Firearm Related Studies Vol 1: Gun Ownership as a Risk Factor for Homicide in the Home

Great Debates
1

The idea behind this thread is to discuss one specific firearm related study. Often threads about firearms meander a bit but for the purpose of this thread I’d like to focus on one specific study. There are many different studies and I’d be glad to start additional threads on those as well, but focusing on one at a time I think can be more productive. I’m starting with one that is often brought up in various discussion – the 1993 Kellerman study, Gun Ownership as a Risk Factor for Homicide in the Home

The study.

Some excerpts:

Some caveats from the study:

The rest of the study material is linked, with underlying charts and tables.

I find the study results poor at best given the methodology employed. In this study several things stood out –
[ul][li]The control group was dissimilar to the test group[/li][li]There was no control for guns brought into the home from another source[/li][li]Several other factors presented a greater risk than firearm existence[/ul][/li]
The one I find most damning is that in this study, there is no effort or distinction made between the gun kept by the homeowner and one that is brought in from outside the home. It’s like saying, you kept gun A, therefore you are 2.7 times more likely to be killed by gun B.

When this study is used to support various arguments about the danger of firearms, these are some of the things that call into question the conclusions of this particular study. Ultimately I don’t think anyone disputes that firearms present a risk - of course they do. However, a more meaningful question is whether those risks are outweighed by the benefits. This study makes no attempt to address that question. As a result, the conclusions from this study are poor, and the usefulness of this study in decision making is of little value.

2

Well that’s a tricky part for extrapolating from the study but not damning to the study itself, istm. Maybe gun owner A only had a gun because he lives in a gun heavy warzone so he of course had a gun at home when he got killed by B. But maybe gun owner B wouldn’t have brought his gun if he didn’t know gun owner A was packing. Maybe gun owner B was there to steal gun A, in which case gun A was definitely implicated statistically in the death. You are saying it’s damning because you assume A has no relation to B.

3

Or they live in higher-crime areas, and are therefore more at risk of getting shot no matter what their psychological proclivities.

That seems like a funny way to express it. If the authors are studying guns in the home vs. no guns in the home, ISTM to be obvious no gun in the home offers no protection at all vs. the limited protection of a gun in the home. If I feel at risk because someone is out to get me, and I don’t have a gun, then the failure of no guns is going to be greater than inadequate.

And correlation isn’t causality. It’s like the guy who tried to carry a disabled bomb onto an airplane, because the odds of two people carrying bombs onto the same plane is much greater than for one.

Regards,
Shodan

4

Sociological data is always going to be rough around the edges. Life is not a laboratory and you can’t control every variable entirely. But you can’t use that as an excuse to dismiss the results entirely.

5

No gun in the home can be some protection, as it doesn’t escalate the situation. Someone who’s out to get you might just want to set your empty parked car on fire. If you come out with your shotgun, it can quickly change from arson to a homicide. I am not in any way discussing what is the proper course of action, simply saying that I don’t think it’s as obvious as you think.

6

I’m wary of “Studies” these days because many of them are for too short a duration of time and cover too small of a sample group. Too often, the results of these “Studies” are tainted by biased interpretation. This often occurs because Universities receive Grants for these “Studies”, and they need to churn them out and produce some kind of meaningful results if they want to keep receiving the Grants. I had a great Probability & Statistics post graduate course where we had a 2-week unit called, “How to Lie with Statistics”. In other words, caveat emptor.

My father owned firearms, and there was never a single problem. First of all, he secured the weapons properly so my brother and I could not access them. Secondly, to address the age old principle of, “forbidden fruit causes temptation”, our father introduced us to firearms formally when we were just ten years old. Before we held a loaded gun, we learned a lot about them including the safety practices necessary to protect the lives of others as well as your own. I carried that learning into adulthood and have never had a problem being a responsible gun owner myself, and neither has my brother.

The issue isn’t responsible gun owners or guns in the home per se, the issue is the lack of controls that allows guns to get into the hands of virtually anyone who wants one no matter what their record or mental state. The NRA panics their constituency into believing that gun control means taking away the guns from responsible gun owners. They fall for it in a big way, and thus they oppose any effort to establish sane gun laws.

