Measures of physical attractiveness are useful to research in several different domains of psychology: social psychology, evolutionary psychology, sexology; perhaps even industrial-organizational psychology. Research on the evolution of mating preferences has shown us that the number one determinant of mate selection for males choosing female partners is physical attractiveness. But how can we be empirical and objective about something that is not only incredibly subjective, but also incredibly personal? The diversity of the literature means that there is no gold standard for measuring such a construct, but it is nonetheless useful to discuss some existing methods for testing how people rate the physical attractiveness of others.

A landmark study on attractiveness found that there was a “golden ratio” when it came to facial beauty, such that ratios of space between specific facial features could be mathematically maximized for peak attractiveness. The researchers used a Thurstonian item response theory (IRT) model to have participants rate the attractiveness of various ratios. In this model, measurements were obtained by means of pairwise comparisons. They therefore represent a relative rather than absolute judgment of preference. Relative judgments may be a better scale to assess attractiveness on, since it seems that we tend to judge attractiveness in a relative manner in the real world anyway. Furthermore, they provide a basis for comparison within the experimental paradigm itself.

To complicate things, researchers have to take into account implicit versus explicit preferences. Explicit attitudes are subject to social desirability bias, and thus need to be supplemented by tests which measure implicit attitudes. One study used the go/no-go association task, which is a type of implicit association task in which participants are asked to judge quickly and without conscious thought whether good/bad words (eg. good partner, intelligent, etc.) are associated with synonyms for physical attractiveness. Surprisingly, they found absolutely no correlation between this and explicit measures of physical attractiveness. Clearly, what people desire in a partner personally is different from what they put across publicly.

A recent study that used people from the website HOTorNOT.com as its sample found that snap decisions about peoples’ “hotness” level correlated greatly (r=.92) with subsequent ratings of more general physical attractiveness. Taken into consideration with the results from the previous study, perhaps we can conclude that physical attractiveness is dissociable from other qualities — we find it easy to judge physical attractiveness independent of intelligence for example, and we don’t necessarily need to find someone intelligent in order to find them attractive.

Of course, this conclusion is problematic because it is confounded by a million other factors — for example, how would our appraisals of attractiveness change the more we get to know an individual? This is no doubt a fascinating area of research, and nonetheless a very complex one. It is clear that there is no right way to measure physical attractiveness, but the best measures will take into account both implicit and explicit attitudes, and will do their best to separate attractiveness from other qualities.