Estimating Likelihood Ratio (LR) for Firearm Evidence Identifications in Forensic Science

Published

September 15, 2020

Author(s)

Jun-Feng Song, Zhe Chen, Theodore V. Vorburger, Johannes A. Soons

Abstract

Firearm evidence identification has been challenged by the 2008 and 2009 National Research Council (NRC) reports and by legal proceedings on its fundamental assumptions and its procedure involving subjective decisions without a statistical foundation for estimation of error rates. To address these challenges, researchers of the National Institute of Standards and Technology (NIST) recently developed a Congruent Matching Cells (CMC) method for automatic and objective firearm evidence identification and quantitative error rate estimation. Based on the CMC method, a likelihood ratio (LR) procedure is proposed in this paper aiming to provide a scientific basis for firearm evidence identifications. The initial LR estimations using two sets of 9 mm cartridge cases with different sample sizes, imaging methods, statistical models and identification criteria showed that for all the identification conclusions of the tested 2D and 3D image pairs, the estimated LRs for the least favorable scenario were well above an order of 10^6, which can provide Extremely Strong Support to identification conclusions. The LR estimations also showed that for all the exclusion conclusions of the tested 3D image pairs, the estimated LRs for the least favorable scenario were above an order of 10^2, which can provide a Moderately Strong Support to exclusion conclusions.

Citation

Forensic Science International

Volume

317

Pub Type

Journals

Download Paper

https://doi.org/10.1016/j.forsciint.2020.110502

Local Download

Keywords

forensics, ballistics identification, likelihood ratio (LR), congruent matching cells (CMC)

Citation

Song, J. , Chen, Z. , Vorburger, T. and Soons, J. (2020), Estimating Likelihood Ratio (LR) for Firearm Evidence Identifications in Forensic Science, Forensic Science International, [online], https://doi.org/10.1016/j.forsciint.2020.110502, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=928524 (Accessed November 21, 2024)