Do Early Morning College Classes Reduce STEM Persistence and Shift Students Toward Lower-Earning Majors?

Yim (2026) asks whether being assigned to early morning college classes changes students’ academic outcomes and long-term academic paths. He analyzes administrative data from Purdue University covering several thousand freshman course enrollments after the university introduced a system that effectively randomized class times across course sections. The study links course schedules to grades, later STEM course enrollment, major choice, and expected earnings by major. Yim finds that assignment to a 7:30 a.m. class lowers course grades by about 0.05 GPA points, reduces the probability of taking another STEM course by roughly 4 percentage points (about 14 percent), and decreases the likelihood of majoring in the same college by about 3.1 percentage points (around 29 percent).

Why This Article Was Selected for The Policy Scientist

The policy implications of course scheduling extend well beyond a single class period. Universities routinely structure schedules around institutional convenience rather than student learning, yet small features of academic organization can shape human capital development and career trajectories. Yim (2026) contributes to this broader discussion by showing that early morning classes influence grades, STEM persistence, and major selection—decisions that affect long-term earnings. The topic is timely as institutions search for low-cost ways to improve student success. The study uses high-quality administrative data from Purdue and leverages a natural experiment created by randomized scheduling, a credible causal inference strategy. While drawn from one university, the institutional setting resembles many large public institutions, suggesting reasonable external relevance.

Full Citation and Link to Article

Yim, A. (2026). How early morning classes change academic trajectories: Evidence from a natural experiment. Journal of Policy Analysis and Management. https://doi.org/10.1002/pam.70088 

Central Research Question

This article examines whether the timing of college classes—specifically very early morning sections—affects students’ academic performance and longer-term educational trajectories. Anthony Yim investigates whether assignment to a 7:30 a.m. class influences grades, subsequent enrollment in STEM courses, persistence within a field of study, and the type of major students ultimately complete. The central research question is therefore whether a seemingly minor institutional feature—class scheduling—can produce meaningful changes in students’ human capital accumulation and academic decision making. The study focuses on outcomes that are directly relevant to long-run economic prospects, including persistence in STEM fields and the likelihood of graduating in majors associated with lower earnings. By analyzing how early academic experiences influence course-taking patterns and major selection, the article seeks to determine whether the structure of university schedules can produce durable effects on students’ educational pathways.

Previous Literature

Prior research has established that early morning work schedules can reduce productivity, increase fatigue, and negatively affect cognitive performance. A number of studies have extended these insights to educational contexts, showing that early class times are associated with lower academic performance and weaker engagement in coursework. However, most existing studies have focused on short-run outcomes such as course grades or attendance rather than long-term academic decisions. In particular, previous evidence on the effects of early class schedules has largely been drawn from military academies, where institutional environments differ substantially from typical universities. In those settings, attendance is mandatory, student schedules are highly regulated, and students face strict behavioral expectations that may mitigate some of the negative consequences of early classes.

Yim’s contribution lies in extending this literature to examine how early morning classes influence broader academic trajectories, including persistence in STEM fields and the selection of college majors with different expected earnings. The study also connects to a broader literature examining how early academic experiences shape long-term educational outcomes. Previous work has shown that exposure to particular instructors, peer groups, or introductory courses can affect whether students continue in a given field. For example, research on STEM persistence has demonstrated that factors such as instructor characteristics, peer composition, and course sequencing can influence students’ decisions to remain in or leave demanding majors. By focusing on course timing as an institutional factor that is largely under administrative control, the article adds a new dimension to this literature.

Data

The analysis uses administrative data from Purdue University, a large public research university in the United States with an undergraduate population exceeding thirty thousand students. The primary dataset contains more than 4,500 student-by-course-by-term observations drawn from the fall semesters of 2018 and 2019. The study focuses on freshman students enrolled in introductory or general education courses that offered multiple sections taught by the same instructor, including at least one section scheduled at 7:30 a.m.

The dataset combines several administrative sources. Course request data record the classes students requested during registration and their preferences for particular sections. Registrar records provide information on finalized class schedules, course grades, student demographics, SAT scores, and major choices. The author also links these records to institutional data on earnings associated with different college majors, which are derived from alumni career surveys conducted by Purdue’s Center for Career Opportunities.

