Zero correlation of Brady Center scores with (preventing) violence

[dnowell]

I read the other thread about the Brady bunch scoring states on gun control. Well, that gives a pretty good starting point for seeing whether their measures work. If the Brady center is right, then states that adopt more of their measures should see violent crime reduced. Note that I'm not graphing gun crime, because I honestly don't care whether I get killed with a knife or a gun. Done either way. I'm interested in the overall effect of Brady legislation on violence. The sources are included in the graphs, which came out a little bit big.

The result: effectively zero correlation. The R2 value shows how strongly they correlate. 1.0 is perfect, 0 is no relationship. The correlation between brady scores and crime rate is .000093, but in the wrong direction: states that have higher brady scores have very very slightly higher crime rates.

Now, what if there was some other thing that might reduce violent crime?

I also graphed % of population with concealed handgun licenses against violent crime. There turned out to be a slight correlation, just under 0.1. It's weak, but arguably present. Both graphs are below. My apologies again for the over-sized images.

[DefensiveCarry.com]

[SamRudolph]

That's what my review of the pertinent literature showed.

There wasn't a very strong correlation either way between crime rates and strict gun control OR liberal carry laws.

[Bandolero]

That's what my review of the pertinent literature showed.

There wasn't a very strong correlation either way between crime rates and strict gun control OR liberal carry laws.

I have spent a lot of time analyzing these types of numbers over the past few years. This is also my conclusion.

What CCW allows is the possibility of one pulling an ultimate ace out of their sleeve, and giving some dirtbag the surprise of their (hopefully about to be short) life.

[automatic slim]

Actually we can plainly see the results of implimenting the Brady's recommendations, Chicago. You'll notice the Brady's aren't bragging about Chicago's "low" murder rate.

[Tom357]

With respect for your labor in pulling this together, a snapshot in time will not tell you how the rate of violent crime has been impacted by either Brady measures or by concealed carry, since it is a discrete point in time snapshot (violent crime in 2007). All this tells you is the rate of crime in states in 2007, and separately, their Brady ratings. There will be statistically insignificant correlation, as you show, because in this light, they are independent facts. If you want to measure the influence of CC or Brady measures on crime, then you would have to look at the crime rates before and after implementation of these measures, in each state, to see how they impacted crime rates in each state. I think isolating the Brady measures and studying their specific impact on violent crime would be a very interesting study. John Lott's research probably comes closest, but even there, I don't think he looked at all the measures encompassed by the Brady ratings.

[beaker]

i agree with the first part of your post Tom. but i disagree with the 2nd part.. i mean, i can see it showing about as much of a correlation as the above graphs do. which, in a way, would more greatly enforce what the graphs are proving if they both showed the same thing.. but on to my point.. there are just to many factors that go into crime rates. crime rates differ every year per state. some states go up and down year to year. showing no real correlation. like recently, crime rates in many states are going up. and its been linked to the economy. people are loosing jobs, loosing money, loosing their homes... not every law abiding citizen is a law abiding citizen forever. some will, and often do resort to crime when they feel it's needed. also, how strictly laws are enforced, and how strict the punishments are and weather or not they are enforced as well are contributing factors. for example, if you make punishments an eye for an eye type punishments, that will no doubt effect crime rates.

not to mention, i really don't think criminals follow laws that closely. say suddenly, california becomes a shall issue state. you think criminals will know immediately? do you think every law abiding citizen will immediately run out to get their permit and a gun? criminals aren't gonna realize right away that every day citizens can and do have concealed guns or other self defense weapons. and not every citizen will run out to get a permit and a weapon. simply put, laws don't have immediate effects. as criminals learn people are starting to carry, they become more careful when picking victims. but ultimately, the great majority of the population still doesn't carry. so criminals are still like to find a victim that doesn't carry. in order for laws to effect crime rates, people will need to first know about them, and than take their personal safety into their own hands.

so therefore, i think graphing crime rates before and after implementation of gun/carry laws isn't a good way to show correlation of crime rates to gun/carry laws. i don't think there is any good way to show a correlation between gun/carry laws and their effect on crime rates. or the brady campaign vs crime rates for that matter.

[Daddy Warcrimes]

Crime causes gun control; gun control does not reduce crime.

Carry is about individual safety rather than community safety. The low probability of an armed citizen encountering a deadly threat prevents it from having any significant impact. The margins that show it reducing crime are well within the acceptable statistical error range.

[Hopyard]

Just a couple of quick technical comments: 1) Isn't the correct correlation coefficient "r" and not r squared? Using r squared instead of r exaggerates how close the Brady score outcome approaches zero ; it also understates the correlation between handgun licenses and crime reduction.

2) How about a test of the significance of r? Is the trend you seem to think present real?

(See a statistics text on how to do that test, or check out some easy to use stat package such as GraphPad StatMate)

However, when you are looking at handgun licenses v crime, r is actually +.313 That makes a nice positive correlation if r turns out to actually be statistically significant. And it probably is, given the relatively high number of data points.

One more technical issue. The correlation coefficient for the license v crime graph may be invalid because % style data isn't normally distributed and must be transformed before r is calculated. I think it is the arcsine transformation that is supposed to be used. A common error btw. But, this needs attention in the analysis especially as the percentages move toward either being very high or very low as in this case.

Oh, very very nice presentation. Thanks for the hard work.

[beaker]

wow. a blast from the past there. Hopyard bringing back things i never thought i would ever need to remember in math. i think you're right about the % thing. for it to be correct the % of concealed carriers would have to be concealed carriers per 100,000 population.

and wouldn't R^2 represent a parabolic graph. its a linear graph so yeah, it should be just 'r'

[Hopyard]

wow. a blast from the past there. Hopyard bringing back things i never thought i would ever need to remember in math. i think you're right about the % thing. for it to be correct the % of concealed carriers would have to be concealed carriers per 100,000 population.

and wouldn't R^2 represent a parabolic graph. its a linear graph so yeah, it should be just 'r'

I like the arcsine transformation because it is quick and easy and some programs will do it for you. You are right I think that rate (carriers per 100,000 population) could be used in stead of %

I don't know about what a graph of r squared would look like, but I am certain that the statistic for correlation coefficient is r and not r squared. This actually makes your results look much better than the way you presented them. An r of .313 ain't bad in real terms if there are lots of 'n's present in the whole population.