{"blocks": [{"key": "0d70c07a", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "a1ebec4d", "text": "Interview for a data-science role on a growth analytics team that needs to estimate treatment effects from observational product data.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "768fd84e", "text": "Question", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "e60013dc", "text": "What is Propensity Score Matching (PSM) and in which business situations would you apply it? 2) List the assumptions required for PSM to yield unbiased causal estimates. 3) How do you assess whether the matching achieved good covariate balance?", "type": "unordered-list-item", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "0a2b6efc", "text": "Hints", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "5d774e14", "text": "Discuss ignorability, common support, balance diagnostics, and potential limitations of PSM.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}], "entityMap": {}}