7

That’s nice but this thread is about a particular study, not “Studies”, nor is it about how everything would be fine if every family did a proper gun education course for their kids.

8

Sure, that is also possible. Maybe your boyfriend is beating you up, and you don’t have access to a gun to shoot him, so he is protected (more or less) from not having a gun in the home. I don’t think you can parse the data that finely.

Regards,
Shodan

9

Or maybe the boyfriend doesn’t shoot her because there’s no gun in the house. I am not even remotely parsing data. I am pointing out that you are making unsubstantiated assumptions.

10

I gave a cursory look at the study, but how is ‘homocide’ being defined. An accidental discharge could be a homocide without anyone being in danger. And you HAVE to count that sort of death and wounding if you’re going to try to study this issue.

Also, I note that the study omits incidents involved people 12 and younger. That’s also going to skew results.

11

As far as I can tell, the study discusses all death from guns, minus suicides. So the dictionary meaning not the legal one.

And no, omitting <12 yr olds doesn’t “skew results”. It omits them from their results.

12

I read the entire study, including the letters to the editor and responses. I didn’t examine it as though preparing for a test, but I did read it all. If I have responses beyond this one, I’ll post them. If not, I won’t.

To answer your questions though, from the report:

So any death ruled a homicide, presumably by law enforcement and/or the ME. I agree that doesn’t seem to be spelled out, because “ruled” could mean by the authors of the study (though I doubt it).

They also don’t explain why the MEs didn’t want victims under the age of 12 to be included.

I’m still interested in the specifics of why this particular study is flawed. When I google “Kellerman Gun Study Flawed,” I get results from firearmsandliberty.com, guns.com, thetruthaboutguns.com, tbuckner.com, and gunowners.org. Truthfully, I haven’t looked through any of those sites, but from the domain names, they’re not very objective in relationship to guns.

Are there any objective sites I can go to that shows why the Kellerman study is flawed?

13

Did you miss the three bulleted items in the first post?

I agree there’s difficulty in extrapolating from this data, and that’s the purpose of the thread. My point isn’t that the contrary position to this study is correct, no. The point is that the data and the conclusions from the study are flawed, to such an extent that relying on them is also flawed. People have used this study to support the contention that the risk of homicide is nearly three times higher in homes with firearms. That assertion is based on this study, and used to extrapolate to all homes across the country. But since the study is flawed, that assertion doesn’t hold water.

14

I’m not accusing you of parsing anything - I am saying the study can’t be parsed. But if someone gets shot and that is a failure of the gun to adequately protect, then if someone gets killed and there is no gun in the house, that is a failure of no guns to protect adequately. It’s a trade-off, and we can’t tell from the study what the numbers are.

Although I think this

is a more cogent criticism.

Regards,
Shodan

15

Well, of course not! Given the massive numbers of yearly home invasions turned aside by vigilant and well armed citizens? Those academic pencil necked weenies don’t want a truth like that to sully their anti-gun biases! That’s why the NRA had no choice but to intervene when it began to look like Congress was going to give those lily-livered wimps funding for one of their “studies”!

Heard that the NRA was in negotiation with the famous Christian rock group Disciple to produce a video What Heat Does Jesus Pack?. One never knows how these internet rumors get started…

16

That study has been debunked by this study:

*http://www.firearmsandliberty.com/kellerman-schaffer.html
Serious Flaws in Kellerman, et al (1993) NEJM
(December, 1993)
by Henry E. Schaffer, Ph. D.
Summary and Overview
The Kellerman, et al (1993) study in the NEJM attempts to use the case-control method (CCM) to show that gun ownership increases homicide in the home. The limitations of the CCM, and serious flaws in the study methodology, result in invalidation of the study’s conclusions.

