Sociology 2205A/B Chapter Notes - Chapter 13: Contingency Table, Scatter Plot, Dialog Box
Document Summary
Up until now, we dealt with one variable at the time we did univariate descriptives. The second step in descriptives is to look at pairs of variables at a time. Often our research question asks how one variable affects or predicts another for instance, how political affiliation predicts voting, or how being unemployed influences smoking behaviors, or how men and women differ in their civic participation. The last question is phrased differently but we could also say that it asks how gender predicts civic participation. We will want to see how the two variables are related in the sample. As with univariate descriptives, the method for bivariate descriptives will depend on whether the variables are continuous or categorical. I am sure you do: frequency tables and bar graphs. With pairs of categorical variables, we just expand the tables and bar graphs to show how the two variables are distributed together (show their joint distribution).