In the previous chapter we discussed the analysis of change over time. The data for this module include information from three samples of adults living in California. The earlier sample was collected in 1985, while the later samples were collected in 1991 and 1995. In this chapter you will use crosstabulation (described in Chapter Three) to analyze trends over time. The exercises compare these cross-sectional surveys at three points in time. This is called a trend study.
EXERCISE ONE One of the questions asks respondents if they agree or disagree that "despite recent gains by women, when all is said and done, it is still essentially a man's world." We want to find out if there has been change in public opinion from 1985 to 1991 and 1995 and which types of individuals were more (or less) likely to change.
Question 1. Find this question in the codebook (Appendix A). Locate the variable name for this question. This is the name the computer recognizes. You will have to use this name when asking the computer to do something for you. What is the variable name?
Question 2. What percent agree strongly and agree somewhat with this statement in 1985? in 1991? in 1995? Let's use crosstabulation to get these figures. You will have to crosstabulate V8 by YEAR to do this. Be sure to ask for the column percents and chi square. What does the chi square allow you to say about the difference between these percentages?
Question 3. Let's combine agree strongly with agree somewhat and also combine disagree strongly with disagree somewhat so that we have only two categories--agree and disagree. You will have to use a recode command to do this. (Your instructor will show you how to do this.) Then crosstabulate V8 by YEAR again. Describe changes from 1985 to 1991 and 1995. Is this table clearer than the previous one?
Question 4. We want to discover which types of people are more likely to change. Let's start by asking if men or women are more likely to change. You will have to crosstabulate V8 by YEAR by V34 (sex), asking for the column percents and chi square. Write a short paragraph describing changes for men and for women. Be sure to cite the appropriate percent differences and use chi square in your analysis.
Question 5. Now, find out if younger, middle-aged, or older respondents are more likely to change and if those with less education are more likely to change than those with more education. Be sure to use the age and education variables which have already been recoded. Write a short paragraph describing the results.
EXERCISE TWO Another question asks respondents if they agree or disagree that "it's more important for a wife to help her husband's career than to have a career." We want to find out if public opinion has changed from 1985 to 1991 and 1995 and which individuals were more (or less) likely to change.
Question 1. Find this question in the codebook (Appendix A). What is the variable name?
Question 2. What percent agree strongly and agree somewhat with this statement in 1985? in 1991? in 1995? Use crosstabulation to get these figures. You will have to crosstabulate V15 by YEAR to do this. Be sure to ask for the column percents and chi square. What does chi square allow you to say about the difference between these percentages?
Question 3. In this exercise, we want to recode so we have only two categories--agree strongly and not agree strongly. Combine agree somewhat with disagree somewhat and disagree strongly so that we have only these two categories. Remember that you will have to use a recode command to do this. Now crosstabulate V15 by YEAR again. Is this table clearer than the previous one?
Question 4. We want to discover which types of people are more likely to change. Specifically, if men or women are more likely to change, if younger, middle-age, or older respondents are more likely to change, and if those with less education are more likely to change than those with more education. Write a short paragraph describing the changes for these groups of individuals. Be sure to cite the appropriate percent differences and use chi square in your analysis.
EXERCISE THREE
Select one of the other opinion variables that was included in all of the surveys. Analyze the changes in opinion for this variable. Look at both the overall changes for the entire sample and the changes for subgroups (e.g., men and women, younger and older). Use chi square in your analysis. Write a brief report explaining what you did and describing the results. Include your tables in the report.
