Mental images and political stories: Tracing the implicit effects of race and gender rhetoric on public opinion.

Abstract

This dissertation explores the ways that people's ideas about race and gender---their race and gender schemas---can influence their opinions on policies that have nothing to do with race or gender on their face. I develop a theory of group implication, which specifies when political rhetoric will mobilize people's race or gender schemas to influence their opinion on a political issue. The most important conditions are that the schema be psychologically accessible and that the rhetoric structure the issue in a way that matches the structure of the schema. When this occurs, people's feelings about race or gender will affect their evaluation of the issue despite the lack of racial or gender content in the issue. I investigate the psychological mechanisms of group implication with a survey experiment run among University of Michigan undergraduates. Participants received artificially-constructed news articles about three political issues. One third received articles subtly structured to implicate racial considerations, one third received gender-implicated articles, and one third received neutral articles. I find that the racialized and gendered articles lead participants to draw on their race or gender predispositions when giving their opinions on the issues, compared with participants who read the neutral articles. Thus, I find that appropriately constructed political rhetoric can lead people to engage their race or gender predispositions in evaluating a range of social welfare policies. In addition, I conduct a series of secondary analyses of nationally representative survey datasets. I find that throughout the 1980s and 1990s, opinion on both welfare and social security is racially implicated among white Americans. Racially conservative whites are less supportive of welfare spending, and more supportive of spending on social security, than racial liberals. I also find that opinion on health care reform becomes implicated with considerations of gender during the 1994 reform debate, leading gender conservatives to oppose health care reform and gender liberals to support it. I conclude the dissertation by speculating on the meaning of my theory and results for American politics, for the study of race and gender, and for the study of public opinion.