Right to Carry vs. crime rates

[illspirit]

I just compiled this image during an argument with a swarm of antis over at my forums:



'Tis a simple mash-up of the animated shall-issue map at Wikipedia and a graph from the DOJ, but I've never seen them animated together like this.

After the lag behind Florida going shall-issue, it almost looks like the crime line is running away from all those blue States, doesn't it?

 

[.cheese.]

while I DO believe there to be a strong inverse relationship between guns and crime, there are two logical flaws with that image:

1) Correlation vs. Causation

2) Unexamined alternatives that would explain why as some states starting becoming shall-issue, initially the rate either remained unchanged or went up, and then down.

Do crimes committed with firearms mirror overall US crime levels? For the most part, yes, - at least the big spike around '93 that you're using to justify the conclusion you want the viewer to deduce. There are alternative theories to it though, such as economist Steven Levitt's well-known and widely published (NY Times best-selling book Freakonomics talked about it) abortion theory (which has come under intense scrutiny just as John Lott's works have).

The large incredibly visible spike in crime I believe is still somehow unexplained. There are a large number of criminologists, economists, etc. who have their own theories.... and I have a feeling the reality was a combination of some of those theories.... but - not just ONE thing - like guns.

I strongly believe carrying a gun helps make myself safer, and that I contribute to deterring crime by being a CCW'er that criminals know are out there..... but there are many other forces out their working their magic to influence otherwise regular people as well as career criminals to do their misguided and sometimes violent deeds.

It's a good image, and thank you for posting it as any work is good work. All I'm sayins is that it would be even better if you could also have it alongside a few other moving graphs that would show how they all contribute to decreasing crime.

Note: I also think that on gun forums, we do very little good to try to explain crime rates solely using ccw and guns. It's the same argument the politicians use in reverse, and frankly it's just as bad. There is not 1 sole factor that controls crime. For the same reason banning guns won't stop crime entirely, if everybody had a gun it still wouldn't stop crime entirely. Both views are equally extreme and equally pointless as well as illogical.

 

[Librarian]

On the other hand, since the crimes did not increase with the increase in CCW, one can say that, whatever was driving the changes in crime, CCW was not making it worse. The time series refutes the rejoinder that 'things might have been better than they actually were had there been no CCW change' - surely some indication of that would have appeared, and it does not.

And 'CCW is not making things worse' is the weak conclusion of
Firearms and Violence:A Critical Review, by the National Academies.

A final lesson to draw from the no-controls dummy variable results is that the results are sensitive to the inclusion of controls. That is, whether one concludes that right-to-carry laws increase or decrease crime based on models of this sort depends on which control variables are included. Such laws have no obvious effect in the model without controls (and therefore no clear level effect in the raw data). Moreover, as demonstrated above, seemingly minor changes to the set of control variables substantially alter the estimated effects. Given that researchers might reasonably argue about which controls belong in the model and that the results are sensitive to the set of covariates, the committee is not sanguine about the prospects for measuring the effect of right-to-carry laws on crime. Note that this is distinct from whether such laws affect crime. Rather, in our view, any effect they have on crime is not likely to be detected in a convincing and robust fashion.

 

[RPCVYemen]

And 'CCW is not making things worse' is the weak conclusion of Firearms and Violence:A Critical Review, by the National Academies.

Thanks for that link. It seems to reinforce an earlier study by the CDC finds no correlation between CCW (or "gun control" laws) and crime rates. I suspect that's correct. The lack of correlation makes for a complicated public policy question. Luckily, the constitutional question is a little clearer.

I do question drawing graphs or cool graphics that seem to draw a correlation between CCW and decreased rates of violent crime. If in fact, there is no correlation, those graphs will be used against us if crime rates rise in the future.

My own guess is that crime rates in general are most strongly correlated with unemployment. In a lecture by the author of Freakonomics, I found out that many of the street corner crack dealers in the gang studied by the author moonlight at McDonald's.

If in fact, crime is correlated most strongly with the economy, and the Great Conservative Spending Spree (aka Bush deficit) does put the brakes on the economy, then we could be forced to argue against the very graphs you are presenting today.

