Heinous.

[frazee]

Pearson Correlations (r), N = 50 (50-state sample):

a) Shooting Deaths – % White non-Hispanic population: r = -0.042 (negligible), p = .774

b) Shooting Deaths – % African-American population: r = 0.181 (weak positive), p = .208

c) Shooting Deaths - % Hispanic population: r = -0.134 (weak negative), p = .354

d) Shooting Deaths – Gini Coefficient (income disparity): r = 0.010 (negligible), p = .947


e) Shooting Deaths – Religiosity: r = 0.518 (strong positive), p = .0001

f) Shooting Deaths – Conservatism (Romney minus Obama voters): r = 0.657 (very strong positive), p < .0001

g) Shooting Deaths – Gun Ownership: r = 0.692 (very strong positive), p < .0001

h) Shooting Deaths – Gun Ownership, controlling for income disparity: r = 0.755 (very strong positive), p < .0001

i) Shooting Deaths – Gun Ownership, controlling for race (index)*: r = 0.837 (ridiculously strong positive), p < .0001


And a Multiple Regression (Predictor Variables – Gun Ownership, Religiosity, Conservatism; Dependent Variable – Shooting Deaths):

R = 0.737 (very strong positive), p < .0001
-no colinearity violations
-residuals acceptably normal

Part Correlations:

Gun Ownership: 0.314
Religiosity: 0.155
Conservatism: 0.081

(This means that Gun Ownership is the biggest predictor of Shooting Deaths. Shocker.)




*((African-American population proportion + Hispanic population proportion) / White non-Hispanic population proportion) x 100

-Data normalized as appropriate


Sources of Data:

Shooting Deaths: http://www.cdc.gov/nchs/data/nvsr/nvsr60/nvsr60_03.pdf

Race: http://www.statehealthfacts.org/comparebar.jsp?ind=6&cat=1).

Income Equality: http://www.census.gov/prod/2012pubs/acsbr11-02.pdf .

Romney/Obama Votes: http://uselectionatlas.org/RESULTS/data.php?year=2012&datatype=national&def=1&f=0&off=0&elect=0

Gun Ownership: http://www.washingtonpost.com/wp-srv/health/interactives/guns/ownership.html

Religiosity: http://www.gallup.com/poll/153479/Mississippi-Religious-State.aspx?utm_source=alert&utm_medium=email&utm_campaign=syndication&utm_content=morelink&utm_term=Politics%20-%20Religion%20-%20Religion%20and%20Social%20Trends%20-%20USA#1

[Atypical Liberal]

1. It's also hard to drown without water.

This is hardly surprising.

If you don't own a gun, it's hard to shoot someone with one, unless you are able to steal one.

[frazee]

6. But the NRA...

...would have us believe that more guns mean more safety. That's my point.

My main criticism of the effort is that I wasn't able to parse out which shooting deaths were defensive in nature with the data accessed.

[AnotherMcIntosh]

27. I'm not familiar with the NRA particular positions. Where's your factual support for your

statement "But the NRA ... would have us believe that more guns mean more safety"?

If that's not a straw-man based upon something that you just made up, where's the link to an official position of the NRA in support of that?

[frazee]

31. This took less than 3 minutes to find.

http://www.nraila.org/news-issues/articles/2006/fables,-myths-other-tall-tales.aspx#FABLE VII:


The it the NRA's lobbying group. You can debate the worthiness of their arguments without me, for now, as I'm headed to work.

[AnotherMcIntosh]

33. Your statement "the NRA ... would have us believe that more guns mean more safety" is not supported

by your link.

Anyone can follow the link to a page which does not say what you said.

Although this is my first time reading the page and I may have read it quickly, I see nothing at all supporting your statement

"the NRA ... would have us believe that more guns mean more safety"

Whether a person owns a firearm or not is an individual choice. If you cannot point to specific language where "the NRA ... would have us believe that more guns mean more safety," you've obviously just that up as a straw-man.

