Gelman post on meta-analysis of the Daryl Bem research on precognition (yes, precognition):
The ESP context makes this all look like a big joke, but the general problem of researchers creating findings out of nothing, that seems to be a big issue in social psychology and other research areas involving noisy measurements. … I have a feeling that the authors of this paper think that if you have a p-value or Bayes factor of 10^-9 then your evidence is pretty definitive, even if some nitpickers can argue on the edges about this or that. But it doesn’t work that way. The garden of forking paths is multiplicative, and with enough options it’s not so hard to multiply up to factors of 10^-9 or whatever. And it’s not like you have to be trying to cheat; you just keep making reasonable choices given the data you see, and you can get there, no problem. Selecting ten-year-old papers and calling them “exact replications” is one way to do it.
I think the parapsychology research is actually extremely useful, especially if one is willing to take as incorrigible the proposition that parapsychological phenomena aren’t real. Because then parapsychology serves as a kind of control group for science practice, and what’s striking about the Bem research is how much it looks like ordinary psychological science–even psychological science that goes above and beyond the norm–and yet the findings are what they are.