The Atlantic: The Geography of Gun Deaths

[Elessar78]

Terrible tragedies like last week's mass shootings in Tucson cause us to search for deeper answers. Many were quick to blame America's divisive and vitriolic political culture for the violence; others portray the shooter as an unhinged, clinically deranged person with his own unfathomable agenda. Arizona has been ground zero for the battle over immigration. Were the state's political and economic travails a contributing factor? There has been some talk about guns, too. Might tighter gun control laws have made a difference?

The map (http://www.statehealthfacts.org/comparemaptable.jsp?ind=113&cat=2) above charts firearm deaths for the 50 states plus the District of Columbia. Note that these figures include accidental shootings, suicides, even acts of self-defense, as well as crimes. As of 2007, 10.2 out of every 100,000 people were killed by firearms across the United States, but that rate varies dramatically from state to state. In Hawaii, at the low end, it was 2.6 per 100,000; in New York and New Jersey it was 5.0 and 5.2 respectively. At the high end, 21.7 out of every 100,000 residents of the District of Columbia were killed by guns, 20.2 in Louisiana, 18.5 in Mississippi, and 17.8 in Alaska. Arizona ranked eighth nationally, with 15.1 deaths per 100,000.

With these data in hand, I decided to look at the factors associated with gun deaths at the state level. With the help of my colleague Charlotta Mellander, we charted the statistical correlations between firearm deaths and a variety of psychological, economic, social, and political characteristics of states. As usual, I point out that correlation does not imply causation, but simply points to associations between variables.
http://assets.theatlantic.com/static/mt/assets/richard_florida/Gun%20ViolenceEDIT.jpg


http://www.theatlantic.com/national/archive/2011/01/the-geography-of-gun-deaths/69354/

[Larry]

"Creative Class"?

[Elessar78]

Musicians, artists, graphic designers . . . you know . . . sissies. The ones who would cry at even the sight of a gun. Unless it was an ironic gun.

[Cept One]

Not sure what to say about this garbage... including justified shootings with illegal uses of firearms renders this info useless. Why just focus on guns? why not look at all violent crime?

[mardi gras skin]

States with the most guns per capita have the most gun deaths per capita. That's not a particularly startling fact.

The list of correlations the writer strung together are irrelevant.

[mjah]

Here's an actual definition of the "creative class," per the author of that article (and the book):


the 40 million Americans – over a third of our national workforce – who create for a living. This “creative class” is found in a variety of fields, from engineering to theater, biotech to education, architecture to small business.

The Rise of The Creative Class (http://www.creativeclass.com/richard_florida/books/the_rise_of_the_creative_class/)

In my personal experience, both engineering and small business types have a notable positive correlation with gun enthusiasm.

---------- Post added January-25th-2011 at 11:36 AM ----------


States with the most guns per capita have the most gun deaths per capita. That's not a particularly startling fact.

The list of correlations the writer strung together are irrelevant.
Statistically they're far from irrelevant. The correlations are what they are.

Only when you start trying to presume causation are some of them destined to become irrelevant.

[mardi gras skin]

Statistically they're far from irrelevant. The correlations are what they are.

Only when you start trying to presume causation are some of them destined to become irrelevant.

And you don't think the author rigged this piece to imply causation? Why did the writer choose to mention that Arizona has a loose gun carry law but he didn't mention Vermont's equally lax gun laws? It didn't fit the author's purpose.

The author said, "I decided to look at the factors associated with gun deaths at the state level," but he neglected the most important factor: states with more guns per capita have more gun deaths per capita. Simple common sense.

Instead, he used data to paint a picture he wants his readers to see and he neglected data that would disrupt his narrative.

[Elessar78]

He does mention that correlation does not imply causation, that it simply points out associations between variables.

[mardi gras skin]

He does mention that correlation does not imply causation, that it simply points out associations between variables.

Sure he did. He presented some facts and withheld other facts and he just let the readers draw their own conclusions. What could possibly go wrong with that?

[RansomthePasserby]

Musicians, artists, graphic designers . . . you know . . . sissies. The ones who would cry at even the sight of a gun. Unless it was an ironic gun.

Not just artists. Engineers are considered part of the "creative class" as well.

[aREDSKIN]

IMO the problem with this "study" is that there is no analysis of the shooter and states variables. The states didn't shoot anyone an individual in a state did and there is nothing that correlates that individual's variables with the states variables.

[Elessar78]

Not just artists. Engineers are considered part of the "creative class" as well.

I was being sarcastic in my definition.

[mardi gras skin]

Louisiana has an outrageously high gun death rate but that's mostly because New Orleans is one of the murder capitals of the world. Louisiana voted for McCain so those deaths go in his column, but New Orleans voted for Obama by a wide margin.

