What is a collider and why is conditioning on it bad?
hardAnswer
- In a DAG A → C ← B: C is a collider (both A and B point into it).
- A and B are marginally independent, but conditioning on C creates a spurious association.
- Classic example: 'talent and beauty are anticorrelated among Hollywood stars, because being famous (a collider caused by both) has been implicitly selected on'.
- Same mechanism drives selection bias, sample-conditioning artifacts, restaurant-only surveys (studies that condition on 'made it to production' etc.).
Check yourself — multiple choice
- Random
- In A → C ← B, conditioning on C opens a spurious A-B association → selection bias; textbook: Hollywood 'talent vs beauty' myth
- Same as confounder
- Not real
Collider: A → C ← B; conditioning on C creates spurious A-B association.
#causal-inference
Practise Statistics Fundamentals
215 interview questions in this topic.