Research in political psychology has not yet reached a definite answer whether there the “thin slices” of information from candidates’ appearances result in a “beauty premium” in elections. We address the question of appearance-based effects on election results using data from the 2013 federal elections in Germany. We confront the outcomes of winning direct candidates with ratings of both their own and their respective runner-up’s perceived attractiveness, competence and likability. To gather these assessments we designed a new online-rating tool which enables us to rate the candidates relative to each other and include ambivalence by measuring decision latency-times. This produces a more realistic comparison situation than absolute ratings of candidate impressions. Using multilevel-regression, we estimate which perceived appearance traits help candidates most while controlling for known other factors of election outcomes such as incumbency status or number of second votes as well as idiosyncrasies of the German states.
