Spring 2021
Class Project VI: Memorandum on Bivariate Data Analysis Part I
Instructions:
Somewhat recent research (see Payne et al., 2004; Unnever & Cullen, 2007) has examined demographic correlates of support for capital punishment, across a variety of contexts. These studies have found that, among others, two of the more salient predictors of support for capital punishment are sex and race. While certainly making consequential contributions to the literature on punitiveness in general and capital punishment in particular, these studies have some limitations. (The former study relies on a nonrepresentative sample, and the latter study employs the use of data over fifteen years old.)
In this project, practicum students will address both of the aforementioned limitations of previous studies in this area by employing recent (2018), nationally-representative data to explore demographic correlates of attitudes toward capital punishment. As a student of a methods/stats course in the social sciences you are well-positioned to investigate this matter by employing the most robust source for trend data on American attitudes across the social/political spectrum: The 2018 General Social Survey
The 2018 GSS contains a host of variables that will allow researchers to assess the null hypothesis that there is no relationship between demographic variables (in this case, sex and race) and support for capital punishment. In particular, you are to investigate the potential link between sex and support for capital punishment (are there any differences between males and females in support for the death penalty), and race and support for capital punishment (are there any differences between whites and nonwhites in their support for the death penalty?). It just so happens that you have recently acquired the requisite skills to properly assess if these relationships are in fact significant – and if so, what is the direction of the relationship (e.g., are males and whites more or less likely than females and nonwhites to support capital punishment?)?
There are three variables that students will use in a condensed version of the 2018 GSS – REDUCED 2018 GSS — that I have created specifically for this project. While I do not specify the independent and dependent variables in this analysis, I would like to think that after nearly one year of research methodology, students should be able to figure this one out on their own. The three variables, which are provided near the bottom of the dataset (variable view in SPSS) are:
- DEATH: dummy variable for support for capital punishment for murder (0 = “OPPOSE”, 1 = “FAVOR”).
- MALE: dummy variable for sex (0 =”FEMALE”, 1 = “MALE”).
- WHITE: dummy variable for white (0 = “NONWHITE”, 1 = “WHITE”).
Given the level of measurement of these three measures (students should know levels of measurement by now, but keep in mind that all three are measured on the same level), your instructor thinks it best if you employ the use of the crosstabulations procedure in SPSS (analyze/descriptive statistics/crosstabs). Based on your analyses, students should answer the following questions in full.
- What are the independent variables in this study? The dependent variable?
- What level of measurement are the variables, DEATH, MALE, and WHITE?
- Are these three variables discrete or continuous? Why?
- After performing the relative analysis (Hint: crosstabulation first between MALE and DEATH) answer the following questions as specifically as possible.
- Is there a significant association between the variables MALE and DEATH — i.e., is there a significant association between a respondent’s sex and their support for capital punishment?
- What information (statistic) did you use to make this determination of statistical significance?
- After viewing the crosstabulation table, what appears to be the relationship between sex and attitudes toward capital punishment – i.e., who seems to demonstrate more or less support (be specific)?
- Students now should observe the association between race and attitudes toward capital punishment, to observe if there is any significant relationship between the two. (Hint: crosstabulation between WHITE and DEATH) Answer the following questions as specifically as possible.
- Is there a significant association between the variables WHITE and DEATH — i.e., is there a significant association between a respondent’s race and their support for capital punishment?
- What information (statistic) did you use to make this determination of statistical significance?
- After viewing the crosstabulation table, what appears to be the relationship between race and attitudes toward capital punishment – i.e., who seems to demonstrate more or less support for the death penalty (be specific)?
- After performing the TWO crosstabulations, discuss your broad takeaway regarding the demographic correlates of support for capital punishment, paying specific attention to the following issues.
- Are males more or less supportive of capital punishment than females? Are whites or nonwhites more supportive of capital punishment?
- Discuss the strength of these demographic correlates in their ability to predict attitudes toward the death penalty.
- Based on the results of the crosstabulations, when it comes to predicting how an individual feels on the issue of capital punishment, which variable appears to be more important in determining an individual’s attitudes – sex or race? Use the results from the tables to make your case here (this question will be graded closely).
- Can we determine causality from these bivariate results? Why or why not?
- In your estimation, what are some other variables that might predict one’s attitudes toward capital punishment. Explain your choices.
- Is this the best way to assess attitudes toward capital punishment and punitiveness in general?
- Are there limitations with the data used?
- Are there limitations with the manner in which these variables were operationalized?
There will be two separate assignment drop boxes created in Canvas: syntax drop box, and Class Project VI drop box. Students should include the syntax files for all procedures used.
Format:
While there is no specific length/page requirement to this assignment, students are asked to answer the questions comprehensively. Assignments MUST be typed, with 1” margins throughout the document. All students must use 12-point, Times New Roman Font. Students should single-space within each answer (including the letters – e.g., 1a, 1b), but double space between answers (e.g., 1, 2, 3).
Due Date: