Abstract Introduction Poor sleep health is common in university students, but the longitudinal impact of sleep habits on academic outcomes remains unclear. We investigated sleep quality, timing, and duration in first year university students enrolled in science, technology, engineering, or mathematics (STEM) tracks in relation to future educational and career-oriented outcomes. Methods First year STEM majors, including Pre-Health track students (N=489, Mage=18.27 years, 72.2% females, 53.8% BIPOC), completed the Pittsburgh Sleep Quality Index, Morningness-Eveningness Questionnaire, Epworth Sleepiness Score, weekday and weekend sleep duration reports, and measures of mental health and fluid intelligence. Cumulative grade-point-average (GPA) and STEM pathway change were collected at the end of Year 2. Analyses were adjusted for age, sex, race/ethnicity, depressive symptoms, trait anxiety, and fluid intelligence. Results Compared to students who left STEM, those who persisted slept 28 minutes/night more at baseline (t(482)=2.73 p=.007, d=.41). Covariate-adjusted analyses indicated that there was a 4.9% reduction in the odds of leaving STEM tracks by the end of Year 2 for every additional 10 minutes of average weekday sleep duration (at baseline). Mediation analysis showed that first year GPA only partially mediated the effect of weekday sleep duration on STEM retention; interestingly, there was still a direct 4.4% decreased odds of leaving STEM for each additional 10 minutes of weekday sleep duration, as well as an indirect association of a 1.2% decreased odds of leaving STEM through first year GPA. Sleep measures were also predictive of overall academic performance: in adjusted analyses, higher global sleep quality at baseline predicted a higher second year GPA (β=-.046). Conclusion Sleep quality and duration in first year students are important predictors of longitudinal academic outcomes, even when accounting for traditional hallmark predictors (demographics, mental health, intelligence). Future research should examine post-graduate career decisions and investigate causal directions, but the current data highlight the likely need for individual- and environmental-level interventions for sleep health in first year students. Support (if any) Support provided by the National Science Foundation (1920730 and 1943323).
Scullin et al. (2026) studied this question.