This is one way these statistics are used to mislead. Another writer trying to drive home a different bias would use the stats differently.

[mcsluggo]

IMO the problem with this "study" is that there is no analysis of the shooter and states variables. The states didn't shoot anyone an individual in a state did and there is nothing that correlates that individual's variables with the states variables.

well... by that reasoning it is invalid to seperate ANY data set by ANY characteristic. Race didn't shooot people, individuals did; Women don't earn less money, individuals do; Afghanistan isn't a violent place... but lots of individuals get dead-ed.

You are basically complaining about the very definition of statistics: making estimates about an entire population based on statistical sampling of a subset of the population, which is only achievable by defining population groupings --- and then analyzing to see if correlations imply that the chosen population grouping makes any sense (ie DOES the population grouping share any characteristics that can be described by data).

[aREDSKIN]

well... by that reasoning it is invalid to seperate ANY data set by ANY characteristic. Race didn't shooot people, individuals did; Women don't earn less money, individuals do; Afghanistan isn't a violent place... but lots of individuals get dead-ed.

You are basically complaining about the very definition of statistics: making estimates about an entire population based on statistical sampling of a subset of the population, which is only achievable by defining population groupings --- and then analyzing to see if correlations imply that the chosen population grouping makes any sense (ie DOES the population grouping share any characteristics that can be described by data).


IMO it would have been a more effective "study" if more thorough analysis was done and some effort was put in to correlate between the shooter and the State's variable(s).

[zoony]

how does one measure "inequality"? :ols:

[Predicto]

how does one measure "inequality"? :ols:

I presume they meant income inequality - the statistical gap between rich and poor varies greatly from state to state. :whoknows:

[Bang]

how does one measure "inequality"? :ols:

Probably by determining the difference of cash in the victim's pockets as compared to the crazed republican prior to the attack.

~Bang!

[AzSkinsFan63]

http://www.americanthinker.com/2011/01/cooking_the_gun_homicide_numbe.html

[mjah]

And you don't think the author rigged this piece to imply causation? Why did the writer choose to mention that Arizona has a loose gun carry law but he didn't mention Vermont's equally lax gun laws? It didn't fit the author's purpose.
Why didn't he mention Alaska, the third state that allows concealed carry without a permit? Alaska is a high-gun-violence state. It would have "fit his purpose" very well to specifically mention Alaska's lax gun laws. Yet he didn't mention it.

He should have, right? Because that would have "fit his purpose?" That's quite the puzzler.

Here's another explanation for why Arizona was mentioned: The article's introduction mentions some noteworthy current events in Arizona involving a concealed handgun that was used to maim and kill. They were literally the first words in the article. That's why the author specifically mentioned Arizona, and their concealed carry law, without invoking either Alaska or Vermont.

Curiously, you didn't mention Alaska either. Did it not "fit your purpose?" :rolleyes:


The author said, "I decided to look at the factors associated with gun deaths at the state level," but he neglected the most important factor: states with more guns per capita have more gun deaths per capita.
That bugs me. Two other things bug me. One is his too-simple conclusion that because 2 of the 3 lax-carry-law states have high gun violence, more gun control statistically must be part of curbing gun violence. The other thing that bugs me is the fact that he asked questions at the beginning of the article which he didn't answer (and then expressly disavowed any intention of answering). It irritates me when an author ruminates "out loud" on topics which he then doesn't attempt to address.


Instead, he used data to paint a picture he wants his readers to see and he neglected data that would disrupt his narrative.
You're seeing that by ignoring things like the Alaska example I gave you. This guy clearly is no academic, but just because his efforts to disclaim politics didn't satisfy you and laid bare some striking correlations, doesn't mean he's simply out to grind an axe.

Agreed on this point though: The article might have been quite a bit shorter if he laid gun deaths per capita next to guns per capita. But the strongest correlation (I presume -- we have yet to see data on that in this thread) isn't the only correlation. And there were some interesting correlations in that article.

[Popeman38]

Why didn't he mention Alaska, the third state that allows concealed carry without a permit? Alaska is a high-gun-violence state. It would have "fit his purpose" very well to specifically mention Alaska's lax gun laws. Yet he didn't mention it.You sure you read the article?
At the high end, 21.7 out of every 100,000 residents of the District of Columbia were killed by guns, 20.2 in Louisiana, 18.5 in Mississippi, and 17.8 in Alaska.

---------- Post added January-25th-2011 at 01:43 PM ----------


http://www.americanthinker.com/2011/01/cooking_the_gun_homicide_numbe.htmlIt is the theme of the liberal. After a tragic event involving a gun, paint the USA, southerners, and whites as having a huge issue with owning guns. Even though a little examination of facts reveal otherwise.

---------- Post added January-25th-2011 at 02:03 PM ----------


What about politics? It's hard to quantify political rhetoric, but we can distinguish blue from red states. Taking the voting patterns from the 2008 presidential election, we found a striking pattern: Firearm-related deaths were positively associated with states that voted for McCain (.66) and negatively associated with states that voted for Obama (-.66). Though this association is likely to infuriate many people, the statistics are unmistakable. Partisan affiliations alone cannot explain them; most likely they stem from two broader, underlying factors - the economic and employment makeup of the states and their policies toward guns and gun ownership.

Firearm deaths were far less likely to occur in states with higher levels of college graduates (-.64) and more creative class jobs (-.52).Funny, he says the stats are unmistakable, implying that Rep leaning states have more gun violence. Then he tries to qualify the stats.

As has been pointed out, LA voted for McCain. New Orleans has a murder rate of 64/100,000, while LA as a whole has a 20.2/100,000. What does that tell you about skewing the stats? McCain picks up the murder tally for the state, while the predominant numbers reside inside New Orleans -- a distinct Obama pocket.

Homicide rate:
L.A. - 9.6/100,000
NYC - 6.3/100,000
DC - 23.8/100,000
Houston - 16.3/100,000

I could go on, but you get the point. Urban areas account for the majority of gun crime. Urban areas are also almost exclusively liberal pockets in even the most conservative states. Each listed above easily went to Obama in the 2008 election.

If you want to profile race:

Homicide rate:
Black - 20.86/100,000
White - 3.11/100,000
Overall - 5.3/100,000

Clearly, Obama carried the black vote, winning ~95% of the vote.

Do we really need to rehash the same, old, tired response offered every time there is a gun tragedy?

[mjah]

You sure you read the article?
You sure you read my post? ;)

I was asking mardi gras skin why the author didn't specifically note that Alaska, a high-gun-violence state, allows concealed carry. That's what I meant by "mention Alaska." Reading his post and my response, it's pretty clear.

Nobody suggested that the word "Alaska" is missing from the article.

[Predicto]

[/COLOR]Funny, he says the stats are unmistakable, implying that Rep leaning states have more gun violence. Then he tries to qualify the stats.

As has been pointed out, LA voted for McCain. New Orleans has a murder rate of 64/100,000, while LA as a whole has a 20.2/100,000. What does that tell you about skewing the stats? McCain picks up the murder tally for the state, while the predominant numbers reside inside New Orleans -- a distinct Obama pocket.

Homicide rate:
L.A. - 9.6/100,000
NYC - 6.3/100,000
DC - 23.8/100,000
Houston - 16.3/100,000

I could go on, but you get the point. Urban areas account for the majority of gun crime. Urban areas are also almost exclusively liberal pockets in even the most conservative states. Each listed above easily went to Obama in the 2008 election.



Umm, I don't understand what you are trying to demonstrate? Yes, there is more gun homocide in cities than in suburbs, but what explains the overall difference in gun homocide rates between the Red States and the Blue States? New York has the Big Apple, California has lots of big cities, but both of them have a much lower rate of gun death than Louisiana. Or, if you prefer not to use Lousiana, use Texas or any other Red State on the map. The urban-rural dichotomy does not explain the difference - if anything, the Blue States are MORE urban than the Red States, yet they have less gun deaths. Am I missing something? :whoknows:

I should also point out that I seem to recall that gun deaths in rural areas are roughly as common as gun deaths in cities. It was suburban areas that pull down the statistics.

[mardi gras skin]

Why didn't he mention Alaska

He didn't need to mention Alaska. In order to bias his article, he only needed to omit evidence that conflicts with his narrative. And clearly, including Alaska would have forced him to include Vermont.


Here's another explanation for why Arizona was mentioned: The article's introduction mentions some noteworthy current events in Arizona involving a concealed handgun that was used to maim and kill. They were literally the first words in the article. That's why the author specifically mentioned Arizona, and their concealed carry law, without invoking either Alaska or Vermont.

Believe me, I do understand what he's doing. He trying to use the high profile shooting to further his agenda.


Curiously, you didn't mention Alaska either. Did it not "fit your purpose?" :rolleyes:

:ols: Love the rolleyes! More guns per capita is the primary reason for more gun deaths per capita. If I were trying to score political points, why wouldn't I massage that? Certainly a wide swath of conservatives would be nervous to let that statement stand. What nefarious purpose do you think I have here?

Alaska has nothing to do with my point. Alaska's gun violence rate doesn't change the fact that the writer avoided Vermont's law while taking a pot shot at Arizon's law. If you're going to suggest one state created the environment for those kinds of crimes with their gun laws, explain how a state with similar laws avoids those problems.

