Causal Inference
Confounder
A confounder is a variable that influences both the treatment and the outcome, creating a fake association between them. Because it feeds into both, a naive comparison mistakes the confounder's footprint for a real treatment effect.
Confounding is the central villain of observational studies. The fix is to adjust for confounders — by controlling for them in a regression, matching, or stratifying — but you can only adjust for confounders you measured. Unmeasured confounding is the reason most observational causal claims stay uncertain.
Age confounds the link between gray hair and heart disease: older people have both, so gray hair only looks risky.