A New York Times article has appeared about a study on the effects of excessive beer drinking on scientific productivity. The study, (Tomas Grim, “A possible role of social activity to explain differences in publication output among ecologists.” Oikos 117 (4), 484-487) done by the aptly named ecologist, Dr. Thomas Grim, claimed that scientific productivity among Czech avian ecologists and behavioral ecologists (as measured by number of publications, citation rate per paper, etc.) dropped according to how many beers the ecologists drank.
Now first of all, I find it kind of annoying that the New York Times, considers “avian ecologists” a proxy for all “scientists.” And, well, to carry about a stereotype who validity is as good as most other stereotypes, I mean, come on, has anyone ever heard of bird watchers being known for their beer drinking abilities? I suspect if I had to pick the group of scientists least likely to be able to take their beer, avian ecologists would be right up there on my list. Show me a study about Czech physicists damaging their publication record by too much beer consumption, and then you’ll get my attention.
I’d also note that the study covers the amazingly huge sample size of less than twenty, that the beer consumption rates are huge for the outliers (Czech, burp!), that there was no description of the methodology for choosing the survey sample (were they his friends, his colleagues? Since the sample was chosen from the author’s field, it sure sounds like it), that any effect, if it is there, is coming from the very high end of the beer consumption spectra (which is fairly spectacular consumption), and certainly a linear regression seems like a spectacularly poor notion of how beer drinking has an effect on scientific output., and that no attempt to separate out the effect of different quality universities and the different geographic consumption levels was made.