Date of Graduation

2020

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Forensic and Investigative Science

Committee Chair

Keith Morris

Committee Member

Jacqueline Speir

Committee Member

Casper Venter

Committee Member

Casey Jelsema

Abstract

The forensic science pattern comparison areas, including fingerprints, footwear, and firearms, have been criticized for their subjective nature. While much research has attempted to move these disciplines to more objective methods, a majority of examiners are still coming to conclusions based on their own training and experience. To compare accuracy between examiners, a method called double-casting was used in this study to create plastic cartridge case reproductions. In the first part of this study, double-cast accuracy was evaluated using two automated comparison systems to quantify the similarity. It was determined that the double-casting method used here produces accurate reproductions with low variability between double-casts of the same master cartridge cases. In the second part of this study, 21 test sets were created to send to firearm examiners for comparison. Double-casts were created of the 21 test sets and mailed to each participant. The double-casts ensured that all participants were comparing exhibits with the same level of detail. Automated comparisons were then performed on the examiner test sets using the NIST toolmark comparison algorithms. The results showed that there are differences in the conclusions made by firearm examiners. Automated comparison systems were found to be complementary to examiners and should be used in combination. A Bayesian network was applied for further analysis of the examiner conclusion data based on likelihood ratios (LRs). Examiners were generally informative towards the true proposition in cartridge case comparisons, and inconclusive conclusions were found to provide evidential value with a LR approach.

Embargo Reason

Publication Pending

Download

Share

COinS