Data analysis report

课堂练习/risk08.sps

* Encoding: UTF-8. get file= 'd:\jacky\risk08.sav'. * 1: Data transformation. recode a00 (missing=0). recode a00 to a89 (99=sysmis). recode a75 to a89 (else=copy) into level major yr age stayhk sex marital kid f.emp m.emp f.edu m.edu parent kin id. recode age (lo thru 15=sysmis) /level (1 thru 3=copy)(else=sysmis) /major marital (0=sysmis) /marital (1 thru 4=copy)(else=sysmis) /yr (20 thru high=sysmis)(0 thru 1=1) /kid (10 thru hi=sysmis) /sex (1,2=copy)(else=sysmis) /f.emp m.emp (1 thru 4=copy)(else=sysmis) /parent (4=sysmis). recode a83 a84 (1 thru 4=copy) into f.status m.status. Compute bachelor=level=1. Compute master=level=2. Compute doctor=level=3. Compute psy=major=1. Compute swk=major=2. Compute soc=major=3. Compute counsel=major=4. Compute crime=major=5. Compute f.boss=f.emp=1. Compute f.sole=f.emp=2. Compute f.hird=f.emp=3. Compute f.home=f.emp=4. Compute m.boss=m.emp=1. Compute m.sole=m.emp=2. Compute m.hird=m.emp=3. Compute m.home=m.emp=4. compute female=sex=2. Compute control=mean(a09,a19,a29). Compute fear=mean(a10,a20,a30). Compute respon=mean(a11,a21,a31). Compute know=mean(a12,a22,a32). Compute disaster=mean(a13,a23,a33). Compute new=mean(a14,a24,a34). Compute detect=mean(a15,a25,a35). Compute unclear=mean(a16,a26,a36). compute risk.soc=mean(a17,a27,a37). compute risk=mean(a18,a28,a38). recode marital (0=1). compute married=marital=2. compute single=marital=1. compute divorce=marital=3. compute widow=marital=4. recode age (lo thru 15=sysmis) /level (30=3) /major marital (0=sysmis) /yr (20 thru high=sysmis) /sex (1,2=copy)(else=sysmis) /f.emp m.emp (1 thru 4=copy)(else=sysmis) /parent (4=sysmis). recode wave2 (missing=0). recode after (missing=0). recode a89 (0=sysmis). Compute bond=mean(v04 to v29). reliability var= a12 a22 a32 /summary all. frequency after to kin bachelor to female married to widow . frequency id. * 2: Subsampling. set seed=55555550. compute random=uniform(1). *temporary. select if (random>.1). * 3. frequency risk know age female /histogram=normal /statistics skew kurtosis all . * 4. npar test k-s(normal)=risk know. npar test binomial(.5)=married(0,1). npar test chisquare=major(1,4) /expected equal. npar test chisquare=major(1,4) /expected .1, .3, .6, .01. npar test k-w=risk by major(1,4). means risk by major /statistics anova. npar test sign=risk with risk.soc. t-test pair= risk with risk.soc. * 5. rank risk /percent into riskp. curvefit riskp with risk. curvefit risk with control /print anova. curvefit var= a17 with a10. means riskp by risk. frequency kid a39 to a44 /histogram=normal /statistics skew kurtosis. rank kid /percent into kidp. frequency kidp /histogram=normal /statistics skew kurtosis. npar test k-s(normal)=kidp. descriptives after to widow /save. graph /scatterplot risk with zrisk. crosstabs f.status by m.status /cell count col /statistics kappa. crosstabs variable= f.status m.status (1,4) /table f.status by m.status /cell count col /statistics kappa. npar test kendall= a09 a19 a29. * 6. descriptives all /statistics default skew kurtosis. Frequency all /histogram=normal. examine risk.soc risk. * 7. t-test testval=5 /var= risk. t-test testval=5 /var= risk.soc risk. * 8. crosstabs f.emp by major /cell row /statistics chisq phi. sort cases by f.emp. temporary. select if (f.emp>-1). split file by f.emp. GRAPH /BAR(SIMPLE)=PCT BY major. * 9. correlation a17 with Age kid f.edu to kin stayhk Bachelor to m.home female . nonpar correlation a17 with Age kid f.edu to kin stayhk Bachelor to m.home female . correlation a17 with a09 to a16. correlation a09 a19 a29. nonpar correlation a09 a19 a29. * 10. reliability var= a39 to a44 /scale(worry)= all /summary all. reliability var= a09,a19,a29 /scale(control)= a09,a19,a29 /summary all. reliability var= a10,a20,a30 a09,a19,a29 /scale(fear)= all /summary all. reliability var= a11,a21,a31 /scale(respon)= a11,a21,a31 /summary all. reliability var= a12,a22,a32 /scale(know)= a12,a22,a32 /summary all. reliability var= a13,a23,a33 /scale(disaster)= a13,a23,a33 /summary all. reliability var= a14,a24,a34 /scale(new)= a14,a24,a34 /summary all. reliability var= a15,a25,a35 /scale(detect)= a15,a25,a35 /summary all. reliability var= a16,a26,a36 /scale(unclear) a16,a26,a36 /summary