What Statisticians Can Do for Forensic Science

Abstract:     
This is the text (with a few added footnotes) of an address given to the Young Statistician’s Section of the Royal Statistical Society in London on September 28, 2011. In it I compare the work of Cedric Neumann and his colleagues on fingerprint identification with the work of Mary and Peter Bush and their colleagues on bitemark identification. These illustrate that good statistical modeling of sound data can lend support to the assumptions of some forensic identification disciplines, undermine the tenability of others, and can sometimes lead the way toward new techniques to supplement or replace human judgment with more objective assessments.

Number of Pages in PDF File: 6

Keywords: evidence, statistics, criminal law, fingerprints, bitemarks

JEL Classification: C10, K14, K41

working papers series