The author is cherry picking stats to tie Republicans to gun violence. A right wing hack could just as easily cherry pick stats to demonstrate that Obama supporters are the primary cause of gun violence and that would be equally irritating.


It irritates me when an author ruminates "out loud" on topics which he then doesn't attempt to address.

Oh, the author definitely addressed his question. He drafted several paragraphs and brightly colored images that made his point abundantly clear. He lead his readers directly where he wanted them to go and then smartly distanced himself from the logical deductions one would make based on the information he presented...and omitted.

[Popeman38]

Umm, I don't understand what you are trying to demonstrate? Yes, there is more gun homocide in cities than in suburbs, but what explains the overall difference in gun homocide rates between the Red States and the Blue States? New York has the Big Apple, California has lots of big cities, but both of them have a much lower rate of gun death than Louisiana. Or, if you prefer not to use Lousiana, use Texas or any other Red State on the map. The urban-rural dichotomy does not explain the difference - if anything, the Blue States are MORE urban than the Red States, yet they have less gun deaths. Am I missing something? :whoknows:First, who said anything about homocide?:ols::ols::ols:

Sorry, couldn't resist. Now back to serious discussion.....

Really? DC has the highest homicide rate per capita. It is the very definition of Blue, and had some of the strictest gun laws in the country. Additionally, the article in question illustrates gun death, not crime. A person acting in self defense counts the same as a summary execution carried out by gangs. A suicide counts the same as a drug deal gone wrong. Accidental shooting counts the same as a revenge killing. All gun deaths are not equal. A simple Google search shows the following:
The majority of gun-related deaths in the United States are suicides.

In 2004, there were a total of 16,750 suicides by firearm. In 2004, there were 11,624 homicides by firearm. As you can see, there were almost 60% more suicides by firearm than there were homicides by firearm. That is more than enough to throw the "statistical analysis" off.

Add to that there were 23,237 accidental, non fatal shootings, double the homicides. There were 649 accidental deaths. All stats taken from CDC.
http://www.cdc.gov/nchs/injury.htm

The guy that wrote this article is paid to do so. I put this response together in roughly 15 minutes. He had an agenda and cherry-picked what to use to point towards the conclusion he wanted to make.
I should also point out that I seem to recall that gun deaths in rural areas are roughly as common as gun deaths in cities. It was suburban areas that pull down the statistics.You can't compare "rural areas" because you would have to include thousands of square miles to encompass 100,000. Urban areas are easy as there is cluster living. Suburbs are easy as well. The VAST majority of homicides by firearm are accounted for in urban areas. New Orleans had ~207 murders in 2008. Using a conservative % using firearms, that puts the homicide by firearm number at roughly 83. I would presume with a little digging you could move that easily past 100. That easily accounts for nearly 25% of the total homicides in LA. Do the same analysis for the other large cities and you will account for well over 75% of the homicides for a state, much less homicides by firearm.

Also, the Homicide rate/100,000 for 2009 was 11.8.

[AsburySkinsFan]

Not sure this is exactly Earth shattering...more guns in an area is going to mean more gun injury. Those statistics might mean something if the Obama and McCain shared the same number of gun owning voters.

Seems to me to be a bit of captain obvious.

[Larry]

Really? DC has the highest homicide rate per capita.

And isn't a state.

Want to see if we can find a single building in Birmingham that has a higher rate?


Additionally, the article in question illustrates gun death, not crime. A person acting in self defense counts the same as a summary execution carried out by gangs. A suicide counts the same as a drug deal gone wrong. Accidental shooting counts the same as a revenge killing. All gun deaths are not equal. A simple Google search shows the following:As you can see, there were almost 60% more suicides by firearm than there were homicides by firearm. That is more than enough to throw the "statistical analysis" off.

Add to that there were 23,237 accidental, non fatal shootings, double the homicides. There were 649 accidental deaths. All stats taken from CDC.

None of which in any way invalidates anything in the study. Those deaths occur in states that voted for McCain in MUCH larger numbers.

Now, if you want to claim that people in McCain states committed suicide in vastly disproportionate numbers, then you might have a point. :)


The guy that wrote this article is paid to do so. I put this response together in roughly 15 minutes. He had an agenda and cherry-picked what to use to point towards the conclusion he wanted to make.You can't compare "rural areas" because you would have to include thousands of square miles to encompass 100,000. Urban areas are easy as there is cluster living. Suburbs are easy as well. The VAST majority of homicides by firearm are accounted for in urban areas. New Orleans had ~207 murders in 2008. Using a conservative % using firearms, that puts the homicide by firearm number at roughly 83. I would presume with a little digging you could move that easily past 100. That easily accounts for nearly 25% of the total homicides in LA. Do the same analysis for the other large cities and you will account for well over 75% of the homicides for a state, much less homicides by firearm.

