There Is No Correlation Between Gun Ownership Rates and Homicide Rates

A few years ago, Professor Carlisle Moody of the College of William & Mary wrote a paper on homicide and gun ownership that showed that there really isn't a correlation between gun ownership and homicide rates. Here's some data that I recreated in graphical form (I also had Excel add a trendline):

As you can see, there are a few countries with far higher murder rates than the US (Estonia, Russia and South Africa) that have far higher murder rates despite far lower gun ownership rates. Russia actually has a 4x greater murder rate despite having one tenth the firearms ownership. Switzerland had double the gun ownership rate of New Zealand yet has less than half the murder rate. It should be no surprise that the R squared statistic (a measure of correlation) for this comparison is 0.006.

I'm sure some will argue that Estonia, Russia and South Africa are outliers and so they should be excluded. Well even if you take those away, the R squared is only 0.2, a weak relationship at best. If you take out the US, which is an outlier on gun ownership, almost double the second highest country, the R squared goes back down to 0.031 for the 22 countries remaining in the dataset.

Some others will probably argue that you can't argue with the fact that we have a higher homicide rate and a higher gun ownership rate than most western countries. That is true, but there are probably other things at play here. I mean, we have a gun ownership rate that is 15 times higher than the UK and a homicide rate that is a bit more than double. That's not a one to one correlation, not even close. Part of it could be a different cultural/ethnic makeup. If you look at the DOJ data on homicide offenders by race, homicide rates among whites is actually much closer to what we see in Europe (3.5 per 100,000 among whites compared to 5.56 for the US as a whole and 2.1 in the UK).