In a first-of its-kind study, epidemiologists at the University of Pennsylvania School of Medicine found that, on average, guns did not protect those who possessed them from being shot in an assault. The study estimated that people with a gun were 4.5 times more likely to be shot in an assault than those not possessing a gun.
source
From the abstract of the paper:
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, PA, 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.
Four plus times more likely is quite a substantial increase in risk. This is what I have always suspected. There are two obvious, opposite recommendations that come to mind: 1) Stop carrying a gun; or 2) Add body armor.
I hope my friend Josh reads this!
Hat tip to Virgil Samms for letting me know about this interesting study.
Source: Branas, C., Richmond, T., Culhane, D., Ten Have, T., & Wiebe, D. (2009). Investigating the Link Between Gun Possession and Gun Assault American Journal of Public Health DOI: 10.2105/AJPH.2008.143099
- Log in to post comments
More like this
I'm bookmarking this one... not that it's going to impact those for whom guns are religious fetishes...
I don't have access to the full paper, but unless the participants were very frank about any criminal activity they were involved in, along with all of the normal sociological questions (how many people do you know who have been shot, etc), I don't see how the authors adjusted for the relative risk of assault ofgun possessors compared to those who don't own guns. In a city with a lot of gun crime, the disparity could be substantial. If they couldn't calculate that, then did the authors ascribe risk of assault to possession of a gun that should instead have been ascribed to the factors that led the person to acquire the gun in the first place?
I think a better study might have been to look at only people who were assaulted with guns, and then try to determine the relative risk of the victim being shot during an assault based on whether or not the victim was armed as well.
It's a shame, and extremely frustrating for many of us, that most people will end up generalizing to dubious conclusions from such studies. What far too many obnoxiously loud voices end up doing is questioning the validity of scientific methods and conclusions instead of considering the possible usefulness of their insights.
Following hibob's comments, I wonder how many of the 677 were carrying a gun legally, i.e., had concealed carry permit or were a LEO.
Obviously then, we should remove guns from the police as well and reduce police fatalities.
But I forgot, it's good for police to carry guns because there is some psychic difference, for them guns help control crime, but for other citizens guns cause crime. Tricky.
I encounted an example of this bizarre double standard when a 'Babies R Us' store where a fully armed uniformed police officer was giving advice on carseats. Now if any other citizen walked into that store armed, people (especially liberals) would be screaming 'think of the children' and declaring that simply the presence of the weapon was tantamount to bad karma and endangered everyone. but if he's got a badge, that bad karma does not apply
How did they correct for the fact that people who, for busines or other reasons, frequently deal with dangerous areas are more likely to be carrying a weapon?
My parents always told me that having a gun in your house made you more likely to get shot, so I'm surprised to find that this is new research.
Jay: I think you are totally missing the point of law enforcement.
Since guns are a phallic symbol, those who carry them are more prone to violence due to their macho behavior and their need to compensate for inadequacies.
"But I forgot, it's good for police to carry guns because there is some psychic difference, for them guns help control crime, but for other citizens guns cause crime. Tricky."
Not tricky at all. Police have extensive training, and daily routine application of that training, more than any civilian can possibly get, not only in the safe technical usage of firearms, but also in the decision making and risk assessment processes involving in even decided when and where to draw the weapon, as well as in dealing with aggressive crime in general, in which the firearm is just one tool out of many.
Anyone using the word "tricky" to describe this is either dim-witted, deliberately deceptive, or has a political axe to grind.
"Obviously then, we should remove guns from the police as well and reduce police fatalities."
Excellent idea. I grew up in the UK where police did not carry guns except in special cases. It made for much a much better relationship between the police and the public and the police had use their heads rather than their guns. They still did a good job, and I suspect they had less fatalities. Of course if you wanted a gun you had to get a license, you could only get a sporting gun not a killing machine, and you were responsible for that gun. so every creep out there was not armed.
I am old, so it may not be that way now.
Somehow I suspect most of their sample are going to be people that are carrying illegally. I can't see the paper right now, but I'm skeptical that this has any bearing on gun control laws (maybe it has some bearing on sentencing for violations though). (And I say that as someone who is pro-gun control.)
What I find interesting is this: If you are pro-gun, all sorts of claims about how gun ownership is necessary and effective as protection from criminals, the gumment, and foreign invaders can come flying like winged monkeys out of your ass all day. Then, if a peer reviewed scholarly paper comes along suggesting the opposite, it is suddenly time to get all critical-thinking and methodological.
this study validates that ol' adage: "he who lives by the sword, dies by the sword".
oh, and jay - seriously, are you obtuse on purpose, or you actually see NO difference between an armed cop and an armed civilian? none? really? just some arbitrary 'double standard'? huh. yeah, like Irene said - you don't "get" law enforcement, do you...
Note; I am not arguing about the study - I rather assumed what the study has confirmed. But...
Not tricky at all. Police have extensive training, and daily routine application of that training, more than any civilian can possibly get, not only in the safe technical usage of firearms, but also in the decision making and risk assessment processes involving in even decided when and where to draw the weapon, as well as in dealing with aggressive crime in general, in which the firearm is just one tool out of many.
Bullshit. Complete and utter bullshit. I have just as much training in the safe technical usage of firearms as your average cop. I also can and have outshot several cops on the range - for that matter, I can outshoot my cousin, who is a master sharpshooter in the marines. And I can take a class at the local gun club, that is just as comprehensive as the class in the law enforcement curriculum at my local community college, where a lot of officers are qualified. I know this, because the guy who teaches that class, also teaches the one at the gun club and uses the same curriculum. He also goes into the same training for non-deadly force techniques and when what course of action is appropriate.
I don't think a lot of you folks "get" law enforcement. Excepting the FBI and some state police forces, there are few law enforcement agencies that require all that much special training that us non-law enforcement types can't easily pick up. Indeed, I have done something that very few of your average municipal cops have ever done and have actually gone through a tactical range. The training most cops get isn't anything exciting, nor is it all that demanding - unfortunately. Or have you missed the myriad stories of fucking stupid cops doing really fucking stupid things?
Go visit the seed blog; "Dispatches from the Culture Wars."
@Greg Laden #13
Not sure if that was aimed at me, but I'm not pro-gun. If a research paper is worth talking about at all, it's worth getting all critical-thinking and methodological while doing so. This is scienceblogs, right?
Greg Laden:
I don't know if I count as pro-gun or not. I don't think guns are effective at protecting against most crime. The ACLU is a far better defense at government overreach. And Americans don't need to worry about foreign invaders. The only people I know who carry guns are LEOs.
OTOH, about half the people I know, from grandmothers to university professors, own a gun of some sort or another. Which is about the estimated average in the US.
Have you missed the myriad stories of fucking stupid non-cops doing really fucking stupid things?
You may have all sorts of training and mad gun skilz, but if you think your purported expertise is the norm, you're nuttier than a fruit cake.
There is a very simple alternative hypothesis here: people who are at greater risk of being assaulted (e.g. who live in high-crime areas) are more likely to decide to carry a gun to defend themselves.
Judging by Science Daily's description of the study's methodology, it does not exclude this hypothesis. So, I think it's a bit of a leap to conclude based on this study that carrying a gun causes you to more likely be shot.
In many homes guns represent the most attractive targets for thieves. Second only to large amounts of cash or narcotics.
"We adjusted odds ratios for confounding variables."
I'm curious...how does one adjust for the confounding variable that people who are worried that they will get shot are more likely to carry guns?
Also, Greg, you're being an ass. If somewhere some pro-gun people are making stupid and unsupported assertions, then they deserve scorn for that. But if your commenters are making specific methodological objections, I think you owe them better.
How would you like it if everyone dismissed what you had to say just because somewhere, some anti-gun person (whom you have no affiliation with) was spouting nonsense? Because it's not that hard to find examples.
Umm, Mike - did you actually read the quote, or were you too excited about jumping on me?
