Books Judged by their Covers: Revealing the Hidden Gender Biases in Impressions of Competence

Abstract

Many of the inequalities in the world today are the result of incorrect, biased perceptions based on appearance during a first impression. This study shows that certain faces are perceived as less competent than others because of their feminine properties. With a statistical face model (Oosterhof & Todorov, 2008; Todorov et al., 2013), we uncover this implicit bias that indicates women are perceived as less competent than men through trait judgments after exposure to faces manipulated on a scale of competence. Experiment 1 re-validates the Competence Model, an existing model of this social dimension (Todorov et al., 2013), as well as tests for the relationship between this model and trait judgments of attractiveness. This allows us to create a new model, a Difference Model, for Experiment 2, which looks to validate a [competence – attractiveness] model, as well as perform additional trait judgments of confidence and masculinity. Finally, Experiment 3 uses both the Competence and Difference Models for a gender classification test, providing strong evidence in favor of the hypothesis. The trait judgment of competence for both the Competence and Difference models demonstrated that the models were accurate in their representations, as the mean ratings increased steadily from the lowest to the highest SD. In addition, the gender classifications yielded results very similar to the competence judgments, with the proportion of faces rated as “Male” rising steadily from the lowest to the highest SD of competence. Most notably, the Difference Model, having removed the effects of attractiveness, shows an even stronger effect from each SD to the next. Therefore, we can conclude that perceptions of competence are highly correlated with gender bias against women– a troubling outcome for society.