- Coursework
2025 · Statistics · Exploratory data analysis · Coursework · ECMM443 Data Science, University of Exeter
Sleep Duration vs Academic Performance
A statistical study of sleep duration and academic performance among university students, with a deliberate focus on the limits of causal inference.
- Python
- Pandas
- Regression
- Statistical analysis
Headline
Association ≠ causation
Confounders and self-report bias analysed
On this page
01 · Problem
Problem
It is easy to find a correlation between sleep and grades. The harder, more honest problem is showing what that correlation can and cannot support once confounders and self-report bias are taken seriously.
02 · Approach
Approach
Self-reported sleep and academic-performance survey data was cleaned and explored, then modelled with correlation and regression analysis, with an explicit confounder and self-report-bias review layered on top of the statistical results.
03 · Results
Results
A regression-backed analysis of the sleep and performance relationship, alongside a clear account of its confounders and limits, and a write-up separating association from causation.
04 · Limitations
Limitations
What it does not show
Self-reported sleep is noisy and subject to recall bias, and the data is observational, so it cannot support a causal story.
Lessons
Keeping an analysis honest means resisting the pull to overstate a causal story that the data cannot support.