A gender bias in reporting expected ranks when performance feedback is at stake

Abstract

We introduce a mechanism for eliciting beliefs that combines the simple use of monetary incentives with the desire to know the own performance. In our experiment, participants performed a real-effort task that naturally reinforced the desire to know their relative performance. In two treatments, differing in the degree of ex-post transparency, we elicited the belief about one’s own standing. In the Baseline, the performance ranking was always revealed. In the Treatment, we deprived the subjects of learning their relative performance if they did not accurately report their actual rank. This simple manipulation creates a bias in behavior that goes in opposite directions for men and women. Under the manipulation, men overplace even more, and women underplace themselves compared to the Baseline.