Does Issue Importance Predict Learning About Candidates?

Abstract

Abstract A long-standing argument is that policy voting is more likely on issues the voter considers subjectively important. However, existing evidence is highly mixed. We leverage panel data from the 2008–2009 ANES Panel to investigate the relationship between subjective issue importance and a key mechanism thought to link it and policy voting: candidate knowledge. Using both lagged dependent variable and fixed-effect models, we find little evidence that subjective issue importance predicts candidate knowledge or learning. Our results suggest that one reason for lack of consensus about whether subjective issue importance moderates policy voting is the lack of a clear connection between it and this important mediator. Our results point to the need for stronger measures of subjective issue importance.