Toward evidence-based medical statistics. 2: The Bayes factor

Authors
Citation
Sn. Goodman, Toward evidence-based medical statistics. 2: The Bayes factor, ANN INT MED, 130(12), 1999, pp. 1005-1013
Citations number
57
Categorie Soggetti
General & Internal Medicine","Medical Research General Topics
Journal title
ANNALS OF INTERNAL MEDICINE
ISSN journal
00034819 → ACNP
Volume
130
Issue
12
Year of publication
1999
Pages
1005 - 1013
Database
ISI
SICI code
0003-4819(19990615)130:12<1005:TEMS2T>2.0.ZU;2-N
Abstract
Bayesian inference is usually presented as a method for determining how sci entific belief should be modified by data. Although Bayesian methodology ha s been one of the most active areas of statistical development in the past 20 years, medical researchers have been reluctant to embrace what they perc eive as a subjective approach to data analysis. It is little understood tha t Bayesian methods have a data-based core, which can be used as a calculus of evidence. This core is the Bayes factor, which in its simplest form is a lso called a likelihood ratio. The minimum Bayes factor is objective and ca n be used in lieu of the P value as a measure of the evidential strength. U nlike P values, Bayes factors have a sound theoretical foundation and an in terpretation that allows their use in both inference and decision making. B ayes factors show that P values greatly overstate the evidence against the null hypothesis. Most important, Bayes factors require the addition of back ground knowledge to be transformed into inferences-probabilities that a giv en conclusion is right or wrong. They make the distinction clear between ex perimental evidence and inferential conclusions while providing a framework in which to combine prior with current evidence.