The Kansas University Medical Center has a statistics journal club and this month, they reviewed an article
- Strasak AM, Zaman Q, Marinelli G, Pfeiffer KP, and Ulmer H (2007) The Use of Statistics in Medical Research: A Comparison of The New England Journal of Medicine and *Nature Medicine. The American Statistician, 61(1): 47-55. Article is behind a pay wall
There was a lively discussion that was difficult to summarize. There is a general consensus that the research problems identified by this article are also true in our personal experience. The problems with research in the basic sciences are probably due to the fact that many basic researchers have been used to doing their statistics on their own and they need to consult more often with statisticians.
If someone comes to you are and presents an approach that you know is flawed, you don’t have to acquiesce just because other people have done it this way and got it published. Just show them this article and point out the Lemming Rule of Research (If all your friends jumped off a cliff, would you jump off also?)
You can find an earlier version of this page on my old website.