ p{@ddddfelƒý hedýýeš™™™™ÙC@ P[è@ à«Î@epeecedeeýofgýeh à-ß@ p{@dfffei•ý33333sL@€ddeýýed�$(¾1Ÿ�A à<à@rcfcieieýpfgýchdŒJê„�u A p{@ffeec¨ý š™™™™yV@hddýýhex�$(¾1Ÿ�A À ò@hdceeieýŒJê„�u Apegýeddd p{@dffgvý‰dÍÌÌÌÌL>@dfýýedpc  ØÓ@  ØÓ@ecieieýp ÀgÉ@egýchdff p{@fec�ý˜geš™™™™�B@dýýefpee à<À@ à<À@cidefýme @É´@gýlhdddf p{@kf¬ý}fdfš™™™™YS@ýýgfv‚ec  Øã@  Øã@edheýpeg ÀgÙ@ýhhdfffe p{@sddwdddý P[è@deeqedcifkeýqeeý à-ß@ p{@eddddffx›ýŠedfýýš™™™™™B@ à<à@ à«Î@gevnfcedieýofgýe @ÉÔ@ p{@hdfffff ý�dddýýgš™™™™ùU@  Øã@  Øã@etofcidieýpfgýfh ÀgÙ@ p{@dfffep�ý333333J@�dedýýgf  Øã@  Øã@tsdcifefýpfgýfdd ÀgÙ@ ˜–@fffeŒ‹ý†ÍÌÌÌÌÌ4@dddýýefp à«Î@ à«Î@gecidieý  ¡Ã@oegýghdf ˜–@ffexŒý‹d33333³H@deýýefrn à<à@ à<à@eceddfýp @ÉÔ@egýieeff ˜–@fenƒý~deš™™™™ÙC@dýýedqce à«Î@ à«Î@cifgeýof  ¡Ã@gýchdddf ˜–@jc�ý�ddˆÍÌÌÌÌLJ@ýýgfvgfc `tÇ@ à<À@idheýmfg @�¾@ýeedddfe ˜–@s—ý†geeý33333³<@ �Ë@ýefp‡fcg �Ë@dgeýnfgý `RÁ@ ˜–@ghdfffe‚”ý†efdýýš™™™™�4@ �Û@ �Û@edpclcefgfýpegýc `RÑ@ ˜–@hffffec’˜Šeedýdg `t×@ducecifeeýdfgých @�Î@ ˜–@deefec›ýš™™™™™B@Ÿdddýýee �Û@ �Û@qhdcgejeýpfgýhdd `RÑ@ ˜–@fffenƒý€ÍÌÌÌÌÌB@dddýdedp à<À@cecifggý @É´@degechdffffc�ý„i33333sE@ddýýedrc P[è@ à<à@fcieggýp à-ß@fgechdfffecdd|defýdefpdf û¹@cidggýde ¡°@gefhddddkd}ý¥ied333333)@ýýefqddc à<à@ à«Î@gfegýofg @ÉÔ@edddfffec|ý�dedý33333³4@ 5¤@ddflgdcifegýdege Þ™@tddddfjg©ý—eedýdÍÌÌÌÌLG@ à«Î@gdwceciedgýdfgec  ¡Ã@hdddfhc„ý�kddýýdš™™™™�8@ �Ë@ 5¤@dlcecifggýgfgech `RÁ@dffefc�ý33333³3@�hddýýef Á¯@ Á¯@pedcifegýiegeddd S¤@ffffe‡ýz33333óE@dfddýdfd �w@cdcifgfdeedýjdfd Ðq@dfhc}ý}f333333)@ddýddfnn Á¯@ecidffýd S¤@fgýfhdef Ðq@efn’ýŠgeš™™™™�N@hýýffrne `tÇ@ `tÇ@cgddfýne @�¾@fýjhdffe Ðq@e‚”ýžedd3333338@ýýeepflc à<à@ `t×@gddfýpfg @ÉÔ@ýehdfffe Ðq@h´ýŠfdÆý333333W@�$(¾1Ÿ�Aýhexnfce À ò@dieýpfgmŒJê„�u Aehdddfff ý™fddýýš™™™™ùU@ @Bù@ À ò@getedceejeýpegmg À*ð@ddfffer¥ýŽfddýýeÍÌÌÌÌ,T@ €¦õ@ à<à@eqlfcedjeýpfgmeh ·ë@dddfex¥ý33333ÓU@›je‚ýýge�$(¾1Ÿ�A P[è@uffcedjeýpegmehdŒJê„�u Afffef™ý—33333SQ@gfÆýýget�$(¾1Ÿ�A ðÄí@gfcedjeýŒJê„�u Apegmhhffffeh­ý f  U@ddýýedpc P[è@ Ð#Ñ@ecefheýo à-ß@egmchdddfic‰ýšddš™™™™YI@ddddeledcefjgddfg¤gdeddfks}ýŠhdi333333)@ýdgdtcfc �Û@efdeýdfd `RÑ@¤chfddeec¤ý³geÆýÍÌÌÌÌ�U@�$(¾1Ÿ�Adegsefcfdgeýdfg¤ŒJê„�u AfhdddfieŒý‹fddýd33333³H@�$(¾1Ÿ�Aeep‚lcedieýdegeeŒJê„�u Ahdfffeh’˜Šfdgýýg P[è@ `t×@etjfcideeýpfgefh à-ß@dfffel�ý33333óC@Ÿfddýded À ò@pclceedeýdegeced €�ç@ddfic‡„œeedýýefp ðÄí@ ðÄí@decgdeeý Ð ã@pegeehdfffef–ý‰d33333³L@ddýýeerg �Û@ �Û@dcifieýp `RÑ@egôfddddfef„ý defffffæA@dýýeexhd  Øã@ �Û@cifheýpf ÀgÙ@gôjddddfigªý„eed33333óT@ýýget‚dc À ò@ P[è@efgeýpeg €�ç@ôfddfffex�ý£jedý33333³<@ à<à@dedqcecefeeýdegô @ÉÔ@cfdddficŸý™fddýýfffffFP@ ðÄí@ P[è@eeqdecedieýpegôg Ð ã@fdddfecdd�fddýdd Ð#Ñ@docecefgeýdfdôch àðÅ@edddkc}ý333333)@¢gfdddefsrdcffgeddfgôidfddejnddzdedýdedr `tÇ@cfcifeeý @�¾@dfgýchdd l”@ddjc�ý|eÍÌÌÌÌÌ4@edýýedpc à<à@  ØÃ@fcefeeým @ÉÔ@egýchdff l”@fecdd{eedýdefphe `t×@cidegýqe @�Î@gýdhdffe l”@en‹ý—fedÍÌÌÌÌìP@ddgdtclcefgedqfdýceefffe l”@cŠýŽgedýÍÌÌÌÌ�J@  Øã@ýeepelcg à<à@dggýpegý ÀgÙ@ l”@gedfffeg¥ýƒeddýý33333³M@ ðÄí@ �Û@efrhecedgeýpedýg Ð ã@ l”@heddfei…ý²ledýdd333333+@ `tÇ@fdcfcfdieýdfdýgh @�¾@ l”@dddficžýš™™™™yT@ˆdehýýhe�$(¾1Ÿ�A €¦õ@usecedgeýpfgýeheŒJê„�u A l”@fffen‘ý¤ffffffC@fedýdget ðÄí@fecedfeý Ð ã@degÁhhdddeir¡ý°eš™™™™9T@eiýdhexd  Øã@fcedheýd ÀgÙ@fgÁghdddfid•ý�feš™™™™™J@dýýfered À ò@ À ò@cgeheýpe €�ç@gÁdddfffen¡ý«gddš™™™™9T@ýdhevedc P[è@eeeeýdfg à-ß@Áeddddfif“ý§fedýffffffC@ €¦õ@dgetnfcedjeýdfgÁ ·ë@ghdddfii¦ýŒfedýýfffff&R@ @Bù@ À ò@hevidceeeeýpegÁf À*ð@ddfffen‘ý�hedýdefffff&C@ P[è@frfdcefieýdfgýgd à-ß@ Pt@dddfknŠ��gfdýýee À ò@ à<à@pxlcedigýpegýhhf €�ç@ Pt@fffek©ýŸÍÌÌÌÌLG@eehýýfer `t×@ `t×@jfcgdgeý @�Î@pfgýehdf Pt@ffeh}ýŒf333333)@ddýýefpe `t×@ `t×@fcedfeýp @�Î@fgýehdff Pt@efdˆý™fdÍÌÌÌÌLF@dýýgetnd P[è@ P[è@cgfdeýpe à-ß@gýhddfff Pt@ex‡„}degýdefseec  Øã@ideeýdeg ÀgÙ@ýehdddfk Pt@e˜ý gedýfffffæQ@ À ò@ýgdtcece  Øã@fieýpegý €�ç@ Pt@chdddfic’˜�eedýý €¦õ@ à<à@edpcecefjeýpfgýc ·ë@ Pt@hdfffecƒý‚dedýýfš™™™™ÙC@ Ð#Ñ@ Ð#Ñ@dtcecifdeýoegých àðÅ@ À‚@dfffec�ý33333óC@Ÿddhýýee�$(¾1Ÿ�A �w@rdecgdgeýeegýfhdŒJê„�u A À‚@efeee‘ýŒ333333?@hddýýefp  Øã@ `tÇ@ddcefggý ÀgÙ@negýfdfd À‚@deex‹ý�dÍÌÌÌÌÌH@edýýgdtc  Øã@  Øã@eciffeýp ÀgÙ@egýchdff À‚@fec™ýªedš™™™™yR@gýdgdtce P[è@cgffeýdf à-ß@gýchdddf À‚@ic–ý�dddfffff&Q@ýdddicec Ð#Ñ@iffeýdfg àðÅ@ýchdfefe À‚@c¤ý²eedýÍÌÌÌÌ�U@ `tÇ@dgdtcfcefieýdfgý @�¾@ À‚@chdddfic—ý”dedýý33333“S@�$(¾1Ÿ�A Ð#Ñ@hdvcfceeieýofg„cŒJê„�u Ahdffefc’ý¤fddýýgÍÌÌÌÌ�N@ À ò@ €¦µ@etgdceeheýkeg„fd €�ç@defeem ý33333óT@•eedýýge�$(¾1Ÿ�Aáî¬��^�At‚dceeieýpfg„eddŒJê„�u AfffesŽý· €:@fddddedrcfcffgeddfg„chdddeic ý¢f33333�S@edýýeese P[è@ `t×@dceffeýp à-ß@fg„eddffefp ýŽdd33333�S@eýýgetje €¦õ@ €¦õ@cidfeýpe ·ë@g„hhdfffeižý¸fdiš™™™™yT@ýdheudec �Û@edfeýdeg `RÑ@„ghfddfikddvdddý�$(¾1Ÿ�Adeepddcifheýdeg„ŒJê„�u AddfdddjfŽý±dd‚ýd33333³U@ à<À@gdwcecifgeýdegec @É´@hdddeicˆýšddgýýefffffæB@ €Ã²@ €Ã²@fsdecgdgeýjegedh �¨@dffef}ªý33333óT@¢fdlýýed�$(¾1Ÿ�Aáî¬��^�AtcfceeheýpegechdŒJê„�u AddfhcŠýœÍÌÌÌÌŒM@gedýýfer ðÄí@  ØÓ@edceefeý Ð ã@pfgeedeffffkŠý–dÍÌÌÌÌŒM@ddýýeep€ P[è@ P[è@dceefeýp à-ß@fgeddfddeef}ý{ddÍÌÌÌÌÌ7@kýýgdtce €¦õ@ 5¤@ciefeýgf ·ë@fechdfffec¡ý¦fedš™™™™ÙT@ýýgdtcec À ò@ À ò@eekeýpfg €�ç@echdddfic©ý§eeddfffffÆV@dhdxceceekedÆegechdddekc‡|•dedýý À ò@ À ò@eerddciejeýpegpg €�ç@ddffeeg•’¹hedýdd P[è@eoddceedeýdegphd à-ß@dddficƒýÍÌÌÌÌÌB@¨fddýýee P[è@ �Û@pdfcedkeýpfgpghd à-ß@fffedŒý£ÍÌÌÌÌL<@hedýýddo  Øã@  Øã@cecegfeý ÀgÙ@pegpchdfffec”ý‹g3333338@edýýgdtc P[è@ Á¯@ecefjeýi à-ß@fgpchdffffc ý½fd33333�S@fýdedqce `tÇ@cffieýdf @�¾@gpchdddfic™ý›fdpÍÌÌÌÌŒM@ýýgeunfc�$(¾1Ÿ�A ðÄí@edieýpegŒJê„�u ApfhdffeexŽý¬dedýfffff¦A@ `t×@dfdscfcieheýdegr @�Î@chdddfic’˜Žfedýý  Øã@  Øã@eeqfdciffeýpegrh ÀgÙ@ddfffex“ý½fddddeffffffC@dpcfcffgeddfgrchdddfkczýÍÌÌÌÌÌ8@‘fddýddf û¹@ocfcidgeýdfgrphd ¡°@ddflc™ý¸33333SQ@jedýdeeq �Ë@fdcffdeý `RÁ@dfgrlddddfidŠ�¯fddýýedpc `t×@ û©@fcgfheýh @�Î@fgrchdddffcƒý¯ee333333D@dddddmcfcifefddeg•cedddfic“ý™dddfffffæL@ýýeepŒfc `t×@  ØÓ@idieýpfg @�Î@•ehdeffe yýœeddý333333+@  ØÃ@ýfeucfci  ØÃ@dheýmfg• Àg¹@eedfffec„ý‡eddýýš™™™™�8@  Øã@  Øã@efqddcifkeýpeg•d ÀgÙ@ddfffed”ý{eedýýe3333338@ à<à@ à«Î@fqddcifefýofg•id @ÉÔ@dfffeyŒýÍÌÌÌÌL<@ydddýýed  ØÃ@ Á¯@pcecifheýieg•che Àg¹@ddffc|ýzffffffA@ddfýýefp `tÇ@ 5¤@eecidifý @�¾@gfg•gldddfel”ý�d P@edýýedsc  ØÓ@  ØÓ@ecifdfýp ÀgÉ@eg•chdfffec|ýžee33333³4@dýýedpce `t×@  ØÃ@cifgfýme @�Î@g•chdffffc€ýœged3333339@ýýedpcec  Øã@ Ð#Ñ@eegeýoeg ÀgÙ@wchdffeec“ý…fddýffffffC@ ðÄí@ýgetidce à«Î@egeýofgw Ð ã@eddddfkl‡ý¬gddýd33333óE@ à<À@ddocfcffgeýdfgwc @É´@hdddfic“ý¦gedýdeffffffC@ À ò@epeeceddeýdfgwff €�ç@dddfif’˜ˆeddýýef Ð#Ñ@ Ð#Ñ@pddcifeeýoegwidd àðÅ@ddfee–ý�fffff&Q@eedýdefp �Û@edcefheý `RÑ@dfgwgddddfjmyr}dehýýefpk `tÇ@ à<À@fcidheým @�¾@egwdhdddfj–‡ý™ee33333óE@dýýdfoid  ØÓ@  ØÓ@cgfhgýpe ÀgÉ@gwhddfffei£µŸdeeýdferdec À ò@edfeýdfg €�ç@wehdddfkyŠý�dddýffffffE@ ðÄí@ýeepieci ðÄí@dheýpegw Ð ã@fhdddfes}ýŒiddddÍÌÌÌÌL7@efodeciddfdqegýf ðs@hdfffeeƒý—dedýýeÍÌÌÌÌÌB@ P[è@  Øã@erndcefdfýpfgýjd à-ß@ ðs@dddfkx“ýš™™™™YQ@¦eedýýhd�$(¾1Ÿ�Aáî¬��^�AxcecifdfýpfgýchdŒJê„�u A ðs@fffecƒýŒš™™™™™3@hedýýefp P[è@  ØÓ@deciddfý à-ß@pfgýfhdf ðs@ffesƒý�gÍÌÌÌÌÌB@ddýdeepd �Ë@eceddfýd `RÁ@fgýihddd ðs@fkfŠý–ee33333³9@dýýeeqkd P[è@ à<à@cefgeýpf à-ß@gufddfffeo¤ý–ged33333óQ@ýýeepfec ðÄí@ à<à@eddeýpfg Ð ã@uffdfffee„ý¢dddýfffffæA@ �Ë@deerefcfdgeýdfgu `RÁ@fhdddfif�ý­eedýd333333J@ `t×@edpcecefdeýdeguc @�Î@hdddfic”ý�eedýýe3333338@ à<à@ �Û@eoedcifdeýpegugd @ÉÔ@dfffek}ýÍÌÌÌÌL7@‹geeýýef à«Î@ à«Î@refcidgeýofguhhd  ¡Ã@fffeddd�gddýddfo à<à@decedfeý @ÉÔ@qfgudfdddlkx”ýŸf ÀL@ddýdeepc Ð#Ñ@dceedeýd àðÅ@fgujddddfkcŒý›defffff&C@gýýeerge à<à@ à<à@cgdgeýpe @ÉÔ@duehdfffeg}ý†fdd333333)@dddddcecfeegdÆfgwchfddfhcžý“ffdýš™™™™yT@ À ò@ýhevndce P[è@fgeýpegw €�ç@fdffffeg¥ý fdxýýÍÌÌÌÌLU@�$(¾1Ÿ�Aáî¬��^�AhexxfceddfýpegwfŒJê„�u Ahdfffep“ý§fddýde33333SP@ `t×@dpcecefieýdfgwcf @�Î@dddficŠý33333³9@�eefýýge @Bù@ �—@tifcedfgýffgwjhd À*ð@ddfknšý”33333sO@fedýýhdv À ò@ P[è@cfceeeeý €�ç@pfgwcfdfffec™ý‚ffffff&Q@eeýýhew‚ €¦õ@ P[è@ecedheýp ·ë@fgwehdfffen€ý fd3333339@dýdeepdd à«Î@cgfdeýde  ¡Ã@gwgddddfic¦ºžfedýýefpddc �Û@ �Û@gekeýpfg `RÑ@gdddfffee˜ý£dedýfffff&N@ P[è@deetddcfgkeýdfgg à-ß@hddddfik‡}�eedýý à«Î@ à<À@edqceceedeýmfggc  ¡Ã@gdfefec”ý¦jedýýeš™™™™�:@  ØÃ@  ØÃ@dpcfceeeeýmfggch Àg¹@dfffec�r»ieddddencfcedjeddedghhdddficŸý›33333SQ@edgýýfes @Bù@  Øã@decedkeý À*ð@pfgggfdddfefŸýšf àR@ddýýeepg à<à@ à«Î@ecedjeýo @ÉÔ@fg¯fedfffer¢ý£fe333333I@dýdedrce @Bù@cefigýde À*ð@g¯cheddfic”ý~ded P@dýegtfdc 5¤@ifgfdgeg¯pddddfh ¬ý§ednýš™™™™YS@ @Bù@dhdvcfcefjeýdeg¯ À*ð@chdddficƒýµhedýdš™™™™ÙC@  ØÓ@eeqidcffheýdeg¯g ÀgÉ@ddddfig™ý­gdfddhš™™™™yR@etrecgdieddeg¯ghdddfig¡ýš™™™™9T@²gedddgdtcfceffeddfg¯chdddfic®ý± U@gedýdheu  Øã@rdcefheý ÀgÙ@deg¯gddddfinŠ�“hefýýferx P[è@ P[è@ecgdheýp à-ß@fgtdheffee‡£ý¢fdš™™™™¹U@hýýhdvce €¦õ@ À ò@ceedgýpe ·ë@gtchefffec¤ý·ddÆ  T@ýdhdvcfc à<à@efheýdfg @ÉÔ@tchdddfkc£ý�fdxýš™™™™¹U@ @Bù@ýgdvcece À ò@fheýpfgt À*ð@chdddfec­¸�dfdýý�$(¾1Ÿ�A @Bù@heu‚dciffeýpeetgŒJê„�u AdffffeŒ™ýšgeiýýgÍÌÌÌÌŒD@ €¦õ@ à<À@etfecedieýmfgtjh ·ë@dddfhjƒýÍÌÌÌÌÌB@�eeiýýge  ØÓ@  ØÓ@t‚fcfdkeýpegtfhd ÀgÉ@fffextý~š™™™™™*@dddýdeeq À ò@ddciffeý €�ç@qegtfddfffff…ý d33333sF@ddýýeepe `t×@ `t×@ecfdgeýp @�Î@ef gedfffep}ý‚fdÍÌÌÌÌL7@dýýefoie Ð#Ñ@ �—@cidgfýfe àðÅ@g khdddfhi¡ý¬fddš™™™™9T@ýdgetcdc  Øã@eedfýdfg ÀgÙ@ hddddfkc‹ý¨ledý ÀE@ �Ë@dddpceceedfýded  `RÁ@cheddfic‡„’eddýý à<à@ à<à@herdeceddfýpeg h @ÉÔ@edddfec‰ýŸmfdýýdš™™™™™2@ à<À@ à<À@fjcdcefggýmeg gd @É´@ffffec”ýš™™™™�:@±ledýded `t×@pcfceedfýdfg chd @�Î@ddfic”ýŸ3333338@gfdýýeer  ØÃ@  ØÃ@fdcgfdfý Àg¹@mfg idfffeeg}ý“eÍÌÌÌÌÌ7@edddeepedcegmfdqfg fddfffen„ý—fdfffffæA@dýýeeqel @Bù@ @Bù@cedefýpf À*ð@g hheffeedˆýŽdddfffffæB@ýýfftedc  Øã@  Øã@gffeýpfg ÀgÙ@�fddffef‚©ý¯gedýfffffÆV@ `t×@dhewedcffjeýdeg� @�Î@gddddeid‡ý¦iddýd33333³7@ `tÇ@efsddcgfjfýdeg�j @�¾@ddddfihšý}edgýýg33333sO@ à«Î@ �—@fvfecedkeýfff�fh  ¡Ã@dddee‡dd†hedýddf à<À@ocdcifggýdfg�ndd @É´@dddkc…ý}š™™™™ÙO@ddnýýfds  Øã@  ØÓ@cecgedeý ÀgÙ@pfg�cheeffec”ý²e P@ddýdeeuj Ð#Ñ@dcffdfýd àðÅ@fg�pdfddfiˆtýyddš™™™™�6@kýýferde À ò@ €Ã²@cidieýje €�ç@g�fhdddffn ý­eed33333�S@ýdgdtcfc �Û@efdeýdfg `RÑ@”chdddficdd�edgý à<À@dffsffcedgeýdfg” @É´@ihfdddkj’˜‡hedýý  Øã@ à<à@efq‚fcedgeýpfg”f ÀgÙ@hdfffee‡„ƒeddýde  ØÓ@fpddchfgeýdeg”ed ÀgÉ@dfffel‡„žfddýded �Ë@qcecffjeýded”chd `RÁ@ddfic¬ý¥š™™™™YS@fddýdeep €¦õ@hfcedjeý ·ë@deg”fedddfif‡}¢lddýýefpe à<à@  ØÓ@dŒefgeýp @ÉÔ@eg”gddddfg‚ddydefýdeepee Ð#Ñ@cidgeýde àðÅ@g”ehddddj–“ýŸfedfffffæL@ýýfdrcec à<à@ à<à@ifdfýpfg @ÉÔ@ýchdfffe u@c€ýydfddš™™™™™5@ýedpcfci €Ã²@fggdjfgý u@chfffffc”ý‘ffdýý3333338@ `t×@ Ð#Ñ@edpclcefgeýofgýc @�Î@ u@hefffec“ý˜fedýýgffffffC@ à<À@ €Ã²@etrdsefieýjfgýfd @É´@ u@gfffgx€ýš™™™™™2@°gddýddd ðÄí@icfcefjeýdfgýche Ð ã@ u@ddficddyddeddedpclcifffddegýchdd u@ddjc�ýŽfÍÌÌÌÌLK@deýýfdrc `t×@ `t×@echfeeýp @�Î@fgýchdff u@fec}ýžgeÍÌÌÌÌÌ7@dýýefsee �Û@ �Û@cgddfýpf `RÑ@gýhhdfff u@ee„ýŽfddfffffæA@ýdgdtcfc  ØÓ@efgeýdfg ÀgÉ@ýcheddfk u@c‘ý§fedý333333?