timezone icon TimeZone (UTC)
Likelihood Ratios for Evidence Quantification in Forensics
30 Minutes
Originally presented at the 2015 Impression, Pattern and Trace Evidence Symposium.

In many forensic applications involving impression patterns, an examiner is asked to compare an artifact or impression recovered at a crime scene with a potential source of the impression recovered from a suspect, and arrive at an opinion as to the likelihood of the suspect being the source of the crime scene evidence. The subjective nature of this practice has come under scrutiny and efforts are under way to develop objective ways of numerical quantification of crime scene evidence. There is currently growing support for the use of a Likelihood Ratio (more properly referred to as Bayes Factor) as a measure of the strength of the evidence rather than making definitive statements such as “In my expert opinion the suspect is the source of the crime scene impression.”

The justification for the use of the Bayes Factor is based on the odds form of Bayes Rule usually written as: Posterior Odds = Bayes Factor × Prior Odds.

The argument in favor of the use of the Bayes Factor is often articulated in a manner similar to the following: the calculation of prior odds, and hence the posterior odds, is outside the domain of expertise of the forensic examiner; however, the calculation of Bayes Factor is within the examiner’s domain of expertise. Knowing the numerical value of the Bayes Factor, the judge and/or members of the jury can each update their own individual prior odds of the proposition Hp that the suspect is the source of the crime scene evidence, by multiplying it by the Bayes Factor and obtaining their own individual posterior odds.

Superficial considerations lead one to accept such an argument for the use of Bayes Factors for reporting strength of forensic evidence. What is not widely known is the fact that there are fundamental flaws associated with viewing the Bayes Factor as strength of evidence. This webinar explains what these flaws are and provides illustrative examples to explain the corresponding consequences. This presentation also points to alternative approaches for quantification of forensic evidence.

This webinar was recorded in its entirety at the time of the Live event in order to capture the one-on-one interaction with the presenter.

Speaker

Hari Iyer
Statistical Engineering Division, Information Technology Laboratory, National Institute of Standards and Technology.
file

This event is funded through a Cooperative Agreement (2011-DN-BX-K564) from the National Institute

of Justice (NIJ), Office of Justice Programs (OJP), and U.S. Department of Justice (USDOJ). The views,

policies, and opinions expressed are those of the authors and contributors and do not necessarily reflect

those of the NIJ, OJP, or USDOJ.