If you have been using the GSS2016x or GSS2016x_reduced data set for the SPSS exercises, you will now need to download the limited data sets designed for this chapter only: GSS2016y or GSS2016y_reduced (for the Student Version of SPSS). Please do this before proceeding. If you have been using the GSS2016 data set, you can just continue to use it.
1. Develop a description of the basic social and demographic characteristics of the U.S. population in 2016. Examine each characteristic with three statistical techniques: a graph, a frequency distribution, and a measure of central tendency (and a measure of variation, if appropriate).
a. From the menu, select Graphs and then Legacy Dialogs and Bar. Select Simple Define [Marital—Category Axis]. Bars represent % of cases. Select Options (do not display groups defined by missing values). Finally, select Histogram for each of the variables [EDUC, EARNRS, INTWKDYH, ATTEND] from the GSS2016 data set.
b. Describe the distribution of each variable.
c. Generate frequency distributions and descriptive statistics for these variables. From the menu, select Analyze/Descriptive Statistics/Frequencies. In the Frequencies window, select MARITAL, EDUC, EARNRS, INTWKDYH, ATTEND, moving each to the Variable(s) window. Then choose Statistics, and select the mean, median, range, and standard deviation. Then choose Continue (and then OK).
d. Collapse the categories for each distribution. Be sure to adhere to the guidelines given in the section “Grouped Data.” Does the general shape of any of the distributions change as a result of changing the categories?
e. Which statistics are appropriate to summarize the central tendency and variation of each variable? Do the values of any of these statistics surprise you?
2. Try describing relationships with support for capital punishment by using graphs. Select two relationships you identified in previous exercises and represent them in graphic form. Try drawing the graphs on lined paper (graph paper is preferable).
3. The GSS2016 data set you are using allows you to easily replicate the tables in this chapter. Try doing that. The computer output you get will probably not look like the tables shown here because I reformatted the tables for presentation, as you should do before preparing a final report. At this point, I’ll let you figure out the menu commands required to generate these graphs, frequency distributions, and cross-tabulations. If you get flummoxed, review my instructions for the SPSS exercises in earlier chapters or go to the book study site and review my tutorials on SPSS.
4. Propose a variable that might have created a spurious relationship between income and voting. Explain your thinking. Propose a variable that might result in a conditional effect of income on voting, so that the relationship between income and voting would vary across the categories of the other variable. Test these propositions with three-variable cross-tabulations. Were any supported? How would you explain your findings?