Though women increasingly participate in mathematics courses, substantial gender disparities persist across math domains, with women consistently underperforming compared to their male counterparts. We argue that these disparities are caused in part by learning climates in math environments that negatively affect female students. We suggest that learning climate emerges out of the dynamic interaction between students' perceptions of themselves and the network of perceptions of each other in the learning context, and, more precisely, students' mutual perceptions of their own and others' value in these settings. As a result of cultural gender stereotypes that portray women as less competent than men in science and math, female students come to view themselves under a cloud of suspicion (Steele, 1997) regarding their ability, and, consequently, expect to be undervalued in math settings. Likewise, many male students expect female students to be underqualified in these domains and consequently undervalue their contributions. We argue that these expectations lead to a host of cascading and detrimental effects for female engagement, learning, and performance and ultimately contribute to climates that perpetuate gender disparities in educational outcomes in mathematics fields. Across 2 studies, therefore, the present work evaluates the extent to which female students expect to be undervalued by others relative to their male peers in a mathematics context (Study 1) and tests an intervention designed to reshape climate by shifting the value students perceive in diversity, and therefore each other, in small mathematics learning groups (Study 2). We do this by providing students with evidence and instructions on how to use peers as a resource for learning and improved performance. Study 2 also evaluates who is in the best position to leverage the intervention to most effectively change group climate. We explore this question by administering the intervention to all students, only to male students, only to female students, or to a no-students (control) in four person (3 male, 1 female) learning group. Results across studies suggest that, in general, women do anticipate being and report feeling undervalued in math contexts relative to their male peers. However, the resource framing intervention does not affect change in self-perceptions of value, regardless of intervention target. The resource framing intervention does influence the extent to which male and female students value their group members depending on who receives the intervention. The intervention also differentially affects the number of times males and females participate in a video-recorded group math task. In addition to providing initial insight into a new intervention with potential to reduce gender disparities in mathematics, the present work highlights the importance of considering who receives an intervention and suggests that truly shifting learning climate may require students to interact repeatedly over time. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]