Do Boys and Girls Perform Better at Math Just Studying More ?

This paper investigates the role of effort on mathematics performance of boys and girls, an aspect that may contribute to our understanding of the gender gap in science, technology, engineering and math (STEM) fields in college. We exploit a remarkably rich primary data set to estimate a simultaneous equations model of mathematics attainment and students’ effort. Our estimation strategy infers causal relations by relying on an instrumental variable approach validated using weak-instruments-robust confidence sets and partial identification techniques. The results show that study effort plays a different role in the math performance of girls and boys. If a boy dedicates one extra hour to study, his math grade increases by 1 point on a 10-point scale. Differently, an additional hour of home study does not have an effect on girls’ math performance, though, in our sample, on average, girls perform significantly better than boys in math. We also examine the role played by peers, the quality of the attended school, and family socio-economic background. These factors mainly affect math achievement only indirectly through student’s effort. Validity tests suggest that our results are not confounded by unobservable heterogeneity. Our findings suggest that asking girls for additional efforts may not be effective to bridge the gender gap in STEM.