ThinkCritical.eu · Manipulation atlas
Flawed argumentsCorrelation ≠ causation
Two things going together doesn't mean one causes the other. Often a third factor is behind both.
When two things rise or fall together, it is tempting to conclude that one causes the other. Often, though, both are driven by a third factor – or it is simply chance.
“In the months when the most ice cream is sold, the most people also drown. We should limit ice cream sales!” In reality both are linked to hot summer weather – people eat more ice cream and go swimming more.
How to spot it
- “People who… have…” followed straight away by a conclusion about the cause.
- Comparing groups who chose for themselves what they do (who buys a supplement, who exercises).
- Nobody asks what else could explain both.
How to check and what to do
- Ask: could something else be behind both – age, income, the weather, lifestyle?
- Find out whether it was an experiment in which people were assigned to groups at random (a randomised trial). Only that can reliably show the cause.
- Try reversing it: could the second thing be causing the first?
Full explanation and exercises:
thinkcritical.eu/en/manipulation-atlas/correlation-causation