Monday, April 8, 2013
Sketchy Stats
Upon reading about the faulty use of statistics employed by Pottie et.al. in their cancer research study, I was somewhat disheartened. The mistakes made by the researchers are of such a careless nature it is truly surprising that the paper was published in the first place. Having mistakes such as "off-by-one" index errors in a paper that was published in Nature Medicine seems completely inexcusable. Perhaps what bothers me most about the mistakes in this paper is how easily they could have been fixed and prevented if the researchers had collaborated with statisticians. As statistical analyses continue to become more and more advanced, it seems inevitable that researchers and statisticians will have to collaborate with one another in order to prevent false conclusions from being drawn.
Subscribe to:
Post Comments (Atom)
I agree with your post, Colleen. Statisticians and researchers should either begin coordinating with one another more closely or researchers should have to take classes or get a degree that deals with the use of statistics. I also found it appalling that the researchers in Pottie et. al. were completely careless in the use of their statistics. In fact, I found it disheartening and worrisome. Also, I think that there should be stricter peer review stages. Maybe then we would have less mistakes published in prestigious medical journals.
ReplyDelete