all. reliability var= a17,a27,a37 /scale(risk.soc)= a17,a27,a37 /summary all. reliability var= a18,a28,a38 /scale(risk)= a18,a28,a38 /summary all. * 11. Factor var= a39 to a44 a45 to a51 /missing pairwise /plot eigen rotation(1,2) /criteria fact(5) /extract ml /rotate varimax. /save reg(all,finan). * Exploratory: financial crisis. Factor var= a09 to a16 /missing pairwise /plot eigen rotation(1,2) /criteria fact(2) /extract ml /rotate varimax /save reg(all,finan). * Exploratory: SARS. Factor var= a19 to a26 /missing pairwise /plot eigen /criteria fact(2) /extract ml /rotate varimax /save reg(all,SARS). * Exploratory: terrorism. Factor var= a29 to a36 /missing pairwise /plot eigen /criteria fact(2) ITERATE(100) /extract ml /rotate varimax /save reg(all,terror). * Confirmatory: risk to society and controllability. factor var= a17 a27 a37 a09 a19 a29 /missing pairwise /plot eigen /criteria fact(2) /extract ml /rotate varimax /print fscore default. * Confirmatory: controllability and fear. factor var= a09 a10 a19 a20 a29 a30 /missing pairwise /plot eigen /criteria fact(2) /extract ml /rotate varimax. * Confirmatory: risk to society and self. factor var= a17 a18 a27 a28 a37 a38 /missing pairwise /plot eigen /criteria fact(2) /extract ml /rotate varimax. * Valdiating risk and newness. factor var= a18,a28,a38 a14,a24,a34 /missing pairwise /plot eigen /criteria fact(2) /extract ml /rotate varimax. * 12. * Main effect. glm risk.soc risk by level major with age female /print parameter opower etasq homogeneity /plot=profile(level major) /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major age female. * Main effect. glm risk.soc by level major a90 with age female /random a90 /print parameter opower etasq /plot=profile(level major) /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major age female. * Main effect. glm risk.soc by level major with age female /print parameter opower etasq homogeneity /plot=profile(level major) /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major age female. * Interaction effect. descriptives risk.soc risk age female /save. glm zrisk.soc zrisk by level major with zage zfemale /print parameter opower etasq /plot=profile(major*level) /emmean table(level*major) compare(level) adj(bonferroni) /emmean table(level*major) compare(major) adj(bonferroni) /design level major level*major zage zfemale. compute major3_level3=(major=3 and level=3). glm zrisk.soc zrisk by level major with zage zfemale major3_level3 /print parameter opower etasq /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major zage zfemale major3_level3. * 13. * Repeated measures. glm risk.soc risk /wsfactor target(2) /print parameter opower etasq. npar test kendall= risk.soc risk. * Repeated measures. glm zrisk.soc zrisk by level major with zage zfemale /wsfactor target(2) /print parameter opower etasq homogeneity /plot=profile(level major) /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major zage zfemale. glm risk.soc risk by level major with age female /wsfactor target(2) /print parameter opower etasq homogeneity /plot=profile(level major) /emmean table(level) compare adj(bonferroni) /emmean table(major) compare adj(bonferroni) /design level major age female. * Repeated measures. glm risk.soc risk by major /wsfactor target(2) /print parameter opower etasq /plot=profile(major) /emmean table(major) adj(bonferroni). glm risk.soc by level major with age female /print parameter /plot=profile(level major) /emmean table(level) adj(bonferroni) /emmean table(major) adj(bonferroni) /design level major age female. descriptives f.boss m.boss f.edu m.edu /save. manova zf.boss zm.boss by level(1,3) major(1,4) with zf.edu zm.edu /wsfactor boss(2) /print parameter homogeneity /pmean table(level major) /design level major . * 14. cluster a17,a27,a37 /plot dendrogram /method ward. cluster control fear respon risk /plot dendrogram /method ward /save cluster(3). quick cluster control fear respon risk /criteria cluster(3) /save cluster(risk3) . quick cluster control fear respon risk /missing pairwise /criteria cluster(3) /print distance anova . means control fear respon risk by clu3_1 risk3 /statistics all. means risk by parent a44 /statistics all. npar test k-w= risk by parent (0,2). crosstabs parent by major /cell row /statistics chisq phi. npar test chisq= female. correlation risk age parent female. correlation risk age parent female /missing listwise. nonpar corr risk age parent female. graph /bar=mean(risk) by parent. * 15. regression /statistics tol /dependent risk /enter control new age. /dependent new /enter yr age. /dependent control /enter yr. /dependent yr /enter age. * 16. temporary. compute inter=control*female. regression /missing mean /statistics default /dependent risk /enter control new age female /enter inter. * Equality of controllability and newness metric effects. temporary. compute both=mean(control,new). regression /missing mean /statistics default /dependent risk /enter both new age female. * Equality of controllability and newness standardized effects. descriptives control new /save. temporary. compute both=mean(zcontrol,2*znew). regression /missing mean /statistics default /dependent risk /enter both new age female. * Equality of controllability, newness, and age standardized effects. descriptives control new age /save. temporary. compute three=mean(zcontrol,znew,zage). regression /missing mean /statistics default /dependent risk /test (three)(new control )(female). * Mulitcollinearity-free moderated analysis. temporary. compute inter=zcontrol*znew. regression /missing mean /statistics default tol /dependent risk /enter control new age female inter. * 17. temporary. compute inter=control*new. regression /missing mean /statistics default /dependent risk /enter control new age female /enter inter. * 18. temporary. compute square=new**2. regression /missing mean /statistics default /dependent risk /enter control new age female /enter square /scatterplot (*pred, new). regression /missing mean /descriptives mean stddev n corr /dependent risk /enter fear unclear /enter soc. regression /missing mean /descriptives mean stddev n corr /dependent risk /enter fear soc. regression /missing mean /descriptives mean stddev n /dependent finan1 to finan7 /enter Ifeelcomfortablearoundotherpeople to Ifrequentlyshowfriendsandfamilywhattheymeantome . regression /missing mean /descriptives mean stddev n /dependent risk worry fear control /enter new /stepwise Age kid f.edu to kin stayhk Bachelor to m.home female. * 19. regression /missing pairwise /descriptives mean stddev /STATISTICS DEFAULTS tol /dependent risk /enter age control new /scatterplot (*zresid,age) (*zresid,control) /partialplot all /residual histogram /casewise cook all /save zresid(riskz). temporaray. compute riskz=age*riskz. graph /scatterplot age with riskz. regression /missing pairwise /descriptives mean stddev /dependent risk /enter age control /scatterplot (*zresid,age) (*zresid,control) /partialplot all /residual histogram /casewise cook all /save cook(c1). temporary. select if (c1<.02 and c2<.02 and c3<.02 and c4<.02 and c5<.02). regression /missing pairwise /descriptives mean stddev /dependent risk /enter age control /scatterplot (*zresid,age) (*zresid,control) /partialplot all /residual histogram /casewise cook all /save cook(c6). * 20. temporary. compute age2=age**2. regression /missing pairwise /descriptives mean stddev /regwgt age2 /dependent risk /enter age control /save resid(risk.r). compute risk.r=age*risk.r. graph scatterplot= age with risk.r . * 21. * Expecting the effects of risk constituents on the risk core and exploring those effects. regression /missing mean /descriptives mean stddev n /dependent a17 a18 /enter a09 to a16 /stepwise Age kid f.edu to kin stayhk Bachelor to m.home female . regression /missing mean /descriptives mean stddev n /statistics zpp /dependent a52 /enter a39 to a44 /stepwise Age kid f.edu to kin stayhk Bachelor to m.home female . * 22. sort cases by female. temporary. split file by female. regression /missing mean /descriptives mean stddev n /statistics zpp default /dependent risk /enter control /enter detect . . regression /missing mean /dependent a17 a18 /enter a12 a22 a32 /stepwise Age kid f.edu to kin stayhk Bachelor to m.home female . * 23. 2sls risk with riskp/riskp with risk /instrument new fear. correlation risk new yr female soc age. 2sls risk with new /new with risk /instrument yr age soc. 2sls risk with new yr /new with risk soc /instrument soc age . 