First off, I don't have "mad gun skillz." I am a really good shot and I had competent instructors when it came to gun safety. Second, the point was that most cops don't have any kind of special fucking skills - which they don't. Anyone who has been through a competent hunter safety course, or who has fired at a gun range, has much the same firearm safety training that cops do.
And yes, asshole, my "purported" expertise is the norm - at least where it counts. Just because you might happen to ignorant in regards to guns, doesn't mean that most people - most of the ones who actually own guns are as ignorant as you are. Most gun owners are all about safety and handling them competently. Just like most cops are. Of course there are fucking morons who do really fucking stupid things with guns - if you weren't rather incapable of actually comprehending written English, you would note that I was pointing out that most cops aren't any more competent than most gun owners, not that cops are somehow less competent.
And for the record, I am all about the notion of only allowing cops with specialized training to actually carry a firearm on their person. I am rather fond of the Brit model...
I probably am being an ass. So I'll ask this: Why does everyone else get to be an ass while talking about guns but me?
I've read the paper, and I'm looking into general treatments of the methods they say they used. Offhand, their conclusions don't appear to clearly follow from their data, methods, and results. Maybe I'm just missing something, but I count myself as reasonably astute, and there are a number of possible problems I'm seeing in the paper. There are many characteristics that they report as significantly different between the case and control groups, and "gun possession" is not one of those in any of the three conditions that they modeled. I think that means that if they bothered to work up a risk assessment for any of those properties, they would come up with risks greater than the 5.45x found for "gun possession" at 0.91% difference between groups. The big one would be "being outdoors" where about a 72% difference between groups held; that should result in a huge risk factor if it is at all proportional to that difference figure. I'm working up a blog post since I started a comment here and it got way too long.
I'll make one more point, and that is about methodology. The dataset is entirely based on shootings. The authors chose a case-control method. But I would have thought a more direct method would have been to compare cases of assaults with firearms where a shooting did not result to those where a shooting did result. Is there reason to believe that a case-control approach would be superior to one based on comparing differing outcomes given the same treatment?
Wesley: Please send a link when your post is done!
It is nice to see critique based on an actual read of the paper! (No offense to those who raised questions from the press report .... but well, it was a press report.)
#19 & #21 beat me to it: many gun-carriers are at increased risk to begin with.
The over-riding concern is that, if you are shot at for some reason, carrying a firearm is the most effective means of enabling yourself to shoot back!
But remember, safety first.
Rt
Let me rephrase my prior snark so it is less snarky and may be more correct.
Prior to this study, the presumption was that carrying a gun was safe, and carrying a gun was the risk-reducing thing to do.
If someone were to say "If you carry a gun around in Philadelphia, you've got an increased chance of getting shot" no gun advocate would expect that or go along with it. In other words, the pro-gun presumption, the null hypothesis if you will, is that carrying around the gun reduces your chance of getting shot, not increases it.
This study may be flawed, and it may be that the control sucks, but given that the null hypothesis would be that packing reduces rather than increases the risk of getting shot, this study's results, even if imperfect, are counter-intuitive.
Now, it may be said, and it is being implied here and there, that the situation among these test subjects is so dangerous that the fact that they are carrying a gun is not the relevant factor in getting shot.
Maybe so, but that statement, the assertion that the baseline is more dangerous and thus the increased risk of being shot, comes from where? So far, it comes form people commenting on a blog post. It is a presumption sans data, a conclusions sans control, a veritable winged monkey flying out .....
... oh, crap, here comes those winged monkeys again. ...
Based on this paper I hope we immediately disarm our soldiers on patrol in Iraq and Afghanistan. Troop deaths should then rapidly decline according to the very wise and insightful Mr. Laden and friends.
DuWayne, you are wildly inconsistent. In comment #15 you say you can out-shoot your cousin, a master sharpshooter in the Marines. I have no reason to doubt that. In comment #23 you said, [And yes, asshole, my "purported" expertise is the norm].
You seem to be saying that the norm for the average gun owning citizen is to be a better shot than a sharpshooter Marine, your cousin. The only way this could possibly be true is if your cousin has that title yet is no longer worthy of it.
For the record, I (1) don't own any guns; (2) want the result of the paper to be true yet reserve judgment until hearing more of how they concluded causation vs correlation; (3) have two grandfathers, two uncles, and one one brother-in-law who were career policemen; (4) believe the US would be a better place if rifles were legal and handguns weren't (except for police/military).
of course jimmy. because U.S. streets are exactly like a war zone.
*facepalm*
"DuWayne, you are wildly inconsistent. In comment #15 you say you can out-shoot your cousin, a master sharpshooter in the Marines. I have no reason to doubt that. In comment #23 you said, [And yes, asshole, my "purported" expertise is the norm].
You seem to be saying that the norm for the average gun owning citizen is to be a better shot than a sharpshooter Marine, your cousin."
Actually, I thought DuWayne was quite clear. His expertise in firearm safety is the norm (you seem to have left off the second part of the quote "my 'purported' expertise is the norm - at least where it counts" [emphasis added]).
His accuracy on the range is likely not the norm, but that is also not the type of training people worry about in letting civilians use firearms. The type of training that cops receive that is available and common to civilian gun owners is basic safety and control issues, not marksmanship.
Frankly, I am neither surprised nor worried that cops are unable to outshoot a civilian enthusiast...I'd much prefer they have a skill set that includes all the ways to defuse a situation before they need to fire a weapon. In fact, I believe I heard the statistic that most police officers have never fired their weapon on the job and, particularly in urban environments, I think that is best. A serious error in accuracy on the range has nowhere near the consequence of a minor error in a densely populated region.
sailor @ 11:
It is still like that in the UK (thankfully).
And remember, kids: Correlation really is causation.
To pick up another fallacy in #15. All the range training in the world doesn't prepare you to deal with real world situations.
You should not assume that just because you have been trained in how to carry a gun safely and hit a target under controlled conditions that you have the skills to carry a gun in public where you must be prepared to handle the situation safely in panicky conditions when you feel under threat.
Said from the relative safety of the UK where thankfully all hand guns are banned.
Jacob writes:
Let me suggest a different "e.g." When I lived in a high-crime area, I never knew anyone who carried a gun because of that. I suspect a more relevant example would be: "those involved in crime."
No Jim B., I am not saying that the norm is for people to be able to be capable of shooting as accurately as I do. I am saying that safety and control is the norm, not my marksmanship. The entire point I have been making, was a direct response to the quote in my original comment, that seemed to make cops out to be some sort of magical beings with magical training that is "more than any civilian can possibly get..." An attitude that seemed strongly reflected by other commenters and an attitude that is directly responsible for cops getting away with some of the most egregious offenses.
To try to make this as clear as possible, marksmanship does not equal expertise. There is only so much of a difference that can be made between how well one shoots the first time they pick up a gun and after they have spent a great deal of time practicing. There are of course exceptions, but in most cases, someone who fails to fire in a tight pattern that first time, is never going to become a sniper no matter how much they practice and train. I mean below a certain level of accuracy, no one should be carrying a gun, but excepting that, it really doesn't matter.
What counts is control and safety - knowing when and when not to use whatever level of force. The other important factor is knowing how they will react under the stress of actual dangerous situations. The fact that general police training doesn't include tactical scenarios is incredibly frightening to me. While they aren't perfect - you still know you aren't actually going to get shot - they are pretty damned intense and can totally make the difference when it comes to shooting the "right" person and basic survival.
Dominich -
That is exactly my fucking point.
"Prior to this study, the presumption was that carrying a gun was safe, and carrying a gun was the risk-reducing thing to do." -- Greg Laden
I know a lot of pro-gun enthusiasts and I have never heard any of them that merely carrying a gun reduced any risk whatsoever.
What it does is possibly increase the chances of a better outcome in a bad situation. It certainly doesn't prevent bad situations from arising. Any suggestion that it does is ludicrous. (I would get an argument from some about that, I'm sure.)
