The Bayes’ factor: the coherent measure for hypothesis confirmation

Abstract

What have been called ‘Bayesian confirmation measures’ or ‘evidential support measures’ offer a numerical expression for the impact of a piece of evidence on a judicial hypothesis of interest. The Bayes’ factor, sometimes simply called the ‘likelihood ratio’, represents the best measure of the value of the evidence. It satisfies a number of necessary conditions on normative logical adequacy. It is shown that the same cannot be said for alternative expressions put forward by some legal and forensic quarters. A list of desiderata are given that support the choice of the Bayes’ factor as the best measure for quantification of the value of evidence.