Speed-dating and simulation data explain the discrepancy between stated and revealed mate preferences.

Abstract

There is little evidence in speed-dating studies that stated preferences – what people say they prefer in a partner – are associated with revealed preferences – what people actually find attractive in a partner. In Study 1, a high-powered speed-dating study (n = 1145) revealed that four out of nine traits provided evidence of a correspondence between stated and revealed preferences. In Study 2, simulations based on the constraints of Study 1’s speed-dating design showed that when attractiveness depends on multiple independent traits, the stated preference for an individual trait can only be, on average, minimally related to the revealed preference for that trait. In Study 3, we investigated methods that simultaneously combine multiple traits when testing the association between stated and revealed preferences (e.g. Euclidean distance, pattern metric). All four omnibus methods indicated an apparent association between stated and revealed preferences in our speed-dating data. However, additional analyses and permutation tests suggest that these significant associations reflect statistical artefacts rather than true correspondences. We conclude that detecting any association between stated and revealed preferences will be difficult under realistic assumptions about the number of traits involved in partner evaluation. In this light, we discuss previous findings and provide suggestions for future studies in this vein.