Mike

 

[fletcher]

I agree that there is not a cause-and-effect relationship from the image. For instance, the number of crimes with a few shall-issue states is higher than that with a couple (first image). Only in the end did it show a decrease, but is still comparable to the original numbers.

Also, converting the "number of gun crimes" to "crime rate" might show a bit more of a difference. There can be a big population change over that period of time.

 

[illspirit]

I agree .cheese., hence why why I said the line almost looks like it's running away. Someone on my forums said something to the effect that legalizing the carry of guns made shootings go up. Now, while one can't really prove shall-issue laws caused the crime to drop, you can kinda prove it didn't cause crime to go up since it didn't.

Well, aside from the crime continuing to go up after the first few States jumped on the bandwagon, but all of those saw either go down or slow compared to the national rate. Adding more details such as that (plus some economic stats and such) to the animation would be a good thing. But my forum is video game related, so I didn't want to get too complicated.

 

[.cheese.]

fair enough then.

Sounds like we're all on the same page.

I would say that "CCW not making it worse." is indeed a weak conclusion, if only reached by looking at the image-series. (Note: I plan to read the texts mentioned and linked to later tonight - so I can't comment directly in regards to them). Based on solely the image-series, you could say that OVERALL CCW factored in hasn't made crime worse in the long-term by total numbers alone, but you can't outright say that CCW itself hasn't "possibly" contributed to crime. In fact, for all we know, there could be some external factor that keeps CCW'ers in check.... maybe the police.

I'm just playing devil's advocate of course here. In reality I don't believe that CCW has done anything but good for crime rates, and as mentioned - like most of us here, I carry daily. However, when it comes to making logical arguments with logical conclusions, it's all in the details, evidence, and assumptions.

Fortunately, we already have multiple well-documented studies that make extremely strong and structured logical arguments in favor of CCW.

 

[JLStorm]

I dont know anymore, I used to preach that as the rule but I live near philadelphia and pennsylvania is one of the easiest states to get a concealed weapons permit in and yet philadelphia's murder rate is sky rocketing. The media says its gun related crime, I dont know what to believe anymore...all I know is either way I am safer if Im packing, especially if the gun related crime is rising since its more likely that the BG's will be packing too.

 

[fantacmet]

I think the main purpose is to gt the anti's to take a second look that guns are NOT the evil living beings that CAUSE crime, as they once thought they were. I believe the argument serves that purpose well.

 

[Zen21Tao]

There is also a phenomenon called "regression to the mean." When something diverts from its average it will reach a high point or low point then must regress right back to the average.

This was the case with the infamous "Sports Illustrated Curse" where it was said that an appearance on the cover of SI leads to poor performance. The truth is, someone has to performing "at the top of their game" to make the sports cover. By the time they make the cover and it is published, they are statically more likely to be regressing back to their average performance than continuing to out perform.

Remember, many gun grabbers during the mid 90's tried to claim it was stronger gun control laws, like the AWB, that caused the decrease in violent crime. This was quickly disproved as it was shown that the crime rate began to drop before the AWB went into affect and that the decrease in crime was not limited to areas that enacted stronger gun control. In fact, its was common for crime to drop faster than the national average in areas that went the opposite way and liberalized gun ownership and/or enacted concealed carry laws.

Personally, I also believe their is a strong inverse correlation (that is causal) between concealed carry laws and crime rates, but to verify this one would have to disqualify all other factors that could be at work.

 

[U.S.SFC_RET]

Bottom line.. Where self protection is needed the worst like Washington DC and Baltimore you can forget packing a gun legally. High crime areas-Get the point? An armed society is a polite society.

 

[MachIVshooter]

I dont know anymore, I used to preach that as the rule but I live near philadelphia and pennsylvania is one of the easiest states to get a concealed weapons permit in and yet philadelphia's murder rate is sky rocketing. The media says its gun related crime, I dont know what to believe anymore...all I know is either way I am safer if Im packing, especially if the gun related crime is rising since its more likely that the BG's will be packing too.

