Positive

a 90% accurate test just called you sick. are you?

the false-positive paradox, run on 1000 people

A test that’s 90% accurate still misfires more than intuition expects once the thing it’s testing for is rare: run it on 1,000 people at a 1% prevalence and you get 9 true positives alongside 99 false positives, 108 flagged in all. A positive result on a rare condition is right only about 8% of the time — the test’s accuracy and the odds that your particular result is correct are two different numbers.

Concepts: probability, statistics.

An interactive you play with one thumb, on Wundr.