Gun Laws and Population Sex Ratio

Back in 2009 when Washington, D.C.'s handgun ban went up before the Supreme Court, I was talking with a colleague about it. He mentioned that D.C. also has the highest ratio of women to men in the nation. I checked his facts and realized that he was not only right, but that the state with the lowest ratio of women to men was Alaska, which has pretty unrestrictive gun laws. I wondered if a surfeit of overbearing women voting against things that scare them could account for these observations, and I decided to see if there was any correlation between the sex ratio of a state and the severity of their gun laws.

The Brady Center to Prevent Gun Violence publishes a "scorecard" that rates states according to the extent to which they control gun ownership. There are a number of gun laws that Brady considers important, and each state gets a certain number of points for each one it has implemented. So this makes a pretty good metric for gun law severity. The data for sex ratio can be easily downloaded from the U.S. Census' website.

Sex Ratio and Brady Center Scores by State

Click image for a PDF version.

I collected the data and plotted it in R. The results are pretty clear. A great majority of the states cluster near the left side of the graph with no apparent trend. Alaska is at the extreme upper left, at 106 males for every female. On the right, there is a handful of states that represent the highest scorers (mostly "blue" states such as New York and Massachusetts), and one state that outscores everything else (California). Within this group, there is a correlation of sex ratio and gun law severity, and it's in the opposite direction I had expected: states with fewer women have stricter gun laws. This is a real trend (Pearson's product-moment correlation: p = 0.0001541) and is still significant even if California is removed from the test.

So does this mean that a stampede of gun-hating men are cracking down on gun ownership in "blue" states? I doubt it. First of all, sex ratio doesn't vary that much. The most you will ever see is six extra people out of a hundred, and that is not really enough to create a huge voting bloc. More importantly, there are so many factors involved in social trends that it's almost impossible for any one factor to make a difference. As they say, correlation isn't causation. So this is an interesting observation, but the best conclusion one can make from it is that lonely single women in DC ought to catch the next plane to Alaska.