SPSS ASSIGNMENT DUE TUESDAY 04/24/18 BY 11:59PM NEW YORK TIME
fhatliesflores
SPSSInstructions_SantaFeGrill_Essn2e_11-28-20101.docPAGE
15
USing SPSS with the santa fe grill database
The SPSS software package is very user-friendly and enables you to easily learn the various statistical techniques without having to use formulas and calculate the results. The approach is a simple Windows-based “point-and-click” process. In this appendix, we provide a brief overview of how to use the package and click-thru sequences for the techniques you will be using. This will be a quick reference point for you to refresh your memory on how to run the various techniques.
When you run the SPSS software, you will see a screen like that in Exhibit 1. The case study database is available from our website at www.mhhe.com/hairessentials3e or from your instructor. You will note the columns are blank because the data has not been entered into the SPSS software.
When you load SPSS a screen labeled in the top left-hand corner Untitled – SPSS Data Editor should be visible in the background. In the foreground is a dialog box called SPSS for Windows Student Version. If you have never run SPSS you will have to tell the program where to find the data. If you have previously run the SPSS program you can simply highlight to location of the database and click on OK at the bottom of the screen. The SPSS Data Editor screen without the dialog box in the foreground is shown in Exhibit 1.
Exhibit 1 SPSS Data Editor Window with No Data
Across the top of the screen is a toolbar with a series of pull down menus. Each of these menus leads you to several functions. An overview of these menu functions is shown below.
MENUS
There are 11 “pull-down” menus across the top of the screen. You can access most SPSS functions and commands by making selections from the menus on the main menu bar. Below are the major features accessed from each of the menus on the Student Version 12 of the SPSS software.
File = create new SPSS files; open existing files; save a file; print; and exit.
Edit = cut and/or copy text or graphics; find specific data; change default options such
as size or type of font, fill patterns for charts, types of tables, display format for
numerical variables, and so forth.
View = modify what and how information is displayed in the window.
Data = make changes to SPSS data files; add variables and/or cases; change the
order of the respondents; split your data file for analysis; and select specific
respondents for analysis by themselves.
Transform = compute changes or combinations of data variables; create new
variables from combinations of other variables; create random seed
numbers; count occurrences of values within cases; recode existing
variables; create categories for existing variables; replace missing
variables; and so on.
Analyze = prepare reports; execute selected statistical techniques such as
frequencies, correlation and regression, factor, cluster, and so on.
Graphs = prepare graphs and charts of data, such as bar, line and pie charts; also
boxplots, scatter diagrams and histograms.
Utilities = information about variables such as missing values, column width,
measurement level and so on.
Add-ons = other functions such as Missing Values Analysis and AMOS.
Window = minimize windows or move between windows.
Help = a brief tutorial of how to use SPSS; includes a link to the SPSS home page
at www.spss.com.
ENTERING DATA
There are two ways you can enter data into SPSS files. One is to enter data directly into the Data Editor window. This can be done by creating an entirely new file or by bringing data in from another software package such as Excel. The other is to load data from a file that has been created in another SPSS application.
Let’s begin with explaining how to enter data directly into the Data Editor window.
The process is similar to entering data into a spreadsheet. The first column typically is used to enter a respondent ID. Use this to enter a respondent number for each response. The remaining columns are used to enter data. You can also ‘cut and paste’ data from another application. Simply open the Data Editor window and minimize it. Then go to your other application and copy the file, return to the Data Editor window and paste the data in it, making sure you correctly align the columns for each of the variables.
Now let’s talk about how to load a previously created SPSS file, such as the one that comes with your text. Load the SPSS software and you should see an Untitled SPSS Data Editor screen. Click on the ‘Open File’ icon and you will get an Open File dialog box. Click on “Look in” to indicate where to look for your file. For example, look on your CD or other storage device. This will locate your SPSS files and you should click on the Santa Fe Grill survey. This will load up your file and you will be ready to run your SPSS analysis.