@ `t×@ddendfcedheýdeg� @�Î@hhdddeidzýwdedýýš™™™™™-@�$(¾1Ÿ�A û¹@eepdecidieýleg�eŒJê„�u Ahdddfjy“–¥hedýde à<à@dpcfceffeýdeg�ch @ÉÔ@dddficŠýÍÌÌÌÌ�J@ˆgedýýfd @Bù@ P[è@sceceeggýpeg�che À*ð@ffeec—ý‰33333óN@jddýýeep ðÄí@ ðÄí@secedjeý Ð ã@peg�ehefffex¢ý¤f333333I@edddgdtcfcefdedqeg�cheddfic“ý…fefffffæL@dýýedsce  ØÓ@ à«Î@cgfieýof ÀgÉ@g�chdfffec¡ý‚eedš™™™™9T@ýýgdtcfc ðÄí@ à<à@efheýpfg Ð ã@�chdfffec’˜–fedý €¦õ@ýeerxdce P[è@fjeýpfgý ·ë@  n@fddfffex¡ý½fddýdš™™™™9T@  Øã@hetcdcfffeýdfgýf ÀgÙ@  n@deddfic™ý¦fedýýeš™™™™yR@ P[è@ P[è@fqiecgdieýpfgýgh à-ß@  n@eddfiy™ýÍÌÌÌÌ�T@ dedýýge�$(¾1Ÿ�A €¦õ@tâdceejeýpegýjddŒJê„�u A  n@fffeÈ�x˜hedýýedq à<À@ à<À@cfceeheý @É´@meg¥chdffeec™ý…d33333SQ@edýýeeqh �Û@ �Û@fcidfgýp `RÑ@fg¥fhdfffe–—ý‘gd33333³<@dýýefpef P[è@ P[è@cgdigýpe à-ß@g¥dhdfffefƒýyddfš™™™™ÙC@ýýedrcfc P[è@ 5¤@ifggýgfg à-ß@¥chdffffc”ý„gddý ÀI@  Øã@ýedrcece �Ë@egeýnfg¥ ÀgÙ@chdddeec�ý²hddýdÍÌÌÌÌL>@  Øã@edpcfcefeeýdeg¥c ÀgÙ@hdddfic™ý�dedýýfš™™™™yR@ P[è@ P[è@drceciffeýpeg·ch à-ß@dfffec˜ýfffff&N@Ÿdedýýgd P[è@ P[è@tcecffdeýpfg·chd à-ß@fffecŠ�„eddýýefr �Û@ à«Î@ddcefjeý `RÑ@oeg·eddfffehŒý�dÍÌÌÌÌŒF@edýýgfte €¦õ@ À ò@lcgdieýp ·ë@eg·eedfffej–ý�ddfffff&Q@dýdffred P[è@ceefgýdf à-ß@g·iddfffeg‚ýŠdddš™™™™�8@ýýefrfdc  ØÓ@  ØÓ@gfjgýpeg ÀgÉ@ýfddddfe }@f�ý„eefýfffffæR@  ØÓ@ýgft�lci Á¯@dfgýifeý ÀgÉ@ }@fhdfffex›ý†ddhýýÍÌÌÌÌLS@  Øã@  Øã@hdtcecifegýpegýc ÀgÙ@ }@edfffecšý¤fdÆýdh33333sO@ À ò@evffcedggýdegýgg €�ç@ }@eddfin”ý3333338@~ddnýdff ðÄí@tkecidieýdff�dhd Ð ã@ffffs’˜‹gddýdgev  ØÓ@edcgeeeý ÀgÉ@dfg�fddddfkeŠýwd33333³9@edýdefqd €¦õ@dcifhgýd ·ë@fgÛfddddfhfŠ�“fedýýeernf €¦õ@ À ò@cedgeýpe ·ë@g�efdfffefšý‹hed33333sO@ýdhevedc�$(¾1Ÿ�AeeieýdfgŒJê„�u A�fddddfkh‰ý€ddgýš™™™™™2@ `t×@ýefqilci  ØÓ@dgeýpeg� @�Î@jhdddfeŒ¤ý‹fddýýÍÌÌÌÌ�U@ ðÄí@ ðÄí@hevfdcgeheýpfg�f Ð ã@ddfffej”ý›gehýýg ÀL@ P[è@ P[è@etxfcgdieýpfg�gh à-ß@dfffei‹ýÍÌÌÌÌìP@«gedddgdtcfcegjedqeg�chdffeec„ý¥š™™™™�C@eedddeepefceddfddfg�ggdddfks�ý¤e333333J@ddýýeere ðÄí@ û©@dceefeýh Ð ã@eg�gddddfif—ý…dd33333óN@fýýgettf P[è@ P[è@cidieýpf à-ß@g�dhdddfek¤ý�degÍÌÌÌÌ�U@ýýeergec ðÄí@ P[è@edfeýpfg Ð ã@�hhdfffejˆý´medý €4@ `t×@deepcdcieeeýdeg� @�Î@lddddfic‰ýªdedýdÍÌÌÌÌŒA@ à<À@fdrcfcifdeýdfd�c @É´@hdddfic˜ý�eddýýgfffffæQ@  Øã@  Øã@etŒdcifdfýpfggfd ÀgÙ@dddfeŒ†ýš™™™™YA@Šjddýddf �w@oddcifgfýdfggddd �m@ddfkc€ý{š™™™™™5@eedýýefn  ØÃ@ û¹@hdcifdeý Àg¹@lfggxddddfllŠ�«dedýdedqc �Û@eciffeýd `RÑ@eegchdddeicŸýƒhdfffffæS@dýýeesxd ðÄí@ Ð#Ñ@ceeefýof Ð ã@gghddddfe–ƒý—gdÆš™™™™ÙC@ýýgeteec  ØÃ@  ØÃ@iddfýmeg Àg¹@gihdfffffŠ�Šfedý �Û@ýfdrcece �Û@feeýpegg `RÑ@chdfffec‘ý•dedýýfffff&C@  ØÓ@  ØÓ@eeqxdcifieýpfghe ÀgÉ@ddfffeiˆý–gddýýeÍÌÌÌÌLF@  Øã@ `t×@epxdcefeeýpfghhd ÀgÙ@dddfex™ý33333SQ@‹feÆýdge ðÄí@tffcedjeýdfmhfhd Ð ã@ddfkiŠý´ÍÌÌÌÌŒM@hdddýeep à<À@elcfdggdmfghlhdddfejƒý–dffffffH@edýýedtc À ò@ P[è@fceejeýp €�ç@eghchdfffec‘ý“ddfffff&C@jýýgdwce à<à@ à<à@cgfeeýpf @ÉÔ@g{chdfffecƒý•eedš™™™™™3@ddedpcfcgfmeddfg{cldddflc”ý€dddý3333338@ û¹@ýefpifci û¹@dheýlfg{ ¡°@ghdfffepŸý½dedýd333333N@ à«Î@egrcdcffdeýded{j  ¡Ã@ddddficˆý†gedýýeš™™™™™8@ `t×@ `t×@dqcfchfeeýpfg{ch @�Î@dffeec‡ýÍÌÌÌÌÌ5@–gedýdef �w@pcdcififýdfg{jdd �m@ddfkc”ý…3333338@gedýýedn `tÇ@  ØÃ@cechfgeý @�¾@mfg{chdddfkc©ý‹eÍÌÌÌÌLG@edýýgeti P[è@ `t×@dceegeýp à-ß@fg³gddfffen•ýœed33333sL@gýýgdtce ðÄí@ `t×@cefffýpe Ð ã@g³chdddeec¡ý”idgš™™™™9T@ýýhevhfc P[è@ P[è@edjeýpeg à-ß@³dhdddfef™ý›gdddÍÌÌÌÌŒM@ýedpceceáî¬��^�Aefedpeg³chdffeec ý“dedýýš™™™™ùU@ à<à@ `t×@eepxdciekeýpeg³e @ÉÔ@ddddfh´‡}˜eegýýe À ò@ P[è@friechdheýpeg³fh €�ç@dfffel�ý €=@zdefýýef  ØÃ@  ØÃ@s{fcidhfýmfgýfhd Àg¹@ ؆@ffff¾¡ý¦š™™™™9T@fegýdhev `t×@ffcfdheý @�Î@dfgýfhdd ؆@dfieddxdedýdedrc  Øã@fcifhgýd ÀgÙ@egýchddd ؆@djc¬ý°ieš™™™™YS@dýdgetfe €Ã²@ceddeýde �¨@dýghdddf ؆@in“ý�fddffffffC@ýýffr}dc `t×@  ØÓ@iedfýpfg @�Î@ýgddddfe ؆@x‡„Žhedý Ð#Ñ@ýddocfce €Ã²@fdfýjfgý àðÅ@ ؆@chdfffec€ý”eddýý3333339@  ØÓ@ Ð#Ñ@ddncfchfeeýoegýc ÀgÉ@ ؆@hdfffecŠý™kedýýeš™™™™™C@  ØÓ@ Ð#Ñ@fp‚ecgddfýofeýfh ÀgÉ@ ؆@deffeeƒýÍÌÌÌÌÌB@‰dedýýgf �Û@ �Û@tsdcifdfýpfgýedd `RÑ@ ؆@ffeen‡„»gedddeerddcfgjeddegýgddd ؆@dfij‘ý‚fÍÌÌÌÌ�O@edýýedrc �Û@ �Û@ecifefýp `RÑ@fgýchdff ؆@fecœý°eeÍÌÌÌÌÌJ@dýdeepef à<à@cedkeýde @ÉÔ@ggjhdddfih”ý‚fef3333338@ýýferjfc Á¯@ 5¤@iddfýgeg S¤@ghhdddfkn‹ýŠfddýš™™™™™8@  ØÃ@ddendecgdjeýdegg Àg¹@jhdddflk|ý‘hdhdý33333³4@  ØÃ@ferdecgdgedmfggfhdfefe‡…ý}dedýýeš™™™™™;@ À ò@ `t×@dqcfciefeýpeggch €�ç@dddfkc}ý333333)@•hedýýee �Ë@ �Ë@pedcgejeýnfmgedd `RÁ@ffefhˆýˆfffffæB@dedýýedp�$(¾1Ÿ�A `tÇ@cfceejeýŒJê„�u Anfggchdffefcƒý•dfffffæ?@ddýýedrc à<à@ à<à@ecgedeýp @ÉÔ@eg�chdddfec‡ý�fd33333óE@dýdeepdd À ò@ceeieýdf €�ç@g�hddddfkf}ý~dddÍÌÌÌÌL7@ýdeesgdc P[è@iefeýdeg à-ß@�jddddffx…ýxdeddffffff6@ýddocfci �—@fdedffg�chdddfjcˆýŽfddýýš™™™™™8@ P[è@ à«Î@fdrceceejeýofg�c à-ß@hdfffec™ý‡fdfýýe33333SQ@ P[è@ P[è@erªfcgdgeýpeg�eh à-ß@dfffen“ýffffffC@£dddýdef �Ë@pcdcgfgeýdeg¢pdd `RÁ@ddfic�ý—š™™™™�B@hddýýefn �Ë@ �Ë@idceffeý `RÁ@neg¢jddddfe‚�ý–f33333³<@ddýdffrc  ØÃ@d}gfgeýd Àg¹@eg¢eddffegc‡}—fedýddfpxd `tÇ@cifgeýdf @�¾@g¢jddddfld©ý§fedÍÌÌÌÌLG@ýýeepedc �Ë@ �w@geeeýefg `RÁ@¢iddddfid¤ý�eegýÍÌÌÌÌ�U@  Øã@ýgftpeci  ØÓ@dgeýpeg¢ ÀgÙ@fhdfffek¤ý‚eedýý  T@ €¦õ@ à<à@gfvgdceffeýpegnf ·ë@ddfffei‘ý�fddýde333333?@ `t×@fnfdcifheýdegngd @�Î@dddfe}�ýÍÌÌÌÌL7@¦dedýýee  Øã@ �w@qiecidfeýeegnlhd ÀgÙ@ddfih†ýwš™™™™™5@dddýýddo `tÇ@ �w@cecifgeý @�¾@effnchdddfec‹ý§eÍÌÌÌÌÌ4@edýdddpc `t×@eceefeýd @�Î@fgnchdddficŠ�›fedýdeepef à<à@cedjeýdf @ÉÔ@gnffdffeex“ýƒdedffffffC@ýýgdxcec P[è@ �Û@efheýpfg à-ß@nchdfffec„ý‹eedý33333³9@ û¹@defqhdcifffýdfgý ¡°@ �r@fddddfhe”ýƒiddýý3333338@ �w@ �w@efpdecidffýefgýd �m@ �r@hdddfjcyr‘gddýde û¹@fpeeciddfýdedýgh ¡°@ �r@dfffe‚ˆý €5@—jddýýef €Ã²@ �—@oddxifdfýfemýpdd �¨@ �r@ffegcdd¬iedýdddo �w@cecgfgfý �m@dfgýchdd �r@ddic”ý�f3333338@edddedpclcefgedÆfgwchdffffc‘¢˜fddýýefppd P[è@ P[è@ceeeeýpe à-ß@gwfddfffef”ý˜fdfš™™™™�:@ýýgetifc �Û@ �Ë@gdefýnfg `RÑ@whhdddfig“ý�fedýffffffA@ ðÄí@ýeepeece  ØÃ@dgeýmfgw Ð ã@hedffffg˜ý}eddýdfffffæQ@ à«Î@edrcfcifheýdfgwc  ¡Ã@hdddfec€ý‹dddýýfš™™™™™5@ à<à@ �Ë@frddcefeeýnfgwgd @ÉÔ@dddffe„ýfffffæ:@}dddýýef  ØÓ@  ØÓ@qcecidgeýpfgýfhd ÀgÉ@ ¼”@fffec™ý�33333SQ@dddýýgdt ðÄí@ ðÄí@cecifieý Ð ã@pegýchdf ¼”@ffec“ý­gffffffC@ddýdefpc à«Î@fcfdgeýd  ¡Ã@fgýhhddd ¼”@ficƒýzddÍÌÌÌÌÌB@dýdefqed Á¯@cefggýdf S¤@gýfdffff ¼”@ffƒý dddÍÌÌÌÌÌB@ýýefsrec `t×@ `t×@iddeýpee @�Î@ýhhdddfe ¼”@”dd~ddhý �—@dhfvfecidgeýdfgý ��@ ¼”@ehddddje‡ý~dedýý3333333@ à<À@ �—@eepgecidgeýfedýh @É´@ ¼”@hdfffff„ýœeedýýefffffæA@ `tÇ@ `tÇ@etdfcgdfgýnfgýeh @�¾@ ¼”@dfffedšý33333sO@�dfdýddd  ØÓ@jcecefdeýÆfdýcee ÀgÉ@ ¼”@ddfkc ýœ33333SP@eddýýefr À ò@ À ò@fdcgegeý €�ç@pfgýdddf ¼”@ffee„ýªgffffff*@ddýdddjc à<À@ecffegýd @É´@fdýchfdd ¼”@fkc€ý§hd3333339@dýddence Á¯@cgdfeýde S¤@dýehdddf ¼”@icƒý…ffdš™™™™™3@ýýdfldfc  ØÃ@  ØÃ@gdgeýmfl Àg¹@ýihffffe ¼”@d|ýˆiedý33333³4@ `t×@dedpclceegeýdfgr @�Î@chdddfkcƒý�dedýýfffff¦K@  Øã@  Øã@eeridcgegeýpegrh ÀgÙ@ddfffeh‰ý„dedýýeš™™™™™2@ `t×@ `tÇ@dpceciegeýnegrch @�Î@dfffec™ýÍÌÌÌÌŒM@ˆdedýýed €¦õ@ ðÄí@pcecgegeýpedrced ·ë@mmfec’ý£33333óC@fddýýedp�$(¾1Ÿ�Aáî¬��^�AcecfedeýŒJê„�u Apegrcheddeecdd•gddýdeepf À ò@dceeeeýd €�ç@fgrfdddddkh�ýzdefffff¦L@dýdgdtcl à<à@‡iegeýde @ÉÔ@g‚chdfffgc‡ý¤fed33333³7@ýýeepfdc ðÄí@ à<à@eeieýpfg Ð ã@‚fddfffef}ýœfddd333333)@ddemeecidgfdqfm‚dhffffedƒý™gedddš™™™™ÙC@get}dcegkedqeg‚fddeeee‚¥ý�fedýýgÍÌÌÌÌ,T@�$(¾1Ÿ�A P[è@dtcfcefieýpfg‚chŒJê„�u Adfffec|ýfffffæ>@¡hddýýde à«Î@ à«Î@oedcgfdfýofgýjdd  ¡Ã@ ð‰@ddfke®ý– U@eedýýhet À ò@ P[è@ndcgffeý €�ç@pegýeddf ð‰@ffex‡ý�i33333óE@edýýefse  Øã@  Øã@dcefdfýp ÀgÙ@egýeddff ð‰@fed‡ýyeeÍÌÌÌÌÌ5@dýdefpdd P[è@cifgfýdf à-ß@gýfddfff ð‰@fn¦ý™fdgÍÌÌÌÌLO@ýýhdxce‘�$(¾1Ÿ�A ðÄí@efdeýpeeŒJê„�u Aýchdffeg ð‰@ctý°fddýš™™™™™*@ �Û@deepcfcfddeýdfmý `RÑ@ ð‰@ihdddfic˜ý¢hegýý33333óP@ €¦õ@ �Û@hevnecedefýpfgýh ·ë@ ð‰@hdffeen‡ý eddýýefffffæ4@ `t×@ `t×@fqeecgddfýpegýie @�Î@ ð‰@dfffeg¬ýfffff�W@�dedýýhe�$(¾1Ÿ�Aáî¬��^�AxfdceejeýpegýfddŒJê„�u A ð‰@fffe‚}ý¢ÍÌÌÌÌL7@jedýýdeg  ØÓ@  ØÓ@cdchfgeý ÀgÉ@pegýddee ð‰@efec}ý�hÍÌÌÌÌÌ7@edýýeepe à«Î@ à«Î@dcgfgeýo  ¡Ã@fgýfddff ð‰@fexŸý¨hdfffffFP@dýdeepcf �Û@cgdeeýde `RÑ@gýiedddf ð‰@ic}ý–jed333333)@ýýedpcfc  Øã@ €Ã²@eejeýjfg ÀgÙ@ýchdffef ð‰@c‡ý‡gddý33333óE@ Ð#Ñ@ýddiclce `tÇ@fggýnfgý àðÅ@ ð‰@chfddfkc‘ý‘eedýdffffffC@  ØÓ@efpcdciffeýdegýg ÀgÉ@ ð‰@ddddfhc|ý�ddeýýe33333³4@ `t×@ `t×@epdeciddeýpegýlh @�Î@ ð‰@dddfe›‡ýÍÌÌÌÌÌ5@~dedýýef  ØÃ@  ØÃ@sddciffeýmegýedd Àg¹@ ð‰@feffn“ý¤ffffffC@fddýdeep  Øã@ddceedeý ÀgÙ@dfgjhddddfilšý¢f33333sO@edýýgetf €¦õ@  ØÃ@dceeieým ·ë@fgjfddffefeŒý›heÍÌÌÌÌL<@dýýdfmcf  ØÓ@  ØÓ@cedfeýpe ÀgÉ@gjihdffeec€ý–ged3333339@ýýdfpcdc `t×@ `t×@gfdeýpeg @�Î@jhddfffecƒý–hedýš™™™™™3@ à<À@ýfdrcecg à<À@fgeýmfgj @É´@chdffffc•ý­fedýdš™™™™YE@ �Û@eepgfcedheýdeg‹f `RÑ@hdddfid‡„¡gedddedpcechededdeg‹chdddficŠý33333³9@ydddýýed 5¤@ 5¤@pcfcifgeýgfd‹chd Þ™@fffec¯ý°33333SV@gddýdeer `t×@edcegieý @�Î@deg‹hddddfii›ý³fš™™™™™Q@dfýdedsc ðÄí@ecefeeýd Ð ã@fg‹chdddffcŒý„ed33333³@@dýdefqge  ØÓ@ciddfýdf ÀgÉ@gýhhdfff ˜ˆ@en”ý™eeg3333338@ýýhfvlfc à<à@  ØÓ@edefýpfg @ÉÔ@ýjhffffe ˜ˆ@l¡ý™fedýÍÌÌÌÌìQ@ �Ë@ýefpkecg �Ë@dfeýnfgý `RÁ@ ˜ˆ@fhdfffepƒýšfedýýÍÌÌÌÌÌB@ `t×@ `t×@efsndcgfdfýpfgýg @�Î@ ˜ˆ@ddfffei”ý¢hedýýd3333338@  ØÓ@  ØÓ@fnddcffefýpfgýld ÀgÉ@ ˜ˆ@dfffef‡„Çmedddemsýdcggje 0�@ddegýÆdd ˜ˆ@ddflý’˜— 0�@ddeýýefr  Øã@  Øã@deciddfý ÀgÙ@pegýehdf ˜ˆ@ffek¥ý„eÍÌÌÌÌLU@e}ýýgfwe  Øã@ Ð#Ñ@ecedefýo ÀgÙ@fgýehddd ˜ˆ@fhŒŠýzddÍÌÌÌÌŒM@pýýeft”e P[è@ à<À@ciddeýmf à-ß@gýdedfff ˜ˆ@f‚Ÿý“gdffffffFP@ýýedpcec�$(¾1Ÿ�A ðÄí@eeheýpegŒJê„�u Aecedfffec�ý§fedý33333óC@  Øã@ýeepedce à<à@gieýpege ÀgÙ@fddffffe€ýxdeddýš™™™™™5@ û©@edpcecieiedhfgechdfffec�ý‘fddýýgš™™™™�B@ ðÄí@ ðÄí@eufecgdfeýpegefe Ð ã@dddfhdŠ�‰dddýýed  Øã@ �Ë@pcecgfdeýnegýced ÀgÙ@ àk@ddfecšý�33333sO@dddýýedr  Øã@  Øã@ceciffeý ÀgÙ@peeýchdf àk@ffecƒý}dš™™™™ÙC@edýdefpc à<À@dcifeeýd @É´@egýedddd àk@fjc}ý�ddÍÌÌÌÌÌ7@dýdefped  Øã@cifefýdf ÀgÙ@gýlddddf @‚@hedd�gddýddfoýdc �w@ 0�@hfegýdfd �m@ýfdfdddk @‚@c}ýydedý333333)@ �—@ddfpddcifgeýdfgý ��@ @‚@hddffefe‰ý•fedýýš™™™™™2@ à<à@ à«Î@efpddcefdeýoegýg @ÉÔ@ @‚@deeffed‰ý—eedýýeÍÌÌÌÌL@@ �Û@ �Û@dqcechegeýpfgých `RÑ@ @‚@dfffecƒýš™™™™ÙC@¦dedýded  ØÃ@pcecfffeýdfgýchf Àg¹@ @‚@ddfic†u‰ffdýýeep `t×@  ØÃ@gdchfggý @�Î@mfgýgdff @‚@ffeŒ‘ý’f333333?