And as he pointed out, the blue states have big cities, too. In fact, they have much bigger ones. But their death rate is lower, anyway.

[RansomthePasserby]

I was being sarcastic in my definition.

? I wasn't going by your definition... I just heard a lecture on the creative class yesterday, and the professor explicitly said "engineers are part of the creative class." No joke.

[Destino]

Really? DC has the highest homicide rate per capita. It is the very definition of Blue, and had some of the strictest gun laws in the country. Additionally, the article in question illustrates gun death, not crime. A person acting in self defense counts the same as a summary execution carried out by gangs. A suicide counts the same as a drug deal gone wrong. Accidental shooting counts the same as a revenge killing. All gun deaths are not equal. A simple Google search shows the following:As you can see, there were almost 60% more suicides by firearm than there were homicides by firearm. That is more than enough to throw the "statistical analysis" off.

Add to that there were 23,237 accidental, non fatal shootings, double the homicides. There were 649 accidental deaths. All stats taken from CDC.
http://www.cdc.gov/nchs/injury.htm

You see to think you're on to something when it looks like you're completely missing the point. Why are you trying to separate accidents, suicides, and murders... when the argument for gun control doesn't single out murder at all and never has. The argument for gun control is that you are more likely to hurt yourself or a loved one if you have a hand gun than a loved one is to get hurt by a "bad guy" if you don't. So you're rant about "all gun deaths are not equal" completely misses the point of this type of data.


(before anyone gets things twisted, I'm hard line pro gun rights and a supporter of castle laws. I'd actually make it straight up legal to shoot thieves that aren't threatening you directly... but that's just me)

[Popeman38]

You see to think you're on to something when it looks like you're completely missing the point. Why are you trying to separate accidents, suicides, and murders... when the argument for gun control doesn't single out murder at all and never has. The argument for gun control is that you are more likely to hurt yourself or a loved one if you have a hand gun than a loved one is to get hurt by a "bad guy" if you don't. So you're rant about "all gun deaths are not equal" completely misses the point of this type of data.


(before anyone gets things twisted, I'm hard line pro gun rights and a supporter of castle laws. I'd actually make it straight up legal to shoot thieves that aren't threatening you directly... but that's just me)My point is simply that every time there is gun incident, this article pops up and the writer highlights only the parts of the data that support their agenda and lead the reader to only one possible conclusion.

Also, the writer used data from a year that supports his claim, even though data from 2009 is available. His data says that in Louisiana, 20.2/100,000 were killed by guns. The murder rate in Louisiana for 2009 was 11.8.

The line:
As of 2007, 10.2 out of every 100,000 people were killed by firearms across the United Statesis misleading. The homicide rate in 2007 was 5.7/100,000. The homicide by firearm rate was 3.6/100,000. There were ~250 "accidental" shooting deaths (negligible statistically). Suicide by firearm rate was 5.9/100,000. Simple math tells us the death by firearm number sits at 9.5/100,000.

Bottom line is this story is grossly inaccurate. And misleading. And the theme is tired.

[mardi gras skin]

Bottom line is this story is grossly inaccurate. And misleading. And the theme is tired.

I don't think the information provided is inaccurate. And I'm less interested in arguing that the article is misleading than I am in saying that it is manipulative. If a person agrees with the direction the the author is taking the statistics, they won't feel the article is misleading. They will appreciate the direction the writer is leading.

But whether or not I agree with the direction of the article, the tactics used here always anger me. Its the same abuse of power the Glen Beck's of the world employ by withholding information that conflicts with their world view. The writer is manipulating his audience and I hate it. If he is too lazy or too frightened to address information that conflicts with his narrative, he should find another occupation. Preferably one that doesn't involve ideas.

[Popeman38]

I don't think the information provided is inaccurate. And I'm less interested in arguing that the article is misleading than I am in saying that it is manipulative. If a person agrees with the direction the the author is taking the statistics, they won't feel the article is misleading. They will appreciate the direction the writer is leading.

