Statistics: base rate fallacy, part 1: false-positive paradox

Tutoring statistics, you encounter day-to-day ideas. The tutor mentions the false-positive paradox.

false-positive paradox: In a population where condition A is extremely uncommon, most tests for it that yield “positive” may likely be false.

The false-positive paradox is easy to imagine with diseases: suppose any person, showing no symptoms but with known exposure to disease A, carries it with probability 0.002. A blood test to detect its presence will give a false-positive for 5% of non-carriers. Then, from a population of 1000 people exposed to A, only 2 people will actually carry it, but 50 non-carriers will test positive for it.


Jack of Oracle Tutoring by Jack and Diane, Campbell River, BC.

Leave a Reply