The Medical College Admission Test (MCAT) was examined for sex bias. Data for student samples from two midwestern state-supported medical schools were analyzed. The criteria included average course grades in each of the first two years of medical school. Tests comparing various regression models based on male and female subgroups generally failed to find significant differences in standard errors of estimate, regression slopes, or intercepts. The few exceptions to this overall finding centered on two regression models which included scores on the physics subtest, and to a lesser extent, the chemistry subtest. In conclusion, this study failed to find great support for the assumption of widespread bias in predicting the medical school performance of females from MCAT scores, as long as undergraduate grade point average and Skills Analysis: Reading scores were considered in the selective index. Caution was also advised in the weighting of physics scores. (Author/GDC)