But whether or not I am in agree with the direction of the article, the tactics used here always anger me. Its the same abuse of power the Glen Beck's of the world employ by withholding information that conflicts with their world view. The writer is manipulating his audience and I hate it. If he is too lazy or too frightened to address information that conflicts with his narrative, he should find another occupation.HE didn't use accurate numbers. He used old numbers that fit his argument. The new numbers paint a different picture. Look up the list of most dangerous cities in America:

1. St. Louis, MO
2. Camden, NJ
3. Detroit, MI
4. Flint, MI
5. Oakland, CA
6. Richmond, CA
7. Cleveland, OH
8. Compton, CA
9. Gary, IN
10. Birmingham, AL
11. Baltimore, MD
12. Memphis, TN
13. New Orleans, LA
14. Jackson, MS
15. Little Rock, AR
16. Baton Rouge, LA
17. Buffalo, NY
18. New Haven, CT
19. Hartford, CT
20. Dayton, OH
21. Kansas City, MO
22. Washington, DC
23. Newark, NJ
24. Cincinnati, OH
25. Atlanta, GA

The bolded cities are in areas that he paints as safe. More than half the list don't support the "facts" of his article.

[mardi gras skin]

The bolded cities are in areas that he paints as safe. More than half the list don't support the "facts" of his article.

He looked at states and you're looking at cities. And he did say what year he got his stats from so, as far as I've seen, he wasn't inaccurate. Again, not inaccurate but obviously biased and manipulative.

[Destino]

My point is simply that every time there is gun incident, this article pops up and the writer highlights only the parts of the data that support their agenda and lead the reader to only one possible conclusion.
Which is fine and I agree but you're doing the same thing. You're arguing about cities while he's talking states. You're trying to single out murder alone for no reason. Deaths aren't limited to murder and that's not the only concern with bringing a gun into a home.

[aREDSKIN]

I don't think the information provided is inaccurate. And I'm less interested in arguing that the article is misleading than I am in saying that it is manipulative. If a person agrees with the direction the the author is taking the statistics, they won't feel the article is misleading. They will appreciate the direction the writer is leading.

But whether or not I agree with the direction of the article, the tactics used here always anger me. Its the same abuse of power the Glen Beck's of the world employ by withholding information that conflicts with their world view. The writer is manipulating his audience and I hate it. If he is too lazy or too frightened to address information that conflicts with his narrative, he should find another occupation. Preferably one that doesn't involve ideas.


Why use "PER CAPITA" death STATS vs TOTAL death STATS in a given State? The more accurate & telling STAT would have been totals per state & not per capita.

[December90]

Why use "PER CAPITA" death STATS vs TOTAL death STATS in a given State? The more accurate & telling STAT would have been totals per state & not per capita.

picking arbitrary states and fake figures to make the point

Wyoming (population approx 530,000)
Texas (population approx 23,900,000)

Now if there were 1,000 incidents in Wyoming and 10,000 incidents in Texas, you could ,by your measurements, conclude that Texas was 10 times worse that Wyoming as measured by these incidents.

On the other hand 1,000 incidents in Wyoming means that 1 in every 530 people has an incident while 10,000 incidents in Texas means that 1 in every 2,390 people had an incident.

This is why you use "per capita" instead of "total"

[Predicto]

I still don't understand your point Popeman.

Camden New Jersey may be the second most dangerous town in America, but it doesn't change the fact that the average resident of New Jersey is far less likely than the average resident of Louisiana to die from a gun. Or Mississippi. Or Texas. Or Arizona. The statistics on that point are not misleading at all. They just aren't the statistics that you want us to focus on.

I can understand if you want to argue that these statistics do not paint a complete picture, or that gun ownership has other positive benefits, or whatever. Feel free to make that point. What I don't understand is your insistence that the statistics are false, or misleading. They aren't. Or at least you haven't showed us that they are. :whoknows:

[Popeman38]

I still don't understand your point Popeman.

Camden New Jersey may be the second most dangerous town in America, but it doesn't change the fact that the average resident of New Jersey is far less likely than the average resident of Louisiana to die from a gun. Or Mississippi. Or Texas. Or Arizona. The statistics on that point are not misleading at all. They just aren't the statistics that you want us to focus on.The bolded is inaccurate. Additionally, suicide will never result in the average New Jersian being killed. Only the person who shoots themselves will be killed. And since over 60% more people killed themselves as killed someone else, taking out suicide paints a much more accurate picture. If I shoot myself in the head, there is a 0% increase in your chances of being the victim of gun violence. That alone makes the article and the stats misleading.

The average resident of New Orleans is more likely to die. I have already illustrated that New Orleans skews the number for the entire state of LA. Add in Baton Rouge, Shreveport, Metairie, Metarie Terrace and Lafayette and you will account for well over 60% of the incidents.


I can understand if you want to argue that these statistics do not paint a complete picture, or that gun ownership has other positive benefits, or whatever. Feel free to make that point. What I don't understand is your insistence that the statistics are false, or misleading. They aren't. Or at least you haven't showed us that they are. :whoknows:The stats he used are OLD. The numbers today are down significantly from his data. The data is available for 2009, which shows a precipitous drop in the violent crime category since 2007. Why not use the most recent data. It is readily available with simple Google searches.

