MEPX
Chapter 6 of 10All chapters
  1. 1What correlation measures
  2. 2The five explanations
  3. 3Confounders
  4. 4Experiments
  5. 5Natural experiments
  6. 6Weighing causal evidence
  7. 7Selection effects
  8. 8Regression to the mean
  9. 9Language and reporting
  10. 10Acting under uncertainty

Chapter 6 of 10

Weighing causal evidence

When experiments are impossible.

Considerations that help

Strength of association, consistency across studies and populations, dose response, correct time order, plausible mechanism, and specificity. None alone proves cause; together they build a case.

  • These are the Bradford Hill considerations, developed for smoking and cancer.
  • They are a guide for judgement, not a checklist to score.

Convergence

Independent methods with different weaknesses pointing the same way is the strongest observational evidence available, which is how smoking was established.