Some Basic Issues for Clinicians Concerning Things Statistical

Abstract

From the title, you may be wondering whether this is another one of those esoteric, unintelligible statistics articles. Or whether the author is another pedantic statistician telling you how much you don't know. I hope that you will find that neither of these fears applies. There are no formulas, derivations, or complex figures here. And I am no “statistician,” at least not in the formal sense, although I do teach statistics. Rather, the focus of this article is to identify and explain, for the clinician consumer of the literature, four basic but often overlooked issues in statistics (as practiced in today's medical and social science journals). Although it is true that most clinicians are misinformed about things statistical (Wulff et al., 1987), it is also true that most academic psychologists and physician researchers—and a surprising number of statisticians—are too (Cohen, 1994; Falk & Greenbaum, 1995; Goodman, 1999a; Oakes, 1986).