Talk:False positive paradox

Comment
I can't be bothered to change it, but this page does not explain the paradox at all. The paradox is that the less people as a whole who have the disease, the likly someone someone who tests positive is not positive. For example if 50% of people had the disease, if you tested positive, you have an extremely high chance of having the disaease. Yet if almost everyone didn't have the disease, chances are you don't have it either.--155.144.40.31 (talk) 04:25, 26 August 2008 (UTC)


 * I am confused by your criticism. The article already states "If a patient received a positive response from the test the odds are ~99.02% (9999/10098) that they are healthy and the test is incorrect even though the test is 99% accurate."  Remember (talk) 10:22, 26 August 2008 (UTC)

merge?
This should be merged somehow with Bayesian theorem, there is similar example. I added at least a link to that 84.16.123.194 (talk) 12:50, 20 December 2008 (UTC)

Not a paradox
Not a fucking paradox. Article should be deleted / merged into base rate fallacy. Articles are exactly the same, even with same example.

The New England Medical Journal of 1987 - talking about AIDS testing - has a very good small graph/chart that visually shows how a false positive can become a likelyhood in a diseaes test ( etc) for a rare disease that is confined normally to a different subpopulation. 159.105.80.220 (talk) 17:48, 10 January 2011 (UTC)

Source
Nice article on the subject: Ocaasi c 05:33, 12 April 2011 (UTC)
 * http://www.genomics.princeton.edu/storeylab/papers/Storey_FDR_2010.pdf

Paradox
It is possible to define a positive test result as a test result that is more likely to happen if the target disease is present. Such a reasoning is very natural, intuitive. Thinking in such way is unfortunately wrong and leads to a paradox. A good definition of a better test result solves the problem. Soete michel (talk) 20:51, 6 January 2015 (UTC)