(Actually, without any particpation of the NRA, it would seem to be a matter of common sense that if you do not live in a gated community and there is a prevalency of home break-ins, those engaged in such criminal activity may desire to practice their trade in safer areas where gun ownership is known to be lower. That's based upon common sense. Any position of the NRA, or non-position, is irrelevant.)

So you found a web page of the NRAILA. So what? You merely found a web page which does not support your representation. At most, you merely found a web page of the NRAILA.

[frazee]

37. If you don't get the implication of their discussions about these "fables,"

I can't help you. Don't much care either.

Now this was easier than I thought, and took all of five minutes. All the links showed up on the first page after a google search.

“More guns, less crime isn’t just ‘quite possible,’ it’s a fact.” ~ Wayne LaPierre, 11-27-12
http://home.nra.org/iphone.aspx/blog/342

“More Guns, Less Crime Again” ~ NRA-ILA, 9-15-12
http://www.nraila.org/news-issues/fact-sheets/2010/more-guns,-less-crime-again.aspx

“States that adopted nondiscretionary concealed-handgun laws saw murders decreased by at least 8%, rapes by 5%, aggravated assaults by 7% and robberies by 3%. The murder rates of women permit-holders fell by as much as five times the drop of their male counterparts.” NRA-ILA,
http://www.nraila.org/gun-laws/did-you-know.aspx

“Coinciding with a surge in gun purchases that began shortly before the 2008 elections, violent crime decreased six percent between 2008 and 2009, according to the FBI. This included an eight percent decrease in murder and a nine percent decrease in robbery.
“Since 1991, when total violent crime peaked, it has decreased 43 percent to a 35-year low. The murder rate, less than half what it was in 1980, is now at a 45-year low.
“Throughout, the number of guns that Americans own has risen by about four million a year, including record numbers of the two types of firearms that the Brady (Campaign) folks would most like to see banned — handguns and the various firearms they call ‘assault weapons.’” ~ NRA-ILA, 12-19-2012
http://www.freerepublic.com/focus/f-news/2593272/posts

“The statistics indicate that between 2008 and 2009, as gun sales soared, the number of murders in our country decreased 7.2 percent.” ~ 5-28-10
http://www.sodahead.com/united-states/nra---fbi-data-again-shows-more-guns-less-crime/blog-334161/

Thank you come again.

[AnotherMcIntosh]

38. So you just made this up about your boogeyman. Got it.

[frazee]

44. No. But in the future,

I'll be sure to bookmark every page I ever visit just in case I happen to get into an internet argument with somebody at some time in the future over some as-yet defined subject.

[gejohnston]

39. they are using FBI statistics

that show mayhem was dropping, that is empirical. Their correlation equals causation is just as full of shit as everyone else's.

[frazee]

41. I think you're arguing against yourself, there.

[AnotherMcIntosh]

42. Actually, he is right on point. Correlation does not equal causation.

[frazee]

43. Good grief.

So correlation = causation if the stats work in your favor, but correlation /= causation if they don't.

Come on, man.

[gejohnston]

45. no,

because they were using correlation in their favor. I called bullshit on them too. I was calling bullshit across the board. I don't get how you came to that conclusion.

[frazee]

47. My apologies for misunderstanding.

Correlations are often indicative of causal relationships, however. An across-the-board dismissal of correlation analysis is not .. can't think of a good word ... advisable, I suppose. As I stated below, correlation should never be accepted as indicative of causation unquestionably.

[DanTex]

2. You might be interested in this study...

They look at changes in gun ownership and changes in homicide rates at the county level, and find that, after controlling for various socioeconomic factors, on average, an increase in 10,000 gun-owning household results in between one and three additional homicides.
http://home.uchicago.edu/~ludwigj/papers/JPubE_guns_2006FINAL.pdf

[GreenStormCloud]

3. From that same study:

Page 9, bottom of page: Table 5 provides suggestive evidence that gun prevalence leads to elevated
rates of homicide through the transfer of guns from "legal" to "illegal" owners, rather
than through increased gun misuse by otherwise legal owners.

Legal gun owners aren't the problem, illegal ones are. Legal gun owners tend not to misuse their guns. IOW, criminals do criminal things.