Welsley R. Elsberry:
Curiously, I was just reading about another politically controversial claim based on case-control studies: that abortion increases the risk of breast cancer. Most researchers seem to discount these studies on that issue, because for a variety of reasons it's hard to know that the control data really is comparable. On breast cancer, we have cohort studies where the alleged correlation disappears.
I guess I should add that not only have I never carried a gun, I also have never had an abortion. ;-)
Donna:
How does "carrying a gun [reduce] risk" not equal, roughly, "increase the chances of a better outcome in a bad situation." unless, perhaps, a "better outcome" is merely "noisier carnage" or "more blood, splattered more places" or something along those lines?
"It certainly doesn't prevent bad situations from arising. " Now, that is a logical statement that I happen to agree with, but I was talking not about what I think or what you think or what the averge gun owner that you or I know thinks, but about the gun nuts. Who very much DO claim that carrying guns mitigates against bad situations.
They say, explicitly, repeatedly, and clearly, that if every teacher was suspected of carrying (which they claim would be the case if many did carry) that Columbine-type events would not occur because the potential perps would think twice, or three times, before pulling out their weapons. And so on.
Again and again gun nuts (and once more, I'm not referring here to all gun owners by any means) claim that if more people carried guns, not only would there be fewer .. not more, FEWER .... people shot, but that the difference in numbers of people shot would be enough to make up for the carnage that occurs from accidental and suicide related shooting.
Once gain, the NRAists and gun nuts, who are not rare, not unusual, not fringe among gun owners, but who seem to be in the majority of voices in support of more or less unregulated gun ownership, claim that more people owing and carrying around more guns = less bad stuff happens with guns. Not more, not less of more not the same, but less, and enough less that the accidental death rate is compensated for. That would be a lot less. Given that, this study should have had very different results.
Instead, this study seems to show that there is a positive, not negative, not neutral, not less positive than previously thought, but positive correlation between gun carrying and getting shot.
The one counter argument (besides the study being flawed, which it may be but which requires looking at the study, not the press release, to determine) that has not arrived yet, but likely will, is that those are different people. The gun nuts will eventually arrive here to tell us that THEY (the gun nuts) can possess and carry safely and to the betterment of all. It's these other guys who can't or should not.
Greg wrote:
Why would it be risk-reducing? What are gun carriers going to do that makes them safer than non-gun carriers? They can't shoot incoming bullets out of the air. If threatened by The Bad Guy With The Bad Gun, they can't do a gunslinger fast draw and shoot first - they're more likely to get shot or stabbed because they unnerved The Bad Guy.
It's very difficult to see under what circumstances the results of firing a gun come out better than not firing it.
Anyway, you Americans love your bloody slaughter of your families and friends, about 10x higher than any other comparable country, isn't it? Safe, wheeeoooo, that's not even on the horizon.
Sam C ... you understand, yes, that in the quote you use I'm referring to a position that I strenuously object to because it is, as you point out, utterly stupid?
... you understand, yes, that in the quote you use I'm referring to a position that I strenuously object to because it is, as you point out, utterly stupid?
AHAHAHAHAHAHAHAHA!!!!!!!!! Why would he think that Greg? A great many people in this discussion seem to be ignoring the bulk of any comments they are responding to and assuming the worse about the person making them.
Greg Laden:
They're not rare. But they are fringe. Even among gun owners. Gun ownership is very common here in the US. Gallup surveys over the years say between 40% and 50% of US households have a gun. The NRAists is that small slice for whom guns constitute a political cause. Someone might argue, "well, if they don't really represent gun owners, then gun owners should form a different political group." The problem is that gun owners generally don't really constitute a well-defined political grouping. It would be pretty much like expecting boat owners to have political cohesion. What has happened is that the right wing has done a great job of leveraging the political issues around guns.
BTW, dominich, if you visit the US, you have a much greater chance of being killed by a car than by a gun. Not that you should excessively worry about that, either. We actually get most foreign visitors back to their homes safe and sound.
Yes, thanks to this study, we now know that guns are made of a startling new substance that attracts and stimulates deranged gunmen. How else to explain a 450% increase in being shot - simply by carrying a concealed weapon, which nobody can see or smell or hear?
Or perhaps it is a matter of pheromones between one gun to another, or even one gun owner to another? What an exciting scientific discovery that would be!
I live in Vermont. Vermont has allowed concealed-carry handgun laws, which require no permit of any kind, for at least 50 years. I believe that this policy has never been changed during the entire history of the state. We are among the safest of states, with a very low crime rate, homicide rate, and gun-related crime rate.
The idea that carrying a gun increases one's chance of being shot does not hold true here. But then again, we don't have a lot of drug dealers plying their trade on our streets.
Urban Philadelphia, on the other hand, can not make such a statement. As I understand it, drug-related crimes account for a very large majority of gun-related crime in this country. If you want to get shot in the US, simply go down to where the kind folks who deal hard drugs work, say the wrong thing, and you would experience a likelihood of being shot that would make the study author's 450% increase look like small potatoes.
This study evidently - I don't have a subscription - compared a population of people who had already been shot and therefore likely to be involved with a drug-related crime with a group of 'controls' who had been .... what? Shot? Not shot? Anybody know what the control group was? Sigh.
Here is the objective of the study:
Based on what we know about drug-related gun crimes, if there was not an increased likelihood that carrying a gun resulted in being shot, then I would be surprised. Most gun-related crime is drug/gang related. While I don't have the study, I echo the sentiments of others here who wonder about validity of the case-control design.
This study really says no such thing - it doesn't examine anything but a situation in urban Philadelphia. The experience in Vermont (and I daresay almost anywhere in the US outside of inner-city urban areas) belies the rather absurd idea that carrying a concealed weapon - in and of itself - increases one's risk of being shot.
Unless, of course, one shoots themselves in the foot - which I think you have done here, Greg.
According to the linked report on the study, there were 677 cases of assault studied. Of those, 6% were "in possession" of a gun (ie, on their person or in their vehicle). 6% of 677 is about 41 people. From this, they make the leap that you're 4 to 6 times more likely to be shot. This seems a bit of a stretch.
When doing any kind research about a social or societal issue, you can't rely on anything as small as 40 individuals. Even just 677 is a bit small for something this wide spread. As well, all individuals were from the same municipality (not to mention the same state, region, country and hemisphere!). It seems to me that it would be somewhat unreliable to use so few cases to make so broad a judgment.
I would very much like to see more research on the topic, and especially research in geographically diverse areas - what about rural areas, for example? Or only upper or lower class areas?
For what it's worth, i suspect what this study actually shows is that in Philadelphia, it's likely that if you're in possession of a gun, you're living the type of life where you're apt to get shot. It's interesting to note that in the study period, only 6 police died from gunfire - but that makes up *15 percent* of the people mentioned in the study of being in possession of a firearm at the time of death.
The Moiety: How have I shot myself in the foot by posting a study for discussion?
Having grown up in sight of Vermont, and lived in some of the most urban areas on the planet, I both understand and would like to correct your (mostly flying monkey) critique. First, you are comparing information you introduce as evidence without any proof or citation. As reasonable as it sounds, you are comparing a winged monkey statement about how gun violence works in vermont with another winged monkey statement about what you think happens in Philidelphia.
Second, yes, there probably is a difference but so what? More people live in Philadelphia than the entire state of Vermont. This study is potentially important even though it might not directly affect you.
Third, the "us" vs. "them" = "It can't happen here" vs. "Let them all die because they're bad anyway" is not very Vermont like. Shame on you.
Regarding the control group: I would like to point out that I've never liked the use of the term "control" for behavioral field studies, as it is generally used. Most of the time, such studies are not "treatment" vs. "control" experiments. They are comparative behavioral experiments To the extent that there are controls, they are best developed in very different ways than is usually done for an experiment. In behavioral field studies, the control group is often a group about which less is known than the focal study group. It should be the opposite!
