"Study" shows carrying a gun = "4.5 times more likely to get shot"

[K-Romulus]

A bit of counter-intel that I hadn't seen here at MD Shooters.

I first saw the news on this "study" last week. Here is a link to the media spin:
http://www.philly.com/philly/hp/news.../63225267.html

Posted on Fri, Oct. 2, 2009

Think a gun protects you in a fight? Think again
(...)
If you have a gun during a fight, think twice about the protection it might offer.

Epidemiologists at the University of Pennsylvania School of Medicine yesterday announced the findings of a study about whether guns are protective or perilous during an assault.

It found that those possessing a gun in an assault situation were 4 1/2 times more likely to be shot than those not possessing one, according to the study's author, Charles C. Branas, associate professor of epidemiology.

It was released online this month in the American Journal of Public Health and will be printed in the November issue. (...)

and

http://www.philly.com/philly/news/local/63656382.html

Posted on Wed, Oct. 7, 2009

Monica Yant Kinney: A downside to carrying a gun?
(...)
"I carry when I go into cities for work or out to dinner with my family," Misus told me, "anyplace where I'm concerned for my safety."

I thought of my phone friend a few days later when University of Pennsylvania researchers released the results of a study seeking evidence that having a gun protects the holder from peril.

To the contrary, the epidemiologists found in the first-of-its-kind investigation: People with a gun on them were actually 4.5 times more likely to be shot than those who were unarmed.(...)

The media spin is just getting started, so expect to see references in the future, esp. at any shall-issue bill hearings.

I found a free link to the study here (saves you paying $30 to AJPH): http://works.bepress.com/dennis_culhane/88/

Abstract

Objectives. We investigated the possible relationship between being shot in an assault and possession of a gun at the time.

Methods. We enrolled 677 case participants that had been shot in an assault and 684 population-based control participants within Philadelphia from 2003 to 2006. We adjusted odds ratios for confounding variables.

Results. After adjustment, individuals in possession of a gun were 4.46 (P<.05) times more likely to be shot in an assault than those not in possession. Among gun assaults where the victim had at least some chance to resist, this adjusted odds ratio increased to 5.45 (P<.05).

Conclusions. On average, guns did not protect those who possessed them from being shot in an assault. Although successful defensive gun uses occur each year, the probability of success may be low for civilian gun users in urban areas. Such users should reconsider their possession of guns or, at least, understand that regular possession necessitates careful safety countermeasures.

******

Basically, the "conclusions" are total spin. Here are the facts as laid out in the study:

1) FACT: the study's data shows that out of the 677 people in the case group who were shot, 94% of them DID NOT HAVE A GUN.

2) FACT: the "control group" for the study was randomly telephoned Philadelphians who were asked "Were you shot on or around [the exact same time that a case group shooting happened in Philadelphia]? Yes or no?"

3) FACT: the study then matched up the "control" to the "gun possession" subset of the case group to get the "4.5x" number.

There are some other details in the study like they matched the case/control by age, race, etc., but that makes no difference in helping support the "conclusion."

"Knowing is half the battle."

More complicated analysis of the study is found here:
http://volokh.com/2009/10/05/guns-di...in-an-assault/

More analysis here:
http://www.skatingonstilts.com/skati...it-sounds.html

[Half-cocked]

Did the study sample a population of law-abiding permit holders, or does it operate "backwards" by sampling a population of people who were shot, AFTER they were shot?

If the latter, the study is totally meaningless, because it's obvious that criminals with a propensity for violence who carry guns, and who frequently put themselves in high-risk situations where violent encounters are likely, get shot at a far higher rate than law-abiding citizens who carry a gun only for self-defense.

[herr.baer]

Studies can show anything you want if you manipulate the data to show your point.

[MikeTF]

Figures never lie, but liars sure do figure! Key problems I see with the study: 1) Urban, specifically Philly, 2) No reference to crime rates in the study area, 3) very small sample size - probably not statistically representative, 4) a college/academic study outside of the field of criminology, 5) etc etc etc.

[sigrug17]

Originally Posted by herr.baer

Studies can show anything you want if you manipulate the data to show your point.

Totally agree. These studies are completely a FRAUD. Not scientific (meaningless control factors), and completely manipulated to show a certain result.

[zombiehunter]

amazon.com
How to Lie with Statistics

Book. Go.

[kohburn]

study seeking evidence that having a gun protects the holder from peril.

To the contrary, the epidemiologists found in the first-of-its-kind investigation: People with a gun on them were actually 4.5 times more likely to be shot than those who were unarmed

I think that is the part that bothers me the most - its presented as if they set out to prove one thing and "found" the other... when the reality is they carefuly manipulated the data to present something that I STILL can't figure out how they got.

677 people in the case group who were shot, 94% were unarmed

this says to me that if you are unarmed you are 16X MORE likely to get shot.

[alucard0822]

Originally Posted by kohburn

I think that is the part that bothers me the most - its presented as if they set out to prove one thing and "found" the other... when the reality is they carefuly manipulated the data to present something that I STILL can't figure out how they got.



this says to me that if you are unarmed you are 16X MORE likely to get shot.

yea, I took it the same way, you only have a few people that were shot during an assault, no telling if they were in the middle of a drug deal gone bad, gang fight, or even a law abiding citizen that stopped the crime, shot the assailant, and then got shot themselves. Like most brady funded BS "studies" it is easily debunked and countered, and serves the purpose of making them look like idiots to anyone that has any concept of scientific principals.

[-Mil-Surp-Phreak-]

There are 3 kinds of lies. Lies, damned lies, and statistics.

One of my favorite quotes.

[fightinbluhen51]

Originally Posted by zombiehunter

amazon.com
How to Lie with Statistics

Book. Go.

More Guns, Less Crime, The Bias Against Guns, GO!

Seriously, this is ridiculous, which can be evidence by just simply doing a study on PA permit holders by themselves. Seems that they probably didn't ask "are you a criminal or have you been convicted of a crime, charged with a crime, arrested, ect" to provide for the fact that criminals have a higher chance of actually being shot. And shot by perhaps, legal gun owners? Yes.