DATA VIEW
When you load up your SPSS file it will show the Data View screen. Exhibit 2 shows the Data View screen for the Santa Fe Grill survey. This screen is used to run data analysis and to build data files. The other view of the Data Editor is Variable View. The Variable View shows you information about the variables. To move between the two views go to the bottom left-hand corner of the screen and click on the view you want. We discuss the Variable View screen in the next section.
Exhibit 2 Data View of the Santa Fe Grill Database
The survey database is set up in columns. The first column on the far left labeled “id” is a unique number for each of the 405 respondents in your database. The remaining columns are the data from the interviews conducted at the restaurant. In the first 4 columns to the right of the id you have the values for the four screening questions. Then, you have the first eleven variables of the survey – the lifestyle variables (X1 – X11). For example, respondent 1 gave the Santa Fe Grill a “7” on the 7-point scale for the first variable (X1). Similarly, that same respondent rated the restaurant a “4” on the second variable (X2) and a “5” on the third one (X3). Exhibit 2 only shows the id, the four screening variables and the first four variables of the survey. But on your SPSS screen if you scroll to the right you will see the data for all of the survey variables.
VARIABLE VIEW
Exhibit 3 shows the Variable View screen for the Santa Fe Grill survey. In this view the variable names appear in the far left-hand column. Then each of the columns defines various attributes of the variables a described below:
Name = This is an abbreviated name for each variable.
Type = The default for this is numeric with 2 decimal places. This can be changed to express values as whole numbers or it can do other things such as specify the values as dates, dollar, custom currency and so forth. To view the options click first on the Numeric cell and then on the three shaded dots to the right of the cell.
Label = In this column you give a more descriptive title to your variable. For example, with the Santa Fe Grill survey variable X1 is labeled as X1 – Try New and Different Things and variable X2 is labeled as X2 – Party Person. When you have longer labels and want to be able to see all of them you can go to the top of the file and click between the Label and Values cells and make the column wider.
Values = In the values column you can assign a label for each of the values of a variable. For example, with the Santa Fe Grill survey data variable X1 –Try New and Different Things we have indicated that a 1 = Strongly Disagree and a 7 = Strongly Agree. To view the options click first on the Values cell and then on the three shaded dots to the right of the cell. You can add new labels or change existing ones.
Missing = Missing values are important in SPSS. If you do not handle them properly in your database it will cause you to get incorrect results. Use this column to indicate values that are assigned to missing data. A blank numeric cell is designated as system-missing and a period (.) is placed in the cell. The default is no missing data but if you have missing data then you should use this column to tell the SPSS software what is missing. To do so, you can record one or more values that will be considered as missing data and will not be included in the data analysis. To use this option, click on the Missing cell and then on the three shaded dots to the right. You will get a dialog box that shows the default of no missing data. To indicate one or more values as missing click on Discrete missing values and place a value in one of the cells. You can record up to three separate values. The value most often used for missing data is a ‘9’. If you want to specify a range of values click on this option and indicate the range to be considered as missing.
Column = Click on the column cell to indicate the width of the column. The default is 8 spaces but it can be increased or decreased.
Align = The default for alignment is initially left, but you can change to either center or right alignment.
Let’s look at the Variable View screen for the Santa Fe Grill database. It is shown in Exhibit 3. To see the Variable View screen go to the bottom left-hand corner of the screen and click on “Variable View.” The name of the variable will be in the first column, but if you look at the fifth column it will tell you more about the variable. For example, variable X1 is “Try New and Different Things” while X2 is “Party Person.” All of the remaining variables have a similar description. Also, if you look under the Values column it will tell you how the variable is coded; e.g., 1 = Strongly Disagree and 7 = Strongly Agree.
Exhibit 3 Variable View of the Santa Fe Grill Survey Data
Click-thru sequences for selected procedures
Mean, Median and Mode – Measures of Central Tendency
The SPSS “click-through” sequence is ANALYZE ( DESCRIPTIVE STATISTICS ( FREQUENCIES. Let’s use X25 – Frequency of Eating at ? as a variable to examine. On the left side of the Frequencies box on the screen is a list of the survey variables. Scroll down and click on X25 to highlight it, and then on the arrow box to move X25 into the Variables box to. Next click on the Statistics box and then click on Mean, Median, and Mode. Now click Continue and OK to get the results. See the results on next page.