These data allow the study to track several key outcomes. First, the analysis examines students’ final course grades in introductory classes. Second, it measures whether students enroll in additional STEM courses within the following two academic terms. Third, it records whether students ultimately select a major within the same college as the introductory course. Finally, the dataset includes measures of expected earnings associated with students’ eventual majors, allowing the author to examine whether early course timing affects the likelihood of graduating in lower-earning fields.

Methods

The empirical strategy exploits a natural experiment created by Purdue University’s batch course registration system, introduced in 2018. Under this system, students submit course requests and the university’s scheduling algorithm assigns them to course sections. When sections are oversubscribed, the algorithm effectively randomizes students across available sections, including early morning and later class times.

Because students’ preferences and course requests could influence their probability of receiving an early morning section, the study employs a control-function approach to isolate the causal effect of early class assignments. The author simulates the scheduling algorithm multiple times using observed course requests to estimate each student’s probability of being assigned to a 7:30 a.m. section. This simulated probability captures nonrandom factors affecting assignment and is included as a control variable in the regression models.

The main estimates therefore compare students who take the same course with the same instructor in the same semester but are assigned to different class times. The models include course-by-instructor-by-term fixed effects as well as demographic controls such as gender, race, SAT scores, and first-generation status. Standard errors are clustered at the student level to account for the individual-level randomization of section assignments. This framework allows the author to interpret differences in outcomes between early and later sections as causal effects of early morning scheduling.

Findings/Size Effects

The results indicate that early morning classes produce measurable changes in both academic performance and longer-term educational decisions. Students assigned to 7:30 a.m. sections experience a modest decline in course grades. The estimated effect corresponds to approximately a 0.05 reduction in GPA points on a four-point scale. While this change is relatively small in magnitude, the results suggest that early class timing reduces students’ academic performance in introductory courses.

More substantial effects emerge when examining subsequent course-taking behavior. Students assigned to early morning sections are about 4.2 percentage points less likely to enroll in another STEM course within the following two academic terms. Given the baseline probability of enrolling in additional STEM coursework, this represents roughly a 14 percent decline in STEM persistence.

The analysis also finds meaningful effects on major selection. Students assigned to early morning classes are approximately 3.1 percentage points less likely to choose a major within the same academic college as the introductory course. This corresponds to a reduction of roughly 29 percent relative to the baseline probability of selecting a related major. The effect is particularly pronounced for STEM fields, where students assigned to early morning introductory courses are significantly less likely to major in those disciplines.

The study also examines the economic implications of these shifts in academic pathways. Students assigned to early morning classes are about 5.1 percentage points more likely to graduate in majors that fall within the lowest quartile of the earnings distribution across fields of study. Among students assigned to early morning STEM courses, the probability of graduating in a low-earning major increases by roughly 7 percentage points. These findings suggest that early class scheduling may indirectly influence students’ long-term economic prospects through its effect on major selection.

Conclusion

The article concludes that early morning class scheduling can influence both short-term academic outcomes and longer-term educational trajectories. Although the direct impact on course grades is relatively modest, the evidence indicates that early morning classes reduce students’ likelihood of continuing in STEM coursework and increase the probability of shifting toward majors with lower expected earnings. These results suggest that small institutional features—such as the timing of introductory courses—can shape academic decisions with lasting consequences.

The study also proposes several mechanisms that may explain these patterns. Survey evidence indicates that students assigned to early morning sections report lower engagement and reduced participation in class discussions. In addition, the author argues that attribution bias may play a role: students who perform poorly in early morning classes may attribute their difficulties to the subject itself rather than to the timing of the course. Finally, institutional constraints such as GPA thresholds for selective majors may amplify the consequences of small declines in course performance.

Taken together, the findings highlight the potential for course scheduling policies to influence student outcomes in ways that extend beyond immediate academic performance. By documenting the broader effects of early morning classes on major selection and educational pathways, the article contributes to a growing literature on how institutional structures within higher education shape the development of human capital.