And the 2009 Junk Science for gun research goes to...

[Geneseo1911]

Surprise surprise, ABC radio picked this "study" up and ran it on their "news" at the top of the hours. ABC radio may be further left than Air America. It drives me nuts as I love our local AM station, but they still insist on carrying ABC.

The contradiction sometimes makes my brain hurt when they deliver the "news" at the top of the hour during Rush.

 

[skoro]

Oh, really...

Penn researchers investigated the link between being shot in an assault and a person’s possession of a gun at the time of the shooting. As identified by police and medical examiners, they randomly selected 677 cases of Philadelphia residents who were shot in an assault from 2003 to 2006. Six percent of these cases were in possession of a gun (such as in a holster, pocket, waistband, or vehicle) when they were shot.

So more than 9 out of ten who were shot during an assault were NOT armed, yet they conclude that those with guns are "4.5 times more likely to be shot."

hmmmmmmm...

I didn't major in statistics and I'm no mathematician, but even I can clearly see a wee bit of a problem with this "study."

 

[RS14]

So more than 9 out of ten who were shot during an assault were NOT armed, yet they conclude that those with guns are "4.5 times more likely to be shot."

hmmmmmmm...

I didn't major in statistics and I'm no mathematician, but even I can clearly see a wee bit of a problem with this "study."

What you think is a problem is not actually a problem. See my previous post here.

 

[Ky Larry]

As Mark Twain said:" There are lies, damn lies, and statistics."

 

[noskilz]

I'm not a professional statistician, but I have a real problem swallowing the results. Their own data provides the following:

Gun Victims Being in Possession of Guns – 5.92%
Control Group Being in Possession of Guns – 7.16%

From their own data, 7.16% of the Control Group had a gun, yet only 5.92% of the victims had a gun. Accordingly, it follows that a higher percentage of the Control Group, who was not shot, was carrying guns than the percentage of gun victims.

Yet, here is their conclusion: "After we adjusted for confounding factors, individuals who were in possession of a gun were 4.46 times more likely to be shot in an assault than those not in possession."

"Adjusted for Confounding Factors." I have to admit to being confounded. :banghead:

Looking at the real data, some real differences were as follows:

High Risk Occupation of Gun Victims – 24.34%
High Risk Occupation of Control Group – 11.40%

Gun Victims Being Outdoors – 83.13%
Control Group Being Outdoors – 9.05

Here's another rather relevant quote from the study: "However, compared with control participants, shooting case participants were significantly more often Hispanic, more frequently working in high-risk occupations, less educated, and had a greater frequency of prior arrest. At the time of shooting, case participants were also significantly more often involved with alcohol and drugs, outdoors, and closer to areas where more Blacks, Hispanics, and unemployed individuals resided. Case participants were also more likely to be located in areas with less income and more illicit drug trafficking."

Funny how this wasn't in the article.

Also this gem: "Our control participants were, however, significantly more unemployed than the general population."

I think they could have found you 4.5 times more likely to get shot if you had a frog in you pocket with application of the right "confounding factors."

 

[Tropical Buzz]

Mathematical accuracy does not always imply honesty

RS14, social statistical conclusions are drawn from the results of a focused study involving the gathering of data from a representative sample of a given population. I am not arguing the mathematical accuracy of the data used to arrive at the study's conclusion. What I dispute is the ignoring of relevant data and contributing factors that effectively skew the gathered data towards supporting a predisposed position or agenda.

If a man shows you an orange and you ask him "How many do you have in that bag?"
He can answer "12." but that 12 could be 7 oranges and 5 lemons. When you asked how many you did not specify how many "what" and therefore his answer is mathematically correct - it's just not honest, that's all.

 

[Zane]

To add to what NoSkilz said, the study barely scraped by with significance in any category and there were marked differences between the study group and the control group. The best part is looking at the mean drug arrests per mile. There are 50% more in the shooting group than in the control group. Additionally, ~24% of the fatal shootings involved drugs, where as it was ~5% in the control. Basically, drugs and crime lead to shootings, yet somehow they try to argue that the mere possession of a gun is what actually led to the shooting. Somehow I think the fatal ones where the individual was ambushed and killed had very little to do with the gun and much more to do with the drugs.

What a load of horse hocky. This one doesn't even pass the smell test. I may have to actually write a letter in response.

 

[Cosmoline]

If this study is correct, and having a firearm makes you more likely to be shot, then why doesn't the Philadelphia police department--who CO-SPONSORED THIS STUDY--order every beat cop to DROP HIS GUN BELT RIGHT NOW!?

If it is safer to be unarmed, then the police should also want to be unarmed.

After all the police in London are mostly unarmed, and they have vastly fewer police killings there than in sunny Philadelphia!