If you want to do a chart, then look at the bottom of the Frequencies box next to the Statistics box and click on Charts. If you select bar charts, and continue, and then OK you will get a chart too.
Range, Standard Deviation and Variance
The Santa Fe Grill database can be used with the SPSS software to calculate measures of dispersion, just as we did with the measures of central tendency. The SPSS click-through sequence is ANALYZE ( DESCRIPTIVE STATISTICS ( FREQUENCIES. Let’s use X22 – Satisfaction as a variable to examine. Click on X22 to highlight it and then on the arrow box to move X22 to the Variables box. Next open the Statistics box, go to the Dispersion box in the lower-left-hand corner, and click on Standard deviation, Variance, Range, Minimum and Maximum. Now click Continue and OK to get the results.
If you want to do a chart, then look at the bottom of the Frequencies box next to the Statistics box and click on Charts. If you select bar charts, and continue, and then OK you will get a chart too.
Cross-Tabulation
To generate the Cross-tab table on your book page 286 Exhibit 11.9, the click-through sequence is ANALYZE ( DESCRIPTIVE STATISTICS ( CROSSTABS. Highlight the dependent variable X31 – AD RECALL by clicking on it and move it to the Row(s) box. Next, highlight X32 – Gender and move it to the Column(s) box. Now click on the Statistics box and then click on Chi-square, and click Continue. Then click Cells tab which is underneath Statistics, check boxes beside Observed, Expected, within Counts window, and boxes beside ROW, COLUMN, and TOTAL in the Percentages window. Finally click Continue and OK to get the results.
Compare Two Means of Independent Samples
The click-through sequence is ANALYZE ( COMPARE MEANS ( INDEPENDENT-SAMPLES T-TEST. Then click variable X22-Satisfaction into the Test Variables box and variable X32-Gender into Grouping Variable Box. For variable X32 you must define the range in the Define Groups box. Enter a 0 for Group 1 and a 1 for Group 2 (males were coded 0 in the database and female were coded 1) and the click Continue. For the “Options” we will use the defaults, so just click OK to execute the program.
Compare Means Using Paired Sample t-Test
The click-through sequence is ANALYZE ( COMPARE MEANS ( PAIRED-SAMPLES T-TEST. Highlight variable X18-Food Taste and click the arrow button, and then highlight X20-Food Temperature and click on the arrow button to move them into the Paired Variables box. For the “Options” we will use the defaults, so just click OK to execute the program.
Compare Two Means Using ANOVA
The click-through sequence is ANALYZE ( COMPARE MEANS ( One-Way ANOVA. Highlight the dependent variable X24 – Likely to Recommend by clicking on it and move it to the Dependent List box. Next, highlight X32 – Gender and move it to the Factor window. Now click on the Options box and then click Descriptives and Continue. Then click OK to get the results.
This shows you if males or females are more likely to recommend the Santa Fe Grill. See results below.
Now try this one more time. Return to the One-Way ANOVA box. Keep the same variable in the Dependent List box. Next, highlight and remove the gender variable from the Factor window. Then, highlight X_s4 – Mexican Restaurant Most Recently Eaten At . . . and move it into the Factor window. Then click OK to get the results – which compare the Santa Fe Grill customers with Jose’s customers.
Compare Three or More Means Using ANOVA
The click-through sequence is ANALYZE ( COMPARE MEANS ( One-Way ANOVA. Highlight the dependent variable X23 – Likely to Return and move it into the Dependent List window. Next, highlight X30 – Distance Driven and the click on the arrow button to move it into the Factor window. Now click on the Options box and then click Descriptives and Continue. Then click OK to get the results.
Follow-up Tests for ANOVA:
A weakness of ANOVA is that the test enables the researcher to determine only that statistical differences exist between at least on pair of the group means. The technique cannot identify which pairs of means are significantly different from each other. To answer this question, we must apply follow-up “post-hoc” tests. These tests will identify the pairs of groups that have significantly different mean groups. The Scheffé procedure is one of these follow-up tests and it is a more conservative methods relative to other follow-up tests, to detect the significant differences between group means.
