More unjustified inferences from limited data in Guyll et al. (2023).

This article critically examines the study by Guyll et al. on the accuracy of firearms examination, specifically cartridge case comparisons, arguing that its conclusions about the method’s high validity are overstated due to methodological limitations. Key concerns include variability in examiner performance, with some practitioners exhibiting high false positive rates, and the use of convenience samples that do not represent the broader population of examiners, many of whom lack certification or accreditation. Additionally, the study’s limited firearm models and failure to include challenging “close non-match” samples introduce spectrum bias, undermining the generalizability of its findings to real-world casework. The article emphasizes that these issues, alongside prior critiques, caution against accepting Guyll et al.’s claims as definitive evidence of firearms examination validity in the criminal justice system.