Mediation analysis — direct vs indirect effects.
hardAnswer
- Path: T → M → Y (with possible direct T → Y).
- Decomposition: total effect = direct effect (T → Y not through M) + indirect effect (T → M → Y).
- Baron-Kenny style regression coefficients or, better, potential-outcomes-based Natural Direct / Indirect Effects (Pearl, Robins).
- Identification requires no unmeasured T-M, T-Y, or M-Y confounders.
- Key application: understanding causal mechanism, not just total effect.
Check yourself — multiple choice
- Not real
- Total = direct (T → Y not via M) + indirect (T → M → Y); use Pearl/Robins Natural Direct/Indirect Effects under no-unmeasured-confounding at T-M, T-Y, M-Y
- Same as regression
- Random
Mediation: total = direct + indirect (T→M→Y); Baron-Kenny / natural effects.
#causal-inference
Practise Statistics Fundamentals
215 interview questions in this topic.