@dddddgnxechdgfddegýkhddd k@fhd‘ý–fefffff&C@dýýedrcf à<à@ �Ë@cgemeýnf @ÉÔ@dýchdddf k@hc¥ýŒhddÍÌÌÌÌ,T@ýýefr dc €¦õ@ €¦õ@eeeeýpfg ·ë@ýfdeddfe k@‡}ý~eddý333333)@ à«Î@ýedpcfch à«Î@fggýofgý  ¡Ã@ k@chffffecŸý·hedýd33333SQ@ à<à@efpcdcefgeýdfgýf @ÉÔ@ k@ddddfkc¡ý²fddýdhÍÌÌÌÌìQ@ ðÄí@fxddcefgeýdegýed Ð ã@ k@dddfipƒ{–fedýddd ðÄí@pcecefeeýdegýcee Ð ã@ k@ddfhc“ýƒffffffC@dedýdedr ðÄí@clceegeý Ð ã@dfgýchdd k@dfjc“ý£ffffffæL@ffýýgfte à<À@ û©@fcgdfgýh @É´@egýehfff k@effdd¯iedýdedpcl @Bù@ceeheýqe À*ð@gýchdfff k@ec|ý†ffd33333³4@dddfeefcidegddfgyghfddfki’˜­gedý  Øã@deepeecedjeýdfgy ÀgÙ@ghdddeid¬ý˜gedýýš™™™™YS@�$(¾1Ÿ�A À ò@hexndceeieýpegyeŒJê„�u Addfffen•ý³hedýde33333³F@ P[è@etkdcgedeýdegygd à-ß@dddfixƒýš™™™™ÙC@›ddfýýge P[è@ `t×@trecedeeýpfgyhhd à-ß@fffeiŠý|ÍÌÌÌÌŒM@dedýýhex P[è@ 5¤@hdcefdeý à-ß@gfgyeddddfjh–¨«hedýdgeth €¦õ@dcefheýq ·ë@eg’gddddfih–¨�edsýýgetsf €¦õ@ ðÄí@cedjeýpe ·ë@g’ehdfffeg–ý™deÆfffff&Q@ýýgetjfc €¦õ@ €¦õ@idieýpeg ·ë@’fhdfffeešýµdedý S@�$(¾1Ÿ�Adgfufdcefdeýdeg’ŒJê„�u Aiddddfi‚‘ý½eddddfffff&C@dflcfcgddeddfd’fhdddfic‹ý†edhýýeÍÌÌÌÌìP@ P[è@ `t×@epeecedieýpfg’eh à-ß@dddeef®ý U@ƒdedýdhd Á¯@wcecifdgýdfg’chd S¤@ddfjc¥ýxÍÌÌÌÌ,T@ddkýýedq ðÄí@ 5¤@cecifheý Ð ã@geg’chdddfjc ýˆe33333�S@eiýýhewk €¦õ@ P[è@ecedeeýp ·ë@fg’ehdfefej…ý…ddš™™™™ÙO@fýýedqce P[è@ �w@ceffeýef à-ß@gýchdddf ˜‚@fc˜ý�ddgš™™™™™T@ýýgftsec à<à@ à<à@ideeýpee @ÉÔ@ýdhdfffe ˜‚@s›ý“ddgýÍÌÌÌÌLS@ à<à@ýhdvceci à<à@feeýpffý @ÉÔ@ ˜‚@cedddfec ý~dddýýš™™™™ùU@  Øã@  Øã@gfuidcifggýpegýd ÀgÙ@ ˜‚@dffffe– ý–fegýýf33333SP@ @Bù@ P[è@ergecedfeýpegýfe À*ð@ ˜‚@dfffek¡ýš™™™™9T@“gdÆdýgd `t×@ucecefeedpfgxchdddfec‚ý“š™™™™™=@gedýdgdt�$(¾1Ÿ�AcfcefjeýŒJê„�u Adegxchdffffc„ý‡dfffffæA@edýýffrf �Û@ �Û@dcieieýp `RÑ@fgxfddfffee¤ýžfdÍÌÌÌÌ�U@jýýgdtcf�$(¾1Ÿ�A  Øã@ceffeýpfŒJê„�u Agxcheffffc ý—ddd33333�S@ýýget‚dc À ò@ À ò@ifjeýpfg €�ç@xgddddfenddÇmedddhmvkdÇmggedqemxfdemmmmk‡ý¢gedýýš™™™™™?@�$(¾1Ÿ�A à<À@edrce}eeieýmegxcŒJê„�u Ahdfffgc�ý¡eedýdgÍÌÌÌÌŒP@ €¦õ@dtceceffeýdfgxch ·ë@dddflc´ý333333W@Œfd–ýýhd�$(¾1Ÿ�Aáî¬��^�AxcecefjeýpegxchdŒJê„�u Addfec“ý¡ffffffC@dddddeeqedcgfefddfgxeddddfif|ý°f33333³4@ddýýefpf 5¤@ �—@ecedeeýf Þ™@egŒjhdfffff ý£eeš™™™™ùU@‚ýýhftje ðÄí@ ðÄí@cgddeýpe Ð ã@gŒdhdfffexƒý¢fedš™™™™ÙC@dýgetddv Á¯@ffdedifgŒiddfffge–¨“gdfý À ò@ýget‚fce  Øã@dheýpfgŒ €�ç@ehdddfex ý„fddýý33333SP@ €¦õ@ ðÄí@hevxdceeieýpegŒe ·ë@ddfffen„ý idddddffffff*@dkcfcefhgddfg�chfddfkcˆý €4@{eddýdee  Øã@pdecidieýded�fhd ÀgÙ@fffef‘¢ˆfddýýefp  Øã@  Øã@ndceeheý ÀgÙ@peg�eddfffer‡ýzd33333óE@dedýeerd �w@ecidfedeeg�jhdddfhi‚ýxdeš™™™™™=@dýdeeped  Øã@cifieýde ÀgÙ@g�fddefffg„ý{dedffffffB@ýdedrcec �w@egkeýdfg �m@ýchdddfh @g@c�ý�eedý33333óC@ `t×@ýefrefcg `t×@degýpegý @�Î@ @g@hheeffei…ý�dedýýffffff6@ `t×@  ØÃ@educecifkeýmegýc @�Î@ @g@hdfffec”ý„dddýýg ÀI@  Øã@ �Ë@fvedciffeýnfgýdd ÀgÙ@ @g@dddfefŠýÍÌÌÌÌŒM@‰ddfýýhf à«Î@ à<À@xfeciddeýmfgýeed  ¡Ã@ @g@fffen‘ýš333333?@jedýdeer à<à@fdceeeeý @ÉÔ@deg‰jddddfkjŠ��deiýýefre �Û@ �—@ecedgeýf `RÑ@eg‰fhdfffek—ý�feš™™™™ÙO@dýýfdrce û¹@ û¹@chfgeýlf ¡°@g‰chdddflc¡ý°kdiš™™™™9T@ýdgetifc  Øã@fdieýdfg ÀgÙ@‰dldddfirŠ��fedý €¦õ@ýeepxece €¦õ@dfeýpfg‰ ·ë@ehdfefei¡ý™gegýýš™™™™9T@�$(¾1Ÿ�A  Øã@geteecedgeýpfg�eŒJê„�u Aedfffen ýƒfddýýgš™™™™ùU@ À ò@ ðÄí@etxdcefheýpeg�ed €�ç@dddfe‚ŒýÍÌÌÌÌŒF@‡jedýýee À ò@ �Û@rffcedieýpfg�dhd €�ç@fffei�ýˆ33333óC@gddýýeer €¦õ@  Øã@ddceejeý ·ë@peg�eddfffef—ýŽfš™™™™ÙO@edýdgdtc�$(¾1Ÿ�AfcefieýdŒJê„�u Afg�chdddfkc‹ý eeÍÌÌÌÌìP@dýýgetsd�$(¾1Ÿ�A ðÄí@cefdeýpfŒJê„�u Ag�eddfffex™ý²hdg33333SQ@ýdeescfc  ØÓ@edheýdeg ÀgÉ@�ghdddfic„ý¬eddýffffff*@ à«Î@dedpceceejeýdfgn  ¡Ã@chdddfic’˜�fddýý�$(¾1Ÿ�A �Û@eepkfcedieýpfgnfŒJê„�u Ahdfffes‡„•fedýýe ðÄí@ à<à@dpcecefjeýpegnch Ð ã@dfffec”ý3333338@~eddýddf �w@nceciddeýdfgnghd �m@ddfjc“ý¡ffffffC@edfýýget  Øã@ `t×@eecgdheý ÀgÙ@pfgnghdddfex�ý…eÍÌÌÌÌŒK@edýýedpc P[è@ `t×@fceejeýp à-ß@fgnchdfffecuý½gdfffff¦F@dýdeepcd �Ë@cgfdeýdf `RÁ@gngddddfic“ý†edhffffffC@ýýgetiec à<à@ à«Î@edfeýoff @ÉÔ@mghdeffes‰ý¥hddýš™™™™™2@ `t×@ýeepefcg  ØÃ@djeýmegm @�Î@fhdddffd…ý‰fddýý33333óJ@ P[è@ `t×@deofecgdgeýpfgml à-ß@hdddfhn ý�eddýýdÍÌÌÌÌŒJ@ P[è@ à<à@docfcgfgeýpegmch à-ß@dfffec‡ý33333óE@�gddýdef P[è@sefcedfeýdfgmfhd à-ß@ddfkx~ý® €<@iddddeephdceedfddegmhddffefe‡„•eedýýedpc �Û@ �Û@ecgedeýp `RÑ@egmchdfffec€ýŠdf3333339@dýýdfigl  ØÃ@  ØÃ@chdggýme Àg¹@gýghffff È‚@eiˆý™ddfš™™™™™8@ýýedtcec `tÇ@ `tÇ@gfdeýnfg @�¾@ýchdddee È‚@c‡ý©eddýÍÌÌÌÌLA@ �w@dedpcfcfffeýdegý �m@ È‚@chdddficddwhmdýd �w@ddlcfcifheýdfdýc �m@ È‚@hfdddlc„ý²eddýdefffffæA@  ØÓ@eredcgfeeýdfgýhd ÀgÉ@ È‚@dddfiiŸýš™™™™yR@›eedýdhe ðÄí@vndcgfdeýdfgýfdd Ð ã@ È‚@fffen€ý�š™™™™™5@dddýdefq à<À@ddciffeý @É´@dfgýhddd È‚@dfkfuý†efffff¦F@dfýýgftr  Øã@ à«Î@ecedfeýo ÀgÙ@fgýghdfe È‚@fex’ý�de33333³N@dýýgeuif û¹@ û¹@cidheýle ¡°@gýghdfff È‚@fpƒý‡defÍÌÌÌÌÌB@ýýgetsec  Øã@  Øã@idkeýpfg ÀgÙ@ýihdfffe È‚@xƒý¯hddýš™™™™ÙC@ P[è@ýeepdlcf à<À@dgeýmfgý à-ß@ È‚@ghdffffe�ý“gedýýš™™™™�B@ Ð#Ñ@ €Ã²@edrcfcefgeýjegýc àðÅ@ È‚@heffefc”ýŒdddýýe3333338@ ðÄí@ `tÇ@dpceceejeýnfgýcf Ð ã@ c@dfffec‡|žeedýýdd  Øã@ `t×@pceceffeýpegýchd ÀgÙ@ c@fffec™ý™33333SQ@fdhýdhdw�$(¾1Ÿ�AceceeeeýŒJê„�u Adegýchdd c@dflcƒýwdš™™™™ÙC@dddýedpc �—@fcifgedfffýchddd c@fec}ý™fd333333)@dddeepedcefeedqfgýfddffe c@es˜ý…ded33333óP@ýýgetxdc�$(¾1Ÿ�A à<à@efheýpfgŒJê„�u Aýeddfffe c@i…ý�dedýš™™™™�G@ @Bù@ýeepdfce ðÄí@dheýpegý À*ð@ c@hgdddfinŠ�~dedýý @Bù@ `tÇ@fgtddcifeeýnegýd À*ð@ c@ddffff¿Œý‹fddýýefffff&C@ ðÄí@ ðÄí@dpcecgfmeýpeg�ch Ð ã@dddfec|ý33333³4@˜dddýýdd `t×@ `t×@mcecgfheýpeg�chd @�Î@fffec“ýœffffffC@eedddefqhecgdeeddfg�khdddfl€„ý›ffffffæA@edýddfod û¹@fchdfeýd ¡°@fg�hhdddfid„ý–dd33333³9@dýdefpdd �w@cifdeýde �m@g�nddddflg‡ý™eed33333³7@ýdfdsclc €Ã²@ifdeýdeg �¨@�chdddflcuý‘dedýfffff¦F@ à«Î@ýhdxcfce �Ë@ffeýneg�  ¡Ã@chdffffc“ý§fddýýfffffæL@ ðÄí@ ðÄí@fesidcgfeeýpegye Ð ã@ddddeesyý«hddýýd333333+@ `t×@ `t×@enddcffgeýpfdyfd @�Î@dfffef”ý33333sB@�fdfýýed ðÄí@ à<à@scecifefýpfgyche Ð ã@ddfic™ý‰333333O@fedýýhdv P[è@ P[è@ceceeheý à-ß@pegychdfffecƒý�fš™™™™ÙC@edýýedpc à<à@ �Û@ecefeeýp @ÉÔ@egychdeefec¤ýžfe33333óQ@dýýhdxce À ò@ À ò@‘gfdeýpe €�ç@gychdfffgc«ý¦ediÍÌÌÌÌ,V@ýýhdtcec €¦õ@ P[è@efdeýpeg ·ë@tcfdddfic¥ý¡dedýÍÌÌÌÌ,T@ À ò@ýfdrcfci ðÄí@eheýpegt €�ç@chdddfic‡ý dddýd33333óE@ `tÇ@efpdecgdfeýdfgth @�¾@hdddfke�ý}dddýdeš™™™™�9@ Ð#Ñ@dpcecifheýdegtch àðÅ@dddflc�rœddqýdef  ØÃ@pcfcidfeýdegtphd Àg¹@ddfic‡„�fddýýefp à<à@ �Û@ddcgfmeý @ÉÔ@pegtfddddfhn�ýŽfffffffL@ddýýhevl P[è@ à<À@dcefdeým à-ß@fgteddddefg¡ý¤fdš™™™™9T@dýýhexhd�$(¾1Ÿ�A P[è@ŒefheýpfŒJê„�u Agwfddfffg‡“ý¥dedffffffC@ýdfereec `t×@gdeeýdfg @�Î@wihdddfih‘ý£fddd333333?@defpslchdgeddegw|hdddfiŒ‡|’eddýd û¹@edmcecgfgeýdfgwc ¡°@hdddflcƒý…defýýgÍÌÌÌÌÌB@ �Û@ �Û@fteecideeýpegwee `RÑ@dfffef—ýš™™™™YR@„ddgýýgf `t×@ `t×@tfecidfeýpegwjhd @�Î@ddfe‘uý fffff¦F@eedýýfdr à<à@ à<à@cecgfeeý @ÉÔ@pfgwchdfffec�ý›gfffffæE@ddýdeept à<À@fcedheýd @É´@fgwfhdffff}ƒý…feÍÌÌÌÌÌB@dýdgetfd �Û@ceffeýde `RÑ@gwgddddfki}ý’gddÍÌÌÌÌL7@ýýedmcec à<À@ û©@gfieýheg @É´@wchdfffec…ý}dedý333333+@  ØÃ@ýdfnddci  ØÃ@fggýmeg‚ Àg¹@fdeffffi¤ý‰fedýýÍÌÌÌÌ�U@ à<à@ `t×@fdscecifmeýpef‚c @ÉÔ@hdddflc®ýˆdddýýg U@ à<à@ �Û@dscecefdeýpeg‚ch @ÉÔ@dfffecœýÍÌÌÌÌ,V@‹eepýýhf À ò@ à<à@teeceddeýpeg‚dhd €�ç@fffendd�ddgýdhdv `t×@cecifegý @�Î@deg‚chfdddjcƒý�fš™™™™ÙC@ddýdfdrc �—@ecifgfýd ��@fg‚chdddffc¡ý}ddš™™™™9T@dýýgftfd  Øã@ `tÇ@cifdeýnf ÀgÙ@f‚eddffffl‡}€dedýýedpcec �—@ �—@efgeýfff ��@pchdffffc”ý²geÆd ÀL@dhevdfcfdgeddegpfhdddfic‚ý‹fddýdš™™™™�8@ @Bù@eepedcefieýdfgpf À*ð@ddddfkg…ý…dedýýgš™™™™ÙO@ `t×@ `t×@etldcifjeýpfgphd @�Î@dfffepŠý ÀN@§fddýdee ðÄí@sfdcffeeýdfgeede Ð ã@ddfie•ý«33333sL@eedýýeds�$(¾1Ÿ�A  Øã@cecffdgýŒJê„�u Apfgecheddeic’˜žeedýýgdwc  Øã@ à<à@ecgedeýp ÀgÙ@fgechdfffec�ý†dfÍÌÌÌÌŒK@dýýgftlf  Øã@ Ð#Ñ@ceddgýof ÀgÙ@gefhffffeŒŠ�Ÿdedýýhewxdc €¦õ@ À ò@efeeýpeg ·ë@eeddffeenddªdedý `t×@dhexedcifeeýdffe @�Î@idddddixˆý§eedýdš™™™™�:@  ØÃ@efpcdcgffgýdfgji Àg¹@ddddficyr–dddýde `tÇ@dpcfcefjeýdegjch @�¾@dddflc™ý33333SQ@§eedýdge  Øã@tddcgfdeýdfgjedd ÀgÙ@ddfid£ý‡ffffffT@fedýdedp  Øã@cecgfgeý ÀgÙ@dfgjchdddfkcŠ�‰hddýýefne P[è@  ØÓ@fcidgeýp à-ß@fgjghdddfej{ý§jdÍÌÌÌÌŒB@dddedqcfcfefgddfdjchdddfic–ý­fddfffff&Q@ýdeepddc `tÇ@ffegýdfg @�¾@jjddddfin”ý�eddýš™™™™™A@ P[è@ýeepdece `t×@dieýpegj à-ß@fgdfffef|ý²eedýd33333³4@  ØÃ@dflcdcifdeýdedýl Àg¹@ Pp@ddddfictýxdeddýeš™™™™�6@ �—@epefcidffdfegýgh Pp@dffffk©ýÍÌÌÌÌLG@†fddýýef  Øã@ Ð#Ñ@qfdcieeeýofgýfdd ÀgÙ@ Pp@fffey€ý´ÍÌÌÌÌÌ2@hddýddeo `tÇ@cecfdeeý @�¾@dfgýhhdd Pp@dfic‡ýœg33333sE@edýdddoc 5¤@fcefeeýd Þ™@fgýchddd Pp@fic„ý§fd333333:@dýddeodl Ð#Ñ@cedgeýde àðÅ@gýihdddf Pp@lcƒýzdedš™™™™™3@ýýdendfc P[è@ à«Î@idgeýofd à-ß@ýghdfffe Pp@h€ý{dedýš™™™™™5@ û¹@ýedqcfci €¦µ@fmfýkfgý ¡°@ Pp@chdffffc‡„ƒfedýý à<à@ à<à@dfmde idgeýpegýk @ÉÔ@ Pp@edfffgx“ýªgedýdeffffffC@  Øã@dpcecffeeýdfgmch ÀgÙ@dddfic¤ýÍÌÌÌÌ�U@›dedddgdtcfcgfeeddfgmchdddfkc®ý„ U@dedýýhdw  ØÓ@  ØÓ@cecifkeý ÀgÉ@pegmcedffffcšýœeÍÌÌÌÌLR@edýýhdvc  Øã@  Øã@fchfgeýp ÀgÙ@fgmchdfffec“ý†fefffffæL@fddgetpecedjedqegmghdfffep¤ý‘gev  T@ddheufecedeedqegmfgefffejƒý fddd33333³A@dedpcecefgedqegmchdffffc�ý‹fedýýš™™™™ÙB@ P[è@ à<à@eepdecedeeýpegje à-ß@edfffee�ý´gddýdefffffæJ@ �Û@erddcegieýdegjjd `RÑ@dddfi‚“ýš™™™™YO@€dedýýfe `t×@ `t×@rxecidfeýpegjghd @�Î@ffffn¤ý�  T@edfýýhex À ò@ à<à@efŒedgeý €�ç@pffjdedfffgj€ý‚g3333339@ddýýednc  ØÃ@ à<À@fcifgeým Àg¹@egjchdfffecƒý³geš™™™™ÙC@ddddelcfcfdeeddfdjfhdddficŸý�fed33333SQ@ýýferedc P[è@ �Û@eeeeýpfg à-ß@jfddffffrˆý’efdýÍÌÌÌÌÌ,@ �Ë@ýeemffcg �Ë@dggýnem� `RÁ@jhfeffed}ý—fedýýÍÌÌÌÌÌ7@ Ð#Ñ@ à«Î@eepddcffdfýofg�f àðÅ@ddfffed‡ýŸhddýdd33333óE@ û¹@flddcgfhgýdfg�ed ¡°@eddfieˆýš™™™™™8@‹dddýdef û¹@pkecidkeýdee�thd ¡°@ddfhfšý—33333sO@ddgýdgft à<À@deciddeý @É´@dff�nhdddflh“ý¦fffffffC@ddýýedpc  Øã@  ØÃ@fcefieým ÀgÙ@fguchdddfic´ý„ee333333W@dýýgdtce €¦õ@  Øã@cefhgýpe ·ë@guchdddflc‘ý›ddd333333?@ýýeepddc À ò@  Øã@eeieýpeg €�ç@ufddddfej–¨¢gddý�$(¾1Ÿ�Aýgdtcfce €¦õ@eeeýpeguŒJê„�u Acheddfecddweddýd�$(¾1Ÿ�AdeoddcieieýdegueŒJê„�u AdddddjgŠý¯gedýddffffffE@ �Ë@dncfceeeeýdeduch `RÁ@dddfic…ý333333+@wdedýdee Ð#Ñ@pffcideeýdfguhhd àðÅ@ffffs‹�–fddýýeep À ò@ �Ë@edceeieý €�ç@nfguedeffeej£¼‰gedýdgetk ðÄí@dceejeýÆ Ð ã@fgigddffffn�ý�ješ™™™™�B@dýýefrdd  Øã@  Øã@cgeieýpe ÀgÙ@gihdeffferyr~dddýýgftddc  ØÓ@ à<À@ifkeýmeg ÀgÉ@idddddfeŒƒýwdedýš™™™™™3@ à«Î@ýedpceci û©@eheýhfgi  ¡Ã@chdffffc}ý�dedýd333333)@ û¹@eepfdcifkeýdfgih ¡°@ddddel|“ý®fddýdeffffffC@ `t×@erfdcgfgeýdfgiid @�Î@dddfifuýfffff¦F@ gddýýed  Øã@  Øã@pcecefeeýpegichd ÀgÙ@ddfec�ý•333333J@fedýýgdt�$(¾1Ÿ�A P[è@cecefheýŒJê„�u Apegfcedfffecƒý�gš™™™™ÙC@dgýdedsc À ò@ecefdeýd €�ç@fgfcfdddflcƒý‹edš™™™™ÙC@dýýefp‚e P[è@ P[è@cgdgeýpe à-ß@gffhdddfeiƒýŸgddš™™™™ÙC@ýdeepdfc�$(¾1Ÿ�AedieýdfgŒJê„�u Afnhdddfkp“–±iddý �Û@deepdfcfdieýdegf `RÑ@dhdddficyr…hedýd `tÇ@dfnfdcifdfýdfghl @�¾@ddfffer‡ý—fedýýeÍÌÌÌÌ�A@ à<à@ û©@epfecgddeýheghfe @ÉÔ@dddfhr‹ýÍÌÌÌÌ�C@­hddddedpcfcefheddeghchdddfic‰ý†š™™™™™2@ddhýdedr ðÄí@cecifeeý Ð ã@deghchdfffecyr‰fedýýgdtc û¹@ û¹@fcgfgeýl ¡°@fghchdffffc�ýŸfd33333³<@dýýdende Ð#Ñ@ Ð#Ñ@cedheýoe àðÅ@ghlhdfffedŒý›eedÍÌÌÌÌŒF@ýýefqdec ðÄí@  Øã@gddfýpeg Ð ã@hfedfffed‘ý¬eddýfffffæM@ `t×@dferddcefkeýdfgm @�Î@fddddfif«ý–ddgýýÍÌÌÌÌ,V@ €¦õ@ à<à@hevfeciddeýpfemh ·ë@hdffeek™ý fedýýf33333SQ@ P[è@ `t×@ersecedieýpfgmgf à-ß@dfffen‘ý333333?