In other words, people who are willing to break the law to own guns might just be more likely to use them to kill people than people who go through legal channels to acquire them.
One thing that is quickly and frustratingly apparent when working on a gun crime/gun control related project is just how shitty the data on gun ownership is. This is because gun owners do not want to divulge any information about their personal store (for both valid and paranoid reasons), and the gun lobby fights for them on this issue. Recall the recent NY shitstorm over this and the fact that Manchin-Toomey (which ultimately lost) had to go above and beyond to quell all fears about a national gun registry.
As a result of this, people hoping to make statistical arguments linking gun ownership or gun rights can just choose a proxy that seems to support the conclusion they want to reach and work backwards from there. This is how we get studies that come to sweeping conclusions like this, when in reality, high gun crime in places like Little Rock, AR and Birmingham, AL (which is responsible for the high numbers attributed to the state as a whole) has a lot more to do with the factors that cause murders in Chicago and Oakland, and not the loose gun laws put in place for the benefit of hunters in rural Southern counties.
Having said all that, I too had no choice but to use proxies for my forthcoming long-form project. Working with state-level data today, I came up with these two charts. The green line is an indexed homicide rate for each state, with the states sorted from most murders to least. The index is based on the number of homicides per 100,000 (from the most recent FBI statistics), with the highest-scoring state (Louisiana) set to 1. All other states are calibrated as a proportion of Louisiana’s total. I used the same indexing method for the proxy I chose for the amount of illegal guns in circulation, which was total weapons arrests per 100,000 people. Illinois was the leader in that category and set as the top of the scale.
For gun laws, I took each state’s rating from the Brady Campaign to Prevent Gun Violence (which is based on a 0-100 scale, with highest meaning more restrictions to gun ownership), divided it by 100 and subtracted that from 1, to get to a scoring system where a higher number would mean easier legal guns and vice versa.
Blue squares represent each state’s “reverse Brady” score, orange diamonds mark the weapons arrests data point. I added linear trendlines in the same colors for clarity. I’m looking forward to getting deeper into city/county level data and of course, many other factors (in the book!), but I thought these two charts were interesting.
Without further comment: