The main objective of this article is to analyze gender differences in terms of student academic performance. The study is based on a data sample of 3,219 students attending the Autonomous University of Barcelona. The data was analyzed using the Tobit estimation model and the Tobit-Blinder-Oaxaca Decomposition Method. According to the results obtained, 67.27% of the differences in male and female academic performance are attributed to observable characteristics. The results obtained also reveal that student intellectual ability, experience with the subject matter, effort made during the course and prior accounting courses taken during high school, all had a significant and positive effect on students of both sexes. On the other hand, scholarships positively influenced the academic performance of only the female students, while the instructor´s professional status and the time of the day at which the class was held a negative impact on the academic performance of males. These results have underlying implications for policymakers and instructors. For example, organizations that offer student financial aid, may increase the number of scholarships available for female students, while instructors can better prepare students by having them perform more practical exercises that simulate real-life scenarios they may encounter when entering the job market.