A more principled use of the p-value? Not so fast: a critique of Colquhoun's argument

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

The usefulness of the statistic known as the p-value, as a means of quantifying the strength of evidence for the presence of an effect from empirical data has long been questioned in the statistical community. In recent years, there has been a notable increase in the awareness of both fundamental and practical limitations of the statistic within the target research fields and especially biomedicine. In this article, I analyse the recently published article (Colquhoun 2017 R. Soc. open sci.4, 171085 (doi:10.1098/rsos.171085)) which, in summary, argues that with a better understanding and thus more appropriate use of the statistic, many of the aforementioned limitations can be addressed. In particular, I demonstrate that the (often implicit) premises of this counterargument are questionable, in some cases arguably inconsistent, and that therefore the counterargument provides little if any justification for the continued use of the p-value. Additionally, my analysis should help researchers seeking to interpret their empirical data by illustrating the nuanced nature and the multiplicity of statistical, methodological and epistemological issues which must be considered in this process.
Original languageEnglish
Article number181519
Number of pages5
JournalRoyal Society Open Science
Volume6
Issue number5
DOIs
Publication statusPublished - 15 May 2019

Keywords

  • Statistics
  • Bayesian
  • Frequentist
  • Empirical
  • Evidence

Fingerprint

Dive into the research topics of 'A more principled use of the p-value? Not so fast: a critique of Colquhoun's argument'. Together they form a unique fingerprint.

Cite this