Shared birthdays in real classes: a maths EPQ idea
A title to start from
How likely are shared birthdays in real classes, given that births are not spread evenly through the year?
Why it works as an EPQ
The classic answer assumes every day is equally likely. Real birth data let you test whether that matters.
Scope and difficulty
Accessible. Accessible. Calculation, simulation and one real data set.
The maths
Builds on these A Level topics: Probability · Hypothesis testing · Sampling and data.
You would learn:
- Complement counting
- Simulation
- Non-uniform probabilities
One possible plan
- Derive the probability of a shared birthday for n people under the usual assumption.
- Find birth counts by day from official statistics.
- Simulate with the real distribution and compare.
- Judge how much the usual assumption matters, and why.
Pitfalls
- Collecting real birthdays from classmates without consent.
- Too few simulation runs.
Where to start reading
- Office for National Statistics births by day of year
- Search for: birthday problem non-uniform distribution
Real data for this idea
- Primary source: US births per day (NCHS and SSA, via FiveThirtyEight): US births by day, 1994–2014. Two CSVs (1994–2003 from NCHS, 2000–2014 from SSA) give year, month, day of month, day of week and births. Licence: CC BY 4.0 (FiveThirtyEight's data repository); cite NCHS and SSA as the original sources.
- Also useful: Office for National Statistics (UK): Births in England and Wales. The release tables include births by month of occurrence. Licence: Open Government Licence v3.0: free to reuse with an acknowledgement.
- Optional tool: RareBirthday. Browse how common each birthday date is; cite the NCHS data, not the tool. Explore only; cite the NCHS data, not the tool.
Frequently asked questions
Is this a good maths EPQ idea?
It can be. We rate it Accessible: it builds mostly on A Level Maths. Calculation, simulation and one real data set.
What maths does this EPQ use?
It builds on these A Level topics: Probability; Hypothesis testing; Sampling and data. You would learn: Complement counting; Simulation; Non-uniform probabilities.
Next step
Turn this idea into your own title, then check it in two minutes.
Check your titleSimilar ideas
- Why do opinion polls miss?Why do opinion polls sometimes get UK general elections wrong?
- Simpson's paradox in real dataWhen can combining groups of data reverse a conclusion, and how should we decide which answer is right?
- 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?
All statistics and data ideas · all 93 ideas
Your EPQ must be your own work. How to use help and AI honestly.