Summary
The episode works through four questions that follow from the first lesson of HSCI 230. It reworks John Snow's 1854 comparison of two water companies as a natural experiment, computing death rates per ten thousand houses and discussing the denominator and possible confounding. It then shows with invented examples how testing many outcomes inflates the chance of a false positive, and how the share of significant findings that are true depends on the proportion of true hypotheses, study power and analytic flexibility. The hosts close with a discussion of the hierarchy of evidence, in which they agree that the question being asked should determine which kind of evidence, method or governance framework applies.