01/07/2014 03:20 pm ET Updated Mar 09, 2014

The Correlation of Gun Laws and Gun Homicide

Previously, I wrote that when we look at all 50 states we find there is a correlation between states with strict gun laws and low rates of homicide. I noted:

For example, the states with the strictest gun laws are among the states that have the lowest gun violence. Also, states with higher gun ownership and weak gun laws lead nation in gun deaths. These are true findings, but these do not prove gun bans work. It could be than laws or it could be the culture in those states; it could be police funding; it could be an excellent mental health or an excellent school system. It could be any number of third variables.

I questioned the assumptions I was working under. As it turns out, when I used my own methods instead of relying on the methods of others, I did not find a correlation.

I based my assumptions on the Brady Campaign to Prevent Gun Violence, which suggests which states with tougher gun laws have fewer gun homicides. The Campaign selects states that make its point. While their intention and heart are in the right place -- to prevent gun violence -- their methodology is not useful or scientifically rigorous.

If we look at all 50 states, as I did using 2010 census data on the homicide rate per state and correlate that with the Brady's state ranking of strict gun laws, we find a correlation of about .05, which is to say there is no correlation. Even when I looked at the five states with the strictest gun laws and the five states with the least strict gun laws, I did not find a correlation. I also did not find a correlation between the 10 strictest and lest restrictive gun laws states. I looked at the extremes and excluded the states in the middle because there may be a tipping point that could result in a correlation. There was not.


Now, it is extremely important to qualify correlation or the lack thereof and what does and doesn't mean.
  • My approach should not be used by gun rights advocates to make any point. My approach is not conclusive, and it is not meant to be. Correlation is not a useful tool to make a point for or against gun laws and their relationship with gun crime.
  • Correlation is not causation. Everyone knows this, yet both sides of this issue like to violate this rule. Just because you have a correlation, it does not mean there is a causal relationship.
  • We can have causation without correlation. Strict gun laws may result in lower gun homicides, or they may not. Just because we don't find a correlation it doesn't mean that the gun laws don't help. Maybe they do, maybe they don't. The point is that correlation is not a useful scientific measurement tool to determine causation or the lack thereof.
  • I only looked at one year's worth of data. This is insufficient to draw any conclusions about a correlation or the lack thereof.
  • I only looked at gun homicide rates. In looking at gun homicide rates, I am not including gun suicide rates or shootings that did not result in a death.

Often times supporters of the Second Amendment like to point out that Chicago has perhaps the strictest gun laws in the nation but one of the highest homicide rates. This is a cherry-picked example. The other side can also point to New Orleans and its very high rate of gun violence and the very loose gun laws. Cherry-picking is done by both sides of this issue and it does not advance gun safety or a better understanding of what works and what doesn't.

The point of this article is two fold: 1) It is important and responsible to question assumptions and revise statements when something new is learned. This is not flip-flopping as opponents like to say. It is a responsible use of facts and research methods. Both sides of this issue need to do this. And 2) I previously urged caution with the use of correlation and I am further urging caution on this. I want to slow down the loose use of correlation as a way to prove or disprove that there is a causal relationship between strict gun laws and gun homicide rates. Maybe there is, maybe there isn't. Correlation is not going to answer this question.

Paul Heroux is a state representative from Massachusetts. He previously worked for a prison and a jail, and he has a Master's in Criminology from the University of Pennsylvania and a Master's in Public Administration from Harvard. Paul can be reached at