But once again, I'll state this: this is a study. It is not a winged monkey of self serving opinion. It might be a flawed study, but there is a difference between collecting data and describing it and making statements with no visible means of support, no matter how logical they may seem. If this study has a valid point to make, it will be in relation to its counter-intuitiveness. Thus, looking at this study and saying "Oh, this is not what I thought must be true, therefore it is not true" is ... OMG, here comes the winged monkeys again ...
"Prior to this study, the presumption was that carrying a gun was safe, and carrying a gun was the risk-reducing thing to do." -- Greg Laden
Who exactly are you attributing this presumption to? Do you have any concrete examples, or just more frantic hand-waving and vague blathering about unspecified 'gun-nuts?'
Rt
DuWayne --
Sorry for missing your point. In one sentence you mention your safety training and in the next sentence how great of a shot you are. Since your point was the amount of safety training you've had vs your average cop, introducing the matter of your shooting abilities muddled your message.
Roadtripper, try this comment thread. Plenty of arguments for the deterence value of gun possession.
http://scienceblogs.com/gregladen/2009/08/ok_time_to_ban_handguns_are_y…
Jim, it's an easy mistake to make. Hang out in enough of these conversations, though, and you'll realize that if you don't say up front that you're comfortable handling guns, people assume you're an effete wuss who's afraid of blued steel.
Road Tripper I have a citation for you:
Go here: http://scienceblogs.com/gregladen/politics/gun_ownership/
Read all of the comments.
Are you actually suggesting that there are few or no pro gun people that think that widespread gun ownership and carrying guns does not make things safer?
Five seconds on the internet gave me these two quick examples:
source
source
And, of course, you can do your own research as well. Click here.
BTW, dominich, if you visit the US, you have a much greater chance of being killed by a car than by a gun.
Not true, Russell:
Car related deaths = 43,664 (2006)
Gun related deaths = 29,573 (2001)
http://www.cdc.gov/nchs/FASTATS/acc-inj.htm
http://www.ojp.usdoj.gov/bjs/glance/tables/frmdth.htm
D.
Daen: That looks backwards from what you said.
Anyway, a hand gun, used properly, vs. a car, used properly:
Car used properly sometimes kills you anyway.
Hand gun used properly (key word here = used!) will at least wound you if not kill you every time.
Daen: That looks backwards from what you said.
It is a greater chance, but not much greater.
Another way of looking at it is if the only two ways of karking it were car or gun, then you'd have a 3/7 chance of dying by gun vs a 4/7 chance of dying by car.
Not such a big difference really.
Another thing: If somebody pulls their gun out, you have a greater chance of getting killed from that than if somebody pulls their car out. Depending on where you are standing, of course.
I am old, so it may not be that way now.
Thankfully, not yet.
Gun deaths in UK in 2009 = 50 (1 in 1,200,000)
Gun deaths in US in 2001 approx 30,000 (1 in 10,000)
http://www.independent.co.uk/news/uk/crime/britain-records-18-fall-in-g…
... people with a gun were 4.5 times more likely to be shot in an assault than those not possessing a gun.
Am I misreading this, or is this not a study comparing armed people being assaulted with unarmed people being assaulted?
If my reading of the summary is correct, then probabilities are >50% that my reading that many of the commenters above are blowing smoke (&/or winged monkeys) out of their asses is also correct.
Otoh...
... the researchers do seem to have used epidemiological procedures in a sociological context, so possibly all the smelly smoke and aviasimians around here originated elsewhere.
Jim B. -
I got very, very tired of people either accusing me of being a gun toting fascist, or being a commie bastard out to steal everybody's guns. I am an ardent supporter of the right to own guns and also an ardent supporter of fairly strict gun control laws. As such, I have found that it helps to be clear, as Stephanie put it, that I am not an effete wuss, who is afraid of things which make a loud bang...
Too, I wanted to make it clear that there is nothing magical about the police in regards to firing guns. While my marksmanship is above average, so is that of slightly less than half of people who handle firearms - whether they are cops or not.
Bottom line, all I have been doing is making the point that cops are not endowed with superior abilities or special training that is beyond your average gun owner. Believe me when I say that I wish it were so, but they simply are not. The problem with people believing they are, is that this makes them seem special or safe, where they are not. It puts them on a pedestal they certainly do not deserve and it makes it easier to excuse them, when their incompetence with firearms leads to unnecessary violence - sometimes death.
Greg:
First, for example, let's say the bad situation is a home invasion. A good outcome to that would be the homeowners tackling, subduing and disarming the criminals, then dialing 911. A bad outcome is the homeowners are killed and the criminals get away with whatever loot they wanted.
A less bad situation is that the homeowners are armed and manage to wound or even kill one or all the criminals. This is less bad even if the homeowners are injured (IMHO, at least.)
The way I see it is if a criminal is intent on doing harm to someone, then the victim should have the right (whether exercised or not) to defend himself. I'm well aware that some will say having a gun in such a situation doesn't mean you'll get a chance to use it in your defense and that it may even end up being used against you.
Second, at the risk of sounding like I'm contradicting myself, I think the "gun nuts" have a point about school shootings. It seems rather silly to make it well-known that NO ONE in whatever place will be allowed to possess a weapon. It's equivalent to a homeowner putting a sign at the end of the driveway saying "There's no alarm system and no dog on the premises."
My agreement is more about the "feel-good" symbolism of "gun free zones" not a desire to see every teacher armed.
I won't make the argument that "those are different people" but I will make the argument that gun regulations that work in one place won't necessarily work in another.
Open carry is an example of that. I know (but don't quite understand why) that the mere sight of a gun not carried by a cop scares some people. It's obvious open carry wouldn't work too well in NYC, but it wouldn't be a problem at all in rural Arizona.
Because of that, I have a problem with one-size-fits-all regulations coming from Washington DC.
I think the study fails to differentiate between lawful gun owners and criminals that carry guns. How many of the people carrying guns had them illegally and were killed because of their ties to gangs and violent crime? On the other hand, how many were lawfully armed citizens acting in self-defense? Given that there are many active gangs in Philadelphia (http://en.wikipedia.org/wiki/Category:Gangs_in_Philadelphia,_Pennsylvan…), and that Philadelphia has violent crime statistics higher than the national average (http://www.cityrating.com/citycrime.asp?city=Philadelphia&state=PA), I think it's an important distiction to make. Of course gang members and criminals are more likely to be armed illegally, and are also more likely to be killed violently. Meanwhile, it's possible that responsible, legal gun ownership actually does make one safer, but the gangsters are driving the average up.
I think the study was poorly done, and it isn't definitive enough to sway my opinion that responsible gun ownership makes one safer. Later research may compel me to change my mind, but this simply isn't good enough due to the presence of what is, in my opinion, a huge confounder.
David, you've read the full study?
DuWayne #15:
You have completely missed my point. I have no doubt you have excellent training in firearm safety, and maybe you are an even better shot than a police sniper, but that is irrelevant.
Do you have training to deal with violent crime? Have you been taught to judge whether or not a stranger reaching into a coat pocket is going for a weapon or a cell phone? Have you had training in conflict escalation, in resolving confrontations without resort to weapons, in deciding at what point it becomes necessary to draw the weapon, and at what point is becomes necessary to fire the weapon? Have you undergone training in targeting and firing your weapon at other human beings? Have you done simulation exercises where other people are firing back at you? With unarmed civilians in the cross-fire?
Do you apply all this training every day in your job, continually practicing and honing your judgment? Are your decisions and actions continuously reviewed by your superiors and evaluated?
If anyone is talking bullshit here, it is you.
Russel #45 (and Daen as well)
Oh dear, way to not to a risk assessment. Try relating the number of deaths to the numbers of instances of use.
I have had the good fortune (if thats the word) to have encountered US homeland security and immigration a number of times and it was bad enough even before 2001 but it has become an onerous enough process enough since to make one think twice before making even a private trip to the land of the "free".
