Eval Academy

View Original

Spurious correlation

A spurious correlation is a situation where two variables are statistically related (they are correlated), but there is not causal link between them. This relationship is often caused by a third variable that is not being accounted for (called a confounding variable).

For example, if the students of a psychology class who had long hair got higher scores on the exam than those who had short hair, there would be a correlation between hair length and exam score. However, there’s no reason to believe there is a causal relationship between hair length and exam score (e.g., that you could grow your hair longer to improve your grades). The real cause might be an unobserved variable: gender. It may be that women usually had longer hair than men, and women performed better on the psychology exam for one reason or another.

 

See also: causation

Return to the Evaluation Dictionary