Originally posted at http://blogs.warwick.ac.uk/simongates/entry/david_colquhoun_agrees/ on 7 April 2014
On the hazards of significance testing. Part 2: the false discovery rate, or how not to make a fool of yourself with P values
http://www.dcscience.net/?p=6518
makes much the same points as I have made elsewhere in this blog, though he doesn’t go as far and recommend Bayesian analyses. But I can’t see how you can sensibly interpret p-values without a prior, and if you’re going to go that far, a fully Bayesian analysis is the natural thing to do surely?