@�dedýýdf à<à@ û©@odlcidgeýhegmghd @ÉÔ@ffefctý­š™™™™™*@fedýýdeo P[è@  ØÃ@cfcedjeý à-ß@megmfedffffc ý‘g33333�S@edýýhevf @Bù@  Øã@fcedieýp À*ð@fgmffdfffee‘ý™eeffffffC@dýýedpcf `t×@ `t×@cegdeýpe @�Î@gmchdfffecŠ�…fddýýedrcfc û©@ 5¤@gfdeýgfg ¡ @mchdddfec€ý�iedý3333339@ Á¯@ddemcdcgfgeýdedm S¤@lddddflc‡ý¦fddýýš™™™™™?@  ØÓ@ à<À@edpcfcggjeýmegmc ÀgÉ@hdddffc„ý’dedýýe333333:@ À ò@ à<à@eqeecidheýpfepfh €�ç@dfffeg|ý333333-@˜fedýýed�$(¾1Ÿ�A Á¯@pceceffeýifgpchdŒJê„�u Affffc¤ý   T@fedýýhdv P[è@ P[è@clcgeeeý à-ß@pfgpcedfffecˆý�dš™™™™™8@edýýedqc �Ë@ `tÇ@ecifeeýn `RÁ@ffpchdffffc|ý¨hd33333³4@dýddenee à<à@cedjeýdf @ÉÔ@gpjhfddffd™ý¢fddÍÌÌÌÌ�P@ddedsceceeheddegpchdddfic�ý¥fedý @J@ À ò@ýeemddce P[è@eheýpegv €�ç@fddfffec‡„¨iedýd  ØÓ@eepdfcedieýdedvh ÀgÉ@hdffffc£ý�eddýýgffffffT@ ðÄí@ P[è@etxdceeieýpegvdd Ð ã@dddfeiƒýš™™™™ÙC@£gddýýee �Û@ 5¤@pddcefdeýgfgvfdd `RÑ@ffffd“ý…ffffffC@eddýýget À ò@ `t×@‚dceegeý €�ç@pfgvfddddfen}ýydÍÌÌÌÌÌ7@edýýeepd P[è@ Á¯@dcifefýi à-ß@egvdddffffm“ý·jeÍÌÌÌÌÌI@dýdeepde �Ë@cfdheýdf `RÁ@gvihdddfke¤ý“ked  T@ýýgdtcfc À ò@ `t×@efieýpfg €�ç@vcedfffec}ýˆfddý333333)@ �Ë@ýdenedci `tÇ@fgfýnfgv `RÁ@fddfffeidd§fddddeepdfcgdgeddedvghddddic“ýµhdeýdgffffffC@  ØÃ@etrfcfdeeýdfgyih Àg¹@dddfie€ý3333339@~deeýýeg à<à@ �w@rfecidheýeegyfhd @ÉÔ@ffffpyý•333333+@hddýdedp  Øã@cecgfheý ÀgÙ@dfgychdddfkc€ýxdš™™™™™5@eddýeeqy €Ã²@ecidiedjfgyfhdddfjlzý|deš™™™™™-@dýdefqdf Ð#Ñ@cedheýde àðÅ@gyfhdfffef…ý›hddš™™™™ÙO@ddeeq‡fcediedqfgyehdddfe¥yý€dedý333333+@ �Ë@ýefmeeci �Ë@dieýnegy `RÁ@hhdfffek•ý²gedýd33333sL@  Øã@eepeecedheýdfgyf ÀgÙ@hdddeik¢ý·gddýýgš™™™™ÙQ@ Ð#Ñ@ Ð#Ñ@euldcgfdeýofgyfd àðÅ@dddfiŒ§ýffffffR@‰de‡ýýgf à<à@ `t×@xjecedkeýpeg�ehd @ÉÔ@ffeesddÇdedddedvcecifgeddff�chddddjc|ýzd333333-@dddýgftg Á¯@dcifgedieg�fddddfjs‡ý›hdÍÌÌÌÌÌ5@dýdgetgd P[è@cfedeýdf à-ß@g�eddddeihŽý‚dfx33333³U@dýhfvgfc  ØÓ@edgfdpeg�dhfddfjxtý’eddýš™™™™™*@ à«Î@ýefnsdcf 5¤@figýgfm�  ¡Ã@hdfffffd™ý‘feddd333333O@getsdcefgedqeggfddfffe}›ý�eddýýeš™™™™™B@ �Û@  ØÓ@epddceegeýpfggfd `RÑ@dfffes„ýš™™™™�C@£fedýdee `tÇ@pddcgegeýdegggdd @�¾@ddfii|ýŸ33333³4@gedýdedm Ð#Ñ@ceceffeý àðÅ@dedgchdddflc€ý�gš™™™™™5@dddddfnfdciehedÆfgghddfffeg‡ý¤eefffffæ4@dýdedpce `tÇ@cgedfýdf @�¾@ggchdddfic‹ýƒdddš™™™™�>@dýeepddc Ð#Ñ@ieiedoegggddfffef�ýŒfdgý @J@ À ò@ýhdtcfch À ò@ejeýpegf €�ç@chdfffec¤ýˆdefýý  T@ `t×@ `t×@heufecedeeýpfgfi @�Î@hdfffe‘•ý—edeýýe33333sL@ ðÄí@  Øã@ergecedgeýpegffh Ð ã@dddeei|ýš™™™™�1@’fedýdee û¹@pddcifheýdfgfldd ¡°@ddfki}ý�ÍÌÌÌÌL7@gedýýeer ðÄí@ à<à@deceddeý Ð ã@pegfgedfffef¤ý‡e  T@dgýýgevi ðÄí@ �Û@fcedeeýp Ð ã@fgffedfefeiŠ�‰eenýýefsje  ØÓ@ à«Î@cidefýof ÀgÉ@gkfhdfffeg�ý¢ddd33333³<@ýýedpcfc à<À@ à<À@eegeýmeg @É´@kchdffeecdd½eedý û¹@ddemcdcffeeýdfdk ¡°@fdddddkc…ýydedýýffffff6@ û¹@ û¹@edqcfcifdgýledkc ¡°@hdddfjc�ý¦gdddde333333B@dpcfcegjeddedkchdddficyý333333+@˜dedýýed P[è@ �Û@qcecgfheýpegkchd à-ß@fffec�m¥eedýdefq  ØÃ@eecgddfý Àg¹@dfgkhhdddfle€ý•g3333339@edýýefod Á¯@ Á¯@dchfdfýi S¤@egkfddffeek“ý�feffffffC@fýýfesfe À ò@ �Û@ceddeýpf €�ç@gjghdfffeŒ�ýšgedš™™™™yS@ýýfesnfc À ò@ À ò@edheýpeg €�ç@jhhdfffef´ý‰eddý333333W@ @Bù@ýhetedce @Bù@eieýpegj À*ð@fddffeei”ý‹gedýý3333338@  ØÃ@ à<À@efpfdcefdfýmegji Àg¹@ddfffex‘ýžhedddffffff&C@eredceejedqfgkeddffeed€ý3333339@ºgfdýdee  ØÃ@pcdcffieýdedggdd Àg¹@ddeic“ý±ffffffC@hedýdedp à«Î@cfcgfheý  ¡Ã@dfggchdddfic”ý•e3333338@edýýeepe  ØÓ@ `tÇ@dceegeýn ÀgÉ@fggiddffffg’˜§fedýýeepef ðÄí@  Øã@cfdheýpf Ð ã@ggfhdfffef•’žeddýdeeqddc  ØÓ@eeieýdeg ÀgÉ@khddddfih”ý feddš™™™™�:@dedpcfcefeedqfggchdddflctýœiddýdš™™™™™*@ �w@efpfdcgfdfýdfggh �m@ddddff}€ý�ddfýýe3333339@ à«Î@ à«Î@ercfcidgeýoegglh  ¡Ã@dfffec‡ý33333óE@«iddýdef `t×@pdfceddfýdedˆfhd @�Î@ddfic–ýº33333“Q@ddgýdgdt ðÄí@cfceejeý Ð ã@degˆchdddfic¢ý„d333333I@dhýýfdrc  Øã@  Øã@fcgfieýp ÀgÙ@egˆchdddfhc}ý¡fd333333)@dýddeocd `tÇ@cffefýdf @�¾@gˆfddddficddwddddddepýdc 0�@ifgfddfgmhdddddjý“ýŠgedd 0�@ffffffC@deesgfcedgfdqfgmehdfffedyr}dddýý €Ã²@ �—@edqcecefgfýffgmc �¨@hdfffecddwdeeddempefcidjedqedmdhdddmjf~ýÍÌÌÌÌL?@©kddýdde Ð#Ñ@ncdceedfýdegmidd àðÅ@ddficyr�gedddedpclcefdedÆfgmchdfffec€ý‘eš™™™™™5@ddýýeend `t×@ €Ã²@dceegeýj @�Î@fgugddffffx�ý¦dd @J@dýdddncf Ð#Ñ@cifdfýde àðÅ@duchdddfic~ý‰ged €<@ýdgfvddc  ØÓ@ifdfýdfg ÀgÉ@ufddddfj ‰ý˜eedýš™™™™™@@ �w@ýedpcfci �w@fdfýefgu �m@chdfffecŽƒ‘dddýý �Û@ `t×@deodfceddfýpeduj `RÑ@hdfffec”ý›eedýde3333338@ à«Î@dpcecefdfýdfguch  ¡Ã@dddficdd‚dddýded û©@pcecifdfýdfguchd ¡ @dddlc’˜�dfdýýded  ØÓ@ Ð#Ñ@ffcedfeý ÀgÉ@oegufhffffef•ýœf33333sL@edddfevfdcegjeddfgieddddeld™ý¡fd33333SQ@ddýeer}f à<à@cedjedpfgiehdddfex©ýˆednÍÌÌÌÌLG@ýýfetefc À ò@  Øã@edieýpfg €�ç@idhdfffej®ý–fddý U@ P[è@ýhexkdce P[è@eeeýpegi à-ß@fddfffek‹ýŸedhýýÍÌÌÌÌìP@ À ò@ P[è@getfecgdfeýpfmie €�ç@hdfffef”ý²gedýde3333338@ Ð#Ñ@dpcecgefeýdfgich àðÅ@dddfic”ýÍÌÌÌÌŒC@€fddýdde P[è@lddceeieýdegpjdd à-ß@ddfid–ý¥fffff&C@fddýýeep P[è@  ØÃ@edceeeeý à-ß@mfgpiddddfig ý˜g33333SP@ddýýeepe P[è@ à<à@ecedheýp à-ß@egpehdfffeg¡ý¨gdš™™™™9T@dýdgetcd `tÇ@cggdeýdf @�¾@gpgddddfkc ýªdde33333�S@ýdeerdec  ØÓ@idjeýdeg ÀgÉ@pnhdddfle”ý¦eddýfffff&K@ ðÄí@dgevrdceeheýdfgp Ð ã@fddddfirŠ�¸gedýd  Øã@edscfceeheýdeg}c ÀgÙ@hdddeic‹ýŸeedýýeÍÌÌÌÌìP@ €¦õ@ ðÄí@epŒecedffýpfg}hh ·ë@dfefedvýš™™™™™;@xddfdýee �w@sfecidgedeff}dhdddejs™ýƒfffff&Q@hedýdffr ðÄí@}fchdgfý Ð ã@dfg}fhdfefe”‡„�jedýýhdwc�$(¾1Ÿ�Aáî¬��^�AfchejfýpŒJê„�u Aeg}chdfffecˆý…ddÍÌÌÌÌ�C@dýdefqxd à<À@cieefýde @É´@g}iddeffelŸý‰fed33333�V@ýýhdvcec  Øã@  Øã@gfgeýpfg ÀgÙ@}chdfffec|ýµjedýfffffæ>@ 5¤@ddflcdchfgfýdfd¿ Þ™@hddddfic‡ýªgedýd33333óE@ à<À@dfocfcfddeýdfg¿j @É´@hdddfic|ý‚eedýýe333333-@  ØÓ@ û¹@fnddcifefýleg¿nd ÀgÉ@dfffe¿|ý33333³4@®eddýddd à<À@gcfchfgfýdeg¿chd @É´@ddfic„ý¥fffffæ:@fddýdddl à<À@cecgfeeý @É´@dfg¿chdddflcdd}dedýdgfte û©@dcifieýd ¡ @eg¿gdddddjn‡ý˜deÍÌÌÌÌÌ5@fýýefrif  ØÓ@  ØÓ@cidfeýpf ÀgÉ@g¿dhdffeei‡ýœdedÍÌÌÌÌ�A@ýdefpcec �—@idgeýdeg ��@§ghdddfkc}ý¸hddýÍÌÌÌÌÌ:@ �Û@dddlcecffjeýdfd§ `RÑ@chdddfic}ý–eedýýÍÌÌÌÌL7@ Á¯@ Á¯@egpdecgdheýieg§g S¤@hdffffcddˆgedýde à«Î@fpcdcefeeýdfg§hd  ¡Ã@edddkc”ýÍÌÌÌÌŒC@}dedýýed `t×@ `t×@rcfceejeýpeg§chd @�Î@fffec„ý­ffffffB@hddýýdfo Ð#Ñ@ à<À@ddŒefdfý àðÅ@mfd§iddfffghƒýzdš™™™™™3@dfýýegpf €¦õ@ €¦µ@ecidgeýk ·ë@fg§fhdfffei™ý‰fdÍÌÌÌÌŒD@iýdgetef  Øã@cedggýdf ÀgÙ@g§hffddfk–”ý’idd33333sB@ýýefpfdc  ØÓ@ à«Î@ifffýofg ÀgÉ@§iddeffef”ý‰gddý3333338@ û¹@ýedocecg û¹@fdfýlfg§ ¡°@chdddfjc‡ý�eedýýÍÌÌÌÌÌ5@ P[è@ Á¯@eeridcefdeýifg§h à-ß@deffffiŠ�’gedýdd �—@eqeecgdgeýdff§eh ��@dddfknddzdddddddocecifgeddeg§chddddkc™ýÇÍÌÌÌÌŒD@fddddedncfcefiedqfg§chdffffc“ý…fffffffC@ddýýeepn �Û@ �Ë@dcefheýn `RÑ@fg§eddfffesŒý‚dd33333³@@hýýgdtce À ò@ �Û@ceeffýpf €�ç@g§chdddfec‘ý�edgfffff&C@ýýfdscfc à<à@ �Û@gejeýpfg @ÉÔ@§chdddfec—ýœfedýš™™™™ÙO@ à<à@ýefqndce Ð#Ñ@fjeýofgý @ÉÔ@ °‰@eddffffj¥ý¼hedýd33333ÓU@ `tÇ@efrfdcffjeýdfgýi @�¾@ °‰@ddddfid‹ý¬dddýýeÍÌÌÌÌ�C@ P[è@ `t×@epfdcggheýpegýhd à-ß@ °‰@dddfinœýÍÌÌÌÌÌJ@šdedýýee  Øã@  Øã@pdfcedjeýpegýihd ÀgÙ@ °‰@fffeg“ý•ffffffC@fddýdfer P[è@}dcgfdeý à-ß@dfgýfddf °‰@ffei“ý”hffffffC@deýdedqc à<à@ecgfgfýd @ÉÔ@fgýchdff °‰@fec‡ý–feÍÌÌÌÌLA@dýýdenfd À ò@ �Û@cgegeýpe €�ç@gýdddddf °‰@hŽ‡ý¬gddffffff@@ýddgocfc  ØÃ@fdfeýdfg Àg¹@ýjhdddfk °‰@c‡ýžfddýš™™™™™?@ à«Î@ýeepdfcg  ØÃ@dheýmegý  ¡Ã@ °‰@fhdddfhfƒý’gedýýš™™™™ÙC@ �Û@ à<À@eeqedcefffýmfgýf `RÑ@ °‰@defffeh›ý„eddýýeš™™™™™B@ à<À@ û©@dqcfcifffýhfgých @É´@ °‰@dddfkcyr}eedddddocecifggddfgýchd °‰@ddfkc•ýž33333³F@gfdýýeep ðÄí@ P[è@edceejeý Ð ã@pegýfdff °‰@ffe‚˜ý d33333óP@edýýhetg P[è@ P[è@dcifjeýp à-ß@fgýeddff °‰@fee”ý�de3333338@dýýefpfd à<à@ `t×@cifdfýpf @ÉÔ@gýfddfff °‰@eg”ý�fdeÍÌÌÌÌŒC@ýýefreec P[è@ `t×@edefýpfg à-ß@ýghdfffe °‰@i ý„feeý33333�S@ �Û@ýefqkeŒg �Û@dgeýpfgý `RÑ@ °‰@fhdfffgŒ’ýšfedýýfffffæM@  Øã@  Øã@edrcfcgfgfýpfgýc ÀgÙ@ °‰@hefffec”ý¬edfýdgš™™™™™<@ À ò@dtcfcffieýdfgxch €�ç@dddfkc´ý333333W@˜fedýýhd�$(¾1Ÿ�A @Bù@xcfceffeýpfgxchdŒJê„�u Afffec‹ý«ÍÌÌÌÌìP@fdeýdgdt�$(¾1Ÿ�AceŒeeieýŒJê„�u Adegxchdffegcdd}iddýdddpc  Øã@ecifgfýd ÀgÙ@fg{chddddkc™ý³gdÍÌÌÌÌŒD@dýdgetdd P[è@cefieýdf à-ß@g{fddddficªý°fdÆ `T@ýdgetdec �Û@edgeýdeg `RÑ@{dhdddfixŠý™eddý33333³9@  ØÓ@ddfoiecedeeýdfg{ ÀgÉ@jhffffei‹ýœeddýýš™™™™™8@ Ð#Ñ@ Ð#Ñ@efpdecgdeeýoeg{h àðÅ@hdddfedŒý~gedýýe33333³@@ à«Î@ à«Î@fpgecidjfýofg{ih  ¡Ã@dfffeh…ý €:@xddeýýed à«Î@ €¦µ@pceciffeýkfg{chf  ¡Ã@fffecˆý©fffffæB@fddýýhew €¦õ@  ØÓ@edceeheý ·ë@pfgnfddddeid¬ýŸfš™™™™YS@dfýýgeu‚ €¦õ@ À ò@ecgdieýp ·ë@egnghdddfei€ý´feš™™™™™5@dýdedrce P[è@ceeeeýdf à-ß@dncheddfic„ýƒfdeffffffB@ýdfetjfc �Ë@edieýdfg `RÁ@Žhhdddfex|ý“fddý33333³4@ à<à@ýedpcfce �Û@fgeýpegŽ @ÉÔ@chmfffecŒý¥dddýdÍÌÌÌÌŒF@ à<à@eeqddcfeeeýdegŽe @ÉÔ@ddddfic‘ý™eedýýefffff&C@ à<à@ `t×@eqfecedefýpfdŽeh @ÉÔ@dfffee’˜³gedddeepcfcfdjeddfgŽhhdddficˆý�ÍÌÌÌÌÌB@hdxddgdtcecefgedqegŽchdddeec‡}»fddýddelc  ØÓ@fcfdkeýd ÀgÉ@edýnhddd  d@fic‡|‡dddýýedncm P[è@ à«Î@ceefgýoe à-ß@gýcheddf  d@ec¡ý�dedš™™™™9T@ýýgdtcec à<à@ à<à@iffeýpfg @ÉÔ@ýchdfffe  d@c“ý�dddýffffffC@ à«Î@dedrcfceejeýÆfgz  ¡Ã@chefffec}ý–fedýdÍÌÌÌÌÌ7@ P[è@getddceegeýdfgze à-ß@ddddfkf™ý�eedýýeÍÌÌÌÌŒD@  Øã@ à«Î@drcfceeieýofgzch ÀgÙ@dffffcŠýffffff5@“jddddedpcfcefdfddegzchdddffc‡ý©33333sE@gddýdeep  ØÃ@dfcfdmeý Àg¹@dfggihdddfif‡ý”d33333óE@edýýdeod à<À@ û©@dcfefeýh @É´@fgghddddfhcdd®hedýdddncf  ØÓ@ceegeýdf ÀgÉ@ggchddddic}ý‹fed333333)@ýýdfnddc à<À@ Á¯@iegeýifg @É´@gfddddfl”ªý¨gdpý33333óT@ `t×@dgetdfcgdheýdegg @�Î@jhdddfik¥ýžfdlddÍÌÌÌÌLU@hdvcfceekedqeggchdddfec ý¬degýýe33333�S@ €¦õ@ ðÄí@esnfcidieýpefghh ·ë@dfffes•ýš™™™™™B@‘fddýded P[è@pcfceejeýdfggchd à-ß@ddfkcœý“fffffæJ@fedýýefp €¦µ@ €Ã²@decidhfý ·«@jfgýfhdf €f@ffee‘ýydÍÌÌÌÌ�O@eeddeeqfecidheddegýehdff €f@fee}ý”geÍÌÌÌÌÌ7@dýdefrfe  ØÓ@ceddfýdf ÀgÉ@fýthdddf €f@k‚©ý¶edeÍÌÌÌÌLG@ýdeepclc �w@gddfýdfg �m@ýÆhdddfi €f@c�ý˜gddýÍÌÌÌÌL7@  ØÃ@ýefqddcg à<À@fdfýmfgý Àg¹@ €f@jddddfeeŽƒ¢eedýý �Û@  ØÓ@delcdceeieýpegýh `RÑ@ €f@ddfffec™ý¼fddýdeš™™™™yR@ à<À@dpcecffieýdfgých @É´@ €f@dddfic†ýffffff;@•dddýddd 5¤@ncecffjeýdfgichd Þ™@ddflc‡„œdedýýeep P[è@ P[è@cfcgdgeý à-ß@pegighdfffec’ý¼g33333óC@edýdedpc à<À@fcfegeýd @É´@egichdddekc™ý©ed33333SQ@dýýedrcf À ò@ à<à@cffieýpf €�ç@gichdfffecŠ�™dedýýeepgdc à<à@ à<à@geieýpeg @ÉÔ@†gddfffelŽý¤iddýfffff¦A@ À ò@ýddncece P[è@eeeýpeg† €�ç@chdddfec€ý�hddýýš™™™™™5@ €¦µ@ Á¯@ddocfcgfeeýifg†c ·«@hdddfec“ý“fedýýeffffffC@ P[è@ à<à@eq‚ecedieýpfg†gh à-ß@dfffen’˜œdddýýed À ò@ à<à@pcfcffieýpfg†chd €�ç@ffeec�ý—ÍÌÌÌÌÌ4@eedýýeeo @Bù@ P[è@ddceejeý À*ð@peg†fddfffee�ý“f33333óC@edýýedpc  Øã@ à<à@lceejeýp ÀgÙ@eg†chdfffec”ý„de  Q@dýýeetfd  Øã@  ØÓ@cifeeýpf ÀgÙ@eýgddfff Ài@ex’ý¤gdifffff¦R@ýýgetgfc�$(¾1Ÿ�Aáî¬��^�AedieýpegŒJê„�u Aýhhdffee Ài@�¡ý�fdiýš™™™™9T@  Øã@ýhdxce‘h  Øã@fffýpfgý ÀgÙ@ Ài@cedddfgc|ýxdedýý33333³4@ �—@ �—@dfpfdcifkfýffgýp ��@ Ài@deddfjn‘ýŸdedýýe333333?