[HailGreen28]

I still don't understand your point Popeman.

Camden New Jersey may be the second most dangerous town in America, but it doesn't change the fact that the average resident of New Jersey is far less likely than the average resident of Louisiana to die from a gun. Or Mississippi. Or Texas. Or Arizona. The statistics on that point are not misleading at all. They just aren't the statistics that you want us to focus on.

I can understand if you want to argue that these statistics do not paint a complete picture, or that gun ownership has other positive benefits, or whatever. Feel free to make that point. What I don't understand is your insistence that the statistics are false, or misleading. They aren't. Or at least you haven't showed us that they are. :whoknows:You know, I'm jumping into this debate too late to catch everything going on. But with Chicago, DC and New York City making their own gun laws, wouldn't it be fairer to look on a city per city basis, if this is about gun control? Or do gun rights vary more per state?

[RansomthePasserby]

Also, the writer used data from a year that supports his claim, even though data from 2009 is available. His data says that in Louisiana, 20.2/100,000 were killed by guns. The murder rate in Louisiana for 2009 was 11.8.

Exactly. If you look at Alaska, the murder rate in 2009 was 3.1/100,000 (firearm or otherwise) compared to a suicide rate of 24.6/100,000 (again, firearm or otherwise.)

3.1 to 24.6. Sure, that's going to give you a high "gun death" rate since guns are the suicide method of choice. Murder by gun rate? Not so much.

http://www.disastercenter.com/crime/akcrime.htm

[PeterMP]

HE didn't use accurate numbers. He used old numbers that fit his argument. The new numbers paint a different picture. Look up the list of most dangerous cities in America:

1. St. Louis, MO
2. Camden, NJ
3. Detroit, MI
4. Flint, MI
5. Oakland, CA
6. Richmond, CA
7. Cleveland, OH
8. Compton, CA
9. Gary, IN
10. Birmingham, AL
11. Baltimore, MD
12. Memphis, TN
13. New Orleans, LA
14. Jackson, MS
15. Little Rock, AR
16. Baton Rouge, LA
17. Buffalo, NY
18. New Haven, CT
19. Hartford, CT
20. Dayton, OH
21. Kansas City, MO
22. Washington, DC
23. Newark, NJ
24. Cincinnati, OH
25. Atlanta, GA

The bolded cities are in areas that he paints as safe. More than half the list don't support the "facts" of his article.

These numbers don't change much over a few years. From 2007, the top 5 are the same. Just the order has changed:

http://www.infoplease.com/ipa/A0921299.html

I seriously doubt there has been a major change in the correlations in 2 years.

[Predicto]

The average resident of New Orleans is more likely to die. I have already illustrated that New Orleans skews the number for the entire state of LA. Add in Baton Rouge, Shreveport, Metairie, Metarie Terrace and Lafayette and you will account for well over 60% of the incidents.



Yes, and the average resident of Camden, New Jersey, or Newark, New Jersey, is more likely to die from a gun than the average resident of Princeton, New Jersey. Why doesn't that skew the results for New Jersey in the exact same way that you claim it skews the results for Louisiana?

Let's put it this way - California has 5 major cities and a much higher percentage of its residents in urban areas than Arkansas does - why isn't California's gun death rate higher than Arkansas? Under your theory, it should. Put another way, Louisiana is not some outlier - if you look at the whole map, there basically are higher gun deaths in every "Red State" compared to equivalent "Blue States."

There may be plenty of reasons to explain this, but I don't see how the reason is "because this study is lying."


The stats he used are OLD. The numbers today are down significantly from his data. The data is available for 2009, which shows a precipitous drop in the violent crime category since 2007. Why not use the most recent data. It is readily available with simple Google searches.



There is always a lag in statistical studies. Do you really think that something changed between 2007 and 2008 to completely change this picture?

---------- Post added January-26th-2011 at 01:55 PM ----------


Exactly. If you look at Alaska, the murder rate in 2009 was 3.1/100,000 (firearm or otherwise) compared to a suicide rate of 24.6/100,000 (again, firearm or otherwise.)

3.1 to 24.6. Sure, that's going to give you a high "gun death" rate since guns are the suicide method of choice. Murder by gun rate? Not so much.

http://www.disastercenter.com/crime/akcrime.htm


To me, that is a much more significant point that the one Popeman is trying to make.

[Burgold]

I remember producing a feature based on a report examining the most "infectious" states. I had to spend most of the feature examining the weaknesses in methodology and then analyzing how and where the study's data could be useful. I don't take studies like this at face value. That said, everybody knows that Republicans like to shoot off their mouths without thinking more than anybody else. Why would it be surprising that they wouldn't treat their guns the same way?

