Evaluating methods to estimate the effect of state laws on firearm deaths (306456)Matt Cefalu, RAND
*Beth Ann Griffin, RAND Corporation
Andrew Morral, RAND
Terry Schell, RAND
Keywords: gun policy; causal effects; difference-in-differences; simulation study
There is an urgent and growing scientific literature investigating the effects of gun policies on firearm deaths. Unfortunately, reviewers of this literature have concluded that strong claims about the effects of most gun laws cannot be made because estimates of the effects appear to be especially sensitive to statistical modeling choices. Different modeling choices typically imply different assumptions about the data. However, no study to date has comprehensively examined which assumptions might be most appropriate for the type of data being examined in gun policy research. Dr. Griffin will present findings from extensive simulation studies on commonly used methods for estimating the causal effects of gun laws on state-level total firearm deaths and their statistical properties. Dr. Griffin will highlight the optimal methods identified, and additionally discuss the use of Bayesian methods to improve presentation of policy effects so they better capture the amount of uncertainty that is inherent in these estimates. The talk will also discuss RAND’s latest findings on the effects of stand your ground, child-access protection laws, and concealed carry laws.