C · DecisionsAdvanced30 min read
6. Experiments & causality caution
Data Literacy · 32 pages source format
Correlation ≠ causation. Novelty effects, seasonality, confounds. QA can challenge “the A/B won” narratives.
What you'll learn
- Confounds
- Peeking
- Practical significance
Challenge a claim
Do this now
Take one causal claim at work. List 3 alternative explanations.
Clear?