The Sleep-Success Link: How Rest Patterns Shape College GPAs

By shrijeetverma13 · July 27, 2026

This analysis of 634 college students reveals that sleep duration and consistency matter for academic performance, though their effect is modest. Students…

Yes, students who sleep under 6 hours show a lower GPA change on average. Their mean GPA change was -0.124, compared to +0.012 for students sleeping 6 or more hours — a difference of 0.136 points in favor of better-rested students.

There's a weak positive relationship between average sleep duration and term GPA among the 634 students analyzed. The correlation coefficient is 0.202, meaning students who sleep more tend to have slightly higher GPAs, but sleep alone doesn't strongly determine academic performance. Students sleeping 7-7.5 hours had the highest average GPA (3.55), while those sleeping less than 5.5 hours had the lowest average GPA (3.17), a notable gap of 0.38 GPA points between the best and worst sleep brackets.

Sleep and GPA both vary modestly across universities and semesters. University of Notre Dame has the highest average term GPA (3.665) with 6.42 hours of sleep, while Carnegie Mellon University has the lowest average GPA (3.353) despite similar sleep (6.45h). Across all universities, average sleep ranges narrowly from 6.42 to 6.86 hours. Looking at semesters, Spring 2016 stands out as the strongest term with a 3.665 GPA, while overall semester GPAs range from 3.279 to 3.665.

Yes, there is a weak but consistent negative relationship. Students with more variable bedtimes tend to have slightly lower term GPAs (correlation of -0.187) and are also somewhat more likely to see their GPA decline (correlation of -0.110). While these relationships aren't strong, they point in a consistent direction: less consistent sleep schedules are linked to worse academic outcomes.

First-generation and underrepresented students sleep about the same number of hours as their peers, but tend to have lower GPAs. First-gen students average 6.65 hours of sleep versus 6.62 hours for non-first-gen students, but their term GPA is lower (3.30 vs 3.48). Similarly, underrepresented students sleep slightly less (6.56 vs 6.64 hours) and have a notably lower GPA (3.27 vs 3.49) compared to their peers. Underrepresented and first-gen students also show slightly higher rates of getting under 6 hours of sleep and lower bedtime variability, suggesting more consistent (but not necessarily longer) sleep schedules.

Taking more term units is mildly associated with getting less sleep, but has almost no direct relationship with GPA change. Specifically, term units and sleep hours have a correlation of -0.241, meaning students with heavier course loads tend to sleep somewhat less. However, term units barely correlate with GPA change (-0.036), and the standardized term load measure shows almost no relationship with either sleep (-0.034) or GPA change (0.002). Interestingly, sleep hours themselves have a small positive correlation with GPA change (0.152), suggesting that sleep may matter more for GPA outcomes than course load alone. A scatter plot comparing term units to sleep hours (colored by GPA change) and a grouped chart comparing sleep and GPA change across term load quartiles were both generated to visualize these patterns.

A correlation matrix and heatmap were successfully created for the 8 variables: avg sleep hours, daytime sleep minutes, bedtime variability, sleep midpoint minutes, term units, term load z, prior gpa, and term gpa. The heatmap visually displays how strongly each pair of variables relates to one another, using a color scale from -1 (strong negative) to +1 (strong positive). You can review the corresponding data tables for the exact correlation values between every pair of metrics, including term gpa's relationship with sleep patterns and academic load factors.

Using the IQR method on average sleep hours and daytime sleep minutes, 35 students were identified as outliers. These students have unusual sleep patterns that fall well outside the typical range seen in the rest of the group. On average, these outlier students have a term GPA of 3.26, which is lower than the overall average GPA of 3.45. This suggests that students with atypical sleep habits — whether sleeping much more or much less than their peers, or having unusual daytime sleep patterns — tend to perform slightly worse academically.

Comparing 104 first-generation students to 526 non-first-generation students shows some meaningful differences. First-generation students had nearly identical average sleep hours (6.65 vs 6.62), but noticeably more consistent bedtimes, with bedtime variability 44.29% lower than their peers. They also tracked their sleep slightly more often (4.32% higher tracking fraction). However, their average term GPA was 5.27% lower (3.30 vs 3.48) compared to non-first-generation students. Two charts and a detailed summary table were generated to visualize these comparisons.

When ranking universities by average GPA change, the University of Notre Dame shows the largest positive average GPA change (+0.027), with an average tracking consistency of 0.889. On the other end, Carnegie Mellon University shows the largest negative average GPA change (-0.042), with a tracking consistency of 0.855. A bar chart ranks all universities by their average GPA change, and a scatter plot compares tracking consistency (nights tracked fraction) against GPA change at the university level, sized by student count, to help visualize whether more consistent tracking relates to better academic outcomes.