Gender bias in risk aversion: evidence from multiple choice exams

We provide evidence on a gender bias in risk aversion among students of economics in Spain. In a sample of 1947 multiple choice exams with penalization for errors, we find that women consistently answer less questions, while differences in marks are not significant. These empirical results are consistent with a recent theoretical prediction of the effect of risk aversion in test exams. This finding shows that women can suffer a disadvantage with this kind of exams widely used in education.