SEX DIFFERENCES IN PERFORMANCE ON THE MATHEMATICS SECTION OF THE SCHOLASTIC APTITUDE TEST: A BIDIRECTIONAL VALIDITY STUDY1

Abstract

We examine the average mathematics scores on the SAT for almost 12 thousand men and women within five specific types of first-year math courses and at each grade level achieved in those courses. We find that after retrospective matching of both performance and course, women score about 35 SAT points lower than do their male counterparts. In prospective regression analyses we find somewhat larger sex effects in the same direction. We conclude that these findings provide evidence of sex differences in this measure of the validity of the SAT-M. Although the data do not allow any conclusions about cause, we believe that exploratory analyses like these can point the way toward the construction of formal models that can more adequately describe the observed phenomena. Pending the availability of such models, we support recommendations made by Petersen and Novick (1976) regarding efficacious ways to remedy possible inequities.