Empirical claims about gun policy often enter courtrooms framed through p-values and confidence intervals—conventions not designed to answer whether a proposition is more likely than not helpful, or how stable that conclusion is under small data perturbations. This article proposes two diagnostics for firearm policy evidence: the Percent Fragility Index (PFI) and the Risk Quotient (RQ). PFI reports the minimum percentage of outcomes that would need to change to reverse statistical significance. RQ reports the minimum percentage needed to eliminate any claimed policy benefit, independent of p-values. Together these measures translate statistical uncertainty into legally relevant information about reliability and probative value, giving courts and policymakers clearer guidance for admissibility and weight assignment.
Thomas F Heston (Fri,) studied this question.