[PeterMP]

I had to drop FL out of the analysis because I couldn't find 2009 data for it, and only looking at states (so DC was dropped out). Murders by hand gun are correlated with McCain vote share 0.24470675 and negatively correlated to Obama vote share -0.21268168.

http://www.guardian.co.uk/news/datablog/2011/jan/10/gun-crime-us-state#data
http://en.wikipedia.org/wiki/United_States_presidential_election,_2008

And if you look at the data in general, it is very similar. The value for La is only 10.46/100,000 people, but it still the #1 state.

**EDIT**
The p-value for getting such a spread in the correlation is appears to be about 0.061 so not quite statistically significant, with respect to this set of data.

[Popeman38]

Yes, and the average resident of Camden, New Jersey, or Newark, New Jersey, is more likely to die from a gun than the average resident of Princeton, New Jersey. Why doesn't that skew the results for New Jersey in the exact same way that you claim it skews the results for Louisiana?From my response to you:
The bolded is inaccurate. Additionally, suicide will never result in the average New Jersian being killed. Only the person who shoots themselves will be killed. And since over 60% more people killed themselves as killed someone else, taking out suicide paints a much more accurate picture. If I shoot myself in the head, there is a 0% increase in your chances of being the victim of gun violence. That alone makes the article and the stats misleading.Why is suicide included. That was the original point I was making. The suicide by firearm rate is more than 60% higher nationally than the homicide by fire arm. No one is worried about an individual killing themselves, because there will be no physical harm inflicted on them. That is like comparing snow to a river. Yes, snow is made up of water, but it is not the same as a river.
Let's put it this way - California has 5 major cities and a much higher percentage of its residents in urban areas than Arkansas does - why isn't California's gun death rate higher than Arkansas? Under your theory, it should. Put another way, Louisiana is not some outlier - if you look at the whole map, there basically are higher gun deaths in every "Red State" compared to equivalent "Blue States."

There may be plenty of reasons to explain this, but I don't see how the reason is "because this study is lying."

There is always a lag in statistical studies. Do you really think that something changed between 2007 and 2008 to completely change this picture?There is a significant difference in data from 2007 and 2009. Everything went down. Violent crimes, homicide, suicide. And just as a note, violent crimes maps illustrate a starkly different picture.
http://upload.wikimedia.org/wikipedia/commons/thumb/7/74/US_Violent_Crime_2004.svg/250px-US_Violent_Crime_2004.svg.png
Some of the higher "violent crimes" states are quite blue. Guess they must have more lax laws against murder and rape. Or is there not a corollary that can be drawn based on a simple study of stats. Because I have looked at the stats, and the higher violent crime states have the same restrictions on the crimes listed above as the non-violent states. My point isn't that the data was lying. My point was the author steered the article using stats that supported what he wanted the conclusion to be. If this article had come from Fox News, I doubt you would be defending it so vehemently.
To me, that is a much more significant point that the one Popeman is trying to make.Umm, I have included the whole suicide issue since my first post in this thread.

[Predicto]

If this article had come from Fox News, I doubt you would be defending it so vehemently.



I haven't been defending the study vehemently. I haven't been defending it at all. I have been questioning one of the points that you seemed to be trying to make, that the statistics are misleading because of differences between urban and rural crime rates. As far as I could tell, that point is meritless, because there are urban and rural areas in all states, and some of the states with the most gun deaths have mostly rural populations (Arkansas), while some of the states with the fewest gun deaths are very urban (New York or Massachusetts).





Umm, I have included the whole suicide issue since my first post in this thread.

Well, I'm sorry. It got obscured in all of your talk about how Obama strongholds are the source of all crime. :poke:

(I'm kidding, I just missed that part because I was focusing on the other point - really, the only thing I have talked about in this whole thread). Anyhow, it is a good point in the abstract, although PeterMP's post seems to indicate that the statistical correlation remains even if you are only talking about handgun murders.

Finally, I still tend to see a similar correlation in that map of violent crime you posted (and that map is from 2004, so why are you using it? :silly: )

[DRSmith]

Could you not then study the gun deaths by nations and then see what similarities there are

[Burgold]

Could you not then study the gun deaths by nations and then see what similarities there are

There would be a ton of confounding variables. Comparing Canada and Sudan is not really reasonable. Gun violence with a country at war or a nation that has no effective police force or is rife with instability is bound to invalidate the results

[DRSmith]

There would be a ton of confounding variables. Comparing Canada and Sudan is not really reasonable. Gun violence with a country at war or a nation that has no effective police force or is rife with instability is bound to invalidate the results

So you would say Sudan has not gun laws

You should be able to compare and countries and then compare them to the states