The Illusion of Political Tolerance: Social Desirability and Self-Reported Voting Preferences

Abstract

Tolerance for diversity in America is often indexed by direct measures, such as self-reported “willingness to vote” polls. However, pressures to be or appear unprejudiced may bias such estimates, yielding misleading and overly optimistic inferences about tolerance in America. The current research investigated the degree to which direct and indirect measures of political candidate preferences converge and diverge across six target groups varying in acceptability of stigmatization (atheists, African Americans, Catholics, gay men and lesbians, Muslims, and women) and across relevant participant demographics. Overall, participants (N = 3,000, nationally representative) reported less willingness to vote for target groups when measured indirectly, relative to directly. Additionally, the divergence between the direct and indirect measures was especially evident for social groups for which overt stigmatization is normatively inappropriate. This research provides a vital benchmark that quantifies the gulf between the direct and indirect measures of tolerance for various oft-stigmatized groups in America.