Why doesn't correlation imply causation?
mediumAnswer
- Two variables can move together because A causes B, B causes A, both are caused by a third variable Z (confounder), or by pure coincidence.
- Establishing causation requires either a randomized experiment (breaks confounding by design) or careful causal inference methods (DAGs, matching, instrumental variables, difference-in-differences, regression discontinuity, do-calculus).
Check yourself — multiple choice
- Randomization breaks confounding
- Correlation always implies causation
- Observational data proves causal effects directly
- Confounders help identify causation
Randomized experiments are the gold standard because they eliminate confounders.
#causal-inference
Practise Statistics Fundamentals
215 interview questions in this topic.
Related questions
- What is Simpson's paradox?
- Users who adopt feature X churn less. Can you say the feature reduces churn?
- Potential outcomes framework — Rubin causal model.
- What is SUTVA?
- How do you handle confounders in observational studies?
- An aggregate trend reverses once you stratify. Which conditioning set is correct?