AFTE Store - An Assessment of the Foundational Validity of Firearms Identification Using Ten Consecutively Button-Rifled Barrels

Using ten consecutively reamed and button-rifled Thompson/Center Arms G2 Contender barrels, fired bullets were collected and 50 comparison test kits distributed to firearm examiners in three countries, including 21 states within the United States. Of the 66 enrolled participants, 44 returned results. Each test kit was comprised of 15 open-set comparisons. Five questioned bullets from each test kit were damaged by being fired into one each of glass, drywall, sheet metal, wood, and Kevlar to replicate damage seen in casework. Participants were unaware that the test kits they were asked to compare were comprised of bullets from consecutively manufactured barrels. Test kits were coded to establish a declared study with double-blind elements. Participants were asked to simply compare the provided knowns to questioned samples, using their laboratory’s policies and procedures to include any quality control measures, and render a result. Allowing participants to use their standard operating procedures examines the entire system and gives insight into realistic error rates that could be expected in real casework. The false identification rate was observed to be 0.455 ± 0.139% (1/220) while the false elimination rate was 1.82 ± 0.555% (8/440), both at the 95% confidence interval. The overall total error rate was found to be 1.36 ± 0.414% (9/660) for 660 total comparisons.