Two competing giants: An experimental test of spatial and valence models of voting

Abstract

Rational choice theories offer one of the major explanations for citizens’ voting behavior. The most prominent models within the rational choice framework – spatial and valence voting – share the same roots but differ in central aspects, emphasizing ideological congruence and competence evaluations, respectively. In this note, we provide a direct test of the models’ relative strength. Does a party’s issue congruence with a voter (spatial) or its effectiveness in handling important issues (valence) have a larger effect on vote choice? Our results from lab-like conditions using a novel conjoint experiment show strong evidence for both models, but valence evaluations consistently trump spatial considerations. These results have important implications for research on electoral behavior, party politics, and democratic representation.