Sleep and Circadian Predictors of Academic Performance and Retention in STEM Pathways: A Longitudinal Study in University Freshmen

Poor sleep health is common amongst university students, but there are diverging viewpoints on whether their sleep loss helps, harms, or has no impact on academic performance. We investigated whether sleep health markers in first year university students predicted longitudinal academic outcomes, when accounting for key variables. First year university students who were pursuing a science, technology, engineering and mathematics (STEM) career pathway (N=489) were recruited to complete a baseline session that included measures of global sleep quality, chronotype, daytime sleepiness, social jetlag (change in sleep timing from weekdays to weekends), demographics, mental health, and fluid intelligence (reasoning). At the end of Year 1 and Year 2, we extracted data on cumulative grade-point-average (GPA), academic major change, STEM pathway change, and institutional withdrawal. After adjusting for demographic, mental health, and fluid intelligence factors, we observed that worse global sleep quality, evening chronotype, and worse social jetlag independently predicted Year 1 GPA. Global sleep quality also predicted Year 2 GPA, even when accounting for prior academic performance. Students with shorter sleep durations were more likely to switch from their STEM career pathway, even when accounting for academic performance, demographics, mental health, and fluid intelligence. In conclusion, sleep health markers are predictive of better future academic performance and retention in STEM pathways. There is a need for individual and environmental interventions to improve sleep health in first year students and to determine causal direction.