Objective
In this article, we address two major gaps in the understanding of the relationship between candidate attractiveness and electoral success. With the assistance of the Victoria Police Criminal Identification Unit in Melbourne, Australia, we show how good‐looking candidates look like by building the faces of six “ideal candidates” in terms of physical attractiveness. Utilizing our “ideal candidates,” we then investigate whether candidate attractiveness can actually sway electoral results.
Methods
We proceed in four distinct steps, using data from the 2008 U.S. House of Representatives elections. First, we collect data on candidate attractiveness. Second, we build our “ideal candidates” and obtain their attractiveness ranking. Third, we model the effect of candidate attractiveness on candidate vote margins. Fourth, we run four hypothetical scenarios that assess whether candidate attractiveness can sway the electoral results in marginal seats.
Results
About two‐thirds of marginal races would trigger a different winner if the actual loser looked like our ideal candidates. In addition, virtually every single marginal race would have had a different outcome if the unsuccessful candidate looked like our “ideal candidate” and the successful candidate was very unattractive.
Conclusion
Candidate attractiveness can sway electoral results, provided that elections are competitive.