@ `t×@ `t×@eqýdcifm 0�@eýpegýgd @�Î@ Ài@dfffeý¡ý 0�@ÍÌÌÌÌìQ@�gddýýgf P[è@ à<à@tfdcgfefýpfgýfdd à-ß@ Ài@fffef›ýzš™™™™™B@dddýýgft à<À@ à<À@kdcifdfý @É´@mfgýfddd Ài@dfe‚“ý‹h33333SP@edýýhfv– P[è@ P[è@dcgekfýp à-ß@egýfddff Ài@fex‡}‚eddýýeep}d  Øã@  Øã@cifefýpf ÀgÙ@gýeddfff Ài@e}”ý�ged3333338@ýýefredc �Û@  ØÓ@gfhfýpfg `RÑ@ýhddfffe Ài@fƒýxdddýš™™™™™3@ @Bù@defrhdcifhfýdegý À*ð@ Ài@jddddfjpddŠfedýd `t×@ddlcecgfdfýdfgýc @�Î@ Ài@hddddjc¡ý»fddýdhš™™™™9T@ à<à@evddcfedfýdfgýhd @ÉÔ@ Ài@dddfid…ý333333+@žhddýded  ØÃ@ocfcffefýdfgýchd Àg¹@ Ài@ddflc€ý†š™™™™™5@fdddýedp 5¤@cecefmfdgfgýchdf Ài@fffcuý§hfffff¦F@ddýdgetd à<à@dcefdfýd @ÉÔ@egýhdddd Ài@fik|ý gdš™™™™�1@dýddelcf û¹@cgdfeýdf ¡°@glihdddfkcˆý“ded €4@ýdeepfdc �Ë@gegeýdeg `RÁ@lpddddflýddzdedý 0�@ �—@defpdecideeýdffl ��@ihddddlfŸý˜eedýý àR@ @Bù@ P[è@hexndceejeýpeglf À*ð@ddfffex|ý‚eedýýe33333³4@ à<à@ à<à@dpcecifefýpfgých @ÉÔ@ `f@dfffecŠýÍÌÌÌÌ�J@¶eedýdee P[è@redceeieýdegýhdd à-ß@ `f@ddfid€ý¨š™™™™™5@gddýdeep à«Î@cdcgfgfý  ¡Ã@Æfgýdddd `f@dffc”ý’h3333338@ddýýeepd Ð#Ñ@ `tÇ@dcefdfýn àðÅ@egýjdddd `f@feeddzfddýdefpie  ØÓ@cidfeýdf ÀgÉ@gýghdddd `f@knƒý}hddš™™™™™3@ýýddncfc û©@ û©@ifefýhfg ¡ @ýchdddfh `f@cxý„dddý333333%@ �w@ýedpcfci �w@fefýeeeý �m@ `f@chdddfhc™ý fedýýÍÌÌÌÌŒM@�$(¾1Ÿ�Aáî¬��^�Agdvcecefeeýpeg„cŒJê„�u Ahdffeec©ý†fdhýýhfffffÆV@�$(¾1Ÿ�A ðÄí@exnfcedgeýpeg„ehŒJê„�u Adfffe}¤ý  T@§feªýdge ðÄí@ukfcedfeýdfg„hed Ð ã@ddfkr“–Œfedýýget €¦õ@ €¦õ@xdcgfieý ·ë@peg„eddfffek©ý¢ffffffÆV@edýýheti�$(¾1Ÿ�Aáî¬��^�AdcefdeýpŒJê„�u Aeg„eddfffee’˜¯hddýdedocf  ØÃ@cfeheýdf Àg¹@gjchdddelcƒý‚eddš™™™™™3@ýýefpidc  ØÓ@  ØÃ@efdeýmeg ÀgÉ@jhddffffi„ý„degýfffffæ:@ à<à@ýefrdeci à<à@digýpegj @ÉÔ@fhdfffeg£ý•fdfýdÍÌÌÌÌŒO@ `t×@getefcedfeýdegzf @�Î@hdddfid”ý€eddýýfš™™™™™<@ `t×@ Ð#Ñ@frddceeieýofgzfd @�Î@dfffee”ý33333sB@�fedýýff `t×@ `t×@v‚fchdffýpfgzfhd @�Î@fffe–Šý’ÍÌÌÌÌŒM@eedýýhfv ðÄí@ ðÄí@idcgfheý Ð ã@pegzeddfffek‘ýŽifffff&C@ddýýgftj ðÄí@ à<à@dcfegfýp Ð ã@fgzeddddfee‹ý£geÍÌÌÌÌìP@dýýedrcf Ð#Ñ@ à«Î@chfegýoe àðÅ@gýcedddf (‡@ic‡ýšdedfffffæ4@ýdeepcdc  ØÃ@ifeeýded Àg¹@ýhddddfh (‡@cŠý¨fddýš™™™™™C@  ØÓ@dedpcfcefjeýdfgý ÀgÉ@ (‡@chdddfic„ý°hedýdfffffæA@ à«Î@eepkfcedieýdfgýf  ¡Ã@ (‡@hdddfkf˜ý§hddýdefffff&N@ P[è@epeecfdfeýdfgýhh à-ß@ (‡@dddfip‡ýfffffæ4@�efdýýef û¹@ û¹@pcfchdegýlegýihf ¡°@ (‡@fffec–ý¤33333³L@edfýýedr P[è@ P[è@cecefeeý à-ß@pegýcedd (‡@dfec´ý®f333333W@ddýýhexk�$(¾1Ÿ�Aáî¬��^�AdceffgýpŒJê„�u Aegggdeddfipƒý�gdš™™™™ÙC@dýýgetgd�$(¾1Ÿ�A €¦õ@ceejeýpeŒJê„�u Aggeddddfhh¡ý©eegš™™™™9T@ýdhevffc À ò@edjeýdfg €�ç@gfhdddfii˜ý—gdfýfffffæQ@�$(¾1Ÿ�AdgdtcfcefieýdfggŒJê„�u Achdddfkc›ýºfdgdý33333³N@ 5¤@hevcfcedeedgfdgfhdddeic¤ýžgedýdfÍÌÌÌÌ�U@�$(¾1Ÿ�AesedcefjeýdfgggdŒJê„�u Adddfks—ýfffff&J@²fddýdhd�$(¾1Ÿ�AtcecffdeýdfggchdŒJê„�u Affffc‹ýžÍÌÌÌÌÌH@fedýýedp �Û@ �Û@clcggheý `RÑ@pfgfchdfffec‡„�gedýdfdrc `t×@eceegeýd @�Î@egfchdddfkc‡„–dddýdeeqde `tÇ@cidheýde @�¾@gfghdddfhf‹ý†gddÍÌÌÌÌ�C@ýýeepdec û©@ û©@hdfeýhff ¡ @fdhdddfhc“ý¦iddý33333SP@�$(¾1Ÿ�Aýhevedce €¦õ@efeýpegfŒJê„�u Agddddefd™ý˜dedýýÍÌÌÌÌŒM@ à<à@ à<à@edrcecifheýpegfc @ÉÔ@hdfffec|ý˜gedýýe33333³4@ `tÇ@ �—@fpdlcgdeeýfegfhh @�¾@dfffed›ýÍÌÌÌÌLS@œfdgýýhd P[è@ �Ë@wcfceeieýnfgochd à-ß@ddffcƒý}š™™™™ÙC@eedýýeer à<À@ à<À@ddciegeý @É´@mfgohddddfk|¬ý™eš™™™™YS@edýýhdxc�$(¾1Ÿ�Aáî¬��^�AecefdeýpŒJê„�u Aegochdfffec‘¢�ddeýýgetie À ò@ À ò@cidieýpe €�ç@goehdfffel„ý~dddfffffæ:@ýýefrddc �Û@ �Ë@iefeýneg `RÑ@ofddfefeiˆý hddýÍÌÌÌÌLF@ `t×@dedrcecgfjeýdfgƒ @�Î@cedddficˆý™dddýdš™™™™™8@ `tÇ@eepddciefeýdfeƒf @�¾@ddddeli”ý«gedýdeš™™™™™A@  Øã@dpcfcefeeýdegƒce ÀgÙ@dddficˆýš™™™™™8@´dddýdee �Û@pddceedeýdfgƒgdd `RÑ@ddfkd¡ý«š™™™™9T@ddeýdgft à«Î@dfciddeý  ¡Ã@dfgƒlhdddfid‡„‡fedýýgetd  ØÓ@  ØÓ@dchfieýp ÀgÉ@egƒfddddfhr‡„„eedýýgetfd À ò@  Øã@ceejeýpe €�ç@gƒfddfffe–“ýƒfedffffffC@ýýferŒec �Ë@ Á¯@edgeýifg `RÁ@gffdfffei‹ýµgdddÍÌÌÌÌ�C@ddemdfcedjeddeggihdddfic¤ý~ddhýýÍÌÌÌÌ�U@ À ò@  Øã@getdecedfeýpfgge €�ç@hdfffehƒýœfddýýeš™™™™ÙC@ €¦õ@  Øã@epfdceeheýpfgged ·ë@dffeef–ý @G@²fedýdee à«Î@pkfcgdfeýdfggghd  ¡Ã@ddfik©ý‘ÍÌÌÌÌLG@eddýdefq �Ë@‚dcefheý `RÁ@dfgggddddfkf£ý¥h33333ÓS@ddýdgetf €¦õ@dcgejeýd ·ë@egggddddeis˜ý²id33333óP@dýýgdtcf ðÄí@ à<à@cfedeýpf Ð ã@dgchdddfic…ý—gedš™™™™�G@ýýfeq‚dc ðÄí@ `t×@efifýpeg Ð ã@gfddfffed}ý{dddýÍÌÌÌÌL7@  Øã@ýedqceci 5¤@fgeýgegg ÀgÙ@chdddfecŠ�Œdddýý ðÄí@ �Û@eepefcedeeýpeggf Ð ã@hdddfee¡ý—depýýf33333³R@ ðÄí@ ðÄí@ergfcgdheýpeggih Ð ã@dfffeh­ý  U@³iedýdge �w@tidcegigýdegggdd �m@ddfii“ý¡ffffffC@ddfýýedr P[è@  ØÃ@cecefdeý à-ß@mfgtchdddffcƒý´fš™™™™ÙC@ddýdedpc `t×@fceedeýd @�Î@fgtchdddfkc‡ý§ee333333A@dýdeercd P[è@ceedeýde à-ß@gthddddfic¤ý¦gehÍÌÌÌÌ�U@ýdefscfc  ØÓ@gdfeýdfg ÀgÉ@thhdddfic‘¢šhdpý P[è@ýeesnecg `t×@dieýpfgt à-ß@fhdfefen�ý–fdfýdÍÌÌÌÌLK@�$(¾1Ÿ�AgdtceceedeýdfgncŒJê„�u AhdddfkcŒý�fddýýeÍÌÌÌÌL<@ €¦õ@ ðÄí@erddceeheýpegned ·ë@dfffen„ýfffffæ:@…fddýýef €¦õ@  ØÃ@pedcefeeýmfgnfdd ·ë@ffffsŽý£fffff¦A@eddýýeep  Øã@ à<à@ddceeeeý ÀgÙ@pegnjddddfecƒý�eš™™™™™3@edýýgeug P[è@  ØÃ@fcedeeým à-ß@fgnefdfffevˆý­edš™™™™™8@dýdeepdf  Øã@cedieýdf ÀgÙ@gnghdddeifyý¯hdd333333+@dddencecfdfedÆfgnehdffffcŸý§fedý33333�V@ ðÄí@dhdvcecefgeýdfgn Ð ã@chdddficŠý·hedddffffff<@edpcfcfeeeddegncedddficŒýšfedýýffffff&C@  Øã@ à<à@dtcfceffeýpegnch ÀgÙ@efffec‡„¤eddýdde  Øã@oddceeheýdegnfdd ÀgÙ@ddfil‰ýŽš™™™™™2@fedýýdfo �Û@ `t×@edchfieý `RÑ@pegnfddfffek›ý‘gš™™™™™B@edýýedsc à<à@ `tÇ@e„efieýn @ÉÔ@fgpchdfffgcŒý‡fe33333³@@eýdefpdf €¦õ@cedeeýdf ·ë@gpgfdddfke�ýŒged333333J@ýýeeridc�$(¾1Ÿ�A  Øã@eeieýpegŒJê„�u Apgddfffeidd–eedý à<à@deepedcgffeýqegp @ÉÔ@gddfffeg¡ý¹hegýdÍÌÌÌÌìQ@�$(¾1Ÿ�AgevgfcfdieýdfgpgŒJê„�u Ahdddfif•ý±hddýýe33333sL@ À ò@  Øã@eqcdceldeýpfgphd €�ç@dffefc™ýfffff&Q@�fddýýge P[è@ P[è@tsdcgfeeýpfgpfdd à-ß@ddfef‡„¼iedýdeeq �Û@ddceedeý `RÑ@degplddddfid˜�®ddxýýgete à<à@ à«Î@fcideeýo @ÉÔ@fgpghdffeff”ý…dd ÀI@dýýeephd�$(¾1Ÿ�A `t×@cefieýpfŒJê„�u Ag§fddddeeŒƒý…dedš™™™™ÙC@ýdedpcec �w@ifefýdeg �m@§chdddfkc‹ýˆfddýÍÌÌÌÌ�C@  Øã@ýefpedch �Ë@fgeýnfg§ ÀgÙ@fddfffex–¨�eedýý ðÄí@ à<à@gdtcecefdeýpfg§c Ð ã@gdfffecdd{dddddeepddcifgeddfg§fdddddlý®ý 0�@ U@…dedýýee P[è@  ØÓ@pddceeeeýpfg§fdd à-ß@fffelƒý�š™™™™™3@eddýdedp Ð#Ñ@cecefeeý àðÅ@dfg§chdddfkc‡„¢fddýdedpc �Ë@ecgeeeýd `RÁ@edjchdddfic‹�Žeddýýdfmed `tÇ@ `tÇ@cgfdeýne @�¾@gjgdefffee‡ý¤ged33333óE@ýdfdqcfc  Øã@egdfýdfg ÀgÙ@jcgdddficƒ‹�fddý À ò@dedpclceeieýdeg| €�ç@cgdddfic£ýœfddýý33333ÓS@�$(¾1Ÿ�Aáî¬��^�Ahewrdcefkeýpeg|eŒJê„�u Addddeeh…ý{defýýe33333sF@ ðÄí@ à«Î@friecideeýoeg|gh Ð ã@dfffe|ƒý33333³A@‚dedýýge  Øã@  ØÃ@undceejeýmeg|edd ÀgÙ@ffeehŸýˆ33333�V@dedýdhdw À ò@ceceffeý €�ç@dfg•chdddflcŸý„d àR@edýýhfxp à<à@ à<à@dcifeeýp @ÉÔ@fg•fddfffesdd}dedýdhftgd �w@ciffeýde �m@g•ededddjsƒý®fdlš™™™™ÙC@ddgdtclcefgeddfg•chdddfic ý®gedý33333�S@ à<à@dgetedcgfdeýdfg• @ÉÔ@fddddfih¡ýœeeŒýýš™™™™ÙT@�$(¾1Ÿ�A ðÄí@hexifcedjeýpeg•eŒJê„�u Ahdffeen„ý|dedýýefffffæ:@ à«Î@ €Ã²@fuldcifheýjeg•hd  ¡Ã@dddfj€dd|defýdgf �—@tiecidfeýdfg•fhd ��@dddjndd dedýdddn û¹@cecifgeý ¡°@ded•chddddic€ý‰dš™™™™™5@edýýefpn 5¤@ 5¤@ecidheýg Þ™@eeslfdddfhsŠ�“edeýýffrne @Bù@ @Bù@cgdieýpe À*ð@gsfhdfffefƒýxdddš™™™™™3@ýýedrcfc�$(¾1Ÿ�A Á¯@iejeýifgŒJê„�u AschdddfjcŠ�ƒeddý P[è@defpfdceegeýdegs à-ß@fddddfen|ý�hddýýffffffA@ `t×@ `t×@eerddcefheýpfgsg @�Î@ddfffeÈ’˜©dedýdg  ØÓ@etidcgffeýdfgshd ÀgÉ@dddfij�ýš™™™™�9@�iedýýff Á¯@ 5¤@vfdcgfeeýgfgghdd S¤@fffes›ý‹š™™™™™B@fedýýedp �Û@ �Û@cecgefeý `RÑ@pefgchdfffec”ýyd P@edýdeer�  Øã@dcieheýd ÀgÙ@fgghddfefeŒ„ýšeefffffæA@dýýfdtcf  Øã@ à<À@cefieýmf ÀgÙ@ggchdfffec¤ý…eedÍÌÌÌÌ�U@ýýgdtcec €¦õ@ P[è@eefeýpfg ·ë@gchdfffec‚ý eddý €4@ �Ë@ýedpcecg �Ë@feeýnfgg `RÁ@chdddfec¡ý�eddýýš™™™™9T@ �Û@ �Û@getkdcgfdeýpfggf `RÑ@ddddfek„ý¬gddýýefffffæA@ P[è@ Á¯@ergdcefdeýifg‰hd à-ß@dffffiddxeedýddg û©@mddcifdeýdfg‰ddd ¡ @dddki‡ý�33333³7@dddýýefp  ØÃ@ 5¤@dfcidieý Àg¹@geg‰dhdddfld¬ý´hš™™™™YS@edýdhewd à<à@dceejeýd @ÉÔ@eg‰gddddfic¢ýœedš™™™™ÙQ@dýýhevhd À ò@ à<à@ceeieýpf €�ç@g‰hddfffei„ý�fddfffffæA@ýýeeuedc à<à@ `tÇ@eeheýnfg @ÉÔ@‰gddfffeyƒýƒfddýš™™™™ÙC@ Á¯@ýfftddcg Á¯@fefýifm‰ S¤@|ddddfhŒdd•gedýd €¦µ@edncfcgfjeýdfg‰c ·«@hddddkcdd‰gedýdh ðÄí@etgdceffgýdfg‰gd Ð ã@edddkr�ý333333J@¸gddddedpcfceeieddem‰chdddfic”ý§33333sB@eddýdedq `t×@cfceeheý @�Î@dfg‰chdddfic”ýžf3333338@ddýdeepc�$(¾1Ÿ�AdceeheýdŒJê„�u Afg‰jddddfkc–ý™gdfffff&Q@dýýedpce @Bù@ `t×@ceeieýpe À*ð@f‰chdfffecšý‹gdf33333sO@ýýedscec ðÄí@ Ð#Ñ@efkeýoeg Ð ã@‰chdddfkcŠý½eedýfffff&H@  ØÓ@defqcfcfdeeýdeg‰ ÀgÉ@hhdddfic—ýšdedýýš™™™™ÙO@  Øã@ à<à@gftgdceedeýpfg‰e ÀgÙ@ddfffeeŠ�ƒgedýýe `tÇ@ `tÇ@fr|dcifeeýnff‰hd @�¾@dfffev™ýš™™™™yR@­fddýdee �—@qedcheieýdeg‰hde ��@ddfieŽý§š™™™™�;@eddýýeep  ØÓ@  ØÃ@ddceejeý ÀgÉ@meg‰hddddefc’˜“hddýýhevl�$(¾1Ÿ�A `t×@dceffgýpŒJê„�u AfglddeddfkxŠý²hdfffff&H@dýdeepdd Ð#Ñ@ceejeýde àðÅ@gliddddfic ý›edeš™™™™ùU@ýdgetifc�$(¾1Ÿ�AedjeýdfgŒJê„�u Allhdddfkx}ý�fddýÍÌÌÌÌÌ7@ à<à@ýeep}fcf à<à@dfeýpfgl @ÉÔ@ghdfefen„ý¼hddýdffffff*@ `tÇ@ddocfcffgeýdfglc @�¾@hdddficzý’hedýýeš™™™™™-@ À ò@ à<À@epkecgdgeýmfglff €�ç@dfffe~‡ý33333óE@³fedýddd  ØÓ@ncfceedeýdeglced ÀgÉ@ddficŠ��dedýýedp  ØÓ@  ØÓ@ceŠifieý ÀgÉ@pfglchdfffgc�ý¤gš™™™™�9@edýýeepd�$(¾1Ÿ�A P[è@dcfegeýpŒJê„�u Aeglhddfffeg‹�¡dedýýedpce�$(¾1Ÿ�Aáî¬��^�AceefeýpeŒJê„�u Aglchdffeec…ýŒgdd33333sF@ýdeepdlc €¦õ@edkeýdfg ·ë@gehdddfkn}ýwdedýÍÌÌÌÌÌ7@  Øã@ýdeoddci �—@ejeýfegg ÀgÙ@eddddfjj¡ý‹fddýýš™™™™9T@ À ò@ à<à@hevidceejeýpfggf €�ç@ddddfkk”ý�eddýýe ÀL@ P[è@ P[è@eqxdcgfjeýpfggfd à-ß@dfffef‰ýš™™™™™2@�dedýýge €¦õ@ à<à@tedceeheýpeggedd ·ë@fffei­ý�  U@gddýýget €¦õ@ À ò@gdceeieý ·ë@peggeddfffei–ý›ffffff&Q@edýýefqx @Bù@ @Bù@fcedjeýp À*ð@egggedfffenƒý’geÍÌÌÌÌÌB@dýýfeqpd €¦õ@ à<à@ceeieýpf ·ë@ggeddfffel’˜�fddýýfdrcec ðÄí@  ØÓ@eeieýpfg Ð ã@gchdddfec…ý‰fdfýÍÌÌÌÌLG@ �Û@ýeereecg �Û@dfeýpeg• `RÑ@dhdfffe‚•ý»gedýý33333sL@  Øã@ à<à@eeplecfdjeýpfg•g ÀgÙ@hdddeid}ýŽgedýýeÍÌÌÌÌL7@  Øã@  Øã@dpcfcggjgýpeg•ch ÀgÙ@dfffecŒý33333³@@‚feeýýef €¦õ@  ØÓ@siecgdgeýpfg•fhd ·ë@fffeh’ýœ33333³N@dddýýget P[è@ P[è@sdciffeý à-ß@pfg•gddfffef…ý‹f333333+@ddýýgetf P[è@ �—@dcefgeýf à-ß@fg•eddddfjnƒý„deš™™™™ÙC@eýýferdl  Øã@ €Ã²@cidheýje ÀgÙ@f•ehdddfjj¤ý–eddÍÌÌÌÌ�U@ýýgevddc  Øã@  Øã@gfgeýpfg ÀgÙ@•fddfffed‡ýšfddý33333óE@ û¹@ddfodecgdgeýdeg• ¡°@jhdddflk ý�dedýý33333SP@ À ò@  Øã@ferhfcedjeýpeg•e €�ç@edfffep•ý´iedddefffff†P@dpcfcffjeddfg•chdddeic‡ýÍÌÌÌÌÌ5@—fedýýeg €Ã²@ �w@pedcgegeýefg•fdd �¨@ffefk‡„«gedýddep à«Î@cdceeeeý  ¡Ã@ded•jddddfic ý‹hš™™™™ùU@ddýýeerr P[è@ P[è@dceeheýp à-ß@eg•eddfffep„ý¸gdfffffæA@dýdeepcd `t×@cffjeýdf @�Î@g•hddddfkc�ý�gddš™™™™�9@ýýedpcec ðÄí@  Øã@eegfýpeg Ð ã@•cgdfffec•ýªfeed33333sL@deerdfcfdieddeg•ghdddeig�ýžfedýý33333³;@ À ò@ �Û@eepdfceddeýpfg•i €�ç@hdfffef“ý¶hedýdeffffffC@ P[è@epefcedjeýdfd•gh à-ß@dddfif¬ýš™™™™YS@‰dedýýgd €¦õ@ À ò@tceceeheýpeg•chd ·ë@ffeec€ý‚š™™™™™2@eedýýefp Ð#Ñ@ Ð#Ñ@decgdegý àðÅ@oeg•dheefffh©ýžgÍÌÌÌÌLG@diýýhewn�$(¾1Ÿ�A À ò@eceddeýpŒJê„�u Afggehdddfejdd½fddýdedpcf à«Î@ceeeeýde  ¡Ã@dgchddddic„ý´dedfffffæA@ýdeepclc �Û@idieýdfd `RÑ@ghhdddficyr¦hddý Ð#Ñ@ýdfmefcf Á¯@dgeýifgg àðÅ@ghdffffe‡}‚geeýý P[è@ à<À@geueecedheýmfgge à-ß@hdfefef‘ý�fddýýefffff&C@ À ò@  Øã@dpcfcefieýpfggch €�ç@dfefec‘¢›gedýýhd�$(¾1Ÿ�A ðÄí@vcfceejeýpeggchdŒJê„�u Affeec‘¢¤gedýýgdt ðÄí@ P[è@cfceegeý Ð ã@peggchdfffec‡„ždedýýedqc�$(¾1Ÿ�A ðÄí@ecegheýpŒJê„�u