Understanding and Applying Research Design

7

CHI-SQUARE AND CONTINGENCY TABLE ANALYSIS¹

In a previous section, we introduced and discussed some examples of using General Social Survey (GSS) data to help test whether or not hypotheses about relationships among variables are supported. Figure 2.2 shows how voting behavior differs by social class identification, for example. This analysis made use of a contingency table, which displays data in rows and columns so that you can identify potential patterns in the relationships. As we saw in Figure 2.2, the analysis revealed a pattern whereby a greater percentage of those in middle and upper classes voted versus those in lower and working classes.

In this section, we take a closer look at contingency tables and some statistical measures that can help us to have a clearer picture of the extent of the relationships among study variables. First, we learn how to set up a contingency table, and then we use SPSS to see whether or not the data reveal meaningful patterns of relationships.

CONTINGENCY TABLES

Contingency tables are created by presenting data in rows and columns according to the independent and dependent variables of a study. In this way, the researcher can easily see how the data are arrayed across the categories of the variables. These tables of data containing the frequencies are called contingency tables because the data in the row cells are contingent upon or are connected to the data in the column cells (independent variable). Statisticians and researchers often refer ...