Exercise 6. Attitude toward gay marriage and presidential vote
Another issue that might have influenced how people voted in the 2004 presidential election is the issue of whether marriage between gays should be permitted. This had not been a salient issue in previous presidential elections, but it was in 2004. Bush clearly opposed gay marriage, and he supported a constitutional amendment to prohibit states from allowing it. Kerry also said that he opposed gay marriage, but he did not favor amending the constitution to prevent states from legalizing it. To examine whether attitudes on gay marriage affected voting, we can look at a table that relates attitude on this issue (V125) to presidential vote. For the reasons suggested in Exercise 1, you should use the recoded version of V002 that you created for that exercise, so that you examine only the major-party vote (i.e., only the Bush and Kerry voters).
After examining Table 6A, you should conclude that those who favored legalizing gay marriage had a far greater propensity to for Kerry than those who opposed gay marriage. However, as we learned from Exercise 5, this does not necessarily mean that attitudes on this issue really had a significant effect on the vote. The relationship in Table 6A could be the result of the actions of an extraneous or confounding variable.
As in Exercise 5, party identification is a possible extraneous or confounding variable, which should be examined to better understand why our independent and dependent variables are related. To do so, one needs to construct a three-variable table that shows the relationship between attitude toward gay marriage, presidential vote, and party identification. To ensure that you have a sufficient N for each column, you should recode V125 so that it has just two categories (favor and oppose) and use the recoded version of party identification that you created for exercise one (Democrats, independents, and Republicans).
In this example, the relationship between attitude toward gay marriage and presidential vote is weakened once we control for party identification, but the two variables remain associated even with this control. This suggests that party identification is a confounding variable that inflated the relationship between attitude gay marriage and vote, but it still appears that attitudes on this issue had some effect on how people voted.