AfggchdffeecŽý�gdÍÌÌÌÌÌC@dýýefpse €¦õ@ à<à@cedeeýpf ·ë@ggihdddfe”�rƒdddýýefpdfc  Øã@  Øã@idgeýpeg ÀgÙ@ghhdddfeg¦Àƒdedý�$(¾1Ÿ�Aýhdvcfce €¦õ@ejeýpfggŒJê„�u Achdfffec•ý¤fedýý33333sL@ €¦õ@ €¦õ@edpceceejeýpeggc ·ë@hdffeec¥ý�eddýýeÍÌÌÌÌLU@ �Û@ �Û@dtcfcefefýpegých `RÑ@ @c@eddfjc¤ýÍÌÌÌÌ�U@Ÿfedýýfe  Øã@  Øã@redcggheýpegýgdd ÀgÙ@ @c@fffee”ýŠ3333338@fedýýffu P[è@ Á¯@ddceffeý à-ß@ifgýfddd @c@dfkg}ý�dÍÌÌÌÌÌ7@ddýýeepe  Øã@ Ð#Ñ@dcefieýo ÀgÙ@egýhddff @c@fe–…r¯fedddedpcfceeeeddegýchdddf @c@ic„ý‘hedfffffæ:@ýýedrcec à<à@ à<à@ifheýpfg @ÉÔ@ýchdfffe @c@cƒý�gedýš™™™™™3@ `t×@ýedpcecg à<À@fheýmfgý @�Î@ @c@chdffffc€ý�ffdýýÍÌÌÌÌÌ2@ P[è@ `t×@dfjcdcifggýpfgýf à-ß@ @c@dffffecŒý³gedýdeÍÌÌÌÌŒF@  Øã@dpcfceeieýdegých ÀgÙ@ @c@dddfic|ýš™™™™�1@†dfdýýgf À ò@  ØÓ@sjfcidggýpegýfef €�ç@ @c@fffehŠý¹ÍÌÌÌÌ�J@lddýdeer à<à@edceefeý @ÉÔ@degýhddd @c@dfie´ý¦i333333W@edýýhexm�$(¾1Ÿ�Aáî¬��^�AdceegeýpŒJê„�u Aegýdddff @c@fejŸý~de33333SQ@dýýfdwce à<à@ à<à@ciffeýpf @ÉÔ@gýchdfff @c@ec€ý¦hddš™™™™™5@ddefÆddcefgeddfgýhddddfi @c@n“ý¥fddýffffffC@ à«Î@ýeep‚dcg à«Î@efeýofgã  ¡Ã@fddddfe}Šý¥fefýýš™™™™™C@  Øã@  Øã@eesffcgdjeýpfdãg ÀgÙ@edfffee‰ýœeedýýdš™™™™™@@ à<à@ �Û@fpndcieheýpegãfd @ÉÔ@dfffeg¢ý333333I@œdedýdef �w@pddcifjeýdegãudd �m@ddfid}ý†ÍÌÌÌÌL7@eddýdddo û¹@cfciffeý ¡°@degãchdddfhc}ý›dÍÌÌÌÌL7@ddýýefpe û¹@ û¹@dcifdfýl ¡°@eeãhddddfheŠý™fdš™™™™™C@fýdgevil ðÄí@cedjeýde Ð ã@gãehdfffesŸý¶gde àR@ddhewdecfdfeddegãfhdddfid|ý�gddý33333³4@ à<à@ýefpdece û¹@deeýlegã @ÉÔ@ehdfffef¡ý˜fedýý33333³R@  Øã@  Øã@gduclcgfgeýpegãc ÀgÙ@hdeffec}ý{dedýdeÍÌÌÌÌL7@ �Û@dqceciffeýdegãch `RÑ@dffffc ý33333SP@…fddýýee À ò@ à<à@pgdceejeýpegãfdd €�ç@ddfenŸýœš™™™™9P@fdnýýget €¦õ@ €¦õ@‚ecedkeý ·ë@pegãghdddfe}”ý€f3333338@edýýgdtc à«Î@ à«Î@ecifheýo  ¡Ã@egãchdfffec“ý°hdffffffC@dýdeepff  Øã@cedieýdf ÀgÙ@gggfdddfig}ý…eedÍÌÌÌÌL7@ýýedrcec P[è@  Øã@eeheýpeg à-ß@gchdfffec}ý¨fedýÍÌÌÌÌÌ7@ ðÄí@deepedceejeýdegg Ð ã@gddddfidƒý�fedýýÍÌÌÌÌÌB@ P[è@ `t×@edsceceefeýpeggc à-ß@edfffec”ý®iddýdd3333338@ ðÄí@dmcfceejeýdfggch Ð ã@dddfic ý33333�S@›eedýýgd P[è@ �Û@tceceefeýpfggchd à-ß@fffecƒý§š™™™™ÙC@fdfýýeds ðÄí@ à<à@cfceejeý Ð ã@peggcheddeic¤ýŸfÍÌÌÌÌ�U@diýýhdvc�$(¾1Ÿ�A P[è@fceefeýpŒJê„�u Afggchdfffec‹ý›gdÍÌÌÌÌ�C@dýýeepdd à<à@ �Ë@cifefýnf @ÉÔ@gƒdddddfec€ý‘gedš™™™™™5@ýddfnddc à<À@ifffýdfg @É´@ƒnddddflfŽýœdedýfffff¦A@ à<à@ýdfmceci à<à@difýpedƒ @ÉÔ@ihdfffecŠ��eedýý Ð#Ñ@ Ð#Ñ@efrkdcifjeýofgƒf àðÅ@ddffeesŠ�†gedýýd `t×@ à<À@docecefgeýmfgƒch @�Î@dfffec£ýÍÌÌÌÌŒO@¡dddýdee `t×@ufdcheffýdegƒddd @�Î@ddfiu™ý‚ÍÌÌÌÌŒD@dddýýgfu À ò@ à<à@edcefdfý €�ç@pfgƒfddddfesªý}d33333óT@deýýgftn �—@ �—@fcidigýf ��@egƒehfddfeg‹ýœeeš™™™™™8@dýdefphd `tÇ@ciedfýde @�¾@gƒjddeffee¥ýždedÍÌÌÌ̬U@ýýeepndc  Øã@  Øã@fejeýpeg ÀgÙ@ufddfffex}ýŠgedý333333)@ P[è@ýedpcfcg Á¯@egeýifgu à-ß@chdffefcƒý~eedýýš™™™™™3@ �—@ �—@edscf‚ifefýffguc ��@hdfffgcŠý«fddýdeš™™™™™C@ à<à@etcdcefieýdfdugd @ÉÔ@dddfic…ýÍÌÌÌÌL5@�iedýýdf à<à@ �Ë@ondcgfdfýnegÝhdd @ÉÔ@fffedŠ�‡dedýýget à«Î@ à«Î@fdcifefý  ¡Ã@ofgÝgddfffee™ýŠfÍÌÌÌÌŒD@diýýhdvc ðÄí@ à<à@fcefdfýp Ð ã@fgÝcedddfec…ý™ge333333+@dýýdfmce �Ë@ �Ë@cfdjfýnf `RÁ@dÝkhdfffec”ý€fedš™™™™�4@ýýefqifc `tÇ@ `tÇ@hdgfýnfg @�¾@Ýfhdfffen�ý¸dedý @J@ �—@deencdcffkeýdedÝ ��@fddddfic€ý–eddýý3333339@ à<À@ à<À@eepdfcgddgýmedli @É´@hfffefl…ý¥gedýýd333333+@ û©@ 5¤@djcfceedfýgfdlch ¡ @dfffec‘¢~gddýýed P[è@ `tÇ@pcecieheýneglchd à-ß@ddeec¡ý˜š™™™™9T@gddýýedp  Øã@ à«Î@cf‘eejeý ÀgÙ@ofglchdfffgc|ýŸd33333³4@edýdedpc û¹@ecifdfýd ¡°@eglchdddflcˆýšde33333³E@dýdedqcf €¦µ@cgffeýde ·«@glchdddfhc�ý�fddffffffL@ýýedocfc�$(¾1Ÿ�A À ò@eejeýpfgŒJê„�u Agchdddeecdd§lddý `tÇ@ddelcdcefgeýdfgg @�¾@ldddddlc”ý£hddýý3333338@ `tÇ@ `tÇ@efrgdcgefeýnfggf @�¾@ddffefh¡ý�dddýýgš™™™™9T@ �Û@ �Û@ftxdcifieýpfgggd `RÑ@dfffek¤ý  T@ gedýýhd À ò@ à<à@xcfceeieýpeggchd €�ç@fffec•ýx33333³F@dddýýedp ðÄí@ û©@cecieieý Ð ã@heggcldfffecƒý¥lš™™™™ÙC@ddýýedpc û©@ �w@fcgegeýe ¡ @fg�chdfffec”ý�gd ÀL@dýdeepee à<À@cedggýde @É´@g�lhdddfhi|ý™fdd33333³4@ýdedqcfc `tÇ@ifgeýdfd @�¾@�chdddfkc”ý§fddýš™™™™�:@ P[è@ýeepdfcg à«Î@ddeýofg� à-ß@fhdfffedˆý‚eefýýš™™™™™8@ `t×@ `t×@ffsdecgdieýpff�f @�Î@hdffee€”ýwdddýýe3333338@ €¦õ@ 5¤@eqddciefeýgfg�fd ·ë@dddfel ý33333SP@³fedýdee �Û@rcfcedieýdeg�jhd `RÑ@ddfic•ý…š™™™™™J@gedýdefq ðÄí@fdcgfeeý Ð ã@deg�eddfffed�ý¯gš™™™™™N@ddýýdepd �Ë@  ØÃ@dcffgeým `RÁ@eg�gdeddffcƒý³feš™™™™ÙC@dýddeocd à«Î@ceejeýdf  ¡Ã@g�iddddfic•ý„hedš™™™™™<@ýdddocec �Û@efheýdfg `RÑ@�chdddfkc®ýªgddý U@ �Û@dhevedcgefeýdfg� `RÑ@fddddfir‡|židdýý `tÇ@ `tÇ@dfocechdheýneg�l @�¾@hdffefcŠý­iedýddffffff<@ �—@elcfcgdggýdel�lh ��@dddficŠýffffff<@‘gedýýdd  Øã@  Øã@ocfcgffeýpeg�chd ÀgÙ@ffeec…ý…333333+@hddýdddo `tÇ@cfcefgeý @�¾@dfg�chdfffec‡„´gddýdddkc à«Î@fceffeýd  ¡Ã@eg�chdddfic€ý¤le3333339@dýdddicf  ØÓ@cefdgýde ÀgÉ@g�cheddfic”ýšfdd3333338@ýýeerjec `t×@ `t×@idgeýpfg @�Î@�fhdfffe€‹ý™gedýÍÌÌÌÌìP@ P[è@ýedrclcg P[è@eheýpeg{ à-ß@chdffeecddŽgddýd P[è@fdsclceegeýdfd{c à-ß@hfdddkczýƒiedýddš™™™™™-@ �w@emedcifgeýdfd{fd �m@dddfki–ý33333³L@šfddýýee €¦õ@ À ò@pªdceejeýpeg{gdd ·ë@ddfe‡ƒý²š™™™™™3@fedýddej  ØÓ@cfcfdgeý ÀgÉ@dfd{ghdddfkc‰ý€fÍÌÌÌÌL@@edýdedpc €Ã²@fcifggýd �¨@fg{chdddfkcƒý‰gffffffæ?@dýdgfsdd �Û@cifeeýdf `RÑ@d{fdfddfkf›ývdedš™™™™™B@ýdeeqedc  ØÓ@ifieýdef ÀgÉ@Èdddfffen¡ý‰fedýš™™™™9T@  Øã@ýgeuldci  Øã@fjeýpfgÈ ÀgÙ@eddfffen…ý�gddýýÍÌÌÌÌL5@ ðÄí@ ðÄí@delddcefggýpegÈe Ð ã@ddddfecddwdeddddendfcidggddfgÈihddddji•’‹gedýýee�$(¾1Ÿ�Aáî¬��^�ApldcefkgýpegÈgddŒJê„�u Afffeh}ý˜333333)@fddýýdeh  ØÓ@  ØÓ@cdcefggý ÀgÉ@pfdÈddfddfec}ýzfÍÌÌÌÌÌ7@ddýýddmc à<À@ �w@fcifegýe @É´@edÈchdddfhc‘¢‡gddýýedpce à<à@ Á¯@cieggýif @ÉÔ@gÈchdffffcŠ�}dedýýedncec @Bù@ à<à@efgeýpfe À*ð@Èchdfefec•ý—hddýš™™™™™J@ €¦õ@ýeepeece  Øã@dfeýpedÈ ·ë@nhdfffeddd‚hddýd Ð#Ñ@deokfciddeýdfgÈg àðÅ@hddddkx…ýªjddýdd333333+@ �w@ejcdchfgeýdfgÓgd �m@fddfkcƒýÍÌÌÌÌÌB@‚ffdýýef à<à@ û¹@pdfchdeeýlfgÓfhf @ÉÔ@fffedƒýŸš™™™™™3@hedýýedp à<À@ �—@cexgfdfý @É´@ffgÓchdfffgc}ý•d333333)@edýýdfnc 5¤@ 5¤@eciddfýg Þ™@edÓihdffefcŠ��ddfýdeerel�$(¾1Ÿ�AcedieýdfŒJê„�u AgÓfhdddeegdd‚eedýdgdtcec À ò@efieýqeg €�ç@Óchdfffec™ý¢eedý33333SQ@ €¦õ@ýdddcfce `t×@ejgýpfdÓ ·ë@cheffffc›ý‡fedýýš™™™™™B@ P[è@  ØÓ@fesndcefjeýpfdÓe à-ß@ddfffes‹ý›eejýýgÍÌÌÌÌìP@�$(¾1Ÿ�Aáî¬��^�AetifcedieýpegÓghŒJê„�u AdffeegŠ�±fedddfesdecfdieddfdÓfheddekc¡ý²š™™™™9T@gedýdhdv  Øã@cecffdeý ÀgÙ@dfmýchdd ��@dfic•ý½f33333sL@ddýdeerd 5¤@fcfdieýd Þ™@fdýehddd ��@ficžý�fdÍÌÌÌÌ,R@gýýhdvce  Øã@  Øã@cgfgeýpf ÀgÙ@gýchdfff ��@ec¢ý™fed333333I@ýýfdrcec À ò@  Øã@efieýpeg €�ç@ýchdfffe ��@c”ý®fdddš™™™™�4@dhevedcffjeddfgý ��@gddffefd”ý²ieddd ÀL@elrgecfdgeddegýf ��@hdddeig„ý’eddýdeš™™™™�C@�$(¾1Ÿ�AerfdceeeeýdfgýgdŒJê„�u A ��@dddfklƒýš™™™™ÙC@ eedýdgf û¹@rcecgdfeýdedýhhd ¡°@ ��@ddfic”ý¡š™™™™�4@gedýýefp Ð#Ñ@ Ð#Ñ@ndcgfigý àðÅ@ofgýidde ��@eeeh–ý”eÍÌÌÌÌL@@edýýefsd à<à@ à<à@dcffgeýp @ÉÔ@egýgdeff ��@feŒ‰ýydfš™™™™™2@ddýedrcf �—@cifdgdfegýchffff ��@fc¤ý±hedÍÌÌÌÌ�U@ýdgdtcec �Û@gffeýdfg `RÑ@ýchdddfi ��@c¡ý}dejý @N@ à<À@ýefsgeci à<À@dheýmfgý @É´@ ��@hhdddfjl•ýˆeddýý33333³F@ `t×@ `t×@efodecidgeýpegýg @�Î@ ��@hdfffeyŠ�„gedýýe à«Î@ �Ë@drcecgfgeýneeých  ¡Ã@ ��@dfffecddwdddýddg �w@lddciffeýdfeýedd �m@ ��@dddj€§ý�fffff†U@dedýýhfw P[è@ P[è@‚dcifeeý à-ß@pfgýfdde ��@efen}ý dÍÌÌÌÌÌ7@ddýýefsf Ð#Ñ@ Ð#Ñ@dcifjeýo àðÅ@egýndddd ��@fh‚‡ý†ddÍÌÌÌÌÌ5@dýdefpdd �w@ciffgýdf �m@gýhdddde ��@ef‘ý±dedfffff&C@ýdeeqkfc `t×@gdgeýdff @�Î@ýihdddfi ��@dšý�eddý33333sO@ à<à@ýhfvkdcg à<à@fdeýpfgý @ÉÔ@ ��@fddddfedˆý|dddýdÍÌÌÌÌÌB@ �Û@gfuidcifeeýdfgýg `RÑ@ ��@ddfffeŒ ýƒdedýýf33333SP@ P[è@ à<à@dsceceegeýpegýce à-ß@ ��@dfffec}ý333333)@£eddýded �—@pcfcgfeeýdfgýchd ��@ ��@ddfkc¡ý²š™™™™9T@jdjýdhet  Øã@kfcedieý ÀgÙ@dfg¼fedddfid¡ýºhš™™™™9T@edýdgexd  Øã@dcefgeýd ÀgÙ@fg¼gddddfictýydeš™™™™™*@dýdeepdl Á¯@cidieýde S¤@d¼ghdfffff˜ý‹fddfffffæQ@ýdget‚fc  Øã@edjeýdeg ÀgÙ@¼ghdfffee‹ý¡eddýÍÌÌÌÌÌ4@ à«Î@ýedpcfcf à«Î@fgeýofg¼  ¡Ã@chdfffec|ýªgedýd33333³4@ Á¯@ddmcfcgfegýdfg¼c S¤@hdddfkc¤ý�gdxýýh  T@ à<à@ à<à@ftgecgdfeýpfmjfe @ÉÔ@dfffei ýš™™™™ùU@�fedýýge À ò@ ðÄí@txdcgfigýpfgjhde €�ç@fffei€ý¡š™™™™™5@fdfddgetkecfdjeddfgjghdddfkk¡ýždš™™™™9T@ddýýhftx Ð#Ñ@ Ð#Ñ@dchffeýo àðÅ@egjgddffffn£ý†deÍÌÌÌÌìR@nýýhexdf @Bù@ ðÄí@cidjeýpf À*ð@gýfhdfff z@ee‡ý|ddg33333³7@ýýgdtcec ðÄí@ `tÇ@iekgýneg Ð ã@ýchdddff z@c—ý›dddý33333“S@ @Bù@ýeepxece €¦õ@dfeýpegý À*ð@ z@ghdddfex£ý~dddýýš™™™™yU@ �Û@ �Û@efpldcifefýpegýh `RÑ@ z@ddddfeª‹ý™gedýýeÍÌÌÌÌìP@ ðÄí@ Ð#Ñ@drcfceejeýofgých Ð ã@ z@dffeec«ýÍÌÌÌÌ,V@œgeiýýhe�$(¾1Ÿ�A ðÄí@xfecedfeýpfgýehdŒJê„�u A z@fffeh‹ý ÍÌÌÌÌìP@fddddeerddceffeddfgýeddf z@feee�ý…fš™™™™�B@ehýýgftf  Øã@  Øã@echdgeýp ÀgÙ@egýdhddd z@fhn�ýŠddÍÌÌÌÌLJ@dýýefrdd à<à@ à<à@cifgeýpf @ÉÔ@gýeddfff z@ee}ý¡gdd333333)@ýýedqcfc à<À@ à<À@ffefýmfg @É´@ýchdefff z@c|ý¥jddý33333³4@  ØÓ@ýeguedcg  ØÓ@egeýpegý ÀgÉ@ z@eddffeee‡ý”fddýýfffffæ4@ à<à@ à<à@eepddcgffeýpegýh @ÉÔ@ z@ddfffed�ý„dddýýd33333³<@ P[è@ P[è@fmndceehfýpegýid à-ß@ z@dfffes‹�{dddddednceciffedÆegýchd z@ddffc”ýŠš™™™™™<@hedýddfl `tÇ@ddcifgeý @�¾@dfgýgddf z@ffekŸý¡eÍÌÌÌÌlS@edýýferc  ØÓ@  ØÓ@dchfgeýp ÀgÉ@egýgddff z@efc}ý”hdÍÌÌÌÌL7@dýýefpdd  ØÓ@  ØÓ@cifdfýpe ÀgÉ@gýhddddf z@edˆý}fddfffffæB@ýýefpdec P[è@  ØÓ@hddeýpfg à-ß@ýehdffff z@}‹ý±hddýÍÌÌÌÌìP@ P[è@deesedcefheýdegý à-ß@ z@hddddfie€ý�eddýd3333339@  ØÃ@edqcfcefieýdegýc Àg¹@ z@edfffec�ý¤fddýýeÍÌÌÌÌLK@ P[è@ �Û@epsecedheýpfgýfh à-ß@ z@dfffep‡ýÍÌÌÌÌÌ5@£kedýded �Û@pcfceeieýdfgýchd `RÑ@ z@ddekc”ý{ P@dddýýefq �—@ �—@ldcifheý ��@ffgýiddd z@dfeŒ�ý§i33333sC@ddýýedpc  ØÓ@  ØÓ@ecffdeýp ÀgÉ@fgýchdff z@ffcŠý€hdffffff<@ddddfoddcefgedqeglgddddfefddzeddýdedqcec Á¯@ifheýdeg S¤@lchddddkcdd�hfdý P[è@dddjcfcefegýdfgl à-ß@chfdddkc}ý‚gddýd333333)@ €¦µ@deoefcedgeýdfglf ·«@hfffefe–ý�gedýýe @G@ ðÄí@  Øã@dpcecefeeýpeglch Ð ã@dfffec‘ý333333?