[DanTex]

5. Umm, yeah, but as the authors point out, guns get transferred from legal to illegal owners.

Which means that, increases in civilian gun ownership overall results in higher homicide rates. Obviously some people are more likely to misuse guns than others. But overall, the statistics show that more guns = more homicide.

[GreenStormCloud]

8. Most such "transfers" are theft.

It takes only a small percentage of legal guns to be stolen to fully arm the criminal community.

[DanTex]

10. Speculation.

In fact, the evidence that most of the effect of gun ownership on crime is through transfers to illegal gun owners is also only suggestive, as the authors point out.

What the evidence shows conclusively is that higher rates of gun ownership results in higher homicide rates. There are many possible means by which this effect take place. The whole idea that the world can be divided neatly into law-abiders and criminals is fallacious.

[GreenStormCloud]

24. Then why do we have so many folks in prison?

Criminology is a well studied field. The vast majority of those who misuse guns already have police records of violent behavior. It is very rare for a law-abiding person to misuse a gun. It does happen sometimes, but it is rare.

[Shadowflash]

26. If the guns aren't there to begin with

Then, legal or illegal, they cannot be stolen and then used in crimes.

Sounds easy to me.

[GreenStormCloud]

30. So you want to ban all civilian ownership of guns.

That won't fly.

[Shadowflash]

32. I know it won't fly.

and, unfortunately, 20 kids and 6 teachers were the latest victims to pay the price for that 'freedom'.

Our society's priorities are SO screwed up.

[MightyMopar]

28. Yeah, guns are such a big deterrent to crime, they're one of top theft targets!

[GreenStormCloud]

29. Mos burglars make a strong effort to make sure no one is at home.

I the UK, they don't seem to care if anyone is at home. Their hot burglary rate is much higher than ours. Our burglars know that there is a high risk of getting shot if someone is at home.

[frazee]

7. Thanks. I look forward to reading it. nt

[jpak]

4. The gun nuts will not like this thread

nope

[ManiacJoe]

9. Is there a link to the complete report/study?

The OP leaves me wondering how to interpret the numbers.

OK, I have the Pearson explanation:
http://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient
http://changingminds.org/explanations/research/analysis/pearson.htm

So, is the rest of the report/study available?

[frazee]

11. Sorry, the stats are mine...

based on the data gathered at the linked websites.

The Pearson's r is a measure of how strongly two variables are correlated. If one goes up/down, does the other tend to go up/down? A causal relationship should not be inferred unquestionably.

The p-value is commonly called significance. It is commonly explained as the probability (that's where the "p" comes from) that the results of a statistical test could have arisen by chance alone. A score of 0.0001 means there is a 1 in 10,000 chance that the observed associated appeared just by chance; such a low p-value suggests that the association is "real." A positive correlation means that when one variable goes up, the other also goes up. A negative correlation means that when one variable goes up, the other goes down.

Multiple regression can be thought of as similar to a Pearson correlation, but instead of a one-to-one correlation, one is testing to see how several predictor variables might influence a dependent variable.

Part correlations are a bit tricky conceptually. When predictor variables are correlated not only to with a dependent variable, but also with each other, things get confusing. Basically, it's an effort to disentangle some of the intercorelations to see which predictor variable is having the most effect on the dependent variable.

Hope this helps.

[ManiacJoe]

13. Thanks for the explanation.

[AnotherMcIntosh]

34. The stats are yours, as is your claim that "the NRA ... would have us believe that more guns mean

more safety".

Why not take credit in the OP for making up the stats and the straw-man which refers to the NRA as a boogeyman.

As explained in Wikipedia,

A bogeyman (also spelled bogieman, boogeyman or boogieman) is an amorphous imaginary being used by adults to frighten children into compliant behaviour.

http://en.wikipedia.org/wiki/Bogeyman

[discntnt_irny_srcsm]

12. Please excuse me for being blunt but...

...are you saying that folks who have no guns don't shoot anyone?

[Tuesday Afternoon]

14. really confused, especially considering post #11 whereby he states that the stats are his

wtf.

confused.

