Female classmates, disruption, and STEM outcomes in disadvantaged schools: evidence from a quasi-experiment

Recent research shows that female classmates make classrooms more conducive to learning. We identify the effect of a higher share of female classmates on students’ disruptive behavior, engagement, test scores, and major choices in disadvantaged and non-disadvantaged schools. We exploit the random assignment of students to classrooms in early high school in Greece. Using administrative and hand-collected student-level data, we contribute by (a) leveraging a rich sample of schools with diverse characteristics and (b) measuring disruption and engagement using misconduct-related (unexcused) and excused class absences rather than self-reports. We find four main results. First, a higher share of female classmates improves students’ test scores in STEM subjects and increases STEM college participation, especially for girls. Second, it reduces disruptive behavior for boys and improves engagement for girls. Third, effects are concentrated among students in disadvantaged contexts, who experience the largest gains in learning productivity and STEM outcomes. Fourth, females in disadvantaged contexts exposed to more female classmates are more likely to choose higher-paying college degrees. These findings suggest that interventions that reduce disruption and improve engagement may be especially effective in disadvantaged environments.