Gender and the self-presentation of academic achievement

Two experiments were conducted in which first-year male and female subjects (75% white; 25% minorities) at a highly selective liberal arts college predicted their first-semester grade point averages (GPAs). The first experiment varied whether predictions were given publicly or privately when an incentive for accuracy was either present or absent. Whereas the actual GPAs obtained by women did not differ from those obtained by men, women predicted that they would receive lower GPAs than men predicted, but only in the public condition. Whether incentive was present or absent did not affect predictions. The second experiment varied whether predictions were given publicly or privately as well as the perceived achievement of the experimenter. Once again, women's actual GPAs did not differ from men's. However, women predicted that they would receive lower GPAs than men predicted, but only when their estimates were given publicly to a low-ability experimenter. Although previous research has concluded that gender differences on performance estimates is attributable to a lack of confidence among women, these results suggest that the gender difference may be (at least partly) attributable to self-presentational style. It appears that women in this study adopted a modest self-presentational style that was motivated by attempts to protect the self-esteem of the experimenter.We are especially grateful to Rob Abel, David McGill, and Melinda Varn for their help in data collection.