HETEROGENEITY AND CERTAINTY IN CANDIDATE EVALUATIONS

While perceptions of candidate character traits are believed to be an important pre-dictor of vote choice, very little research attempts to explain where these perceptions come from. There is also a remarkable inconsistency in the literature about which voters use traits and which use issues in voting—with completely contradictory results common. The solution to the inconsistency developed here is to focus on the rela-tionship between issues and traits as opposed to just these attitudes ’ direct effect on candidate evaluations. By positing that the relationship between issues and trait per-ceptions depends on the citizen’s certainty about the candidate’s issue positions the theory develop here answers questions about the nature of trait perceptions and the heterogeneity of voter decisionmaking. The theory is tested using evaluations of ten different presidential candidates from five National Election Studies. Key words: candidate evaluations; certainty; trait perceptions; voter heterogeneity. Voters evaluate candidates like they evaluate other people they encounter. We form impressions of candidates ’ abilities, intelligence, integrity, and hon-esty, and we rate and evaluate the candidates in part based on our assessment of what they are like as people. These perceptions of candidate character traits are one of a few central components of vote choice, on par with partisanship, issue positions, or assessments of the economy. In the literature, trait per-ceptions are often connected to two questions. Are traits more important than issues? And, perhaps more commonly, which voters rely on traits and which rely on issues when voting? Unfortunately, both questions have produced very inconsistent answers. For the instance, Miller, Wattenberg, and Malanchu