Interview: after segmentation, one segment has 60% women but each region shows 40%. Why?
hardAnswer
- Simpson's paradox — different regions have different sizes / proportions in that cluster.
- Verify with contingency table: within each region, the segment may be 40% women, but the sample-size-weighted overall proportion is 60% because certain regions dominate the segment.
- Fix by post-stratifying — report within-region proportions if that's what stakeholders care about.
- Also check for sampling bias / SRM in segmentation.
Check yourself — multiple choice
- Bug
- Simpson's paradox: aggregate 60% ≠ within-region 40% due to region size imbalance; post-stratify to report within-region proportions; check for SRM / sampling bias
- Random
- Not real
Simpson's paradox in segmentation: aggregate ≠ within-region; post-stratify.
#interview#causal-inference#clustering
Practise Unsupervised Learning
214 interview questions in this topic.
Related questions
- Interview: 'you're asked to segment 10M customers — how?'
- Interview: cluster 5M e-commerce customers for a marketing campaign.
- Interview: how would you group log messages from a large distributed system?
- Interview: business asks 'why is customer X in Cluster 3?' — how do you explain?
- Interview: 'you have to pick k for the executive team — walk me through it.'
- Interview: how would you cluster 1B rows on a budget?