Analytics and experiments

Causal inference interview questions

When you can’t run an experiment, someone still wants to know whether the thing worked. Causal inference questions give you an observational comparison, like a feature launched in Melbourne but not Sydney, and ask whether the claimed lift holds up.

Expect to be asked about confounders, mediators and colliders, Simpson’s paradox, difference-in-differences and its parallel-trends assumption, propensity score matching and uplift modelling. The strongest answers say exactly what would have to be true for the estimate to be right, and how you’d check it.

Try one now

Standard

You launched a new checkout flow in Melbourne but not Sydney. Over four weeks, Melbourne's orders rose 15% and Sydney's 6%. The product team claims a 9% lift. Do you agree?

Free, no sign-up. The interviewer asks follow-ups on your reasoning, then rates your answer.

What the 10 questions cover

  • Correlation vs causation8
  • Difference-in-differences4
  • Confounders, mediators and colliders2
  • Simpson's paradox1
  • Propensity score matching1
  • Synthetic control1
  • Instrumental variables1
  • Regression discontinuity1
  • Uplift modelling1

A question can cover more than one topic, so the counts add up to more than 10.

How the practice works

  1. The AI interviewer asks the question and follows up on your answer, the way an interviewer would.
  2. Stuck? Ask for a hint. Each one gives a little more away, so you can get unstuck without being handed the answer.
  3. Get a written debrief. Four ratings, each from Needs work to Interview-ready, tied to what you actually said, and what to study next.

More analytics and experiments questions

See all categories