Intellectual Instinct

FINDING OUT

Experiments

The base-rate trap

A test can be accurate and still be hard to interpret. If a condition is rare, false positives can outnumber true ones. Change the inputs to see why.

CHANCE OF CONDITION AFTER A POSITIVE

50.0%

Per 10,000 people: 99 true positives and 99 false positives.

Bayes' rule: P(condition | positive) = prevalence × sensitivity / [prevalence × sensitivity + (1 − prevalence) × (1 − specificity)]. A toy model, not medical advice. No data leaves this page.