Inconclusives, errors, and error rates in forensic firearms analysis:Three statistical perspectives

Area of Science:

• Forensic Science
• Statistical Analysis
• Human Factors in Forensics

Background:

• Published error rates for forensic firearms examiners are very low (<1%).
• These low rates are challenged for not fully accounting for "Inconclusive" findings.
• The interpretation of "Inconclusive" results and their impact on error rates is a significant debate.

Purpose of the Study:

• To critically examine the impact of "Inconclusive" results on reported error rates in firearms examination.
• To offer new statistical perspectives on evaluating examiner performance.
• To assess the reliability of current error rate estimates and propose improvements for future studies.

Main Methods:

• Review of existing viewpoints on "Inconclusive" results in firearms examination.
• Application of three novel statistical perspectives: ideal objective measurements, basic sampling theory, and human study experimental design.
• Analysis of how "Inconclusive" findings in studies may differ from casework.

Main Results:

• "Inconclusive" results can be interpreted as errors, correct findings, or justified outcomes depending on the perspective.
• Even without counting "Inconclusive" as errors, doubt is cast on current error rate assessments.
• "Inconclusive" findings in studies may mask potential errors in actual casework.

Conclusions:

• Current studies do not allow for trustworthy error rate estimates; only reasonable bounds can be established.
• Potential error rates are likely much higher than those reported in existing studies.
• Improved study designs, incorporating objective measures or blind proficiency testing in casework, are essential for sound error rate estimation.