Donna, have you killed anyone? Probably not, me neither. I know a lot of people who have and it's generally pretty devastating. Some people actually prefer they died rather than killed. I'm not sure if after the fact everyone would agree with your categories. A lot of people would prefer to put several of those outcomes into the "Unthinkable" category and do all that can be done to avoid them. Like a general disarmament of the population.
My neighbor killed a 14 year old boy last year. The boy was in my neighbor's house around 10 or 11 PM on a dare, thinking it was an abandoned home (his 'friends' told him that). My neighbor was in the habit of sleeping with his shotgun loaded next to the bed. The boy was killed with one shot, even though my neighbor was aiming at a shadow that he knew was a person (but not whom). Pretty good shooting for an old guy who didn't even bother putting on his glasses.
I guess that was a good outcome because nothing was taken and the home invader was killed. And that was a good example of a successful use of a gun, the gun owner did not kill or harm himself or anyone who was not the home invader.
Some of my other neighbors were very happy. They said "if you break into someone's house, and you get killed, even if you broke in only as a joke, you were asking for it. This will reduce the chance of home invasions happening again in our neighborhood" ... or at least, the ones interviewed for the TV said that, or words to that effect.
David[61] I think the study fails to differentiate between lawful gun owners and criminals that carry guns.
Do you know that to be an aspect of the study? Or is this a winged monkey 'factoid' that you are using to criticize the study? (I'm not saying it is not a factor, I'm just asking, and asking how you know.)
Is it OK for someone to die because of their ties to gangs? You do know, right, that such a determination writes off a huge chunk of the youth of our nation who live in a situation where they don't have uch choice.
It amazes me that only the report of "findings" is here. Victims hmm - he was dealing drugs and in possession of a gun, a rival gang member shot him. Conclusion Possessing a Gun doesn't protect ? Hmmm Some people will buy into anything with out checking the facts. No wonder our country is impoverished and in 12 trillion of Debt, soon to be 24 trillion. Is it any wonder the Nigerians write spam mail here - for all of us; the incredibly Gullable.
Katherine - having a Gun in your house makes you more likely to get shot ? SHOT by WHO ? Would you like to know - the answer to that - not by the criminals coming to rob you. But by the People in your own home, the people who freak out when they hear a noise, the people who are itching to yank that thing out, by the kids who had access to it because it was not properly locked away. By the people who refuse to teach their kids about Guns - THAT THEY ARE DANGEROUS. NOT TOYS. By people who do not know how to deal with their own angers and faults and shortcomings. But then again Anger , Rage, Hatred, we do not do enough to teach people to resist those things. BUT since our society loves the criminals and imprisons the victims, people keep Guns. Guns do not Kill - PEOPLE DO. If they don't have Guns - Criminals will have them, that includes the Government. A Government finds itself in a very difficult situation when everyone in the nation potentially has a gun! Why do you think it is in the Constitution ? We have not evolved - just look at our streets..and look at our Government today.
This is exactly the issue with guns, right there. Someone breaks into your house, and well, time to kill them! Let's bring deadly force in. I understand protecting oneself and your family. But in a case like an intruder, you're better off calling 911 and trying to remain safe. Going out and confronting a intruder is just a bad idea. You pull your piece, and guess what? The intruder probably does too. Now your are in a confrontation. Wait safely, call 911 and let the authorities get there. And if they get away? So what. Possessions can be replaced. Life can't.
Now I understand the whole "what if they come to get us" type mind set. Luckily most intruders are not there to physically harm, they are there to steal. Killing someone is a much bigger deal than breaking and entering. Notice that you often hear of shootings (non gang related) as robberies gone wrong. What do you think happened? Confrontation.
I understand sporting riffles. I'm all down with hunting. I can not for the life of me understand the need for the civilian public to need to have a concealed handgun.
-JJ
I looked over the methodology of the study and they did not do anything to distinguish between those who are legally carrying a firearm versus those who were unlawfully carrying.
This study was done in Philadelphia. A study which, in spite of many good things going for it, has a huge problem with gang related violence. Gang members tend to assault one another quite often and I think that we can all agree that they are also more likely to shoot one another than the general population is.
I would wager that many of these people who were carrying firearms when they were shot were gang members who illegally possessed the weapon. A single factor, which is being a member of a gang, makes an individual both more likely to be carrying a firearm and more likely to be shot in the course of an assault.
This increased risk factor of having rival gang members want to murder you over crack does not typically apply to a law-abiding citizen who applies for a concealed weapon permit, takes the formal training classes and purchases a weapon after a thorough background check.
Therefore I believe it is irrational to declare the lawful possession of a firearm to be especially risky based on this particular study.
Greg (#65) -- strawman argument maybe? I didn't think the discussion was about dealing with killing someone.
However, I still think a dead bad guy is a slightly better outcome than a dead good guy.
Your anecdote is actually very tragic where both involved were basically good guys. It's a very different situation. But that said, I think the shooter made a mistake in shooting at a shadow.
I find it very odd that you know a lot of people who have killed someone. My husband and I just had a hard time coming up with one person that we knew who had killed someone and that guy was a Marine sniper in Vietnam.
Maybe we just live sheltered lives.
By titling the post as "Carry a gun = you get shot more often" when one can not draw such a conclusion from the study.
And by demanding citations from posters who disagree with your position when it appears that you yourself have not read the study nor have provided any citations to rebut.
I simply can not take the time to hunt down all the statistics. It would have been nice if you had done it to accompany the report, though, since your assertion about what it says (and what it seems to conclude)is patently ridiculous at face value.
I also think that it is going to be extremely difficult, if not impossible, to find data in order to answer the questions we reasonably can pose at this time. It doesn't look like gun-related assaults, shootings, homicides and drug use can be satisfactorily teased apart, as sufficient data appears to be absent, or impossible to gather. Non reporting issues, for example.
Nevertheless, I did spend a little time, and here is a report from the DOJ looking at drug-related gun crime which, I think bolsters my argument about inner city gun crime being substantially related to drug use:
http://www.ojp.usdoj.gov/bjs/pub/pdf/drrc.pdf
But unfortunately this report is not detailed enough.
Please trust me on the Vermont gun data - we read it all the time up here, so, like the magazine article about the study, it must be accurate!
I see your point about Vermont being smaller than Philadelphia, but at least the data for this sort of thing is not expressed as absolute numbers but rather is relational.
What makes you think I care that the study doesn't affect me? I only care that the study doesn't make much sense, and I am frustrated to not see the study design and Methods sections. And the fact that you are making conclusions based on something on what the apparently lousy study doesn't support. To further your own philosophical leanings whilst denigrating those on the other side of the issue as having "winged-monkey" assertions. You haven't read the study!
WTF are you talking about??? I didn't say or imply anything of the sort! Have you been talking to those winged monkeys you see so often? :D
Here is the crux of the problem! The idea that simply having a gun in one's pocket all in itself - as opposed to an ipod or a phone in your shoe(?) - makes one 4.5 times more likely to be shot is completely absurd. And evidence-based conclusions must take into account whether the study data a/or conclusions are believable in relation to common experience of the world, not to mention the body of data that exists previously. And this doesn't pass the smell test.
The onus is on the authors to provide remarkably robust data, and on those who care to make sweeping statements about gun safety, and disparaging remarks to those who are simply properly skeptical of the study, to have done their homework as well.
amphiox -
Most of what you list, I could get trained in, for very little cost. Gun clubs all over the country, as well as some police auxiliary and other community organizations offer such training for anywhere from free, to a few hundred dollars. And If I wanted the whole shebang I could spend a lot of money and go through one of a myriad training centers that offer tactical assault and urban combat training. Those centers mostly train private security contractor employees, but some of them are specifically oriented to (usually wealthy) civilian adventurer types.
Bottom line, I haven't, but easily could get the sort of training you list. I could probably get a substantial discount (base price $180) on the training course offered at the gun club around the corner, because the guy who teaches the course (same curriculum wannabe cops get locally) is a friend. He was my hunter safety instructor and was on the very first hunt I ever went on, when I was twelve. He also teaches wannabe cops at my school and we had lunch fairly regularly last winter semester. And for a few hundred bucks or less, you could get the same sort of training and probably not all that far from where you live.
