Pascale Lane reviews a paper about retraction rates in top journals: "Papers 'Not Meant to Be Factual'".
Rigorous peer review may help uncover fraud or fabrication, but, as the editor of Science wrote, "It is asking too much of peer review to expect it to immunize us against clever fraud."
I'm just noting this because, on the subject of my previous post about grant applications, the rate of fabulism must be much higher in that system than in the top journals. Nobody will retract your money if your "preliminary results" fail to play out. And the top journals (or "glamour mags" as many call them) have a shockingly high rate of retractions.