This project was designed to examine whether and how achievement goals contribute to the effect of gender stereotypes on women's low expectancies for success on math tasks. Study 1 confirmed predictions from the Stereotyped Task Engagement Process (STEP) model (Smith, 2004) by demonstrating that, compared to a counter-stereotype situation, women reminded of the gender-stereotype endorsed performance-avoidance goals. Study 2 tested whether performance-goal adoption differed between men and women. Results showed that, compared to men, women in a stereotype salient math situation were more likely to endorse performance-avoidance achievement goals. This performance-avoidance goal adoption in turn, accounted for the negative relationship between participants' gender and performance expectations for a standardized math test.Math-related occupations afford great economic advantages and shape much of the future of the USA information society. For these reasons, women's (actual or perceived) difficulties in mathematics have long been of interest to researchers (Hyde & Kling, 2001). The current project was designed to examine how women's motivational experiences contribute to gender differences in math performance expectancies. Much is known about stereotypes' effect on performance, but less attention has been paid to the effect of stereotypes on expectancies for performance. The Stereotyped Task Engagement Process (STEP; Smith, 2004) suggests that individuals who are subjected to a competence-related stereotype experience different motivations (i.e., goals) for task engagement than do individuals who are not targets of the stereotype. The aim of the current project was twofold: