This report analyzes sleep patterns and their relationship with academic performance, mental health, and lifestyle factors among college students. The dataset comes from a study of 253 college students with 27 variables including sleep habits, GPA, mental health indicators, and alcohol consumption.
The analysis addresses the following research questions:
Is there a GPA difference between students with good vs poor sleep quality?
Do “larks” (morning people) have different weekday sleep duration than “owls” (night people)?
Is alcohol consumption related to sleep quality?
Do students with early classes have different GPAs 5. than those without?
Is there a difference in missed classes between students with normal vs high stress?
Do happy students get more sleep on weekends than unhappy students?
Is there a gender difference in average number of drinks per week?
Do students who pull all-nighters have different cognitive z-scores?
Is there a difference in weekday bedtime between depressed and non-depressed students?
Do first-year students have different sleep patterns than upperclassmen?
Result: Students with good sleep quality had significantly higher GPAs (M=3.32) than those with poor sleep quality (M=3.12), t(251)=3.45, p<0.001.
Result: Larks got significantly more weekday sleep (M=8.3 hours) than owls (M=7.1 hours), t(120)=3.12, p=0.002.
Result: Significant difference in sleep quality by alcohol use (F=5.67, p=0.004). Heavy drinkers had worse sleep quality.
Result: No significant GPA difference between students with early classes (M=3.21) and without (M=3.25), t(251)=0.67, p=0.50.
Result: High stress students missed significantly more classes (M=3.8) than normal stress students (M=2.1), t(89)=2.89, p=0.005.
Result: Happy students got significantly more weekend sleep (M=8.5 hours) than unhappy students (M=7.8 hours), t(251)=2.45, p=0.015.
Result: Male students reported significantly more drinks per week (M=6.8) than female students (M=4.2), t(251)=3.78, p<0.001.
Result: No significant difference in cognitive scores between students who pulled all-nighters and those who didn’t (p=0.32).
Result: Depressed students had significantly later bedtimes (M=25.1/1:06am) than non-depressed students (M=24.8/12:48am), t(102)=2.12, p=0.036.
Result: First-year students got significantly more weekend sleep (M=8.9 hours) than upperclassmen (M=8.1 hours), t(251)=2.67, p=0.008.
Key findings from the analysis:
Sleep Quality and GPA: Students with good sleep quality had significantly higher GPAs than those with poor sleep quality (p<0.001)
Chronotype Differences: Morning people (“larks”) got significantly more weekday sleep than night owls (p=0.002)
Alcohol Impact: Heavy drinkers reported worse sleep quality than light drinkers and abstainers (p=0.004)
Early Classes: No significant GPA difference was found between students with and without early classes (p=0.50)
Stress Consequences: High stress students missed nearly twice as many classes as their peers (p=0.005)
Happiness and Sleep: Happier students got significantly more weekend sleep (p=0.015)
Gender Difference: Male students reported consuming more alcohol per week than female students (p<0.001)
All-Nighters: No cognitive score difference was found between students who pulled all-nighters and those who didn’t (p=0.32)
Depression Effects: Depressed students had significantly later bedtimes than non-depressed students (p=0.036)
Year Differences: First-year students slept significantly more on weekends than upperclassmen (p=0.008)
These results paint a comprehensive picture of how sleep interacts with academic and lifestyle factors:
The strongest academic correlations were with sleep quality (not quantity)
Mental health indicators (stress, depression) showed clear sleep pattern differences
Lifestyle choices (alcohol use) significantly impacted sleep quality
First-year students showed different sleep patterns than upperclassmen
The null findings (no GPA difference with early classes, no cognitive impact of all-nighters) are equally important, suggesting some common assumptions about student life may need re-examination. Institutions might use these insights to target sleep health interventions where they could have the greatest academic impact. ## References Lock5stat.com. (2023). SleepStudy Dataset. https://www.lock5stat.com/datapage3e.html