The economic returns of taking math and science courses in high school are estimated for women who do not go on to college and for women entrepreneurs. A human capital model is used to estimate returns for respondents drawn from the National Longitudinal Survey's New Youth Cohort. Wage rates in 1990 of women who were ages 14-21 in 1979 were related to courses in math and science that were taken in high school, as recorded on a respondent's transcript and coded in standard year-long units. Little direct effect was found for the influence of high school curriculum in subsequent wage rates, either for women or for men. For women, courses in science and math did, however, significantly improve the probability that they would gain postsecondary training or go on to college. These indirect effects argue for a multi-equation model that estimates effects for women of all educational levels. Even when considering women of all educational levels, the study did not show any significant effect of science and math courses on the earnings of the self-employed. (SW)