State gun laws, gun ownership, and mass shootings in the US: cross sectional time series
BMJ 2019; 364 doi: https://doi.org/10.1136/bmj.l542 (Published 06 March 2019) Cite this as: BMJ 2019;364:l542
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Hello,
Is there a list of the mass shootings used in your study as well as those rejected for inclusion in your study? The public has seen before where the actual incidents included people would hardly think of as mass murders.
Further what most people would think of as mass murders, such as Virgina Tech, are still rare beasts. Just having one or two, while tragic skew results significantly. I also question how you accounted for states that allow local preemeption and how many of the mass murders included/rejected were in those zones.
Competing interests: No competing interests
The authors address an important and neglected area. The issue of gun control has been part of the political agenda in many countries.
The article focuses on mass shootings, although it only mentions briefly the issue of suicides resulting from firearms. I would like to draw the readers' attention to Army XXI, an initiative in Switzerland aimed at gun control, which subsequently led to a 22% reduction in firearm-related suicides. This was not substituted by other suicide methods.
Policymakers should take into account that gun control policies may have an impact not only on mass shootings but also on overall suicide rates.
Competing interests: No competing interests
Naive. Fails to control for the proportion of people who actually have guns in their household and relies on nonsense judgements about permissiveness of gun regulations. Instead, see: Handwerker, Our Story (Routledge, 2015), esp chapter 7:
“An Armed Society Is a Polite Society” (Robert Heinlein)
In short, yes, deterrents consist of an ability to inflict evolutionarily significant consequences on anyone who fails to hold up his or her end of the human bargain to treat others fairly. But. Deterrent credibility (and effects) varies with whether people frame choices as gains or losses. Figure 7.4 shows violent crime rates for the United States between 2001 and 2004, given specific values for variables that bear on the evolutionary significance of a consequence and its credibility. If evolved properties of mind provide the means by which we respond to the world of experience, the immediacy with which a consequence bears on survival, eating well reliably, and reproductive success should determine the design of choice alternatives and the costs of their consequences, which we ordinarily call the balance of power in a relationship. Significant and quick consequences should deter violence. A low risk of significant and quick consequences should increase violence. Relatively insignificant and slow consequences should also increase violence. These behavioral trends should reflect the degree to which population frame their choices as losses. (The Appendix at the end of this chapter provides details of the variables and the modes of analysis.)
Here’s the short story. We measured Choices Framed as Losses with information on the size of populations plausibly subject to exploitative dependencies. We measured the likelihood of consequences with information on small-world properties of specific
populations (h/t Russ Bernard) and the cost of consequences with information on the availability of means of defense useful even for small, physically weak, elderly, or disabled people. We checked these results with information on things like house hold income, Gini coefficients,and the size of the local prison population.
The finding? Violent crimes go up when the likelihood and cost of consequences—as measured bysmall-world properties andthe availability of guns—goesdown. Study the effect of the prevalence of loss choice frames, shown as high by large symbols and low by small symbols. Note that a high prevalence of of small world properties and effective means of defense.
Competing interests: No competing interests
Re: State gun laws, gun ownership, and mass shootings in the US: cross sectional time series
Dear Editor,
While Reeping et al. make an interesting contribution to the literature, the methodology of the study has many major flaws and its results ought to be taken with caution.
First, their use of an index is problematic and, for this reason, is rarely used in the quantitative fields of economics or criminology. Such indices are necessarily subjective rather than objective measures of gun control stringency; this is why economists and criminologists study specific laws. They say their index is correlated with other indices, but since the use of an index is always problematic, that doesn't really solve the issue.
Second, the authors fail to discuss the possibility of reverse causation. It has been demonstrated by multiple research studies that gun ownership may rise in response to high levels of crime (see, e.g. Kovandzic et al. 2012, and Kovandzic et al. 2013, Kleck and Patterson 1993). This may also be true of mass public shootings; usually gun sales go up after these incidents, so it is very possible.
Third, the study does not mention clustering their standard errors to correct for serial correlation, which almost always is a problem in longitudinal studies.
Fourth, they should have included state fixed effects in their adjusted models, which is standard in the criminological and economic research using panel data.
Sixth, in the adjusted models, you can't use firearm suicide as a proxy for gun ownership. Kleck (2004), their source #17, explicitly argues that while the percentage of the suicides using guns is a good proxy for gun ownership between different geographic areas, it has little to no correlation with gun ownership longitudinally. He comes to the same conclusion in two newer papers, Kovandzic Schafferand Kleck (2012 and 2013). This means the paper didn't actually look at changes in gun levels.
Competing interests: No competing interests