Have you undergone training in targeting and firing your weapon at other human beings? Have you done simulation exercises where other people are firing back at you? With unarmed civilians in the cross-fire?
First, with the exception of the FBI, possibly other federal law enforcement, some state police agencies and a minuscule number of municipal forces, regular cops don't get any of that kind of training. Hell, few enough SWAT officers get that. Municipal law enforcement pretty much just have to qualify at the range. SWAT usually have a little more intensive training and qualifications, but not generally as much as you are describing.
And for the record, I have gone through a tactical range - though while I didn't get killed (i.e. I got all the "bad" guys), I did shoot a hostage in the head. But that was one time through - not great, but actually not that bad. If you lived in the Salem/Portland, OR area, you could too - for $350 a run - three grand for ten. I did a bunch of work for the guy who owned it and was a little bit cocky about my ability with firearms, so he thought it would be interesting to see how I would do.
You have a way inflated belief in what cops are actually trained for. It isn't fucking magical and they aren't either.
me, Myself and I @ # 68: A Government finds itself in a very difficult situation when everyone in the nation potentially has a gun!
Did you know that just about every household in Saddam Hussein's Iraq owned one or more guns - and that he handed them out like car-wash coupons when a foreign invasion loomed?
By many accounts, Qaddafi's Libya is the same way, maybe more so.
Those who claim widespread distribution of small arms is an effective defense against tyranny are so naive, it's almost cute.
#69, jj -- my example was a home invasion, not an intruder. Home invasion implies the use of force and assault coupled with the awareness that someone was at home at the time.
Daen, the majority of those deaths are suicides. In 2006, there were 10,177 murders by firearm.
So, I think it's a bit of a leap to conclude based on this study that carrying a gun causes you to more likely be shot.
It's not that much of a leap: if you have a gun, and count on it (or think you can count on it) to make you safer, then you will be more likely to pull it, or threaten to pull it, when you feel threatened; and your adversaries will then be more likely to pull theirs in response. And such things are especially likely to happen when the gun-toting civilian in question is letting fear, hate and gun-fetishizing ideology trump sound tactical thinking, as too many seem to be doing here.
Meanwhile, it's possible that responsible, legal gun ownership actually does make one safer, but the gangsters are driving the average up.
You really think guns make individuals safer against ARMED CRIMINAL GANGS, like those you speak of in PHILADELPHIA? Are you fucking kidding me?
@Greg:
Greg, you made an argument using a study YOU didn't read, YOU just cited the abstract and the press release. I wasn't too lazy to read the paper - IT's FIREWALLED. You read those items uncritically, some of us did not. To me, the details of the methodology that are available at the links YOU provided suggest that the authors may be conflating several risks with gun ownership in their analysis. Without having access to the original paper, it's certainly as valid a conclusion as your own.
Stephanie Z #62. I didn't have time to read the whole study prior to posting due to time, so I just read the linked article and the abstract to the study. I have now read the whole study and stand by my opinion. The study did control for prior arrests, neighborhood income, and history of drug trafficking, but it did not say how many of the victim's guns were legally acquired and how many were not. This would have been relatively easy to do. From the study: "We coded control participants as in possession if they reported any guns in a holster they were wearing, in a pocket or waistband, in a nearby vehicle, or in another place, quickly available and ready to fire at the time of their matched caseâs shooting." In Philadelphia, a combination of state and local law dictates that all guns of that type must be licensed with the state (http://www.pafoa.org/law)(http://www.pafoa.org/law/carrying-firearms/open-carry). The only exceptions to that law are if the victims were at home or at work. Few employers allow their employees to be armed, so we can pretty much discount this. Also, given that the vast majority (83%) of victims were outside, we can assume that very few were at home. Therefore, at least 83% of the victims' guns could have been checked to see if they were carrying legally.
I hypothesize that relatively few of the victims were carrying legally, which would make them less likely to follow other laws, more likely to engage in violence, less likely to train themselves at a shooting range, and less likely to seek professional training. Because of these differences, I believe that statistics pertaining to criminal gun carriers are not applicable to legal gun carriers. Criminal status of carrying is, in my opinion, too big a confounder and failure to address this in the study makes it fail to convince me.
Greg #66. When did I say that I was writing off the people that carry guns illegally or that they deserve to die? I was saying that the subject population was actually made of two very distinct populations. I hope that the explanation earlier in this post will show you that I didn't just make this objection up.
David: I'm not sure how the researchers would be able to determine the legality of the guns so easily. Technically it might be OK but asking questions that lead a research to understand a person's criminal or legal status raises ethical questions and would almost certainly be left out of a survey for IRB related reasons.
Is it OK for someone to die because of their ties to gangs? You do know, right, that such a determination writes off a huge chunk of the youth of our nation who live in a situation where they don't have uch choice.
It is unfortunate, but that is not the point. The point is that it makes the study rather disingenuous. The implication of the study - at least according to the article (and it seems this is how someone involved in the study plays it), *I* am considerably more likely to be shot, if I have a conceal and carry permit and walk around armed. If, as David makes a reasonable argument for (I haven't read the paper - would be happy to if it just happened to fall into my inbox - hint to Greg's readers with access), they did not factor those legally carrying, then as far as implications for legal gun owners, this study is useless.
To be clear, I already assume that even carrying legally increases the odds of being shot. I just have a problem with citing this article as evidence of that, if illicit firearms were included. And honestly, if illicit firearms were included, it is really a no brainer. Because of course the odds of being shot increase, if one is involved in illicit activities that require they carry a gun. I am not exactly inclined to judge this without reading the study myself, but if David is correct this study is pretty much fucking useless. I certainly won't be citing it until I have - and unless David's rationale seems to be way off the mark after I read the paper, I am pretty sure I won't be then.
Technically it might be OK but asking questions that lead a research to understand a person's criminal or legal status raises ethical questions and would almost certainly be left out of a survey for IRB related reasons.
Why exactly? There are studies conducted about illicit drug use all the time that ask questions about similar legal status issues. I am really not sure what would make this all that different. It isn't terribly complicated to anonymize survey calls in a call center. After the initial contact is made by name, the call if flipped to someone else in the center (or even another center) without any identifying information connected to the results.
The problem here, is that without information regarding the legality of the firearms in question, the study is bloody well useless. Especially given it was conducted in an especially high crime community.
Why would past history of arrest not be a good proxy for the (il)legality of the guns, particularly in an area with gang activity? You know who the cops are going to come down on for any infraction, and you know who's going to have difficulty getting the guns legally.
DuWayne, it is extra hard to get IRB approval where you are delving into individual's illegalities. We can theorize all we want about how it can be done easily, but every additional item on the checklist of IRB review adds significant additional requirements for process, documentation, and possibly steps in review, and of course time. For each research project the researchers have to make decisions about where to stop, post-study quarterbacking about what might have been easy to do notwithstanding.
Greg -
I understand that and sympathize, but not having that information makes the study rather useless.
And I should have been clear that "easy" is a relative term. I have been delving into the world of IRB's through bloggers who regularly have to deal with them and understand just how difficult they are. They are totally big fun when you are doing psychological research.
As for post study "quarterbacking," is it unreasonable to point out significant - even fatal flaws? This is not an insignificant factor we're talking about. The legal status of the weapons in question make the difference between this study being of any use whatever, in determining whether or not it is a useful in determining whether or not carrying a legal firearm puts one at significant risk. Nor do I think it is unreasonable to put out ideas for determining the legal status, should someone want to do a similar study in the future.
I happen to think that a reasonably produced study that deal with that factor would be useful. I suspect rather strongly that these folks have a reasonable hypothesis. But the gun nuts are going to tear into any flaw they can find and this is a fucking huge gaping wound, before they get their paws on it. Leave no room - or as little room as possible for error. It is never going to change the minds of the gun nuts, but it may well change the general thrust of public opinion.
In that vein, another method would have been to make initial contact by phone, in an attempt to garner cooperation and then send the subjects a scantron survey in the mail.
Stephanie -
The problem is not determining legal status of a known participant - it would be possible to just run public records of licenses. The problem is that it is unethical to do so. If asking is going to be considered unethical, going behind their backs is far worse. It is breaking their trust - keeping in mind that researchers are legally required to debrief participants afterward; and it is potentially damaging to the participant, should that information get out. And yes, in this context getting arrested for being a criminal is harming a subject. At the same time, having information about criminal activity is also an ethical conundrum.
To give you an idea of just how difficult IRBs can be, I have an instructor who is doing research into depression. He had to revise the methods that dealt with questions about suicide 17 times. First, because they were concerned that the mere questions might cause someone to do it. Then there was concerns about making sure that information was kept confidential. Finally, there were concerns about it being too confidential, in the event that someone indicates a strong likelihood of committing suicide. I would note that the committee came down heavier on the side of confidentiality, than they did on making it easy to identify participants who were at significant risk of suicide. Basically, the identifying information was put in the hands of an independent committee and any request for identities would have to be run by them, while making sure that the coded files were kept closed, so they wouldn't have access to the actual identities - only the PI would be allowed to have that information.
That is just a relatively minor example of how difficult it is to pass things by these boards. In the social sciences, it is virtually impossible to get research involving human subjects passed on the first five runs - usually it is considerably more (this varies some from institution to institution - board to board). The one at my instructor's school isn't all that difficult, compared to others (in part because it has to pass the department chair, before it goes to the review board).
Table 1 in the paper has simple numbers for a variety of conditions that they took note of or adjusted for. I'm working on calculating the odds ratio/estimated relative risk for all the conditions based on the raw numbers. This will be useful to compare to the relative risk numbers they published based on the "adjusted" data, plus it should be informative about what sort of numbers you get out of case-control methods in general.
Unfortunately, their description of the process of adjusting the numbers is opaque -- without their particular dataset, model parameters, and conditional logistic regression algorithm, that part of what they did is still a black box. I hope to be able to show the extent of "adjustment" from the raw numbers, though, by assuming the adjustment is done to the control group numbers alone.
This takes, needless to say, some work. I'll leave another comment when I finish off my post.
DuWayne, the researchers (according to David; I don't have access) already controlled for prior arrests. That's the data to which I was referring.
DuWayne, you are only reinforcing my point. Even if such training is available, it is not typical for civilian gun owners to have it. You yourself do not have it. Even if you had it, you are not going to be applying that training as a routine activity, day in and day out. You do not go to work every day knowing that you may be called upon to use that training, as part of the routine performance of your professional duties (unless your profession is one in which you are required to shoot other human beings).
I do not have an inflated opinion of police training. I am well aware of how atrocious it can be (and the answer to that is to train the police better, not to make snide remarks about taking away their guns). But no matter how little training or however poor quality it is, all police get it - they are required to get it, and they are required to practice it. The average civilian gun owner has no such training whatsoever.
Stephanie #84. Prior arrests are not a good proxy for legal status of guns because not all arrests are for crimes that would discount one from being able to legally buy a gun. My understanding is that the only crimes that preclude you from buying a gun are violent ones and things involving the narcotics trade. Prior arrests are a poor proxy because someone that was arrested for underage drinking decades ago but has been an upstanding citizen ever since would be automatically assumed to have an illegal gun.
Greg, I don't know that the information would be so hard to find. There's no way that the Philadelphia PD gave the researchers real names. I would assume that the cops said they had X cases (3000ish if I remember correctly) and the researchers used a random number generator to pick as many as they wanted (650ish if I remember correctly). The PD then got a list of case numbers and gave them the information of gun carrying status, age, race, location, etc. I guarantee you that the cops ran the serial numbers of the guns they got from the scenes, and it's probably in the case file whether it was legal or not. One more tidbit of info on top of the set they're already getting doesn't seem unreasonable.
And even if I'm wrong about the ease of procurement, which I could be, that still doesn't make it any less important of a flaw and, in my opinion, any more valid of a study.
amphiox -
If your average gun owner doesn't have it, then how is it that it is that there are so very many of these programs, with so many of them busy as all hell? Yes, there are varying degrees of training, but the fact is that a lot of gun owners get that training.
And I wasn't being snarky, when I suggested taking guns away from them. There are a lot of cops who really have no reason to have a gun on their person. What the hell is wrong with leaving a shotgun in their car and otherwise have them follow the UK model? I understand this would not work in a lot of locations, but in a lot of locations it would be just fine.
The problem is that it is expensive to offer them reasonable training. So why not focus those funds on areas that need cops to carry, while areas that don't, don't have as many cops with guns? It really isn't as bad an idea as people would have you think. I know that in my area, for example, there is no reason that every cop on the street needs a fucking gun. Having a couple of officers per patrol area who are armed and the rest get shotguns in their car would be perfectly adequate.
That way the cops who carry are reasonably well trained and we can be reasonably well assured that they are of sound mind and responsible enough to handle it.
And btw, most cops outside highly urbanized settings, aren't getting much chance to reinforce that training either. I doubt even half a dozen cops here in Portage have pulled their gun on duty, in the last ten years. The sheriff deputies have been involved with more raids, so they probably have a few more, but still a small percentage. Kalamazoo is a little more exciting, as it were, but even there it is a rarity. The worst that most of them have ever dealt with, are minor assaults and domestic violence. Go figure, I have intervened in those sorts of problems myself, being rather handy for talking drunken jackasses down from acting stupid. I expect that most of us are pretty capable of that.
Ok, I've read the paper and it seems to me that what it really has to say, is that if you carry a gun in a ghetto that has a lot of illicit drug activity, you are more likely to be shot than if you don't carry a gun. I am not even sure that having information about how many were illegal would make a substantial difference.
Color me completely unimpressed. I will definitely not be citing this paper as evidence that carrying a legal gun under most circumstances is more likely to get you shot. Though if the discussion is carrying a gun in a high-drug trafficking area, I will be all over mentioning what a bad idea it is.
Jimmy@29 said:
So what you're saying is that the guns protect them against roadside bombs, IEDs, suicide bombers and so forth? Frankly, if you want the troops to be safer, getting the hell out of there sounds a much better approach.
Re: home invasions ... DonnaB@60
You left out the possibility that amongst the things they might steal and/or use against you would be the guns you were counting on to deploy against the invaders, and likewise the collateral damage to other members of your houshold directly or indirectly resultant upon you drawing your weapon.
You also overlook the potential use of this weapon in domestic violence -- which is very common. You can't just have the gun when it is ideal to have it, and hope it all goes as anticipated. If you know that you don't have a gun, you may well seek a less risky strategy.
OK, for better or worse I've posted my article about the study.
http://austringer.net/wp/index.php/2009/10/05/gun-possession-and-assaul…
My conclusion:
#93 - Home invaders (a very specific group) by definition come with their own arms and intent to harm. I did not suggest (and do not!) that using a gun is the best thing to do in every instance involving a crime.
I merely stated that sometimes it can lead to a bad thing having a better outcome than otherwise.
As for domestic violence, it is most often committed without a weapon. Please correct me with statistics if you have them, as I'm simply going on personal observation from having worked with a YWCA shelter.
Potential confounds I see here:
Are people more likely to carry guns who live/work/hang-out in dangerous neighbourhoods? I heard in Philadelphia, in particular, notoriety between adjacent streets can vary radically with respect the their reputation for crime.
Also, are people who carry guns of a more violent nature, and therefore could escalate a situation into a shooting incident? Irrespective of this, does the mere sight of a weapon on another provoke a violent offender to be more likely to shoot?
Are people who walk around with a gun - a device designed to kill - more likely to be mentally unsound in whatever way? Are they more likely to have poor inter-communication skills that could have otherwise been utilised to diplomatically disarm a tense situation?