2sls risk with new /new with risk /instrument control yr . 2sls risk with control new /control with new risk /new with age yr control /instrument age female yr. 2sls control with risk yr /instrument new age female. 2sls new with risk yr age /instrument control female yr age. 2sls risk with new control fear /new with age control /instrument age control fear /print cov. 2sls risk with new control fear /instrument new control fear /print cov. 2sls risk with new /instrument yr age . 2sls risk with new /new with risk /worry with risk new /instrument age female /print cov. 2sls risk with age /instrument age /print cov. * 24. frequency disaster /histogram=normal. quick cluster disaster /criteria cluster(2) /missing pairwise /print anova. quick cluster disaster /criteria cluster(3) /missing pairwise /print anova. quick cluster disaster /criteria cluster(4) /missing pairwise /print anova. quick cluster disaster /criteria cluster(5) /missing pairwise /print anova. quick cluster disaster /criteria cluster(6) /missing pairwise /print anova. quick cluster disaster /criteria cluster(7) /missing pairwise /print anova. quick cluster disaster /criteria cluster(3) /missing pairwise /print anova /save= cluster (disasc). means disaster by disasc /cell all. correlation control to unclear. quick cluster disaster control /criteria cluster(3) /missing pairwise /print anova. cluster risk risk.soc /plot hicicle /method ward. cluster a09 to a16 /plot dendrogram /method ward. proximities a17 a27 a37 a09 a19 a29 /view variable /matrix out(*). cluster all /plot dendrogram /method ward. cluster all /plot dendrogram /method single. cluster all /plot dendrogram /method complete. descriptives finan1 to terror2. descriptives age kid f.edu to kin stayhk Bachelor to m.home female . factor var= f.boss finan1 a18 /missing pairwise /print corr default. * 25. hiloglinear major(1,5) marital(1,4) female(0,1) level(1,3) /method backward /print default association. * 26. genlog major marital female /design major marital female major*marital marital*female /print estim. genlog major marital female /design major*marital marital*female /print estim. loglinear major(1,4) marital(1,2) female(0,1) /print estim /design major marital female major*marital marital*female /contrast(major)=deviation(1) /design major marital female major*marital marital*female /design. frequency major marital female. * 27. temporary. compute married_male=(marital=2)*(female=0). genlog major marital female with married_male /design major marital female married_male major*marital /print estim. * 28. temporary. compute married_male=(marital=2)*(female=0). compute married_female=(marital=2)*(female=1). genlog major marital female with married_male married_female /design major marital female married_male married_female major*marital /print estim. * 29. temporary. compute married_female=(marital=2)*(female=1). genlog major marital female with married_female /design major marital female married_female major*marital /print estim. * 30. temporary. compute match=(f.emp=m.emp). compute bread=(f.emp<4 and m.emp=4). genlog f.emp m.emp with match bread /design f.emp m.emp match bread /print estim. * 31. temporary. compute match=(f.emp=m.emp). compute bread=(f.emp<4 and m.emp=4). compute higher=(f.emp<m.emp). compute wealth=(f.emp<3 and m.emp=4). genlog f.emp m.emp with match bread higher wealth /design f.emp m.emp match bread higher wealth /print estim. * 32. genlog level /print estim. * Log rate model: weighting by the reciprocal of level. temporary. compute level0=1/level. genlog level /cstructure level0 /print estim. * 33. temporary. compute riskier=(risk>risk.soc). logistic regression riskier /enter control new age female. temporary. compute risk=(risk>7). compute control0=control/10. compute new0=new/10. compute age0=age/32. logistic regression risk /categorical level /contrast(level)=indicator(1) /enter level control new age female. descriptives age. * 34. genlin a18 by level with age0 female single /model level age0 female age0*female single distribution= poisson link=log /repeated subject=a90 corrtype=unstructured. genlin a18 by level with age female single /model level age female age*female single distribution= poisson link=identity /repeated subject=a90 corrtype=unstructured. genlin a18 by