Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 134 - Statistics for Strengthening Inferences from Forensic Evidence
Type: Topic Contributed
Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
Sponsor: Advisory Committee on Forensic Science
Abstract #322599
Title: Characterizing Variability in Forensic Decision-Making with Item Response Theory
Author(s): Amanda Luby*
Companies: Swarthmore College
Keywords: Forensic Science; Item Response Theory; Bayesian Statistics
Abstract:

Black Box studies have been instrumental in providing overall error rates in forensic science, but we generally cannot directly compare individual examiners’ performance, since each participant typically is asked to evaluate a random subset of items and some items are more difficult than others. IRT estimates proficiency among participants while simultaneously accounting for varying difficulty among items. Using an IRT-based analysis, we find that the largest variability in fingerprint examiner decisions occurs in print quality assessments and inconclusive decisions. We also find some participants were likely to over- or under-report difficulty even after accounting for other factors; and examiners who report items to be more difficult perform similarly to examiners who report items to be easier. Furthermore, variability exists in where examiners mark distinguishing features of the print, which are used to determine matches both in comparison with candidate prints and sometimes as inputs into fingerprint databases that generate those candidates. These results underscore the importance of better understanding the cognitive factors involved in latent print examination decisions.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program