Student Research Projects
Review of Pseudoreplication in the Biomedical Literature
|Student Name(s)||Francis Manno III|
Dr. Maynard Schaus|
|Course||BIO 489: Research in the Natural Sciences|
AbstractPseudoreplication is a statistical error caused by improper use of statistical tests. Description and explanation of this error are lacking in medicine. There are four types of pseudoreplication: simple, temporal, pooling, and implicit. A review of fifty articles total, ten each from five prominent journals: JAMA, BMJ, The New England Journal of Medicine, The Lancet, and The Annals of Internal Medicine, were analyzed for the occurrence of pseudoreplication and in particular which statistical test was utilized most resulting in pseudoreplication. Multivariate analyses were carried out to determine the correlation of the typologies of pseudoreplication to particular journals and the degree to which particular statistical tests resulted in pseudoreplication.
Pseudoreplication occurred an average 37.4% (SE=5.0) of the time in the studies analyzed irrespective of type, and in AIM 30.0%(SE=12.2), BMJ 30.0% (SE=5.7), JAMA 60.0% (SE=13.7), Lancet, 27.6% (SE=7.2), and NEJM 45.0% (SE=17.3). Temporal pseudoreplication occurred the most frequently in 64% of studies. Temporal and implicit pseudoreplication were significantly related to the first statistic, F=3.45, df=6,43, P<0.007 and F=5.32, df=6,43, P<0.0004, respectively and simple pseudoreplication was significantly related to the second statistic utilized, F=2.81, df=5,44, P<0.03. The number of M.D.’s on a study was significantly related to pseudoreplication error as determined by Roy's Largest Root = 1.2, F=2.18, df=13,23, P<0.05. Pseudoreplication is a serious statistical problem related to the improper implementation and analysis of statistics. Physicians need to be cognizant of the manner in which the error occurs to prevent the inevitable effect on Type I error probabilities.