A study involving 716 male and 1,113 female first-year students admitted to the University of Delaware in the fall of 1985 was conducted to investigate the differential performance of men and women on the mathematics section of the Scholastic Aptitude Test (SAT). Data collected included SAT math scores; gender; and number of years of high school mathematics, physics, and computer science courses taken by the students. To investigate the effect of the number of years of coursework, unstandardized regressions were estimated. Results indicate that approximately 25% of the difference between men and women on SAT math scores is due to the difference between sexes in the number of years that mathematics, computer science, and physics courses were taken. The remaining approximately 75% difference between SAT math scores of men and of women is due to differences between sexes not investigated in this study. A further study is recommended that would control for math curriculum and student grades to determine whether other variables (such as sex-role socialization, item bias, and academic self-concept) are involved in the score differences. Three data tables are included. (TJH)