Problem
Scenario
You work on a consumer fintech app and want to test a growth change in the referral and invite experience. The treatment could affect not just the exposed user, but also the friends they invite and the households they belong to, so you are worried that a naive user-level split may contaminate results.
Question
How would you think about the right unit of randomization for this growth experiment? Walk through how your choice would affect metric definition, power, and the validity of the test.
What This Tests
- Choosing between user, household, and cluster randomization
- Recognizing network interference and SUTVA risk
- Defining a primary metric with guardrails and MDE
- Reasoning about power loss from clustered designs
Practicing as: Data Scientist interview at RobinhoodHi, I'll play your Robinhood interviewer for the Data Scientist role. Candidates describe these interviews as mixed and moderately difficult, so expect me to be professional and fair. Take your time with the question above and answer like we're in the room.
You are practicing as a guest. Sign up free to get your answer graded with AI feedback. Your draft stays right here.
