Department of Statistics Unitmark
Dietrich College of Humanities and Social Sciences

Bayes Factors: What they are and what they are not

Publication Date

January, 1997

Publication Type

Tech Report

Author(s)

Michael Lavine and Mark J. Schervish

Abstract

Bayes factors have been offered by Bayesians as alternatives to P-values (or significance probabilities) for testing hypotheses and for quantifying the degree to which observed data support or conflict with a hypothesis. Schervish (1996) showed how the interpretation of P-values as measures of support suffers a certain logical flaw. In this paper, we show how Bayes factors suffer that same flaw. We investigate the source of that problem and consider what are the appropriate interpretations of Bayes factors.