Presented by Henry Swofford, PhD, Lead Scientist with the Forensic Science Research Program in the Special Programs Office at NIST
In recent years, there has been controversy relating to the treatment of inconclusive decisions in forensic feature comparison disciplines (e.g., friction ridge, firearms and toolmarks, footwear, tire tracks, and handwriting). While such discourse has spotlighted an important issue in forensic science, it has also conflated several issues and created ambiguity and disagreement within the forensic science community regarding how inconclusive decisions ought to be treated when considering the reliability of a method or outcomes of the method.
In this presentation, the presenter offers a brief review of the various viewpoints and suggestions that have been recently put forth, followed by a path forward that we believe more effectively addresses the issue surrounding the treatment of inconclusive decisions. Specifically, the presenter distinguishes between the contexts of “method conformance” and “method performance” as two distinct concepts but both of which are necessary for the determination of reliability.
Method conformance relates to assessments of whether the outcome of a method is the result of adherence to the procedure(s) that define the method. Method performance relates to measures that reflect the capacity of a method to effectively discriminate between different propositions of interest (e.g., between same source and different source comparisons). The speaker then discusses the different circumstances for which these contexts are applicable (method development, method validation, performance monitoring, and reporting results) and implications for the forensic science community.
Learning Outcomes
By the end of this webinar, you’ll better understand the:
- controversy surrounding inconclusive decisions in forensic science and their impact on error rate calculations.
- the concepts of method conformance and method performance and their role in determinations of reliability.
- way inconclusive decisions should be treated and how measures of performance should be presented.