Race, Republicans, and the Return of the Party of Lincoln

5. Seeing Is Believing?: Reactions to the 2000 Republican National Convention

5 Seeing Is Believing? Reactions to the 2000 Republican National Convention THE THEORETICAL MODEL of party image change depicted in ‹gure 2 illustrates that reshaping partisan stereotypes is a function in part of individuals’ predispositions and the media’s framing of the party. The results presented in chapter 3 suggest that individuals vary in their perceptions of political parties: African Americans’ pictures of the two parties’ images along racial lines are more crystallized than those of whites. Furthermore, blacks’ threshold for what constitutes change is also set higher than that of whites. Thus, the Republican Party’s ability to reshape its image with respect to race should prove more dif‹cult among blacks than whites. Likewise, reshaping the Republican Party’s racial symbolism should also be harder when individuals encounter media messages that highlight aspects of the party that have not changed or contain no discussion of the elements of the party that have been altered. The data presented in chapter 4 show that citizens might have stumbled onto several alternative versions of the 2000 Republican National Convention. People could have read a basic account of the convention that included a description of the black attendees, media coverage that completely omitted mention of blacks at the convention, or an account that juxtaposed the attendance of many black speakers and entertainers with the fact that the party did not change is traditional conservative platform. Therefore, the media had the potential to serve as an obstacle to the Republican Party’s attempt to modify its image with respect to race. Chapter 5 seeks empirically to test these relationships. First, using survey data, I gauge general reactions to the 2000 Republican National Convention as well as the reactions of whites and blacks separately. 103 The data used in these analyses come from the Post–GOP Convention Poll, which was conducted by the Gallup organization shortly after the close of the convention. This poll seemed particularly well suited to test the correlation between convention exposure and perceptions of the Republican Party. First, the Post–GOP Convention Poll oversampled African Americans, allowing me to examine black/white differences in reactions to the convention. Second, the Gallup poll was conducted almost immediately after the convention, minimizing the possibility that exogenous events could explain changes in perceived Republican Party images. Third, any other events that occurred between the convention and the survey would have muted rather than ampli‹ed the effect. Finally, there was no reason to believe that convention watchers and non–convention watchers would be affected differently by some intervening occurrence. Still, using survey data rather than experimental data poses a potential measurement problem. In the real world, people self-select themselves into watching the convention. As a result, other unmeasured motivating factors may in›uence both convention watching and evaluations of the Republican Party. To overcome this problem inherent to using survey data, I conducted the 2002–2003 Party Image Study, which incorporated an experiment into its design. In an experiment, the researcher can control who is exposed to the treatment—in this case the convention—and who is not. Further, an experiment allows the researcher to control what type of information people receive about the convention. The 2002–2003 Party Image Study was, of course, conducted a few years after the convention, and numerous events occurred in the interim, including the contested 2000 election and the 2001 terrorist attacks on New York City and Washington, D.C. Consequently, convention events should be displaced in citizens’ minds by other more recent salient activities of the Republican Party. Therefore, the ability to observe an effect of the experimental manipulations should be more dif‹cult. Nevertheless, I used data from the 2002–2003 Party Image Study to establish the causal link between convention exposure and perceptions of the Republican Party’s image with respect to race. Moreover, I utilized these data to determine how subtle changes in the media’s interpretation of the convention might have moderated individuals’ reactions to the convention. To do so, I replicated some of the frames used by the news media when describing the convention. In the experiment, subjects were 104 Race, Republicans, & the Return of the Party of Lincoln exposed to one of four scenarios: (1) a control with no mention of the Republican convention; (2) a race-neutral description of the Republican convention with no references to the presence of African American attendees; (3) a description of the...