← Checking What You Hear, in Practice
Numbers and charts · 38 of 73
Ask what else could explain it
Expert guidance
Do this
- See “A goes with B”, or “people who do A have more B”? Don’t jump to “A causes B”.
- Ask: could B cause A?
- Ask: could a third thing cause both?
- Could it be chance?
- Ask whether anyone ran a fair test.
How the trick works
The Australian Bureau of Statistics gives the case of ice cream and sunscreen: sales rise together because of the hot weather, not because one drives the other. It says causation is commonly misunderstood and misused, and that controlled studies are the most reliable way to test it. A plausible story isn’t enough. Expert guidance.
Try it today
Find one “linked to” headline. Name a third thing that could explain it. Three minutes.
For example
Students who go to tuition score higher, Rekha reads. Does tuition do it? Maybe. Or maybe families who can afford tuition can also afford books, quiet and time. Only a fair comparison could say.
Where these facts come from
- Correlation and causation, Australian Bureau of Statistics.
- Oxman, A. D. and others, Informed Health Choices Key Concepts, Informed Health Choices Group, 2022.
General information, not professional advice. No page can fit every situation. You use it at your own risk: see our terms. Spotted a mistake, or know something better? Tell us.