level with age female single /model level age female age*female single distribution= negbin(1) link=log /repeated subject=a90 corrtype=unstructured. genlin a18 by level with age female single /model level age female age*female single distribution= negbin(mle) link=log. sort cases by a90. genlin a18 by level with age female /model level age female age*female distribution= binomial link=identity /repeated subject=a90 corrtype=ar(1). sort cases by a90. genlin level by major with age /model major age distribution= poisson link=log /repeated subject=a90 corrtype=ar(1). * 35. sort cases by female. split file by female. proximities a09 to a16 /view variable /measure correlation /matrix out(*). proxscal all /proximitites similarity /model generalized /criteria dimension(2). factor var= a09 to a16 /plot eigen /format sort /criteria factor(3) /extract ml /rotate varimax. tsplot /var= risk. graph /errorbar=risk. examine var= age. frequency var= disaster /histogram=normal. * 36. * Visual Binning. *disaster. RECODE disaster (MISSING=COPY) (LO THRU 6=1) (LO THRU HI=2) (ELSE=SYSMIS) INTO disaster.b2. VARIABLE LABELS disaster.b2 'disaster (Binned)'. FORMATS disaster.b2 (F5.0). VALUE LABELS disaster.b2 1 '' 2 ''. VARIABLE LEVEL disaster.b2 (ORDINAL). EXECUTE. *frequency all /histotram=normal. twostep cluster /continuous var= disaster /print ic count summary /save cluster var=disaster.c2. crosstabs disaster.c2 by disaster.b2 /statistics kappa. means disaster by disaster.b2 disaster.c2 /statistics all. frequency m.edu /histogram=normal. twostep cluster /continuous var= m.edu /print ic count summary /save cluster var=m.edu.c4. * Visual Binning. *m.edu. RECODE m.edu (MISSING=COPY) (LO THRU 4.5=1) (LO THRU 8=2) (LO THRU 12=3) (LO THRU HI=4) (ELSE=SYSMIS) INTO m.edu.b4. VARIABLE LABELS m.edu.b4 'm.edu (Binned)'. FORMATS m.edu.b4 (F5.0). VALUE LABELS m.edu.b4 1 '<= 4.50' 2 '4.51 - 8.00' 3 '9.00 - 12.00' 4 '13.00+'. VARIABLE LEVEL m.edu.b4 (ORDINAL). EXECUTE. crosstabs m.edu.c4 by m.edu.b4 /statistics kappa. means m.edu by m.edu.c4 m.edu.b4 /statistics all. twostep cluster /continuous var=risk risk.soc /print ic count summary. twostep cluster /continuous var=risk a39 /print ic count summary /save cluster var=risk_a39. twostep cluster /categorical var=major /continuous var=age female /print ic count summary /save cluster var=back. * 37. *mixed-effect modeling. mixed zrisk by level major with zage zfemale /fixed level major zage zfemale /random intercept | subject(a90) covtype(vc) /print solution testcov. mixed zrisk by level major with zage zfemale /fixed level major zage zfemale zage*zfemale /random intercept level major zage zfemale zage*zfemale | subject(a90) covtype(vc) /print solution testcov. * 38. * Ordinal logislic regression analysis. plum a18 by level with age0 female new0 control0 /location level age0 female new0 control0 /print fit tparallel parameter summary . plum a18 with age female single /location age female single /print fit tparallel parameter summary . regression /dependent a39 to a44 /enter age female a12 a22 a32 a52 to a57 . * 39. compute one=1. temporary. compute r18=10-a18. cox regression r18 /status one(1) /categorical level /enter level age female single. temporary. compute r18=10-a18. cox regression r18 /status one(1) /categorical level /enter level age female age*female single. * /strata a90. temporary. if (a18=0 or a18=10) one=0. compute r18=10-a18. cox regression r18 /status one(1) /strata a90 /categorical level /enter level age female age*female. cox regression r18 /status one(1) /strata a90 /categorical level /enter level age female age*female a10. coxreg one a18 by level with age female /variable status=one(1) /model level age female /testassumptions prophazard. * 40. nomreg f.emp with age female f.edu /model age female f.edu /print summary parameter fit association cps step mfi. nomreg f.emp with age female f.edu /model age female f.edu age*female /print summary parameter fit association cps step mfi. **************************************. frequency age. frequency after to widow. regression /missing pairwise /criteria pin(.049) pout(.050) /dependent a58 to a74 /stepwise bachelor to female age stayhk married to widow. /dependent a01 to a08 control to risk /stepwise a58 to a74 /stepwise bachelor to female age stayhk married to widow. correlation know a70 swk. factor var= know a70 swk /print corr. variable labels after 'After the financial tsunami' /wave2 'Repeated survey'. 75 Level of study: 1 = bachelor's level, 2 = master's level, 3 = doctoral level 76 Major field of study: 1 = psychology, 2 = social work, 3 = sociology, 4 = counseling, 5 = criminology 77 Year of study in the current program 78 Age 79 Years in Hong Kong 80 Sex: 1 = male, 2 = female 81 Marital status: 1 = unmarried, 2 = married, 3 = divorced/separated, 4 =widowed 82 Number of your children 83 Father's major employment status in life: 1 = employer, 2 = self-employed, 3 = employee, 4 = not working 84 Mother's major employment status in life: 1 = employer, 2 = self-employed, 3 = employee, 4 = not working 85 Number of parents living with you 86 Number of other relatives living with you 87 To let us avoid duplicated cases, please indicate your student ID ********************************************. input program. numeric a00 a a01 to a57 b a58 to a74 c a75 to a89. end file. end input program. save trans outfile= 'd:\jacky\risk08.xls' /replace /fieldname. Introduction 1 Your endorsement for the popular belief that a financial crisis is under political control 2 Your suspicion about the popular belief that the government has a lot of top-secret information 3 Your belief in government warning about financial crisis 4 Your suspicion about the government description of the financial condition as a bubble 5 Unreliability of government information 6 Your belief that the government helps people avoid the financial crisis 7 Importance of seeking information from government 8 Your suspicion about possibility of government warning about financial crisis 9 The controllability of financial crisis by government 10 The dread (fear) of financial crisis 11 The personal responsibility for financial crisis 12 Knowledge about financial crisis 13 The disastrousness of financial crisis 14 The newness of financial crisis 15 The detectability of financial crisis 16 The uncertainty of financial crisis 17 The risk of financial crisis to society 18 The risk of financial crisis to you 19 The controllability of a SARS-like epidemic by government 20 The dread (fear) of a SARS-like epidemic 21 The personal responsibility for a SARS-like epidemic 22 Knowledge about a SARS-like epidemic 23 The disastrousness of a SARS-like epidemic 24 The newness of a SARS-like epidemic 25 The detectability of a SARS-like epidemic 26 The uncertainty of a SARS-like epidemic 27 The risk of a SARS-like epidemic to society 28 The risk of a SARS-like epidemic to you 29 The controllability of terrorist attack by government 30 The dread (fear) of terrorist attack 31 The personal responsibility for terrorist attack 32 Knowledge about terrorist attack 33 The disastrousness of terrorist attack 34 The newness of terrorist attack 35 The detectability of terrorist attack 36 The uncertainty of terrorist attack 37 The risk of terrorist attack to society 38 The risk of terrorist attack to you 39 Worry about your financial condition 40 Worry about your health 41 Worry about your safety 42 Worry about your social relationship 43 Worry about your family 44 Worry about children 45 Unwilling to support government welfare for unemployed people 46 Willing to support the government's using tax revenue for healthcare 47 Unwilling to contribute to public insurance for healthcare 48 Willing to support government 's combat against terrorism even though the combat is inconvenient to you 49 Unwilling to support government assistance to new arrivals 50 Willing to urge the government to increase assistance to welfare recipients 51 Unwilling to support time-consuming inspection to control terrorism in the airport and other entrances to Hong Kong 52 Distrust of government policy 53 Trust in the government of Hong Kong 54 Distrust of government officials 55 Trust in legislative councilors 56 Trust in the laws of Hong Kong 57 Distrust of political parties B. Please indicate the number of times of the following discrete events in the past year 58 Being sick 59 Staying in hospital for treatment 60 Being robbed 61 Being beaten 62 Losing $100 or more 63 Donating $100 or more to charity 64 Receiving