Bayes' theorem: a positive test is not a diagnosis
A population of 200 people, 5% sick, is tested with a 90% sensitive, 90% specific test. The animation counts out 9 true positives and 19 false positives among those who test positive, then builds Bayes' formula step by step to show P(sick | positive) is only about 32%. Paced deliberately with clear pauses between counting, grouping, and the equation, it helps students see why base rates matter before trusting a positive result.
16:9 · Preview before teaching · automatic layout checks do not establish subject accuracy
The prompt that made it
Apply ALL of the following review notes precisely; keep everything else identical. This video is for social media and must be impeccable. - The idea of the video is good, however it is lakcing a bit better flow and timing, a lot of information and equations are being thrown very rapidly
Make the next one in this series
- Show how P(sick | positive) changes as disease prevalence drops from 5% to 0.5%
- Compare two tests with the same sensitivity but different specificity side by side
- Animate a tree diagram version of the same 200-person example for visual learners
