Simpson's paradox in real data: a maths EPQ idea
A title to start from
When can combining groups of data reverse a conclusion, and how should we decide which answer is right?
Why it works as an EPQ
A short piece of arithmetic with big consequences for medicine, admissions and sport, and a judgement about causation at the end.
Scope and difficulty
Accessible. Accessible. Two or three real cases explained carefully.
The maths
Builds on these A Level topics: Sampling and data · Probability.
You would learn:
- Weighted averages
- Confounding variables
- Simple causal diagrams
One possible plan
- Build a small example where a reversal happens and explain why.
- Analyse real published cases.
- Show how the choice between pooled and split data depends on the causal story.
- Write guidance for reading such statistics.
Pitfalls
- Explaining the arithmetic but not the causal question.
- Using a case whose data you cannot find.
Where to start reading
- The Art of Statistics (David Spiegelhalter)
- Search for: Simpson's paradox kidney stones Berkeley admissions
Similar ideas
- Does sacking a football manager work?Does sacking a football manager improve results, or is it regression to the mean?
- P-values and the replication crisisWhy do so many published results fail to replicate, and how much is the p-value to blame?
- Can people be random?Can people generate random sequences, and how can statistics tell?
- How accurate are weather forecasts?How accurate are UK weather forecasts, and what is the fairest way to measure accuracy?
All statistics and data ideas · all 93 ideas
Your EPQ must be your own work. These pages coach: ideas, structure, checklists and planning. Submitting text, proofs, code or analysis written by someone else or by an AI tool as your own is malpractice. How to use help and AI honestly.