[discntnt_irny_srcsm]

15. Yeah, me too.

He seems genuine but I always approach conclusions relying on statistics with a certain skepticism.

(Figures lie and liars figure.)

[frazee]

16. The intent of this post was to counter something...

...that I've heard with increasing frequency over the past four days -- that more guns is the answer. That guns make us safer. These stats dispute those notions. Perhaps I should have put them into some context, but I'm not much of a writer, as far as entertainment value goes.

I might have chosen a variable other than "shooting deaths," which encompasses not just murder and non-negligent homicide, but also justifiable homicide and suicide. But I'm willing to bet that the vast majority of cases are murder and non-negligent homicide.

For what it's worth, using the 2011 FBI stats, gun ownership is positively correlated with murder/non-negligent homicide when controlling for race (r = .455, a moderate-to-strong correlation, p = .001). Also with rape (r = .409, moderate, p = .003).

Point being: Guns do not make us safer. (Generally speaking, of course.) Now this would seem obvious to many of us, but to a huge portion of the populace, this is heresy.

[discntnt_irny_srcsm]

17. Not to sound like a broken record but...

...correlation doesn't equal causation. Eliminating a means while not addressing the causes will probably not change things overall.

(btw: I've hated statistics since my thermodynamics class.)

[frazee]

18. Correlation doesn't necessarily imply a causal relationship,

but it is often indicative of just such a relationship.

For grins, google "no correlation" and "gun" together and feast upon multitudes of people stating, in no uncertain terms, that there is no correlation between guns and violence.

One can often fiddle with the data (especially in the social sciences) to get them to say what one wants. I've tried to be as straightforward as possible, even offering the source data.

You are welcome to draw your own conclusions.

[discntnt_irny_srcsm]

19. In no way...

...am I saying that you are manipulating data or being dishonest in any way. I'm rather certain that there are various groups (mostly right wing folks) who not only think the only answer is more guns but that the real issue is that there isn't a problem.

Clearly, at least for me, there are (major) issues in addition than gun ownership.

[frazee]

20. There can be no doubt.

Social patterns are notoriously messy, stats-wise.

I often wish that I had remained a chemistry major.

[discntnt_irny_srcsm]

21. I hated chemistry more than thermo.

But maybe the best answer is just to wait and do the opposite of whatever the Republicans suggest.

[frazee]

22. Fair enough.

Incidentally (going back to your correct assertion that other factors are at play), even though I've characterized the correlation between shooting deaths and gun ownership as very high, the r value (properly, the r-square value) suggests that gun ownership "explains" less than half of the variance, leaving plenty of room for other factors. Even when I throw in a couple other possible causal factors, only a bit over half of the variance is "explained."

Of course, few if any patterns are ever thoroughly "explained" in the social sciences.

Bed now.

[discntnt_irny_srcsm]

23. Food for thought

Stay safe and have a good night.

[AnotherMcIntosh]

35. Yea, some people have created "more guns is the answer" as a straw-man

to advance their own agenda of knocking down their straw-man.

[PavePusher]

36. Except "more guns" isn't what is being said, so your hypothesis is false from the begining.

What is being said is "guns in the proper hands at the right times and places".

Very different things.

[frazee]

40. The survey on gun ownership was conducted

by the Behavioral Risk Factor Surveillance System. Now, I have no idea of the ratio of "proper hands" versus improper ones among the respondents, but I'm guessing that not too many owners with "improper hands" answered in the affirmative, especially since the group has a spooky word like "surveillance" in their name. If true, then what is being measured may be what you call "guns in the proper hands," whatever that means.

[GreenStormCloud]

25. You don't seem to have controlled for having a criminal record.

You seem to have lumped law-abiding folks in with the criminals. Texas has some published stats that you may be interested in.

http://www.txdps.state.tx.us/RSD/CHL/Reports/ConvictionRatesReport2011.pdf

It compares the conviction rates of Concealed Handgun License holders to the general public. The CHLers have a far lower conviction rate than the general public.

[OneTenthofOnePercent]

46. So... in summary... people need guns to produce shooting deaths.