Why Most Published Research Findings Are False

Not new (2005), but certainly still newsworthy. Too many scientists, too few first-rate problems to attack, and too many small studies of marginal topics whose statistical basis is more questionable than the attention they get from a breakthrough-hungry popular science press.

I find the premise depressing, and maintain that this doesn't happen in hard sciences and engineering except anecdotically, and also know of researchers in biomed who are irreproachable and would find the gross generalization taken as far as here simply untrue. It's not that the research activity is perfect, but serious debate needs to be more sophisticated than 'most restuls are false'.