How Do Unrepresentative College Grades Shape Race and Gender Gaps in the STEM Pipeline?

While course grades are a primary way instructors communicate with students about their academic performance, there is less research on how grades are interpreted differently by college students. As a result, differential responses to grades may affect which students pursue STEM degrees and may therefore be an important contributing factor in the well documented gender and race/ethnicity gaps in STEM degrees. Prior studies examining this hypothesis, however, have been largely limited to analyzing data from individual courses or from single colleges and therefore cannot rule out alternative explanations based on other aspects of a student's academic record or aspects of the institution where the student is enrolled. Likewise, previous studies have not been able to account for factors influencing students' final grades in a course, such as a personal hardship. This project proposes to address these challenges by examining student grades across all STEM courses at all campuses in the University of North Carolina system. The study will include analyses of the Spring 2020 disruption caused by COVID-19 and will consider how the potentially unrepresentative grades students received in Spring 2020 may have shaped subsequent choices to pursue STEM coursework. This project intends to generate generalizable estimates of how students' differential sensitivity to grades may affect the size of race and gender gaps in the STEM pipeline. Understanding these responses has the potential to allow schools to devise fairer grading systems and better tailor advising programs and resources to support students pursuing STEM majors.

The project team will employ a mixed-methods approach to better understand whether unrepresentative grades change students' trajectories through the STEM pipeline, whether these changes vary with students' race/ethnicity or gender, and how departments advise students when they receive unrepresentative grades. The team will use student transcript data from 15 campuses in the University of North Carolina system from the academic years 2015-16 through 2023-24 to compare the trajectories of students in STEM majors based on whether they received uncharacteristically high or low grades in STEM courses in Spring 2020 compared to what might be expected based on the performance of prior cohorts of students. In addition, the team plans to interview academic advisors from a sample of STEM departments in the UNC system to better understand the guidance students receive about how to interpret their grades when deciding whether to pursue a STEM degree. One potential broader impact of this project is inform knowledge about grading policies and academic advising with an aim to reduce achievement gaps in STEM participation. The team plans to disseminate results in academic journals and distribute findings among local and regional stakeholders in the form of policy briefs and presentations. This project is funded by the EHR Core Research (ECR) program, which supports work that advances fundamental research on STEM learning and learning environments, broadening participation in STEM, and STEM workforce development.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.