@‹fddýýed à«Î@ à«Î@rcechfgeýoeglchf  ¡Ã@ffeec“ýœš™™™™YA@fedýýdfm P[è@ à<à@cfcedeeý à-ß@peglfhefffec|ýºk33333³4@ddýddeoc Ð#Ñ@fceddgýd àðÅ@fglggdddfic�r�hfdýýddjce `tÇ@ `tÇ@cefggýne @�¾@glchffffecŒýªked33333³H@ýdedpclc �Ë@feegýdfg `RÁ@lchdddfic•ýŸfddý33333sL@ P[è@deeredcefgeýdedý à-ß@ ð~@ideddelp¡ýÇgdfddš™™™™9T@gdtcfceeeeddfgýc ð~@hdddfic—ý§gedýdeÍÌÌÌÌŒD@ €¦õ@drcfceeheýdfgých ·ë@ ð~@dddfic„ýfffffæ:@�dedýýff @Bù@ `t×@sedcifdeýpegýfdd À*ð@ ð~@ffffn•ý¢š™™™™™<@dedýýgdt @Bù@ À ò@cecefeeý À*ð@pegýchdd ð~@dficŠ��fegýdeese P[è@fcedjeýd à-ß@fgýefddd ð~@fkx„ý½idfffffæA@dýdeepcf `tÇ@cfdeeýdf @�¾@gýlhdddf ð~@ic ý‡gdd33333�S@ýýeerfdc P[è@ Ð#Ñ@eejeýofg à-ß@ýeddfffe ð~@h€ýŽhedýš™™™™™5@ à«Î@ýefrhdci à«Î@fdfýofgý  ¡Ã@ ð~@eddeffej‡ý‰eedýý33333óE@ à<à@ à<à@edqcfcefggýpegýc @ÉÔ@ ð~@hefffeczýˆmmdddeš™™™™™-@fpýdcefg 8�@eddfdýed ð~@mmmffýdd 8�@‡fddddedpcecigkgddegýchd ð~@dddkc—ý~š™™™™ÙO@ddnýýhdv €¦µ@ €¦µ@cecgffgý ·«@keeýchfd ð~@dffc|ý°d33333³4@edýddfhc 5¤@dcifdgýd Þ™@egýhdddd ð~@fic‹ý…defffff&N@dýýefrne à<à@ à<à@cideeýpe @ÉÔ@gýhhdfff ð~@e}€ý¦hfdš™™™™™5@ýdddlcfc û¹@gfgeýdfg ¡°@ýchfddfi ð~@cdd‰kddý �—@defpddchfggýdfgý ��@ ð~@|dddddkidd§eddýd à<À@fftcdcffefýdegýh @É´@ ð~@dddddic€ý|eedýdd3333339@ €Ã²@fondcifgeýdegývd �¨@ ð~@dddfhg�ý @J@™jddýýdf Ð#Ñ@ Ð#Ñ@odlcgdggýofgýhhd àðÅ@ ð~@fffeeœý‚ÍÌÌÌÌLK@fddýýgev ðÄí@ €Ã²@hdceegeý Ð ã@jfgigddddffgdd�hfdýddejc à<à@dceffeýd @ÉÔ@fdiedfdddkc€ý�heš™™™™™5@dýdefqcd �Ë@cgfegýde `RÁ@gihdeddflc›ý�hdhš™™™™™B@ýdeepdec `t×@gdgeýdfg @�Î@iihdddfhf�ý·hddýffffffL@ Ð#Ñ@deercfcfdeeýdfgi àðÅ@hhdddficƒý°dddýdš™™™™™3@  Øã@demcecedgeýdfgil ÀgÙ@hdddficdd¨hddýde  ØÃ@emdecgdgeýdfgifh Àg¹@ddddii„ýš™™™™�1@�eedýdfe à<À@tcechdgeýdfgijhd @É´@ddfkc–ý�fffff&C@iddýýddo `t×@ `t×@cfcgfgeý @�Î@pfgicedfffec™ý�fÍÌÌÌÌŒM@edýýedqc  ØÓ@  ØÓ@eceeeeýp ÀgÉ@egicheffeec”ý›ie3333338@dýdddocf �w@cgfieýdf �m@gochdddfic™ý¤eedš™™™™yR@ýdeeqfec à<à@gddeýdfg @ÉÔ@ojedddfin®ý‘eddý U@ À ò@ýhexidce ðÄí@efeýpego €�ç@eddddfel¥ýšdddýdÍÌÌÌÌLU@ �Û@geundcefdeýdeeog `RÑ@ddddfknŠý‚edgýýgš™™™™™C@ Ð#Ñ@ û¹@ftdecedeeýlfgohh àðÅ@dddfj}Œýfffff&C@¼fddýddd �Û@mceceggeýdfgochd `RÑ@ddfic‡ý£33333óE@gddddeemkdceffeddfgogdfddfkgƒýšdÍÌÌÌÌÌB@ddýýeepe  Øã@  Øã@dcfgheýp ÀgÙ@efohddddfe–ŽýÇdefffff&I@dddgdtclcefggddfmºchdffefc‘ý—gedfffff&C@ýýeerndc €¦õ@ `t×@eefeýpfg ·ë@ºeddfffe “ý�gdiýfffffæL@ €¦õ@dgetllcedieýdfgº ·ë@gfdddfkr˜ý˜fdhýý33333óP@ À ò@ ðÄí@eesnecedfeýpegºf €�ç@hdddfes�ý�dejýýg333333=@ €Ã²@ 5¤@fxgecideeýgffºih �¨@dddfheƒ{”eddýýef `t×@ `tÇ@pddceeieýnegºgdd @�Î@fffex›ýªÍÌÌÌÌLO@gedýdedr�$(¾1Ÿ�AcecfeheýŒJê„�u Adfgºchdddfic•ý¢f33333sL@deýýedrc�$(¾1Ÿ�Aáî¬��^�AeceeheýpŒJê„�u Afdºchdffeec¤ýƒddÍÌÌÌÌ�U@jýýfeugf À ò@ Á¯@cidieýif €�ç@gºfhdfffef›±šfdhddgduceceeieddfgºchdddfkc‘ý‹fddýÍÌÌÌÌ�O@ À ò@ýgetsdce  Øã@gjeýpfgº €�ç@dddffefn’ý¼iddýd33333óI@ à<à@eeqefcedgeýdegºg @ÉÔ@hdddfir™ýˆddfýýgš™™™™yR@ ðÄí@ ðÄí@fv‚eceddeýpegºhh Ð ã@dfffes‹ýÍÌÌÌÌìP@¥gddýýee�$(¾1Ÿ�Aáî¬��^�ApndceggeýpegºfddŒJê„�u Affeef“ý�ffffffC@fedýdeep  Øã@elcedieý ÀgÙ@dfgºffdddfkf…ý…d333333+@dgýýfdrc à<à@ Ð#Ñ@ecifgeýo @ÉÔ@fggchdddfecƒý|deš™™™™ÙC@dýdgdtce `t×@cieggýde @�Î@ggchdddfjc�ý|ddf333333;@ýýedqcec  ØÃ@  ØÃ@iejeýmeg Àg¹@gchdddfjcyýxdddý333333+@ û©@ýffsfdci û©@ffeýhfgg ¡ @eddddfjr¬ý¥fedýýš™™™™YS@ P[è@ P[è@gdtclceejeýpeggc à-ß@hdffeecˆýžiedýdd €4@ à<À@dlcfciejeýdeggch @É´@dddehc’ý33333óC@³fedýýee À ò@ à«Î@sddceeieýoeggedd €�ç@ffefc�ý³33333³<@hddýdefp `tÇ@cdcfeieý @�¾@deg{gddddfic•ýŒe33333sL@e‚ýýgeth à<à@ à<à@ecfdgeýp @ÉÔ@eg{fhdfffen›ý�fdš™™™™™B@dýýfer‚d  Øã@ �Û@chfgeýpf ÀgÙ@g{gddffffn‡ý®ied33333ó@@ddedpcecegjeddeg{chdddficŠý¨fedý33333³9@ Ð#Ñ@ýegpddcg Ð#Ñ@eheýofd{ àðÅ@fddffffl„ý�dddddfffffæA@eepddcifgedqfd{edefffec“ý¬edgýýf33333SP@ à<à@ à<à@ewdlcgdgeýpfgýfh @ÉÔ@ j@dddeeg“ý33333SP@“fddýýge  Øã@  Øã@tfdcgeheýpegýfdd ÀgÙ@ j@fffejƒý‹š™™™™ÙC@ddfýýeer  ØÓ@  ØÓ@eecgdgeý ÀgÉ@pegýdhdf j@ffe ˜ýˆfš™™™™™H@edýýfete  ØÃ@ à<À@dciffgým Àg¹@fgýeddff j@feh�ý…fdÍÌÌÌÌLK@dýýeernd P[è@  Øã@cgegeýpe à-ß@gýfddfff j@eŒ’˜{dedýýfdscec `tÇ@ `tÇ@iffgýneg @�¾@ýchdddfj j@cŒýŒdedý33333³@@ `t×@ýgetxdcg `t×@fdeýpfgý @�Î@ j@fddfffek¥ý�dddýýÍÌÌÌ̬U@ à<à@ à<À@gfvndcefdeýmegýg @ÉÔ@ j@ddddfk}®ý‰eedýýh U@  Øã@  Øã@dxcecefeeýpegých ÀgÙ@ j@dfffecƒýš™™™™ÙC@~dddýýff Ð#Ñ@  ØÃ@rfdciefeýmegýedd àðÅ@ j@ddfjp•’§lddddeevedceggeddedýeddd j@deid”ý†e3333338@edýýefse Ð#Ñ@ à<À@dcifieým àðÅ@fgýfddff j@fe ƒý�eeš™™™™™3@dýýedpcf `t×@ `t×@cifheýpe @�Î@gýchdfff j@ec}ý‘hed333333)@ýýdfoflc �—@ �—@edgeýfff ��@ýhhdddek j@i­ý„edpý33333sU@ ðÄí@ýgdtcece à<à@eeeýpege Ð ã@chdfffec‰ý„dddýýš™™™™™@@ �Û@ �Û@efrodcifkgýpegee `RÑ@ddfffei€ý|dedýdfš™™™™™5@ �—@frfdcifggýdegedd ��@dfffeeƒýÍÌÌÌÌÌB@�dedýdgf  ØÃ@tŒdcifgeýdegeedd Àg¹@fffee�ý› @J@hddýýefr À ò@ ðÄí@idceefeý €�ç@pegefddfffen‹ý‰dÍÌÌÌÌìP@edýdgdtc �Û@ecifeeýd `RÑ@egechdfffec”ýªee ÀI@dýdeerdd `t×@chedeýde @�Î@gegddddfic…ý™ded333333+@ýýefqcec à<À@ à<À@ideeýmeg @É´@ehhdddfhc–ýŸdddýfffff&Q@ à«Î@defpkfcfdkeýdege  ¡Ã@lhdddfik„ý£dddýýfffffæA@ P[è@ à<À@edpcfceeieýmfgec à-ß@fdddfic–ý™fddýýe @G@ ðÄí@ ðÄí@epdlcgdeeýpegefh Ð ã@dfffed‡ý33333óE@ydeddýed €¦µ@pclcilggdkegecedfffec­ý¡  U@dedýýget À ò@ À ò@‚dcieheý €�ç@pedeiddffee´‡ý«ffffffæ4@edýdeepc �w@lcfdeeýd �m@egehheddficdd¥dedýdedpce à<À@cifgeýde @É´@gechddddlc‡„•dedýdegrcdc P[è@eefeýdfg à-ß@edddddekc‚ý¥dddýš™™™™�4@  ØÃ@dedpcecffgeýdege Àg¹@chdddfic‘ý„dedýdÍÌÌÌÌ�O@ �Û@gftgdceeheýdfgkg `RÑ@ddfffftddšfddýde ðÄí@dpcfceejeýqegkch Ð ã@dffeec™ýÍÌÌÌÌŒD@«hddýdhe À ò@vjdcfekeýdfgkfdd €�ç@ddfig}ýŠÍÌÌÌÌL7@dfdýddfk �w@dfcidhgý �m@degkhhfddfhd}ý¼i333333)@edýddegc ðÄí@dceedeýd Ð ã@fdkjddddeic‡„°gddýdeencd à<à@cfefeýde @ÉÔ@gklddddfic|ý„gddš™™™™�1@ýdegqdec �Û@edhgýdeg `RÑ@khhdeffe‘dd~fddý `t×@dddocecefieýdfdk @�Î@chddddkc®ý†edfýý U@�$(¾1Ÿ�A €¦õ@hexefcedeeýpfgkfŒJê„�u AgdfefesŽƒ fedýýe À ò@  Øã@gpcfcedfeýpegkih €�ç@dffefcdd™hedýdef  ØÃ@mgdcefgeýqeg~gdd Àg¹@ddfhd”ýˆ3333338@gegýdefm €Ã²@dfcgdgeý �¨@dfg~dhdddfks�ý}eš™™™™™N@edýýedqc Ð#Ñ@ à«Î@fceeieýo àðÅ@fg~chdffefc®ý™fe U@dýýhdvce  ØÓ@ à<À@cefjeýmf ÀgÉ@g~chdffeec…ý†gfd333333+@ýddfmdfc �Ë@edgeýdfm `RÁ@~ghfddfkfyrœfedý  Øã@ýefoddce Ð#Ñ@eieýofg~ ÀgÙ@jddfffee‡}‡dddýý  ØÃ@ 5¤@efpcecfdegýgfg~g Àg¹@hdfffec¢ý‚dedýýf333333I@  ØÓ@  ØÓ@drclcefgfýpeg~ch ÀgÉ@efffec}ýÍÌÌÌÌÌ7@“iedýýed Ð#Ñ@ �Ë@oceceffeýnfg~chd àðÅ@fefec�ý~33333³<@dddýýefo `t×@ û¹@|fcidfeý @�Î@leg~fhdffeeg¢ý…d333333I@edýýefpd `t×@ `t×@ecedgeýp @�Î@eg~ehdfffecœý„deÍÌÌÌÌÌJ@dýýedpce �Û@ `t×@cgeheýpf `RÑ@g~chdfffec®ýŽfed U@ýýhfxndc €¦õ@ ðÄí@efeeýpfg ·ë@kdddfffen�ý�edjý333333B@ à<à@dfdsceceeieýdfgk @ÉÔ@chdddfjc¢ý~fddýý333333I@ `t×@ `t×@edmcfceffeýpegkc @�Î@hdfefecªý�ffdýdhfffff†U@ ðÄí@fvefcedhgýdfdkeh Ð ã@fddfkf¢ý333333I@€fddýýed à<à@ à<à@tceceejeýpegkchd @ÉÔ@ddfec¢ý�333333I@eedýýffs  Øã@  Øã@fdcifjeý ÀgÙ@pfgkfddfffex’ýˆg33333³N@ddýdefpe  Øã@dceefeýd ÀgÙ@fgkfddddfke¢ý”ge333333I@dýýgfv–d €¦õ@ �Û@cefdfýpe ·ë@gkjddfffek¢ý€feˆ333333I@ýýgftjec ðÄí@ ðÄí@edigýpeg Ð ã@kehefffexžý‰fdiý `R@ P[è@ýgftnlce P[è@dheýpegk à-ß@fhdfffe‡}ý¡fdddd333333)@eepedcgfgedÆfg~fddfffff”ýŒgedýýe3333338@ P[è@ P[è@dsceceegeýpeg~ch à-ß@dfefecŠ�|dddýýef �Û@ û©@qhdceedeýhfg~edd `RÑ@fefes|ýy33333³4@dddýddfp  ØÃ@edcifefý Àg¹@deg~hddddfhf|ýžd33333³4@edýýeepc  Øã@ �—@lcideeýf ÀgÙ@ed~ihdffffcŒý“eeÍÌÌÌÌŒF@dýýferfd à<à@ à<à@cieheýpe @ÉÔ@g~idefffeeddƒdddýdeeqgdc Ð#Ñ@efieýqfg àðÅ@~gddfffekŸý�gddýfffffFP@ €¦õ@ýedpcece €¦õ@eheýpegp ·ë@chdddfhc ý�dedýý33333SP@ ðÄí@  Øã@getedceeheýpedpe Ð ã@ddfffer¡ý¥ddjýdg33333³R@ à<à@etdecidfeýdegplh @ÉÔ@dddfid”ý ÀL@Žgedýýee ðÄí@ €¦µ@qedceffeýkfgphdd Ð ã@ddekgŠýƒÍÌÌÌÌŒM@ddiýýhev €¦õ@ ðÄí@xfcedgeý ·ë@pfgpehdfffes¬ý¦dš™™™™YS@ddýýhexg  Øã@ à<à@dceeheýp ÀgÙ@egpeddddfes§ý“feffffffR@dýýeerfd  Øã@ �w@cefgeýef ÀgÙ@gpfddffefmƒý•fddš™™™™ÙC@ýýdepddc `t×@ `tÇ@eggeýnfm @�Î@ghddfffekŽƒ˜fedý `t×@dddocecefeeýdegg @�Î@chdddflc�ý³fedýdš™™™™�6@ Ð#Ñ@edpcecfeieýdedgc àðÅ@hdddfic“ý¤fdsýdfffffffC@ ðÄí@dscfceeheýdfggch Ð ã@dddfkcŠ�Žhedýddd  ØÃ@oclcefgeýdfggchd Àg¹@ddfkc‡„ždddýýeem  Øã@  Øã@ddceegeý ÀgÙ@pegggddddfei‡„Ÿgddýýeepe À ò@ À ò@dceeheýp €�ç@eggfddddekf—ý›ke33333“S@dýýeeped Ð#Ñ@ û¹@ceegeýlf àðÅ@ggjddffefx€ý‡iedš™™™™™5@ýddenddc  ØÓ@gffeýdfg ÀgÉ@ggddfffeg�ý�eddýš™™™™�B@ ðÄí@ýeepheci �w@dgeýeegz Ð ã@ihdffeen‹ý�fedýdš™™™™™8@ `tÇ@eepnecedjeýdegzf @�¾@hdfffedyr}dedýýe �Û@ €Ã²@dpcecieieýjegzch `RÑ@dddfhc}ýÍÌÌÌÌL7@žgfdýýdd P[è@ à«Î@lcfcefeeýoegzchf à-ß@fffec™ý¢33333SQ@fegýdget À ò@iecedgeý €�ç@degzihdddfis•ý”gš™™™™YE@edýýgdtc�$(¾1Ÿ�Aáî¬��^�AfcefieýpŒJê„�u Aegzchdfffec™ý¤fdfffff&Q@pýdgdtcf À ò@ceegeýde €�ç@gzchdddfic‹�•fegýýeerdec À ò@ P[è@edheýpeg €�ç@zgfdeefedŸýŠfedý àR@�$(¾1Ÿ�Aýgdtcece  Øã@fheýpegzŒJê„�u Achdfffec¡ý…ddxýd @N@ ðÄí@gevgeceddeýdfgšf Ð ã@fdddfjl¡ýžfdgýýgš™™™™9T@ À ò@ à<à@etefcedjeýpfgšge €�ç@dfffen•ýš™™™™™<@�gedýýfd P[è@  ØÃ@tcfcefjeýmfgšchd à-ß@ffffc¢ýª333333I@eedýýhex P[è@ P[è@edcgfheý à-ß@pfgšfddfffer‡„œfefýýeeqd P[è@ à<à@ecedjeýp à-ß@egšhhdfffen‡ý¢he333333A@dýdedpcf  ØÃ@cgeieýde Àg¹@gechdddfic¢ý¢fed333333I@ýýgetedc À ò@ ðÄí@eedeýpeg €�ç@enddddfid‡ý£iedý33333óE@  ØÓ@ddepcdceedeýdege ÀgÉ@ededdfic—ý‚edeýýÍÌÌÌÌ�M@ à<à@  ØÓ@eerefcedeeýpegee @ÉÔ@hdfffe‡ƒý’fddýýdš™™™™™3@ 5¤@ 5¤@epefcedeeýgffgeh Þ™@dfffed‰ýš™™™™™@@–eedýdee  ØÃ@pcdcfeifýdfgggdd Àg¹@ddfkc¦ý‚ÍÌÌÌÌLO@fedýýedp à<à@ à<à@cfceeieý @ÉÔ@peggchdfffec‹�¨gddýddeoc à«Î@ecfdfeýd  ¡Ã@fggfhdddeic™ý fd33333SQ@dýýeepxd €¦õ@  Øã@ceeheýpf ·ë@gggddfffen‡ý�gddÍÌÌÌÌÌ5@ýdeepedc  Øã@eedeýdfg ÀgÙ@ghddddfki ý‰fedý33333�S@  Øã@ddeodecedheýdfgg ÀgÙ@ghdddfkf”ý“gddýý3333338@ P[è@ à«Î@edqceceedfýofggc à-ß@hdddfec’ý—iedýýe33333óC@�$(¾1Ÿ�Aáî¬��^�AepddªeejeýpegghdŒJê„�u Adffegn¡ýš™™™™9T@‹efdýýgf �Û@ à<À@unfcidggýmfgiehf `RÑ@fffeg�rªhedýddeo €¦µ@dfcfdeeý ·«@dfdijhdddficƒý~gš™™™™ÙC@edýýedpc à<à@ à<à@echegeýp @ÉÔ@fgichdffffc¤ýžfeÍÌÌÌÌ�U@dýýeeqnf P[è@ P[è@cedeeýpf à-ß@gighdfffen‡„žfddýdeeocdc û¹@gfjeýdeg ¡°@ihddddflcªý¢fedý33333óT@�$(¾1Ÿ�Aýhdxcecgáî¬��^�AekeýpegiŒJê„�u Achdfffec™ý�edfýý33333SQ@ À ò@ P[è@getnecedfeýpegkf €�ç@hdddfex”ý”hedýýf3333338@ `t×@ à«Î@drcfcefgeýofgkch @�Î@dfffecdd…dedýdfd  Øã@scfcefieýdfgkchd ÀgÙ@dddkc–¨—gedýýfet  Øã@ à<à@eeŒgdgeý ÀgÙ@pegkhhdfffgn¦ý¢gÍÌÌÌÌLO@dnýýhdtc  Øã@ à<à@fceeheýp ÀgÙ@egkchdddlec‡ýŒfdffffff@@dýýefpgf Ð#Ñ@ Ð#Ñ@cidfeýoe àðÅ@gkhhdfffeguý�fdifffff¦F@ýýeesffc  Øã@  ØÓ@edeeýpfg ÀgÙ@kehdfffep’ý¥eddý33333³N@ à<à@deerhfcgdheýÆfgk @ÉÔ@dhdddeel¤ý‹geiýýÍÌÌÌÌ�U@ À ò@ P[è@fesdfcedjeýpfged €�ç@hdfffef¤ý•fddýdgÍÌÌÌÌ�U@ Á¯@etcdcfedeýdfgedd S¤@dddfkc¬ýš™™™™YS@°gedýdhe ðÄí@vfdceejeýdegefdd Ð ã@ddfin“ýª33333SP@fedýdefp  Øã@cdceeeeý ÀgÙ@dfgedddddficƒý™gš™™™™™3@edýddeof  Øã@dceegeýd ÀgÙ@fdedddddfkd’ý›ee33333óC@dddeepdecgdkedqfgefhdmmeee}ýŸddd333333)@ýýeepedc  ØÓ@ û¹@eedeýlfg ÀgÉ@dfddffefgŠ�­hedý à<à@dednceceefeýdegd @ÉÔ@cedddfic€ý²fddýdÍÌÌÌÌÌ2@ `tÇ@edpcfceedeýdfdec @�¾@hdddfkcdd„gedýde  Øã@fpdfcedgeýdfgeeh ÀgÙ@ddddkv©ýÍÌÌÌÌLG@�gddýýge `t×@ `t×@tddcgfdeýpfgeddd @�Î@ddfed‘ý¢ÍÌÌÌÌLG@fddýýedp P[è@ P[è@cfceeieý à-ß@pegechdddfec›ýžgš™™™™™B@edýýeepe P[è@ `t×@fcedgeýp à-ß@fdefhdfffee•ý‹gdš™™™™™J@dýýeepff À ò@ À ò@cedieýpe €�ç@geehdddfeg}ý�gddÍÌÌÌÌL7@ýdeepeec û¹@gdgeýdeg ¡°@edhdddfke¢ýzdedý333333I@ �Û@ýeeqfdce �Û@ekeýpegh `RÑ@gddfffep£ý…dedýýÍÌÌÌÌŒO@  Øã@  Øã@edpcecififýpeghc ÀgÙ@hdfffec|ý�dedýýgfffffæ>@ Ð#Ñ@ Ð#Ñ@ftedcifeeýoeghjd àðÅ@dffffŒ¡ýš™™™™9T@’edgýýgf à<à@ à<à@thfciddeýpfgheed @ÉÔ@fffeh‡„¢dddýýgdt `t×@ `t×@cfcieheý @�Î@peghchdddfec”ý€fÍÌÌÌÌŒC@ddýýefnp à<à@ à<à@dcifefýp @ÉÔ@edheddddfeh|ý¯hd33333³4@dýddeldd à«Î@ceeeeýde  ¡Ã@ghhddddfie–ý�iedfffff&C@ýýdfprdc P[è@ �—@efieýffg à-ß@rhddffffrˆýƒediýÍÌÌÌÌÌB@  Øã@ýfdtcece  Øã@eheýpegr ÀgÙ@chdddfkc}ý„dedýdÍÌÌÌÌÌ7@ Ð#Ñ@getddcifefýdefrf àðÅ@ddffffsŽý{dddýýefffff¦A@ @Bù@  ØÓ@dpcfcigieýpegrch À*ð@dddfhc“ýffffffC@}fedýded�$(¾1Ÿ�ApcfcefkeýdfgrchdŒJê„�u Addfkc„ý‘fffffæA@gefýýhev ðÄí@ `t×@xfcedifý Ð ã@pfgrfhdfffe‚ ý†g33333SP@ddýdefse  Øã@dceeieýd ÀgÙ@fgrhddddfjp’˜�hdeýýferef à<à@  ØÓ@cedfeýpe @ÉÔ@grihdddfej¢ýªged333333I@ýdferfdc `t×@eegeýdeg @�Î@rgddddfif“ý‹dedý33333SP@  Øã@ýefseece