fallacy

Correlation implies causation

Also called: Cum hoc ergo propter hoc, Confusing correlation with causation

This fallacy concludes that because two things occur together, one must cause the other.

How it works

A correlation can arise from causation in either direction, a third factor causing both, selection effects or pure chance. Treating it as proof of one specific cause skips all of these alternatives.

Examples

"Cities with more police officers have more crime, so police cause crime."

Cities with more crime hire more police: the causation likely runs the other way.

"People who drink coffee live longer, so coffee extends life."

Coffee drinkers may differ in income, health or habits that explain the difference.

How to spot it

"Linked to", "associated with" turned into "causes".
No discussion of a third factor.
The direction of causation is assumed, not tested.

How to respond

Ask whether the cause could run the other way.
Name plausible third factors.
Ask whether there is experimental or quasi-experimental evidence.