Introduction

This report examines patterns of sleep, stress, and academic performance among college students using the SleepStudy dataset. The data includes 253 observations across 27 variables, providing insights into students’ lifestyles and academic habits. The analysis aims to address questions about GPA, sleep habits, mental health, and related factors.

Research Questions.

The report addresses the following questions:

Is there a difference in GPA between male and female students?
Do first-year students have more early classes than later-year students?
Do "Larks" (morning types) perform better cognitively than "Owls" (evening types)?
Do students with early classes miss fewer classes?
Are students with normal depression status happier than those with severe depression?
Is there a difference in average sleep between students who pull all-nighters and those who do not?
Do male students consume more drinks per week than female students?
Is weekday bedtime earlier for students with high stress?
What percentage of students get less than 6 hours of sleep?
Is there a difference in weekend sleep between underclassmen and upperclassmen?

Key Findings

1. GPA by Gender

Unsurprisingly there doesn’t seem to be much difference in grades between male and female students. Both groups perform similarly in academics.

2. Early Classes by Year

The analysis reveals how the proportion of early classes varies across different class years. While the the test indicates a significance (p = 0.0268) I do not see a practical difference.

3. Cognitive Scores: Larks vs. Owls

The data suggests a slight edge in cognitive performance for early risers (“Larks”), though the difference is not statistically significant.

4. Missed Classes by Early Class Attendance

Students with early morning classes actually miss fewer classes. The dots in the graph are outliers in the data.

5. Happiness Levels by Depression Severity

Not surprisingly, students with lower depression levels report higher happiness. Though an interesting note is the higher scores among those students with “Moderate” vs “Severe” depression.

6. Sleep Hours and All-Nighters

Students who pull all-nighters sleep significantly less on average (7.27 hours) compared to those who do not (8.07 hours). The difference, with a p-value of 4.34×10^(−6), strongly supports that pulling all-nighters negatively impacts average sleep.

7. Alcohol Consumption by Gender

There are some differences in drinking habits between genders. Men and women seem to have different patterns of alcohol consumption, with males being heavier drinkers.

8. Weekday Bedtime and Stress Levels

While the graph suggests that high-stress students may go to bed slightly earlier on average, the difference is not statistically significant (p = 0.286). This indicates that stress level does not have a meaningful impact on weekday bedtime in this dataset, but is equally likely that there is insufficient data.

9. Proportion of Students with Limited Sleep

About 2.8% of students report getting less than 6 hours of sleep on average, which is significantly lower than initially assumed.

10. Weekend Sleep Hours: Underclassmen vs. Upperclassmen

Weekend sleep patterns remain consistent across student year groups, with no significant difference between underclassmen and upperclassmen. Despite expectations that upperclassmen might sleep less due to busier schedules, the averages are nearly identical.

Conclusion

This analysis highlights the intricate relationships between sleep, mental health, academic performance, and student lifestyle. Quality sleep is linked to cognitive performance, mental health strongly influences happiness, and early classes may improve attendance. Additionally, alcohol consumption appears to correlate with stress levels, while sleep patterns evolve as students progress through college. To foster better outcomes, students should prioritize consistent sleep, manage stress with healthy activities, approach alcohol consumption mindfully, seek mental health support when needed, and maintain a balanced lifestyle. However, some data elements, like the “Happiness” score, lack clear definitions and may rely on subjective reporting, emphasizing the need for cautious interpretation and further exploration.

Notes