Likelihoodism, Bayesianism, and a Pair of Shoes

10 Pages Posted: 25 May 2013

See all articles by David H. Kaye

David H. Kaye

PSU - Penn State Law (University Park); ASU - College of Law & School of Life Sciences

Date Written: 2012


In R v. T, [2010] EWCA (Crim) 2439, a footwear analyst followed recommendations of the Forensic Science Service in testifying to the weight of the evidence according to a standardized table for characterizing likelihood ratios, reporting that the evidence established “a moderate degree of scientific evidence to support the view that the [Nike trainers recovered from the appellant] had made the footwear marks” in question. The Court of Appeal for England and Wales offered a variety of reasons for holding that this testimony should not have been received. Although the opinion can and should be read narrowly, the apparent preference for traditional opinion testimony about the source of such trace evidence is unfortunate. This essay adds to previous criticism of that aspect of the opinion by distinguishing between likelihood and Bayesian theories of inference. It argues that courts should receptive to the efforts of forensic analysts, guided by either of these theories, to avoid source attributions and to direct their testimony to the strength of the evidence with respect to competing hypotheses.

Keywords: weight of evidence, forensic science testimony, inference, interpretation, Bayes' rule, likelihood ratio, source attrribution

JEL Classification: C10, C11

Suggested Citation

Kaye, David H., Likelihoodism, Bayesianism, and a Pair of Shoes (2012). Jurimetrics, Vol. 53, No. 1, Fall 2012, Penn State Law Research Paper No. 15-2013, Available at SSRN:

David H. Kaye (Contact Author)

PSU - Penn State Law (University Park)

Lewis Katz Building
University Park, PA 16802
United States


ASU - College of Law & School of Life Sciences ( email )

111 E Taylor St.
Phoenix, AZ 85004
United States


Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics