http://aje.oxfordjournals.org/content/160/10/929.full
It seems to me the paper is saying that you are at least twice as likely to die from a homicide if there is a gun in the home, or 3 times, depending on what sentence you read.
“far more likely”, though is pretty clear.
I guess I would point out that you are assuming that having a gun causes you to be more likely to be murdered (sometimes with a gun and sometimes not with a gun). Why isn’t it just as valid to assume that people who are likely to get murdered frequently buy guns?
The Newtown killer killed his mother and broke into her gun cabinet.
I’d like to point out that there are some questions pertaining Kellerman’s methodology on that 93’ study. His Wikipedia page says as much and a quick google search reveals a number of editorials and threads critical of his study.
Ironic that the link to that 2004 paper was posted just yesterday. I say that because today, in the more than credible Annals of Internal Medicine, a major new study came out.
HERE is the link to this new study.
It is an important addition to the literature since it represents a scientifically credible (and I think successful) attempt to analyze virtually all previous research studies on the question of whether ‘guns in the home’ are associated with suicide and murder and, if so, to what extent.
This type of study is called a meta-analysis and is being used increasingly to try to get answers to tough questions, especially when earlier studies gave conflicting results. Still, it’s not a foolproof method and has its unique liabilities, e.g. failure to include “negative studies” (which are less likely to have been published in the first place and thus may not have been included in the meta-analysis), and assigning too much weight to a few of the studies being analyzed (with studies weighted on the basis of how many patients were enrolled, appropriateness of the study design, . . .).
In any case this, new article not only summarizes succinctly much of the previous research in the area, but extends and solidifies it. I think it is likely to become an oft-cited bit of research in the ‘guns in the home debate’.
Oh, what were its results? You can look for yourself or take my word for it when I say that it demonstrated that guns in the home are associated with an approximately tripled risk of suicide (compared to homes without guns) and an approximately double risk of being murdered (compared to homes without guns).
Actually I doubt so…your run of the mill burglar is neither particularly clever nor particularly well prepared / knowledgeable and their crime is generally not very targeted beyond “hey - that house looks easy to break into” (there is much anecdotal evidence to support this view)
If somebody is specifically going to target your (the general you) house, with forethought and knowledge of what they expect, and have researched etc etc - I would rather NOT be armed.
If somebody has gone to that level of research and preparation, then intuitively they are also going to prepare themselves to face your resistance / over come and subdue you. And we know for a fact that when it comes to guns - the best advantage is preparation. If they already come into your house with the thought of subduing someone with a gun, and have prepared themselves for that, they have a huge advantage over you.
And this is the problem I have with defensive gun use - I’d think that burglars generally fall into two broad categories
a) those that want to avoid the owner, just get stuff and get out
b) those that are there to harm / subdue the owner, are looking for violence.
In most cases for (a) you don’t need a gun, they want to get out of the place more than you want them gone, but by threatening them with a gun, you are just inviting them to do something rash.
For (b) - if they truely mean you harm - they are more likely to have the jump on you than you are to get on them, and when they see a gun things are more likely to go badly. Besides which - this sort of crime is very very rare.
Some facts and figures for those that didn’t read the study
That suicides are predominantly ‘impulse’ acts (particularly in Western societies) is very well-known. We can not make any meaningful comparison to suicide rates in countries like Japan, for instance, where the radically different views of suicide and (the almost complete lack of) religion skew the data.
We also know that merely making it slightly harder to kill yourself means that people initially thinking of suicide…kinda just go on living. In other words - faced witht hte idea of killing yourself, most people couldn’t be bothered if they actually have to *work *at it. See the example of putting up barriers on bridges, the example of gas ovens in the UK, etc.
Each year thousands of people kill themselves with guns. Would every one of them go on living if guns weren’t so readily available? No - I’m sure some would find other ways…*but those ways are already available to them anyway. *I suspect that a conservative estimate would see - at a minimum - an immediate drop of 20-30% in suicides.
Obviously I doubt we’ll ever see that in my lifetime.
Bruce Wayne observes for us that in response to the published article, people on the internet have written criticisms.
Thanks for the insight! Wow!
Two of those “people” were Gary Kleck and Don Kates. Both criminologists, I believe the former played some role in the heller case.
From the study:
The second sentence caught my attention - not sure how they synthesized this (and I will admit that I do not know the math involved). A challenge that comes up in some of these studies is that homicide tends to occur in households that have had prior run-ins with the law. Controlling or reporting on that has an impact on the final numbers.
More on Homicide limitations:
But in the end doesn’t it have the same failure to prove any sort of causation (at least on the murder front).
Please cite an example of a study that proves causation of any thing. Please. Any study. Any thing.
Are you saying that the study that all the hoplophobes are waving around doesn’t prove a goddam thing (well, I guess it is trying to prove correlation)? Or are you saying that this study gets close enough that we can draw conclusions because there aren’t any glaring flaws in the study?
Just look a few posts up to Algher’s posts.
So, you still struggle to understand the basic elements of science and how it is conducted? Dumbass.
And if you have a pitbull you’ll be even safer! 
Actually, any dog will do in most cases, even just a yappy ankle-biter, but I just had to get in that dig because I’m evil. Sorry.
So still no response to Algher’s points? Or are you still too busy being an asshole?
Authors discuss the potential limitations of their data and analyses in papers. It’s standard.
Oh, sometimes I forget how fucking ignorant and stupid you are. You probably didn’t know that and can’t figure out what they mean.
Have you found a study of anything that proves causation yet? Can you figure out the issue, or are you too stupid?
Those are just potential limitations? They look like flaws that make the study meaningless in a policy setting. Thanks for playing. Good luck with repealing the second amendment.
If your only arguments at this point are pettifoggery and insults, then you are probably wrong. I mean if your only participation is to chirp in from the peanut gallery to point out that I used the word “prove” in a way that everyone else understood but you took issue with, then what sort of point do you think you are making?
At this point you are just bitter that the world does not bend to your will. You should do something about that, it will affect other areas of your life.
So, your view is that the “Limitations” section of articles - a standard section of published scientific articles - reflect flaws that make the study meaningless? You are suggesting that peer reviewers and the editor would agree on the publication of an article that in its own text includes information that renders the article meaningless?
You are a mouth breathing moron and I grow increasingly curious how you are able to even turn on a computer, let alone type out messages. It does help to explain why you are completely unable to change your message board behavior after people repeatedly tell you that walls of text response make your contributions less than useful. Imbecile.
You fucking idiot. The problematic word was not “prove.” It was “cause”. The point is that there are no studies of anything on any subject that prove causation. When you challenge research on this particular topic by observing that it does not prove causation, you’re lodging a complaint that is true about all science.
You’re a person who is just not very smart at all. I don’t doubt that my informing you about stats, or about standard elements of publications like the limitations section, or about “cause” seem like pettifoggery to you, but that’s only because you lack the cognitive capacity to understand what I am saying.