Inconclusives in firearm error rate studies are not ‘a pass’

Abstract

One question Professors Arkes and Koehler (2022) (hereinafter ‘A&K’) ask in their thoughtful paper is ‘What role should “inconclusives” play in the computation of error rates?’ (p. 5) The answer to this question is vital because the number of inconclusives in firearm error rate studies is staggering. For example, firearm examiners in the FBI/Ames Laboratory study made 8,640 comparisons, of which 3922 (45%) were deemed inconclusive (Bajic et al., 2020, Table V). The most recent firearm error rate study reported that 51% of all comparisons were deemed inconclusive (Best and Gardner, 2022). Determining how to count half of the responses is the critical—perhaps even decisive—factor in interpreting the error rates from the study. A&K assert that inconclusives should play no role in computing error rates. They ‘argue that inconclusives should not be coded as either correct or incorrect when tabulating forensic error rates’ (p. 1). Furthermore, A&K write, ‘in the more usual sense of the meaning of error, an inconclusive is not an error. It is a pass. An inconclusive means that the examiner offers no judgement about whether two [cartridge cases or bullets] do or do not share a common source’ (p. 9, citing Koehler, 2008).

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 14 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.