cousin_it comments on Open Thread: March 2010, part 3

[cousin_it]

I still fail to see how Bayesian methods eliminate fluke results or publication bias.