You're going to see that anywhere when you look at big cities. While I believe that is in part due to the fact that a much larger percentage of folks living in major metropolitan areas choose to depend on others for their safety (read; they're unarmed and/or unwiling to defend themselves), I suspect that simple population density/demographics and higher rates of unemployment/poverty that are associated with major cities are probably larger factors. I would love to believe that our zero percent homicide rate in my county is due entirely to ~85% armed population, but I suspect the average household income of $69K (2004) and population density of 10.7 per square mile (2000) probably read more into that when compared with Denver at $41K (2004) household income and 3,625 (2000) people per square mile.

BTW, Denver has a 10.5 per 100,000 homicide rate (national average is 6.9), and is the only city in the state able to enforce their own stricter gun laws. Denver's Violent crime overall is about 40% higher than the national average.

 

[JLStorm]

The first time I went to a gun range my father took me, and after we were shooting for a while he puts his hand on mine and makes me lower the gun and puts his finger up to his lips as a signal to be quiet and after a few seconds he says "what dont you hear" I had no clue and just looked at him with a bewildered look and he said "you dont hear anyone arguing". That always stuck with me.

 

[LaEscopeta]

It's a good image, and thank you for posting it as any work is good work.

+1.

However, I also think a problem with the graphic in the original post is “crimes committed with firearm” shown in the chart are not necessarily evenly distributed across the map shown.

A classic example of this fallacy is a chart showing amount of fish consumed per person in different countries vs. breast cancer rates in those countries. It is almost a straight line relationship; the more fish eaten the lower the breast cancer rates. There is also an almost a perfect inverse relationship for beer consumed vs. breast cancer rates. The problem? The statistics for fish/beer consumed are for the entire population and the cancer rates are for only part of the population. Who says women eat fish or drink beer at the same rates as the general population?

Same as in the original post; do the “crimes committed with firearm” numbers drop in the states after they turn blue (become shall issue) faster the nation as a whole? I don’t know, you can’t tell from the graphic.

There is also a phenomenon called "regression to the mean." When something diverts from its average it will reach a high point or low point then must regress right back to the average.

This is also called the “law of averages.” Unless you are talking about a self limiting process, I would advise thinking hard about this before you believe it. If you flip a coin 5 times and get 5 heads in a row, does this mean the odd of getting tails on the next flip are greater then 50-50? If so, how does the coin know what the last 5 flips were? Or what force is there that could remember the last 5 flips and could influence the next flip to be tails?

 

[Zen21Tao]

If you flip a coin 5 times and get 10 heads in a row, does this mean the odd of getting tails on the next flip are greater then 50-50? If so, how does the coin know what the last 5 flips were? Or what force is there that could remember the last 5 flips and could influence the next flip to be tails?

In short, the 50/50 average of a coin flip is the same for each individual flip since each trial is independent (not dependent upon the outcome of the previous or subsequent trials). But, if you flipped a coin an infinite number a times (or a very a large number of times) then yes, you would expect to have runs of 10 heads and 10 tails in a row. So, it wouldn't be unexpected to have 10 heads or 10 tails in only 10 flips. However, you can't expect such an outcome any more than you can expect to have 5 heads and 5 tails.

Such an expectation is a fallacy called "the law of small numbers." This is a play on the valid "law of large numbers." Basically, it is incorrect to think that what happens over a large number of trials will also happen under a small number of trials. That is, one could expect to have roughly a 50% average of heads or tails if flipping a coin 10,000 times but not 10 times. As mentioned above, short lived runs would not be uncommon given a large number of flips.

 

[Novus Collectus]

I blame Maryland's and Washington DC's high violent crime rate on neighboring Virginia's shall issue laws!
I mean after all, where would violent criminals rather commit ther crimes; where the victims are armed, or where they are defenseless?

 

[Double Naught Spy]

After the lag behind Florida going shall-issue, it almost looks like the crime line is running away from all those blue States, doesn't it?

Typical of wanting to believe something even if it isn't actually shown or supported. The crime line isn't shown running away from anything, nor does it almost look like it.