The Effectiveness of Implicit and Explicit Racial Appeals in a “Post-Racial” America.

Abstract

This dissertation examines various racial appeals used by black and white candidates, Democrats and Republicans. Moreover, I examine the possible effects of these appeals on black and white respondents, as well as racial liberals and racial conservatives. Using two experiments and five case studies of Congressional campaigns, I develop and systematically test a theory of racial signaling. My theory of racial signaling states that candidates who are courting the electoral support of White Americans, have an incentive to “signal” either implicitly or explicitly that they are not beholden to “black interests,” or that they will not govern in such a way as to show favoritism towards African Americans. However, they must also demonstrate that they are not racially insensitive. Therefore, the latitude that candidates have to use the various types of appeals is contingent on the race and partisanship of the candidate in question. This study builds on previous work and contributes to the extant literature on racial appeals in a number of ways. First, while the racial priming theory argues that explicit racial appeals are less effective than implicit racial appeals, I theorize and demonstrate that not only are explicit racial appeals effective under certain circumstances, but that candidates have an incentive to use them. Since the theory of racial priming has largely been tested in the context of White Republican candidates, the full landscape of racial appeals was previously unknown. Secondly, this study tests the racial priming hypothesis among African American respondents. While African Americans are often the subject of racial appeals, few studies examine how they react to these appeals. I find that African Americans punish the use of explicit appeals, regardless of the race of the messenger. Finally, I evaluate the effects of racial appeals on politically salient subgroups, such as racial conservatives, who although they constitute a significant subset of the American population, have typically been underrepresented in the convenience samples of earlier experimental work.