| Suzanne Graham
Harvard University
The exodus from mathematics: When and why?
FINAL REPORT:
Mathematics is important for numerous fields of study and occupations, yet few students, particularly women, take mathematics through high school and college. To design interventions to increase mathematics participation, policymakers need to know when and why students first leave mathematics. For my dissertation, I addressed these questions by analyzing seven years of longitudinal data collected as part of the Longitudinal Study of American Youth.
Previous research examining persistence in mathematics is limited by cross-sectional, retrospective designs, longitudinal designs covering limited time periods, an almost exclusive focus on individual characteristics as predictors of student persistence, and a reliance on analytic methods not sensitive to questions about timing. My research overcame these limitations by using discrete-time survival analysis to analyze prospective, longitudinal data covering both high school and college years and by focusing on contextual characteristics as predictors of the risk that students will first stop taking mathematics. I also determined whether the effects of these characteristics differ for females and males.
I found that the risk of first leaving mathematics is especially high during 12th grade and the first two years of college. Nearly half of the mathematics pipeline is drained by 12th grade. From 11th grade until fourth semester of college, women are at an increasingly higher risk of dropping out than men. I also determined that students whose parents have high educational expectations for them are less likely to leave mathematics than students whose parents have lower educational expectations, even controlling for SES, parent educational level, mathematics achievement, and attitudes about mathematics. In addition, I found that, controlling for background, individual, and home characteristics, students whose teachers reported a major emphasis on problem solving are at less risk of leaving mathematics than those whose teachers reported less problem solving emphasis, but this effect existed for only one of the two cohorts. Finally, I found that high-achieving students who felt that their mathematics teacher treated males and females differently are more likely to stop taking mathematics. The effects of all background, individual, home environment, and mathematics classroom environment variables are the same for males and females.
Back to Funded Dissertation Grants Page |