psychological counseling or treatment 65 Joining activities hosted by legislative councilors 66 Joining activities hosting by government officials 67 Joining activities organized by political parties 68 Being injured 69 Doing volunteer work to help the needy 70 Having a self-initiated body check-up 71 Enrolling in courses about safety or self-protection 72 Joining activities to urge for increasing public welfare 73 Talking with needy people and endorsing their demand for more welfare 74 Learning how to protect yourself from assault C. Please write down your background information 75 Level of study: 1 = bachelor's level, 2 = master's level, 3 = doctoral level 76 Major field of study: 1 = psychology, 2 = social work, 3 = sociology, 4 = counseling, 5 = criminology 77 Year of study in the current program 78 Age 79 Years in Hong Kong 80 Sex: 1 = male, 2 = female 81 Marital status: 1 = unmarried, 2 = married, 3 = divorced/separated, 4 =widowed 82 Number of primary school children in your household 83 Father's major employment status in life: 1 = employer, 2 = self-employed, 3 = employee, 4 = not working 84 Mother's major employment status in life: 1 = employer, 2 = self-employed, 3 = employee, 4 = not working 85 Father's number of schooling years (Primary 1 = 1 �K) 86 Mother's number of schooling years (Primary 1 = 1 �K) 87 Number of parents living with you 88 Number of other relatives living with you 89 To let us avoid duplicated cases, please indicate your student ID temporary. compute match=(f.emp=m.emp). compute bread=(f.emp<4 and m.emp=4). compute higher=(f.emp<m.emp). genlog f.emp m.emp with match /design f.emp m.emp match /print estim. genlog f.emp m.emp /design f.emp m.emp /print estim. crosstabs f.emp by m.emp /cell count col /statistics kappa. ROC risk BY single (1) /PLOT=CURVE(REFERENCE) /PRINT=SE COORDINATES /CRITERIA=CUTOFF(INCLUDE) TESTPOS(LARGE) DISTRIBUTION(FREE) CI(95) /MISSING=EXCLUDE. frequency single. temporary. compute risk1=risk.soc>7. roc risk by risk1(1) /print se coordinate. TWOSTEP CLUSTER /CONTINUOUS VARIABLES=know disaster new /DISTANCE LIKELIHOOD /NUMCLUSTERS AUTO 3 BIC /HANDLENOISE 0 /MEMALLOCATE 64 /CRITERIA INITHRESHOLD(0) MXBRANCH(8) MXLEVEL(3) /VIEWMODEL DISPLAY=y EVALUATIONFIELDS=year . compute group=uniform(1)>.5. sort cases by group. split file by group. * Hierarchical clustering for 2 to 9 clusters. cluster risk risk.soc /method ward /print cluster(2,9) /save cluster(2,9). means risk risk.soc by clu9_1 to clu2_1 /statistics all. * select 3 or more clusters to ensure all squared etas to be above .5. split file off. * K-means clustering using hierrchical clustering for Group 1. quick cluster risk risk.soc /initial (5.6 6.5, 7.2 8.2, 3.7 4.4) /criteria cluster(3) /save cluster(quick1). * K-means clustering using hierrchical clustering for Group 2. * Make sure the matching of the following clusers with previous clusters. quick cluster risk risk.soc /initial (5.2 6.0, 7.2 8.0, 2.9 4.4) /criteria cluster(3) /save cluster(quick2). crosstabs quick1 by quick2 /statistics kappa. * Repeat the previous 3 commands for 4 to 9 clusters, with change in initial centroids. * Need to use some new variable names to represent the clusters. * The optimal result will have the greatest kappa. Descriptive Statistics N Minimum Maximum Mean Std. Deviation a00 132 .00 1.00 .5530 .49907 a 0 a01 132 .00 9.00 5.3712 1.78802 a02 132 2.00 10.00 6.3561 1.92664 a03 132 .00 9.00 5.9242 1.70594 a04 132 .00 8.00 5.1515 1.68280 a05 132 .00 9.00 4.8333 1.78700 a06 132 .00 9.00 5.3788 1.86789 a07 132 2.00 10.00 6.1515 1.50564 a08 131 .00 10.00 5.3130 1.54961 a09 132 1.00 9.00 5.2348 1.73803 a10 132 .00 10.00 6.0076 1.76694 a11 132 .00 10.00 5.4015 1.96869 a12 132 1.00 10.00 5.3106 1.78258 a13 132 .00 10.00 6.1136 1.82706 a14 132 .00 9.00 5.2879 1.65557 a15 132 .00 10.00 5.4394 1.80862 a16 132 .00 10.00 5.9697 1.73838 a17 132 1.00 10.00 6.8258 1.63220 a18 132 1.00 10.00 6.1364 1.72884 a19 131 .00 10.00 5.3893 1.73902 a20 132 3.00 10.00 6.0303 1.74277 a21 132 2.00 9.00 5.6061 1.64274 a22 132 2.00 9.00 5.6970 1.53304 a23 132 2.00 10.00 6.2424 1.79533 a24 132 1.00 9.00 5.7045 1.70199 a25 132 1.00 10.00 5.3939 1.72435 a26 132 2.00 10.00 6.1288 1.66878 a27 132 2.00 10.00 6.7045 1.85233 a28 132 2.00 10.00 