News for hackers. Stuff that matters.
Re:The Statistical Crisis in Science by yummyfajitas on Friday January 16, 2015 @09:36AM (#8899109)
No, the problem is that p(reject|null hypothesis) != p(reject|null hypothesis, experimenter saw pattern).

Its true that continuous observation eventually fixes things. In the limit of large N, we can skip the stats and just do division.

There are no comments on this comment yet.