Although the mathematics performance gap between males and females has narrowed over the past decade (e.g., Hall, Davis, Bolen, & Chia; 1999; Hyde, 1997; National Center for Education Statistics [NCES], 2001), there continues to be a gulf between the number of women and men who pursue college degrees in engineering, physical sciences, computer sciences, and mathematics (Bae & Smith, 1996; Higher Education Research Institute, 1996; Stumpf & Stanley, 1996). Furthermore, women who hold bachelor's degrees in science and engineering are less likely than men with similar degrees to actually be employed in those fields; women constitute only 23% of the science and engineering labor force (National Science Foundation [NSF], 2000). The underrepresentation of women is especially evident in the physical sciences, where women comprise only 9% of employed engineers and 10% of employed physicists (NSF, 2000).

In light of diminishing performance differences, the continuing gender gap in math/science educational and career choices suggests that such choices are based on much more than achievement (Linver, Davis-Kean, & Eccles, 2002). Numerous theories dealing with competence, expectancy, and control beliefs provide explanations for performance on different kinds of achievement tasks; however, many of these theories do not systematically address another important motivational question: What makes the individual want to do math or science? Even if individuals feel competent, they may not want to pursue it.