Michael Siegel, Yamrot Negussie, Sarah Vanture, and Jane Pleskunas are with the Department of Community Health Sciences, Boston University School of Public Health, Boston, MA. Craig S. Ross is with Virtual Media Resources, Natick, MA. Charles King III is with Greylock McKinnon Associates, Cambridge, MA, and Pleiades Consulting Group, Lincoln, MA.
Am J Public Health. 2014 Oct;104(10):1912-9. doi: 10.2105/AJPH.2014.302042. Epub 2014 Aug 14.
We examined the relationship between gun ownership and stranger versus nonstranger homicide rates.
Using data from the Supplemental Homicide Reports of the Federal Bureau of Investigation's Uniform Crime Reports for all 50 states for 1981 to 2010, we modeled stranger and nonstranger homicide rates as a function of state-level gun ownership, measured by a proxy, controlling for potential confounders. We used a negative binomial regression model with fixed effects for year, accounting for clustering of observations among states by using generalized estimating equations.
We found no robust, statistically significant correlation between gun ownership and stranger firearm homicide rates. However, we found a positive and significant association between gun ownership and nonstranger firearm homicide rates. The incidence rate ratio for nonstranger firearm homicide rate associated with gun ownership was 1.014 (95% confidence interval=1.009, 1.019).
Our findings challenge the argument that gun ownership deters violent crime, in particular, homicides.
我们检验了枪支拥有量与陌生人及非陌生人凶杀率之间的关系。
利用联邦调查局犯罪报告中的补充杀人报告数据，对 1981 年至 2010 年全美 50 个州的情况进行分析，我们以代理指标衡量州级枪支拥有量，构建了陌生人及非陌生人凶杀率的模型，通过固定效应的负二项回归模型，对各州的观察结果进行了广义估计方程的聚类分析，以控制潜在的混杂因素。
我们发现枪支拥有量与陌生人枪杀凶杀率之间没有明显的统计学相关关系。然而，我们发现枪支拥有量与非陌生人枪杀凶杀率之间存在正相关关系。枪支拥有量与非陌生人枪杀凶杀率的比值为 1.014（95%置信区间=1.009，1.019）。
我们的研究结果对枪支拥有量可阻止暴力犯罪，尤其是凶杀犯罪的观点提出了质疑。
