3 ms·
Suppose a test has a 1/100 false positive rate, and the true incidence of the disease over the population that you're testing is 1/1000 (that is, you expect a t
by gms7777 7y ago
Suppose a test has a 1/100 false positive rate, and the true incidence of the disease over the population that you're testing is 1/1000 (that is, you expect a true positive every 1/1000 times).
In this case, every positive test you observe is around 10x more likely to be a false positive than a true positive.
- andrewseanryan 7y agoGot it. Thanks! I have a masters in public health and still need to get schooled by hacker news! Haha