Excessive Violent Crime in the South

By: Kathy Gerber
Published On: 5/6/2007 10:28:48 AM

The South has more than its share of violent crime in every category.  This came to my attention when reading background on Amnesty International's report about rape and Native American and Alaska Native women (highly recommended press release and full article).  I hope to write more on that, but meanwhile the excessive crime in the South is a topic meriting more attention than its getting.

Using the UCR geographical divisions the Southern region's share of both murder and rape is roughly 50% higher than it should be.
One common response in discussion of crime is to "otherize" the high rates or place them at a geographical remove. While this tactic may be comforting and give rise to a sense of personal agency, it does very little in getting at the core of the problems and reducing the rates.

First, a look at levels in broad regional areas. The color gradation of this FBI regional map does not correspond with gradations in actual rates.

This table shows the proportion of US population and the proportions of categories of violent crime. The South still has more than its share of violent crime in every single category.  For example, with 36.3 percent of the population, a level distribution would show 36.3 percent of the violent crime.  In actuality that rate is 41.9 percent.  Rates higher than the regions population percentage are highlighted in red. Click on it for a more legible version.

Here's a histogram of this data.


Preliminary data for 2006 shows that violent crime rates are increasing nation wide and that the South is the only region to show an increase in every single category.

                                                                                                       
Region
Violent  crime Murder
Rape
Robbery
Assault
Total+3.7+1.4*+9.7+1.2
Northeast+2.9+0.5-2.5+5.8+1.7
Midwest+3.9-2.0+1.0+10.4+1.1
South+3.3+3.3+1.1+8.0+1.5
West+4.7+1.6-1.0+14.6+0.6
* Less than one-tenth of 1 percent.

At the state level here are the states ranking lowest and highest for the 2005 rates for reported rape.
                                                                                                                                             
 

Lowest Ten Rates

 
 

Ten Worst Rates

 
 

State

 
 

Rate

 
 

State

 
 

Rate

 
 

NJ

 
 

13.9

 
 

SC

 
 

42.5

 
 

WV

 
 

17.7

 
 

AR

 
 

42.9

 
 

NY

 
 

18.9

 
 

CO

 
 

43.4

 
 

CT

 
 

20.0

 
 

MN

 
 

44.0

 
 

WI

 
 

20.6

 
 

DE

 
 

44.7

 
 

MD

 
 

22.6

 
 

WA

 
 

44.7

 
 

VA

 
 

22.7

 
 

SD

 
 

46.7

 
 

GA

 
 

23.6

 
 

MI

 
 

51.3

 
 

WY

 
 

24.0

 
 

NM

 
 

54.1

 
 

ND

 
 

24.2

 
 

AK

 
 

81.1

 

Virginia data at the city and county levels are incomplete.  Here's what's available.


Comments



I'd be curious to see... (Lowell - 5/6/2007 11:14:49 AM)
...an overlay of gun prevalance and violent crime rates.  Just from a quick glance at this, it sure looks like violent crime rates are highest where guns are most prevalent, and vice versa.  Do you know the answer to this?  Thanks.


I don't know. (Kathy Gerber - 5/6/2007 11:48:06 PM)
First, how do you measure prevalence? Check out this fact sheet - lots of info there -
http://www.jhsph.edu...

The problem with prevalence is having any idea of how many illegal guns there are.  500,000 per year are stolen. Remember before Virginia's 1-gun-a-month bill that a very large proportion of handguns used in crimes in NY were from Virginia.

There's another problem with the meaning of prevalence that has to do with scale.  At the household level - from the same fact sheet in homes with guns homicide is 3 times as likely and suicide is 5 times as likely. 

Scaling up to communities brings in so many variables.  The problem is complex.  But to go back to the Amnesty study - they cite this article -
http://www.adn.com/n...

That guy probably could have done as much or more damage with a crowbar. One factor there was the lack of law enforcement and that's a recurring problem in remote areas.

It was also a problem in Henry County.  The officers there were involved in the selling of illegal firearms.  And the violent crime rates in the area reflect it.

In both domains there is a mini- ecosystem of criminal behavior that develops.  And rape - which one might expect to be unrelated to drugs, hot guns and corrupt police - was higher than expected in 2005 in Henry County.

Anyway, what I found most alarming were the rates for Native American and Native Alaskan women.  Rape is so prevalent there that the odds of having a family member raped is nearly a certainty.  And one case worker reported that of 77 women she worked with, only three reported.



Thanks for the information. (Lowell - 5/7/2007 7:20:12 AM)
That's horrible about the Native American women...what's being done about that, if anything?


New England has a righ rate of gun ownership and low crime rate (Hugo Estrada - 5/7/2007 10:42:45 AM)
The Post had a survey on gun ownership in 2001.  I tried to find something more recent, but couldn't find it quickly enough with data that I would trust.

http://www.washingto...

If you compare the gun ownership rate, New England has a comparable gun ownership rate with the South, yet the rate of violent crimes involving a firearm is very low there.

The NRA makes a point of using New England as an example on how high gun ownership produces low crime rates.

The South and some Western states support the claim that gun ownership do related with violent crimes.

When looking at this two points, it seems that gun ownership  may not be the sole cause for firearm violence. Sure, you cannot have gun violence without guns, but having them does not necessarily lead to a high crime rate.

I suspect that honor cultures may have something to do with this. Mexico, where honor plays a very important role in human interactions, was a world leader in murders until a national gun ban went into effect. There was a dramatic reduction of murders after this.

Yet, while researching this, I found this recent study supports the idea that more gun ownership equals more murders. I haven't read it because of lack of time, but I am including it for those who are interested in reading the findings.

http://www.hsph.harv...



