Simpson's Paradox
by Anonymous
Shift the patient mix to see Simpson's paradox: the hospital ahead in every subgroup falls behind in the pooled total.
Ready to remix — hit the 🍴 Remix button below. You'll sign up first, then this app is copied into your workspace.
About this app
Simpson's Paradox is an interactive statistics lesson that demonstrates how hospital rankings can reverse when mild and severe patient cases are pooled together. Adjust the share of severe cases at each hospital to see how the mix of patients, rather than the underlying success rates, changes the overall result. Guided tabs and missions explain why subgroup results are often more relevant than a combined percentage.
Use cases
- Students learning Simpson's paradox and weighted averages
- Teachers demonstrating confounding and data interpretation
- Adults exploring critical thinking and cognitive bias
- Anyone comparing subgroup rates with pooled statistics
Features
- Interactive sliders for the severe-case share at each hospital
- Separate success-rate comparisons for mild and severe cases
- Live pooled-rate ranking and reversal feedback
- Explanations of how unequal patient mixes affect totals
- Guided missions about equal mixes, ranking gaps, and valid comparisons
- Quiz-style choices about which rate applies to a severe patient