Girls outperform boys in terms of educational attainment in school and spend more time and effort on schoolwork. The paper presented in this Research Seminar evaluates a randomized information intervention targeted at individual beliefs about the productivity of effort among a cohort of first year university students and investigates whether it can reduce gender differences in academic outcomes. The authors measure individual beliefs about the productivity of effort using subjective expectations about the education production function. Their intervention increases overall first-year GPA by 0.14 of a standard deviation, with similar effect on graduation GPA. The effects are larger for men, such that the intervention significantly reduces the gender gap in academic achievement, which is 0.12 of a standard deviation in the treated group compared to 0.39 in the control group.