The CCM has a number of limitations in what it can accomplish, and has a number of conditions (assumptions) which must be satisfied for it to be able to satisfactorily accomplish even the limited goals for which it is suitable. The biggest limitation is that the CCM can’t demonstrate causation. The CCM finds ‘associations’ between studied factors and the ‘outcome’ which defines the ‘cases’. These ‘associations’ may suggest that there is a causal relationship, and may then be used to justify a study of causal relationships, but it is incorrect to jump from the discovery of an association to a conclusion of causation. Other weak points in the CCM have to do with susceptibility to biases in the selection of the cases, and with confounding factors which can affect the choice of the controls. These can easily lead to spurious associations when there actually are none, or to associations which are reversed in direction from what actually exists.

The Kellerman, et al (1993) study has been widely quoted as demonstrating that there is a causal relationship between handguns in the home and homicides. The paper itself doesn’t go that far, but it uses suggestive language, which suggests that there is more than merely an ‘association’. The flaws in the paper are such as to make the the reader suspicious of the association found. Showing flaws in the methods does not prove that the paper is wrong, but it causes a loss of confidence in the results. Conclusions which are not properly supported must be considered invalid until proper support becomes available, if ever. It is the responsibility of the authors to support their conclusions. It isn’t the responsibility of the readers to go out to collect data to prove that the flaws in the paper lead to incorrect conclusions.

The detailed treatment of these flaws, with supporting data, examples and methods is necessarily quite long, but it does illustrate that the Kellerman, et al paper is based on unsupported assumptions and that the conclusions must be viewed with suspicion or rejected as being unsupported.*

There is also this editorial:
http://www.guns.com/2015/08/24/kellermanns-gun-ownership-studies-after-two-decades/

and this:

http://guncite.com/gun-control-kellermann-3times.html
*In a letter to the editor in the New England Journal of Medicine, “The students of Dr. Mark Ferris’s Mathematical Statistics 460” class ask, “In how many of the homicides was the victim killed with a gun that was kept in the house rather than a gun that was brought to the house by the perpetrator?” The question is a relevant one since, as the letter also notes, the study’s authors had stated in part based on their findings that “people should be strongly discouraged from keeping guns in their homes [p. 1090].” In other words, advising people against keeping a gun in the home doesn’t make sense unless it causes an increase in homicide risk.

…
Kellermann’s own data suggests that for all gun homicides of matched cases no more than 34% were murdered by a gun from the victim’s home. (GunCite’s analysis of Kellermann’s data.) …

Additional analysis of Kellermann’s ICPSR dataset shows that just over 4½ percent of all homicides, in the three counties Kellermann chose to study, involved victims being killed with a gun kept in their own home (see derivation). This supports the conclusion that people murdered with a gun kept in their own home are a small minority of all homicides, precisely the opposite of what an uncritical reader of Kellermann’s study would likely conclude. *

I want to emphasise that last "This supports the conclusion that people murdered with a gun kept in their own home are a small minority of all homicides, precisely the opposite of what an uncritical reader of Kellermann’s study would likely conclude.

17

So, no more than one in three murdered by their own weapon? That’s reassuring.

18

I didn’t miss them. Point one:
-The control group was dissimilar to the test group
How was the control group dissimilar?
From the report (Table 2), I see the gender percentage between the case subjects and controls are the same.

Race is close (32.9, 62.1; 34.5, 61.6 white and black; case and control respectively); age groups are close if not an exact match, it’s only the type of dwelling that seems to differ significantly.

Point two:
-There was no control for guns brought into the home from another source
No comment

Point three:
-Several other factors presented a greater risk than firearm existence
Such as?

I’m not necessarily defending the study, nor am I attacking it. I just prefer my arguments to be reasoned, logical, and free from emotional bias (if that’s at all possible).

19

So, if you take 2.7 times more likely and 1 in three are killed by a gun in their own home, you get “no more likely to be murdered if you have a gun than if you don’t”.

But actually it’s
*Additional analysis of Kellermann’s ICPSR dataset shows that just over 4½ percent of all homicides, in the three counties Kellermann chose to study, involved victims being killed with a gun kept in their own home
*

20

I’m betting on a nail gun.