# SPSS Exercises

For these exercises, you will be using the MIDUS 2 (Biomarker Project, 2004–2009) data set, which is available on the study site. You can download and open the data set in both SPSS and the online analysis tool directions using the same instructions from Chapter 1 SPSS exercises.

SPSS

1. The MIDUS 2 study is unique in that it includes an experimental condition in which participants are subjected to different “conditions” of cognitive stress. In this exercise, let’s see whether random assignment to the two experimental conditions resulted in two groups with equivalent characteristics.
2. Since we are only looking at the Math and Stroop conditions, we need to filter our cases. Go to DATA > SELECT CASES > Click the “If” button > Select the B4VCS1 variable in the left box, and then click the transfer arrow button to move it to the right box. Click the "=" button and then type “1”; now also type “OR B4VCS1 = 2” and then Click OK.
3. Now that we have filtered our cases to the treatment condition we wish to examine, let’s look at some group mean differences for this variable, such as age and gender:
1. Go to Analyze > Compare Means > Means
2. Select B4VCS1 as the Grouping Variable
3. Define Groups as 1 and 2
4. Select B4ZAGE (age) and B1PGENDER (gender) as the Test Variable(s)
5. Click OK
4. What are the means for age and gender in the two groups? Are they similar? Do they differ at a level that is statistically significant (note the “Sig. (2-tailed)” value for the “Independent Samples T Test)?” (Of course, since Gender is a dichotomy, the “mean” value corresponds to the proportion of cases that are in the category with the code of 2 rather than 1, after subtracting 1).

Online Analysis

1. The MIDUS 2 study is unique in that it includes an experimental condition in which participants are subjected to different “conditions” of cognitive stress. In this exercise, let’s see whether random assignment to the two experimental conditions resulted in two groups with equivalent characteristics.
1. At the top of the page, click on Analysis > Comparison of means.
2. Type B4VCS1 into the dependent variable box, and B4ZAGE (age) and B1PGENDER (gender) into the box labeled “row”
3. Under CHART OPTIONS, select “no chart” and check the “show means” box
4. Since we are only looking at the Math and Stroop conditions, we need to filter our cases using the “selection filters” option. Type B4VCS1(1-2) into the box labeled “selection filter(s)”
5. Run the table.
2. What are the means for age and gender in the two groups? Are they similar? Do they differ at a level that is statistically significant (note the “Sig. (2-tailed)” value for the “Independent Samples T Test)?” (Of course, since Gender is a dichotomy, the “mean” value corresponds to the proportion of cases that are in the category with the code of 2 rather than 1, after subtracting 1).