The Nature of the P Value

Authors:
Publisher website
Google Scholar
Abstract
Pocock and Stone (Sept. 8 issue) provide an excellent primer on evaluations of clinical trials and their applications to practice.1 However, early on in their article they make a common but erroneous statement regarding the nature of the P value: “A P value of 0.05 carries a 5% risk of a false positive result.” This misconception is found throughout statistical and medical literature and was recently addressed in the following statement, published earlier this year by the American Statistical Association: “P-values do not measure the probability that the studied hypothesis is true, or the probability that the data were produced by random chance alone.”2 A P value is an expression of the probability of getting results at least as extreme as what was observed, under the assumption that the null hypothesis is true.2,3 The P value as it was originally intended was simply meant to serve as a guide for evaluating whether an experiment was worth repeating.

This publication has 4 references indexed in Scilit: