2 ms·
I don’t follow but am interested in your point. Please elaborate.
by andrewseanryan 7y ago
I don’t follow but am interested in your point. Please elaborate.
- gms7777 7y agoSuppose 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