à«Î@dfeýofgr ÀgÙ@ihdddfht¦ý£eddýdÍÌÌÌÌÌR@ à<à@eeqfdcggjeýdfgrj @ÉÔ@ddddfii|ýŸhddýdd33333³4@  ØÃ@dlcfceeieýdfgrch Àg¹@dddfkc£ýÍÌÌÌÌŒO@žfddýýed €¦õ@  Øã@pcfceeheýpegrchd ·ë@ddfec¢ý‚333333I@fddýdfdr Ð#Ñ@cecgfifý àðÅ@qfmrchdfffec}ýƒfÍÌÌÌÌL7@ddýýefpf à<À@ �—@eceddeýf @É´@egrhhdddfhr|ý�defffffæ>@dýdefxid  ØÓ@cifdeýde ÀgÉ@greddddfjs ý�gedš™™™™ùU@ýýhdxcfc À ò@ P[è@eejeýpeg €�ç@rchhfffecŠýƒdedýÍÌÌÌÌŒM@  Øã@dedqceciejeýdegr ÀgÙ@chdfffec“ý�gddýýffffffC@ €¦õ@  ØÃ@ffr„dcefefýmfgse ·ë@ddfffexuý~dddýdgfffff¦F@ €¦õ@ftsdceeieýdfgsdd ·ë@dfffel¢ý333333I@|fddýýef ðÄí@ �Û@q}dcgfheýpfgsedd Ð ã@fefen¢ýŠ333333I@jedýdefq P[è@edceffeý à-ß@degsfddfffed€ýŠfš™™™™™5@edýdddmc �Û@ecefggýd `RÑ@ffschdddfkc•ý—fd33333sL@dýdeerjd P[è@cefffýdf à-ß@gshddffeeŒ¢ý�gdd333333I@ýýferxdc P[è@ P[è@feefýpeg à-ß@sgddfffei”ý§dedýš™™™™™<@ à<à@ýddlcfcf �Û@fgeýpfge @ÉÔ@chdddfic¥ý¥gedýýÍÌÌÌÌLU@ ðÄí@ Ð#Ñ@eepdecedieýoegeg Ð ã@hdffefd™ý¹fedddeš™™™™yR@epcfcfddeddfgefhdddeic|ý333333-@xeddddedpcecifggdÆfgechdddfjcƒýwš™™™™™3@deddýedp 5¤@ceciffedgfgechdfffecŽý‘hš™™™™�;@edýýddlc Á¯@ Á¯@fcifgeýi S¤@fgechdddfkcyý®je333333+@dýýedpcf €¦µ@ �w@cffmeýef ·«@gechdfffec“ýƒidd33333SP@ýýedpcec `t×@ `t×@effeýpfg @�Î@echdfefec�ýšiddýÍÌÌÌÌ�J@ à<à@ýedpcfcg �Û@fegýpfge @ÉÔ@chdfffec¦À¶jddýd `t×@getddcefieýdegeg @�Î@ddddeie‡„•dddýýe @Bù@ @Bù@epcfcgdgeýpegefh À*ð@dffeec…ý333333+@–fddýýed  ØÃ@  ØÃ@ocfcgffgýmfgecee Àg¹@ddffcˆýyÍÌÌÌÌLF@gddýýedp `tÇ@ `tÇ@cecieggý @�¾@nfgechdddfec€ý‰eš™™™™™5@efýdferc à<À@ecidhgýd @É´@feefhdddfhc|yŽfedýýgdsce à«Î@ à«Î@cgfggýof  ¡Ã@gechdfffec“ý·fedffffffC@ýdedpcfc  ØÓ@ffjeýdfg ÀgÉ@echdddfic”ýºgedý3333338@ Ð#Ñ@defpcdcffjeýdfge àðÅ@dddddfic}ý gddýýÍÌÌÌÌL7@  Øã@ Ð#Ñ@edrceceegeýoegec ÀgÙ@hdfffec¤ý±hegýdeÍÌÌÌÌ�U@  Øã@etdfcedfeýdfgegh ÀgÙ@dddfie�r�hdjýdgf ðÄí@tffcedieýdegddhd Ð ã@ddfeiŠ�šdehýýget à<À@ à<À@dfcidieý @É´@megdfhdffeed‡„”fedýýefpn  Øã@  Øã@dcgfkeýp ÀgÙ@egdeddfffep¤ý°geÍÌÌÌÌ�U@dýýgetfd  Øã@  ØÓ@cffjeýpe ÀgÙ@geeddfffff‡ý¤idd33333óE@ddeepdfŒediedqfgeghdfefgc…ý¤gddý333333+@ �Û@ýedpcecg �Û@ffeýpfde `RÑ@chdfffec•’„dedýý @Bù@ €¦õ@edpclceeheýpegec À*ð@edfffec‡„–fdddýe  Øã@epdfcedgedpemeehdfffee‹ýÍÌÌÌÌÌ4@“iddýýed ðÄí@ €¦µ@pceceeheýkfgechh Ð ã@ddfjcdd†fddýddeo Ð#Ñ@ddcieegý àðÅ@dfgeedddddkd—ýŸfš™™™™YR@edýýfesp  Øã@  ØÓ@ecedeeýp ÀgÙ@fgtghdfefeˆ´ýŒfe333333W@}ýýhexee�$(¾1Ÿ�Aáî¬��^�AcgdheýpeŒJê„�u Adtehdfefeeˆý˜hedÍÌÌÌÌÌ,@ýdedpcfc P[è@eeeeýdfg à-ß@tchdddfkc›ý³geddÍÌÌÌÌLS@dgdtcfcffjeddfdtchdddficƒý|dedýýš™™™™ÙC@  ØÃ@  ØÃ@efsjdciefgýmegtg Àg¹@ddffff|tý|ddfýýgš™™™™™*@ à«Î@ à«Î@ducecifeeýofgÏch  ¡Ã@dfffecŠý33333³9@{dedýýef `tÇ@ 5¤@urdcifdeýgfgtddd @�¾@ddfj‡…ý~ €:@fedýýedp `t×@ €Ã²@cfciefeý @�Î@jffkchdfffec˜�‚eddýýeepe �Ë@ �Ë@dcifeeýn `RÁ@fgkeddfffen�ý€deš™™™™�B@dýýedqce  Øã@ à<à@ciefeýpe ÀgÙ@gkchdddfhc�ý•hdd33333³<@ýýeepffc €¦õ@ ðÄí@edfeýpeg ·ë@kgedfffedyý�dedý333333+@  ØÃ@ýefpdeci  ØÃ@dggýmfek Àg¹@iheddfjmƒýŠfddýýš™™™™ÙC@  Øã@ à<à@gdtcecifgeýpffkc ÀgÙ@hdfffecžý�eddýýeÍÌÌÌÌ,R@�$(¾1Ÿ�A  ØÓ@dpcecifeeýpfgkchŒJê„�u Adddfec™ýÍÌÌÌÌŒM@šfedýýge À ò@ à<à@tldceejeýpfgkgdd €�ç@ffeep¦ý•ÍÌÌÌÌLO@fdeýýhew P[è@ à<à@iecedieý à-ß@pefkfhdddfhxzýŠeš™™™™™-@ddýýeeue �Û@ �w@dcefjeýe `RÑ@fgkkddffff€‡„¡fddýýeepcf à<à@ �Û@cedeeýpe @ÉÔ@gkhhdddfec›ý¤fdeÍÌÌÌÌLO@ýdeeqdlc �Û@edfeýdfg `RÑ@khhdddfif€ý•dedýfffffæ6@ à<À@deerfdciedeýdegf @É´@jddddfh‚‡ý€eedýý33333óE@  Øã@  Øã@efpddciefeýpegff ÀgÙ@ddfffec‡ýŒfedýýf33333sE@ @Bù@ Ð#Ñ@eredcegeeýofgfed À*ð@dfffeg}ýÍÌÌÌÌL7@xdedýýed û¹@ �w@pcfcgfheýeeggchd ¡°@ffffcŒý‡ÍÌÌÌÌŒF@deddýeeq `t×@gdcieeedpeegjddddfh}|ýœe33333³4@edýýeeoe  Øã@ `t×@ecedheýp ÀgÙ@fgfiedfffej‘ýšfd333333?@dýýeepef ðÄí@ à<à@cedeeýpe Ð ã@gffhdddfeg�ý|dddš™™™™�B@ýýeepfec  ØÓ@  ØÓ@idieýpeg ÀgÉ@gffdfffee‘ýŠeedýffffffC@ €¦õ@ýhdvcece à<à@eeeýpegg ·ë@cedfffec–ý“dedýýfffff&Q@  Øã@  Øã@edpcfcieeeýpfggc ÀgÙ@hdfffec¤ý›gdlýýgÍÌÌÌÌ�U@ ðÄí@ �Û@dtcfcefieýpeggcf Ð ã@dddffcŠýÍÌÌÌÌŒM@šdehýýhe à<à@ à<à@vnecedeeýpfggfhd @ÉÔ@fffel“ý©ffffffC@hedýdeep �Û@ddcffeeý `RÑ@dfggjddddfie}ý‚e333333)@ddýýdene À ò@ à<À@ecedheým €�ç@fggffdfffej–ý‹hd33333³L@dýdgeted ðÄí@ceegeýdf Ð ã@gggddddfkk¡ýšdel @N@ýýgetkec À ò@  Øã@edeeýpfg €�ç@fefdfffer—ýžfedd33333³<@dfdrceceemeddfgfchdddfkc­ý¤gedýdfffff†Q@ ðÄí@edqceceejeýdegfc Ð ã@fdddfic‹ýžhdxýýgÍÌÌÌÌìP@ À ò@ ðÄí@dtceceefeýpeffch €�ç@dddfec„ýfffffæ:@~dddýdee�$(¾1Ÿ�AqhdceeeeýdegfeddŒJê„�u Addffx™ý—ÍÌÌÌÌ�P@fddýýfer P[è@ P[è@idceejeý à-ß@pegfiddfffek}ý—e333333)@edýýeepd P[è@ €¦µ@dceeieýk à-ß@fdffddffefeyr�fedýddfmcd û¹@cgfdeýdf ¡°@gfjddddfkc”ýƒfdd3333338@ýýeeocdc à<à@ û¹@iejeýlfd @ÉÔ@fgddffffc‡„�hddý P[è@ýefpdecg P[è@dgeýpegf à-ß@dhdfffef¡ý£geiýýš™™™™9T@ P[è@  Øã@geveecgdjeýpfdfe à-ß@hdfffekƒý�dedýýeš™™™™ÙC@�$(¾1Ÿ�A à<à@dpcecifeeýpfgfchŒJê„�u AdfffecŒýš™™™™�E@”fddýýdf ðÄí@ ðÄí@odfcedgeýpedfghd Ð ã@fffec‘ýž333333M@gddýýedp ðÄí@ P[è@cfceeeeý Ð ã@pfgfcgdddfec­ýƒf  U@ddýýgets�$(¾1Ÿ�A À ò@dceeieýpŒJê„�u Aegfdddddfes‡}‹gedýýeeped ðÄí@ ðÄí@ eejeýpe Ð ã@gfeddfffgk¦ý‚fexš™™™™™O@ýýgeudec À ò@ ðÄí@edgeýpeg €�ç@hegdfffed”ý³iddd3333338@deepcdcfefeddfdheddddfic€ý™fddýd3333339@ Ð#Ñ@ddncechfeeýdedhc àðÅ@hdddehc‡ý”hdgýdeš™™™™™?@ û©@esflcgdgeýdfdhdh ¡ @dfffeuŠýffffff<@©fddddeeredcgejeddeghhddddfid´ý§333333W@ee ýýhex�$(¾1Ÿ�Aáî¬��^�AefcedheýŒJê„�u Apedhfhdffffe€ý»jš™™™™™5@edýdddmc à<À@fcffdeýd @É´@fdhchdddfic¤ý‘gd  T@dýýhewdd ðÄí@  Øã@ceegeýpf Ð ã@gffddddfee‡ýµidd33333óE@ýdeepddc �Û@eeieýdeg `RÑ@eiddddfic’ý�fddd33333óC@deepdfcedjedqedfefdddeed}ý¬jddddÍÌÌÌÌÌ7@ddmcecegeeddfdfchdddfic¤ý�eedýýfÍÌÌÌÌ�U@ ðÄí@ à<à@etddceedeýpfgffd Ð ã@dffffeˆýš™™™™™8@”fedýýed�$(¾1Ÿ�A �Ë@pcfcegjeýnfgfchdŒJê„�u Addfkcdd‡eddýddfo `t×@ddcifgeý @�Î@dedfndddddld‡ý{d33333óE@fddýdfdf Á¯@dcifggdieljedffffefŠ�Žfeeýýffrgl Ð#Ñ@ Ð#Ñ@cgddfýof àðÅ@gjhhdeffehŠ��iddýdedncec à«Î@gfefýdfg  ¡Ã@ichdfffecyrªhedý û¹@ddfocdcgfdfýdfgi ¡°@gddddfic†ý‹hedýdš™™™™™5@ 5¤@efpceciddfýdffjg Þ™@hdddfkc„ýƒfedýdefffffæA@ à<À@fpheciddfýdfgjkh @É´@dddfkh{ýÍÌÌÌÌŒB@Šgddýýdf �Ë@ �Ë@oddcifggýnfgjldd `RÁ@ddfkfyrœiedýdeep à<À@cdcifdfý @É´@dfgjpddddfic“ý»dffffffC@edýdeerc �Ë@fcfdjeýd `RÁ@fdjfhdddficdd†gfdddeepcecedfgddfgjhhfdddkc”ý¢gdd3333338@ýýdelddc `tÇ@ `tÇ@gedfýnfg @�¾@jiddffffc�r¢mfdý �Ë@ýdfdcfci �Ë@dfgýnegj `RÁ@ghffffec–ýŸfddýdfffff&Q@ à<à@deneecedieýdegjh @ÉÔ@hdddfil–ý¢gddýýe @G@  Øã@ Á¯@dpclceedfýifgjch ÀgÙ@dfffec•ýfffffæI@˜dedýýed  ØÓ@  ØÓ@pcfceeefýpfgichd ÀgÉ@effec‡}µeddýddeh à«Î@cfcfddfý  ¡Ã@degifedddfic�ý¸f33333³<@edýdeerc  ØÓ@dcfeeeýd ÀgÉ@egihddddeic}ý~eeÍÌÌÌÌL7@dýýedqce à<à@ à<à@ceeffýpf @ÉÔ@eichdfffec’ý«fed33333óC@ýýferddc ðÄí@ à«Î@eeieýoeg Ð ã@ifddffeecŠ�‰fedý  ØÓ@ddeoeeciddfýdfgi ÀgÉ@hhdfffff�ý®gddýýÍÌÌÌÌŒK@  ØÓ@ `tÇ@edpcfceejeýnegic ÀgÉ@hdfffec‚ý–fddýýeš™™™™™=@ Ð#Ñ@ Ð#Ñ@gpddcgegeýofgidd àðÅ@dddfei•ýfffffæI@Œidfýýgd  Øã@ à<à@tcfceegfýpegfchd ÀgÙ@fffecœý¦ÍÌÌÌÌÌJ@fedýdedp à<À@cfcfeheý @É´@dfgfchdddfic”ýŒd3333338@ddddffrfecedgfddfgflhdddfk|¡ý¡de @N@dýdgeted  ØÓ@ceejeýdf ÀgÉ@gfiddddfif�ý|hdd33333³<@ýýefndec �—@ �—@gdeeýfeg ��@fehdddfff|ýxeddý33333³4@ `t×@ýedpceci à«Î@fgfýoegf @�Î@chdfffec„ýŒfedýýfffffæA@ À ò@ Ð#Ñ@edrcfceeieýofghc €�ç@hdfffec�ýœfedýýe @J@  Øã@  Øã@epndcgejeýpegrfd ÀgÙ@dfffef„ýš™™™™�C@¤gedýýee à<à@ à<à@pdecgddfýpeghihd @ÉÔ@fffed”ý¸3333338@gdddddendfcfdieddfghehdddfic�ýŠi33333³<@ddddeepdecgdjgdqeghihdddeki„ý†fdš™™™™�8@dýýddmce `tÇ@  ØÃ@chfggýme @�¾@ggchfddfec�r‡hfdýýdfdcdc €¦µ@ €¦µ@ifggýkfg ·«@hfdffffec—ý¥gddýš™™™™ÙO@ `t×@dhevcdceedfýdfdh @�Î@iddddfic•ý«hedýd33333sL@  Øã@eeprlcfdeeýdfghj ÀgÙ@hdddeisŠ�Šfdgýýg�$(¾1Ÿ�Aáî¬��^�AdtcecgfieýpfghchŒJê„�u Adfefec…ýÍÌÌÌÌLG@™dedýýfe P[è@ P[è@rdecedfeýpeghfhd à-ß@fffed–ý¤ @G@fedýdedp û¹@cecgfdfý ¡°@deghchdddfic¡ý¬iš™™™™9T@eÆddfewcechdegddfghghdddeic¤ý‰dd33333óQ@dýýhevhd  Øã@  ØÓ@ceedfýpf ÀgÙ@ghhddddfeh…ý€jdd333333+@ýdedmcfc Ð#Ñ@efffýdfg àðÅ@hchdddfkcŒý›eddýÍÌÌÌÌŒF@  ØÓ@ýefodfcg  ØÓ@dieýpfgh ÀgÉ@fhdfffep‡„”gedýd  ØÃ@edpcecgfeeýdfghc Àg¹@hdddekcˆý™fddýýeš™™™™™8@ Ð#Ñ@ Ð#Ñ@dqcecgfjeýofghch àðÅ@dffeecˆýÍÌÌÌÌÌB@‘gddýýde  ØÃ@  ØÃ@kcfcedgeýmedhfhf Àg¹@ddefc˜ýœ33333óP@eeeýýeeq À ò@ Ð#Ñ@}fcedjeý €�ç@ofgegfdeffes“ýºfffffffC@deýdferc  ØÓ@fcfdheýd ÀgÉ@fmeghdddfic}ý’gdÍÌÌÌÌÌ7@dýdeemde �—@ceddfýdf ��@ge|hdddfkf‡ý�edd33333óE@ýdeeoddc Á¯@ifdfýdeg S¤@efddddferŠ�~eddý  ØÓ@ýefqdfci  ØÓ@djfýpfge ÀgÉ@hhdfffet„ý¯hddýdffffff*@ Ð#Ñ@eepdecfdheýdfghi àðÅ@hdddfie~ý’gddýýe €<@  ØÃ@  ØÃ@eoidcifdfýmfghkd Àg¹@dddfe‚Žý33333óM@·dedýdee  ØÓ@okdcfedeýdfghÆdd ÀgÉ@ddfikŠý}33333³9@dedýýefr �Ë@ €¦µ@sdcifheý `RÁ@kfghhddffff‘¤ý›jÍÌÌÌÌ�U@edýýeerd à<À@ à<À@f”hdgeým @É´@fghjhdfffge…ýœee333333+@dýýeepdf �Û@ à«Î@cedheýof `RÑ@dkfhdfffeedd�hedýdedpcfc ðÄí@eeeeýdfg Ð ã@kchddddkc€ý±fddýš™™™™™5@ `t×@deepcfcedjeýdfgk @�Î@ghdddfkc€ý‰gdeýdš™™™™™5@ À ò@efpefcedgeýdfgkf €�ç@fdddfkx”ý§gddýde3333338@ �Û@epkfceddeýdfgkgh `RÑ@dddfiy|ý33333³4@�gddýýed Ð#Ñ@ à«Î@pcfceegeýoegkchd àðÅ@ddfec}ý™ÍÌÌÌÌL7@eedýýeep  ØÓ@  ØÓ@decgdheý ÀgÉ@pefjfhdfffecdd³fedddedpcfcfgmeddfgjchddddkcvý‡geÍÌÌÌÌL>@dýdeepdd P[è@ceegeýdf à-ß@gjeddddfkfˆýžgddš™™™™™8@ýdeepeec û¹@gdeeýdfg ¡°@hhhdddfln¤ý§fedýÍÌÌÌÌ�U@ �Û@dferddceeheýdfgh `RÑ@gddddfigdd¼hddýd à<À@edpcfcfffeýdfghc @É´@hddddkczýšgedýdeš™™™™™-@ û¹@focdcgfieýdfgigd ¡°@dddfkcyrzdedýýdf �w@ �w@nqdcifmeýeedifdd �m@fffeq™‚�iedýýfer P[è@ Ð#Ñ@fdcefgeý à-ß@ofgihddfffe‚ˆýŠfÍÌÌÌÌÌ,@ddýýefpe û¹@ 5¤@fcgdieýg ¡°@ffiehdddfee‡„’eedýýeeped `t×@ `t×@cffdeýpe @�Î@gifddfffeg…ý•gdd333333+@ýdeepdfc  Øã@edheýdfg ÀgÙ@ijfdddfke’˜†fddý à<à@ýfftjfci à<à@dgeýpegj @ÉÔ@fhdfffe yýšfddýd333333+@ à<à@edpcfceegeýdfgjc @ÉÔ@hdddfkc“ý†hddýýefffffæL@ `t×@ �Ë@frxdceefeýnfgjed @�Î@dfffen©ýfffffÆV@©feeýdhe à<à@wrecedfeýdfgjghd @ÉÔ@ddfig–ýˆ @G@iddýdefp `t×@ddcgfgeý @�Î@dfgjdddfffel–ý¥e33333“Q@ddýýheth P[è@  Øã@dcefeeýp à-ß@egjfddfffee™ý”fdÍÌÌÌÌŒD@dýýgeted ðÄí@ Ð#Ñ@ceedeýof Ð ã@gjdddfffedƒýŠfddš™™™™™3@ýýeesdec  Øã@ à<À@edeeýmfg ÀgÙ@jfhdffffeƒý‚dedýš™™™™ÙC@ Ð#Ñ@ýeeredci û©@ffeýhegj àðÅ@fddddfhn¤ý”fedýýÍÌÌÌÌ�U@ à<à@ û¹@gevedceeieýlfgfd @ÉÔ@ddffffi‹�•gedýýe `t×@ `t×@epndcgeheýpfgffd @�Î@dfffex–ýfffff&Q@šiedýýgd ðÄí@ `t×@tceceffeýpegfchd Ð ã@fffec¤ý•  T@fdfýýgdt À ò@ à<à@cfceeieý €�ç@pegfcedfffec˜ý™f33333óP@egýýgetl ðÄí@ à<à@ecedeeýp Ð ã@fgfdhdfffef£ý˜deÍÌÌÌÌŒO@lýýgetie ðÄí@  Øã@cedeeýpe Ð ã@gfehdfffepŽý¶gd‚33333³U@ýdhexgec À ò@fddeýdeg €�ç@ffhdddfie¡ý£fdpdš™™™™9T@dgeteecedfeddfmifhdddfif�ýšfddýýš™™™™�9@ `t×@ `t×@eepddcgeheýpegig @�Î@ddddfec„ý¢fedýýeš™™™™�8@ à<à@ à<à@dpcfceeheýpegich @ÉÔ@dfffec|ý33333³4@Œddgýýee à<à@ �—@relcidieýfegifed @ÉÔ@fffej§ý…ffffffR@gexýdefs P[è@efcedgeý à-ß@dfgidhdddfjpˆý¯gš™™™™™8@ddýdeerc  ØÃ@dcfljeýÆ Àg¹@fgihddddeic¡ý½hdš™™™™9T@dýdffrcd �Ë@cffieýdf `RÁ@gigddddfkc�ýœdddÍÌÌÌÌL7@ýýeepddc à<à@ à<à@gegeýpeg @ÉÔ@ehddfffed‹ý|dddýÍÌÌÌÌ�C@ à<à@ýeercdci à<À@efeýmege @ÉÔ@hddfffecŒý�dedýýfffff&C@ P[è@ �Û@edqcecieeeýpegec à-ß@hdfffec‡ý«fedýýe33333sE@  ØÓ@ à«Î@epddcfeheýofgefd ÀgÉ@dffffe