Predicting sex differences in performance on the SAT I quantitative section: how content and stereotype threat affect achievement

Males consistently outperform females on the SAT-I quantitative section by about one third of a standard deviation. Previous research suggests this is due to a complex mix of biological, sociological, and psychological factors. This study examines 12th-grade male and female performance on NAEP items and uses this data to predict performance gaps on the SAT-I quantitative section. Study results suggest that sex differences in performance are due not only to the construction of the test, but also the environment in which the exam is administered. This research has far-reaching implications for the design and administration of standardized mathematics tests, which have historically exhibited large gaps in performance between the sexes. This research has implications in particular for the SAT, which is used for determining admission to many colleges and the awarding of scholarships.