To run the Scheffé post-hoc test we use the SPSS compare means test. The click-through sequence is ANALYZE ( COMPARE MEANS ( One-Way ANOVA. Then highlight X23-Likely to Return and move it into the Dependent List Window. Next highlight X30-Distance Driven and then click on the arrow button to move it into the Factor window. Now click on the Options tab and then on Descriptive. Next click on the Post Hoc tab and then on Scheffé. Then click Continue and OK to get the post-hoc results.
n-Way ANOVA
If the Santa Fe Grill restaurant owners want to know first whether customers who come to the restaurant from greater distances differ from customers who live nearby in their willingness to recommend the restaurant to a friend. Second, they also want to know whether that difference in willingness to recommend the restaurant, if any, is influenced by the gender of the customers. The database variables are X24-Likely to Recommend, measured on a 7-point scale, with 1 = “Definitely Will Not Recommend” and 7 = “Definitely Recommend”, X30 – Distance Driven, where 1 = “Less than 1 mile,” 2= “1-3 miles”, and 3 = “More than 3 miles”, and X32–Gender, where 0 = male and 1 = female.
On the basis of informal comments from customers, the owners believe customers who come from more than 5 miles will be more likely to recommend the restaurant. Moreover, they hypothesize male customers will be more likely to recommend the restaurant than females. The null hypotheses are that there will be no difference between the mean ratings for X24-Likely to Recommend for customers who traveled different distances to come to restaurant (X30), and there also will be no difference between females and male (X32).
To examine this hypothesis, we first need to separate out the Santa Fe sample. To select the Santa Fe Grill responses (253) from the total sample of 405 responses, the click-through sequence is DATA ( SELECT CASES. First click on the Data pull-down menu and scroll down and highlight and click on Select Cases. You now see the Select Cases dialog box where the default is All cases. Click on the If condition is satisfied option, and then on the If tab. Next highlight the screening variable X-s4 and click on the arrow box to move it into the window. Then click on the equal sigh (=) and 1 for the Santa Fe Grill customers. Now click Continue and then OK and you will be analyzing only the Santa Fe Grill customers. Thus your output will have the results for only one restaurant. Remember too that all data analysis following this change will analyze only the Santa Fe customers. To analyze the total sample you must follow the same sequence and click again on All cases.
After selecting the Santa Fe sample, the click-through sequence is ANALYZE ( GENERAL LINEAR MODEL ( UNIVERIATE. Highlight the dependent variable X24-Likely to Recommend by clicking on it and move it to the Dependent Variable box. Next, highlight X30-Distance Driven and X32-Gender, and move them to the Fixed Factors box. Now click on the Options box on the right side of the SPSS screen and look in the Estimated Marginal Means box and highlight (OVERALL) as well as X30, X32, and X30 * X32 and move them all into the Display Means for box. Next place a check on “Compare main effects” and Descriptive Statistics. Then click Continue and finally click OK.
Pearson Correlation
The SPSS click-through sequence is ANALYZE ( CORRELATE ( BIVARIATE, which leads to a dialog box where you select the variables. Transfer variables X22 and X24 into the Variables box. Note that we will use all three default options shown below: Pearson correlation, two-tailed test of significance, and flag significant correlations. Next go to the Options box, and after it opens click on Means and Standard Deviations and then continue. Finally, when you click on OK at the bottom it will execute the Pearson correlation.
This correlates satisfaction with likely to recommend.
Multiple Regression
The SPSS click-through sequence to examine this relationship is ANALYZE ( REGRESSION ( LINEAR. Highlight X22 and move it to the Dependent Variables box. Highlight X12, X15, X16 and X17 and move them to the Independent Variables box. We will use the defaults for the other options so click OK to run the multiple regression.
This regression tells you which of the four restaurant perceptions variables used in the regression as independent variables can predict satisfaction. X14 (fresh food) is the best predictor and x12 (friendly employees is the second best predictor.
The conceptual model for the multiple regression is below.
Satisfaction
Attractive
Interior
Reasonable Prices
Fresh Food
Friendly
Employees
