Inferences of Competence from Faces Predict Election Outcomes

In Australia, we have a federal election coming up. On November 24th, we'll vote for one of the above men to be Prime Minster.

How will we decide? On policy? Party allegiance? Boondoggling?

Perhaps. However, I recently came across a very cool study (Tondorov, Mandisodza, Goren, & Hall, 2005) that seems to show that elections are determined, to a surprisingly large extent, by candidates' facial characteristics -- yes facial characteristics.

Participants in this study were shown pairs of photographs of politicians. Each pair depicted the winner and runner up of from a US Senate or House of Reps race. Each pair was flashed on a screen for 1 second, and participants were asked "which person is the more competent?". Care was taken that the participants had no prior knowledge of the politicians.

Amazingly, participants' competency ratings predicted the winner about 70% of the time. It suggests that our initial intuitive judgments about politicians (uninfluenced by advertising, policy, social pressures, party endorsement etc.) have quite a large influence on who we vote for.

But could competent people just have competent looking faces? Did Howard tame his eyebrows just as he tamed the economy?

And for any non-Australian readers out there, please tell us, from your naieve standpoint, which of the gentlemen shown above looks the most competent?

More on the Tondorov et al. study here and here.

(And for the stats nerds out there: the correlation between the differences in competency ratings and the difference in the proportion of votes between the winner and runner up was .44 -- equivalent to nearly 20% of the variance in the winning margins). That seems quite impressive to me.

References

Todorov, A., Mandisodza, A. N., Goren, A., & Hall, C. C. (2005). Inferences of competence from faces predict election outcomes, Science (Vol. 308, pp. 1623-1626): American Association for the Advancement of Science.