The purpose of this study was to collect and analyze data on sexual differences in secondary school students' attitudes towards science. Attitudinal differences were also analyzed for the independent variables of science programs and grade levels. Data were collected from 988 students using a modified version of the Fennema-Sherman Mathematics Attitude Scales to represent attitudes toward science. Reliabilities of the modified science subscales were all high ( > 0.83). Multivariate analysis of variance (MANOVA) was used to analyze the data for the main and interaction effects of the independent variables of sex (male, female), grade level (10th, 11th, 12th), and science program (advanced placement, academic, general, terminal). Significant differences (p < 0.05) were indicated for all main effects (sex, grade, science program). Interaction effects were not found. Mean separations for the various levels of sex, grade, and science program were performed for all attitudinal subscales. Females evidenced a significantly more positive attitude (p ⩽ 0.01) than males on three subscales: Attitude Toward Success in Science Scale, Science as a Male Domain Scale, and Teacher Scale. Although not significant, males evidenced more positive attitudes on all the remaining five subscales. Eleventh graders evidenced significantly more positive attitudes than tenth graders on all but the Effectance Motivation Scale. Students in 11th grade had more positive attitudes than 12th-grade students on all scales but Science as a Male Domain Scale; however, these differences were not significant. Tenth graders differed significantly from 12th graders on three subscales; Science Usefulness Scale, Confidence in Learning Science Scale, and Teacher Scale. Positive attitudes decreased from advanced placement to terminal programs. Academic students did not differ significantly from general students except on the Father Scale; however, they were significantly different (more positive) from the terminal students for all subscales. General students were also significantly different from terminal students except on the three subscales of Attitudes Toward Success in Science, Science as a Male Domain, and Effectance Motivation.