Read that post link again ... (loboforestal - 5/7/2007 1:25:01 PM)
If you compare the gun ownership rate, New England has a comparable gun ownership rate with the South, yet the rate of violent crimes involving a firearm is very low there.

Gee, I'm looking at MA and CT and they look very low.
Only ME and VT are sticking out: both rural kind of states.



Yeah, didn't sound right to me at all. (Lowell - 5/7/2007 2:16:55 PM)
I grew up in New England, and I don't remember many guns there.


Rhode Island is low as well (Hugo Estrada - 5/7/2007 2:33:26 PM)
And New Hampshire is moderately high. Thanks for catching that. :)

Still, the New England states with high gun ownership do not share the same rates of gun violence that we see in the South or in the Western states.



Southerners and the Herding Culture (PM - 5/7/2007 5:09:52 PM)
Marc Hauser has written a book called Moral Minds.  Hauser is a professor of Psychology, Organismic and Evolutionary Biology, and Biological Anthropology at Harvard.  Hauser argues that mankind has evolved a universal moral instinct.

One cross cultural variation is the macho culture, and that is part of the American South.  The theory (developed by others) is that "cultures of honor" develop among herders, who had to defend themselves against their animals bring stolen. (Farmers did not have this problem.)  This has happened in herding cultures worldwide, including our South.  The "culture of honor" carried through to other life events as well (fooling around with the herder's woman, e.g.,) and has continued long past its trigger point.

Two researchers, Nisbett and Cohen, tested northern and southern students in a so-called "bump" test (you also tell the person they're an a-hole).  Northerners tend to walk away, southerners become aggressive.  Those researchers believe this behavioral tendency manifests itself in opposition to gun control laws, a preference for laws allowing violence in protecting one's home, preference for using violence in socializing children, etc.

So, there's one theory for the mill.

I have only read a small part of the book.  My wife informed me of this analysis.  From what I can see, Hauser is very easy to read, presents lots of experiments, and heavily footnotes his book.

I was able to find a number of reference discussing the "herding" hypothesis easily through Google.



A little more (Kathy Gerber - 5/8/2007 1:17:18 PM)
I did a couple more scatterplots and so on for this but our connection at home went down.  I'll try to put them here soon.


Video, scatterplots, lineplot (Kathy Gerber - 5/8/2007 10:30:46 PM)
This video tells a bit about what is being done.
http://www.youtube.c...

Here's the regression of the data Hugo found (thanks, Hugo) with 2005 murder rates. DC is the outlier on the left with a murder rate of 35.4.  There is no meaningful correlation, and the outlier was removed.


This is a text book case of uncorrelated data - low R^2, beta near zero and alpha (intercept) close to the mean of the dependent variable.  It isn't all that surprising that ownership and murder rates aren't strongly correlated because we know that there are so many other variables involved.


Here's a scatterplot to show that there isn't any strong correlation between murder and rape. (No news is news).  Alaska is an outlier because of its rate of rape and DC is an outlier because of its rate of murder.


Below are the time series of rape for 50 states and DC.  There was an enormous surge in rape in DC back in the 60's.  DC is the thicker blue line. The DC rate decreased over the next two decades, and DC fell in line with the rest of the states for the next 15 years.  What were the factors leading to this decrease?  If this was the result of law enforcement, why hasn't the same been done in Alaska (the black line)?  Alaska's rape rate led the way for 20 out of 25 years  in the 1980-2005.  Alaska would have led throughout that entire quarter century had not several states had surges (DE, MI, WA and NV).

Why has this been neglected for 25 years? Maybe herding culture is a part of this.



So, the conclusion is.... (Lowell - 5/8/2007 10:36:22 PM)
...that gun ownership is totally uncorrelated with violent crime?  But then how do you explain the low murder rates in countries like Japan, England, and Canada which have much stricter gun laws than we do?  On the other hand, I suppose there are countries like Switzerland and Israel where everyone's armed to the teeth but there are also low murder rates.  Maybe it's cultural?  If so, what is it about U.S. culture that results in a higher violent crime rate than other countries? 


Too strong of a statement (Kathy Gerber - 5/8/2007 11:15:47 PM)
Those plots use one time series of self-reported gun ownership data and a simple regression which gave no results. It's far more complicated than that. The WAPO article opens with this -

In 2001 the Behavioral Risk Factor Surveillance System (BRFSS) in North Carolina surveyed 201,881 respondents nationwide, asking them, "Are any firearms now kept in or around your home? Include those kept in a garage, outdoor storage area, car, truck, or other motor vehicle." Here are the results.

-- Virginia's one-gun-a-month law had a positive impact.
-- Maryland's Saturday night special law had a positive impact (but not right away - there was a brief surge in crime;  people rushed out to buy guns before the law went into effect).
-- Gangs are a factor - don't remember the year but for both LA and Chicago gang related homicides were 1/2 of the incidents.
-- rural vs. urban vs. suburban, south vs. north, lots of meth vs. not so much, socioeconomic status, etc. etc.

By the way the murder rate in Puerto Rico is way lower than any state.

And how do you measure adequacy when it comes to law enforcement?



Creeps (Kathy Gerber - 5/9/2007 6:20:13 AM)
Does corruption have anything to do with crime rates?

http://www.tpmmuckra...

http://www.dailykos....

Maybe there's a corruption index out there somewhere...



If I'm economically disadvantaged and I see the higher ups are stealing (PM - 5/9/2007 8:39:03 AM)
it stands to reason my moral code may be loosened a bit.