6.0303 1.96511 a29 132 .00 10.00 5.0909 2.21189 a30 132 .00 10.00 4.7424 2.39153 a31 132 .00 10.00 3.8106 2.22111 a32 132 .00 10.00 4.7652 2.12890 a33 132 1.00 10.00 5.5985 2.28456 a34 132 .00 10.00 4.9167 2.13313 a35 132 .00 10.00 4.6742 2.14901 a36 132 .00 10.00 5.1894 2.29215 a37 132 .00 10.00 5.2121 2.63572 a38 132 .00 10.00 4.5379 2.78054 a39 131 .00 10.00 6.2061 2.26586 a40 132 .00 10.00 6.1742 2.09870 a41 132 .00 10.00 5.2121 2.22762 a42 132 .00 10.00 5.1894 2.36428 a43 131 .00 90.00 6.0534 7.90162 a44 130 .00 10.00 3.5538 3.15261 a45 131 .00 10.00 4.2290 2.30906 a46 131 .00 10.00 5.6183 2.16069 a47 131 .00 10.00 4.8092 2.09437 a48 131 1.00 10.00 5.1145 1.94399 a49 131 .00 10.00 5.0534 2.24744 a50 131 .00 10.00 5.5725 2.15917 a51 130 .00 10.00 4.4769 2.16102 a52 131 .00 10.00 5.2137 1.79318 a53 131 .00 10.00 5.2824 1.86578 a54 131 .00 10.00 5.3053 1.95288 a55 131 .00 10.00 5.1679 1.99674 a56 130 .00 10.00 6.3538 1.85046 a57 127 .00 10.00 5.5512 1.85895 b 2 5.00 6.00 5.5000 .70711 a58 131 .00 13.00 4.0420 3.17444 a59 131 .00 7.00 .1908 .81443 a60 130 .00 8.00 .1615 1.01006 a61 131 .00 7.00 .1985 .91500 a62 129 .00 100.00 1.4690 9.04843 a63 131 .00 60.00 2.4504 6.27107 a64 131 .00 14.00 .3893 1.61041 a65 131 .00 15.00 .2977 1.46600 a66 130 .00 20.00 .4923 1.93798 a67 131 .00 5.00 .3130 .86899 a68 131 .00 50.00 1.4618 4.85585 a69 129 .00 50.00 4.1744 8.03864 a70 131 .00 36.00 .8092 3.36524 a71 131 .00 30.00 .4656 2.67840 a72 130 .00 10.00 .7538 1.81749 a73 131 .00 50.00 2.7405 6.98745 a74 124 .00 6.00 .4919 1.15115 c 3 .00 5.00 2.0000 2.64575 a75 129 1.00 30.00 1.7054 2.57816 a76 127 .00 4.00 2.0394 .86741 a77 122 .00 27.00 3.3463 4.18570 a78 124 .00 51.00 26.1129 8.66936 a79 123 .00 51.00 23.4976 11.47963 a80 127 .00 44.00 1.9528 3.79402 a81 127 .00 29.00 1.3465 2.52095 a82 126 .00 29.00 .4365 2.64876 a83 124 .00 4.00 3.0242 .93238 a84 125 1.00 7.00 3.3920 1.01518 a85 123 .00 20.00 6.7317 4.47777 a86 123 .00 20.00 6.0163 4.11910 a87 127 .00 4.00 1.2598 .89276 a88 124 .00 5.00 .6290 1.02384 a89 117 165.72 318.95 175.0546 31.12000 after 132 .00 1.00 .3788 .48693 wave2 132 .00 1.00 .3182 .46754 level 128 1.00 3.00 1.4766 .56072 major 126 1.00 4.00 2.0556 .85140 yr 117 1.00 4.00 2.5214 .81577 age 120 20.00 51.00 26.8833 7.66218 stayhk 123 .00 51.00 23.4976 11.47963 sex 125 1.00 2.00 1.6320 .48420 marital 118 1.00 2.00 1.2034 .40424 kid 125 .00 4.00 .2080 .66352 f.emp 123 1.00 4.00 3.0488 .89492 m.emp 123 1.00 4.00 3.3415 .93939 f.edu 123 .00 20.00 6.7317 4.47777 m.edu 123 .00 20.00 6.0163 4.11910 parent 126 .00 2.00 1.2381 .86189 kin 124 .00 5.00 .6290 1.02384 id 117 165.72 318.95 175.0546 31.12000 bachelor 128 .00 1.00 .5547 .49895 master 128 .00 1.00 .4141 .49449 doctor 128 .00 1.00 .0313 .17468 psy 126 .00 1.00 .2778 .44969 swk 126 .00 1.00 .4444 .49889 soc 126 .00 1.00 .2222 .41740 counsel 126 .00 1.00 .0556 .22998 crime 126 .00 .00 .0000 .00000 f.boss 123 .00 1.00 .0813 .27441 f.sole 123 .00 1.00 .1301 .33777 f.hird 123 .00 1.00 .4472 .49923 f.home 123 .00 1.00 .3415 .47614 m.boss 123 .00 1.00 .0976 .29793 m.sole 123 .00 1.00 .0325 .17810 m.hird 123 .00 1.00 .3008 .46049 m.home 123 .00 1.00 .5691 .49723 female 125 .00 1.00 .6320 .48420 control 132 1.67 9.67 5.2361 1.41811 fear 132 2.00 9.33 5.5934 1.47257 respon 132 1.00 8.00 4.9394 1.42546 know 132 2.00 8.67 5.2576 1.26785 disaster 132 2.00 9.67 5.9848 1.51546 new 132 1.67 8.00 5.3030 1.35006 detect 132 1.33 8.00 5.1692 1.36880 unclear 132 2.00 8.67 5.7626 1.43832 risk.soc 132 2.00 9.33 6.2475 1.63228 risk 132 2.00 9.33 5.5682 1.62034 married 118 .00 1.00 .2034 .40424 single 118 .00 1.00 .7966 .40424 divorce 118 .00 .00 .0000 .00000 widow 118 .00 .00 .0000 .00000 Valid N (listwise) 0 TWOSTEP CLUSTER /CONTINUOUS VARIABLES=risk risk.soc /DISTANCE LIKELIHOOD /NUMCLUSTERS AUTO 15 BIC /NOSTANDARDIZE VARIABLES=risk.soc risk /HANDLENOISE 0 /MEMALLOCATE 64 /CRITERIA INITHRESHOLD(0) MXBRANCH(8) MXLEVEL(3) /VIEWMODEL DISPLAY=YES. twostep cluster /continuous var=risk risk.soc age /print ic count summary. /print ic count summary. /save cluster var= c2. /NOSTANDARDIZE VARIABLES=risk.soc risk /HANDLENOISE 0 /MEMALLOCATE 64 /CRITERIA INITHRESHOLD(0) MXBRANCH(8) MXLEVEL(3) /VIEWMODEL DISPLAY=YES. sort cases by year a90 a89. list var= year a90 a89.