hard20 pts · first answer countscausal-inferencebias
A link that only appears in a subset
Among hospitalised patients, two unrelated diseases appear negatively correlated, although they are independent in the general population. What explains this?
Discussion 0
Join free to post in the discussion.
Related questions
- Survey sent only to app usersData Scientist
- Studying only active usersData Scientist
- Coaching the worst storesData Scientist
- Difference-in-differences assumptionData Scientist
- Feature users retain betterData Scientist
- Propensity scores near 0 and 1Data Scientist
- Key assumption of an instrumentData Scientist
- Direction of omitted variable biasData Scientist