Do people carry guns BECAUSE they have warranted fears for their lives - concerns they might get shot?
These are just questions that popped into my head just now after reading your blurb. Maybe not so much confounds as potential pitfalls for cause-effect conclusions being drawn. If you can account for all this somehow, I'd be curious to see the result.
A final thought: ensure proper trigger-discipline in the police force + eliminate guns and gun culture from a society = no illegal shooting.
Daen, the majority of those deaths are suicides. In 2006, there were 10,177 murders by firearm.
OK, Russell. So the new stats are:
Gun related deaths in UK for 2009 = 50 (1 in 1,200,000)
Gun related deaths (murders) in US for 2001 approx 10,000 (1 in 30,000)
There is a 4,000% greater chance of being murdered by a gunman in the US versus the UK. What conclusions can you draw from that?
How many of those victims you worked with died as a result of weaponless domestic volence? And what percentage of victims of domestic violence with a weapon do you suppose die as a result?
Oh dear, way to not to a risk assessment. Try relating the number of deaths to the numbers of instances of use.
Approx 32% of US households (100 million people) owned a firearm, compared to 3.4% of UK households (2 million people) in 2000(http://www.ajph.org/cgi/reprint/96/10/1752.pdf)
So around 50 times as many people have access to a gun in the US compared to the UK. And gun related murders in the US (1 in 40,000) are around 30 times higher per population head than in the UK (1 in 1,200,000). Does access to guns lead to more gun related murders? Seems tempting to draw that conclusion ...
Daen, the statement I made was that someone visiting the US was much more likely to be killed by a car than by a gun. The numbers say about 4 times more likely. If you want to argue with someone about UK gun laws, you need to find someone who will oppose you on that.
David [90]: I'm not saying the info in not available. I'm also not saying it is not potentially important. but there are two things being missed here: 1) (to repeat) getting access to certain information in behavioral studies is costly in all sorts of ways, and IRB issues is a cost. At every juncture every researcher has to make decisions and the hypotehtical ease with which we may think some data are available is a world of difference from the actual on the ground reality. 2) If a study is flawed because a certain angle of research that is interesting and important to the question at hand is not addressed in that study then I have yet to see an unflawed study of anything anywhere. Research proceeds by bits and pieces.
I will add: I find the ghettoizatoin and criminalization and other-ization of the victims of these assaults, and the unfettered comparisons with this concept that seems to be emerging of "normal" gun owners (as distinct from these "inner city" gun owners) to be ... really interesting . (David, I'm not speaking here to your point, but more generally)
DuWayne [92]: Hold on a second, buddy. Before you found the study flawed. Now you find it not flawed, just narrowly defined to a certain circumstance that would not apply to you. Therefore the flaws go away. Interesting. Plrease rethink.
It is, indeed, narrowly defined which is what any study like this should be. I find it hard to believe that you, with your academic orientation, find focus and narrow definition in an academic paper to be a flaw. Have you ever seriously looked at an academic question of any kind? (Yes, you have, I know this for a fact.) Did you find that one paper that is not flawed because it addresses all circumstances, read that, and be done, or did you read a dozen papers that collectively addressed the issue (and even then left open isuses). (Yes, you did.) Please relocate your academic thinking hat from whatever toilet you dropped it in, dry it out, and put it back on.
Now, regarding your general conclusion, am I reading you right on this? As long as this study's meaning about gun-related risk does not directly address your white privileged ass, it is not relevant because you are, like, a different species from these inner city ghetto-troglodytes so the basic principles could NEVER under any circumstances apply to your privileged white ass?
What exactly is it that is different about the circumstance of getting involved in a situation in which people pull out their guns that makes your PWA somehow better able to make effective defensive use of the firearm you are (hypothetically) carrying when push comes to shove that makes a study of circumstances with THESE people (or should I say THOSE people) you so readily disdain irrelevant? You got some kind of special super power or something?
Rethink that too.
daen [97]: Put the suicides back in, until someome comes up with a solid argument that those dead people are somehow less dead, or less killed by a firearm.
Suicides by means other than firearms are much more often failed. Suicide attempt + firearm = person who is dead now but would not have been had there been some sanity in our gun laws . Taking out the suicides is cynical.
Always the same gut "thinking" with USAmericans when talking about guns... And I'm not saying that about you, Greg.
Anyway, for me it's a matter of circumstances and places.
For instance, where I live, in Spain, and anywhere else in Europe, I never had felt the need to carry or use a weapon and probably never will.
But when I was living in the US, I always carried one on me.
In any case, you're certainly right, wearing a body armor would have been more efficient and less dangerous, surely.
Greg -
Hold on a second, buddy. Before you found the study flawed. Now you find it not flawed, just narrowly defined to a certain circumstance that would not apply to you. Therefore the flaws go away. Interesting. Plrease rethink.
No, I still think it is flawed.
It is, indeed, narrowly defined which is what any study like this should be.
The problem is that the study itself is narrow ranging, while the conclusions are not. They are making a very wideranging claim, based on a very narrow band of evidence.
Now, regarding your general conclusion, am I reading you right on this?
No.
As long as this study's meaning about gun-related risk does not directly address your white privileged ass, it is not relevant because you are, like, a different species from these inner city ghetto-troglodytes so the basic principles could NEVER under any circumstances apply to your privileged white ass?
Last place I lived, I was downrange from a ghetto shooting and clamped my hands down on the wound of the women who was hit by a stray. It most certainly has applied to my privileged white ass, because these sorts of places have been where I have spent most of my adult life living. I would also assume, based on this study, that me carrying a gun where I last lived, would also significantly increase my risk of being shot.
More importantly, this is not about motherfucking race. It is about a specific criminal subculture that has little to do with race. While gangs often tend to run down ethnic and racial lines, that is not so true of traditionally black gangs anymore. But more to the point it is irrelevant. It is not the skin color or country of origin causing the risk factor - it is the fact that the people involved are career criminals.
And it isn't that they are irrelevant - it is just that a) there are confounding factors that are not guns increasing that risk of being shot and b) the conclusions of this study claimed something far different than what the study was actually looking at.
What exactly is it that is different about the circumstance of getting involved in a situation in which people pull out their guns that makes your PWA somehow better able to make effective defensive use of the firearm you are (hypothetically) carrying when push comes to shove that makes a study of circumstances with THESE people (or should I say THOSE people) you so readily disdain irrelevant? You got some kind of special super power or something?
For fucks sake Greg, I want to see less guns on the street, not more. I would like to see a study that does show what this study concludes, because I suspect that though this study doesn't reasonably show it, the conclusions are correct.
Ok, very good.
However, I do think many have over-broadened the claims made by the study then yelled about how they are overbroad.
That I would agree with. Though I think that it is important to note that a lot of people are probably basing their argument not on the paper itself, but the conclusions of the article written about it - which are broader yet, than the conclusions in the paper itself.
David, the problem is that you're not using legal status of the guns directly. As someone pointed out snarkily upthread, steel doesn't emit pheromones, and the status of the gun won't change that. You're using it as a proxy for illegal activity more generally, which makes arrest record quite relevant.
Would the legal status of the guns be useful information? Yes. Does the lack of it invalidate this study? Not in the least.
The same is true for Wesley's critique of the study. Would it be useful to compare assaults involving a gun without shootings to those with shootings? Of course. Does the fact that this study is not that study make this study useless? Of course not.
And suggesting that the authors of the study set out to confirm their own biases based on the fact that the detailed statistical methods used were not presented in the paper is absurd. A quick glance at the table of contents for the AJPH shows that this article is already longer than most in the journal. That's just reaching for reasons to blow off the article.
The point is that after adjusting for the involvement of drugs, location, age, high-risk occupations, education and prior arrest record, the possession of a gun was not found to be protective. In fact, the more of these variables the researchers adjusted for, the greater the risk of being shot among those in possession of a gun.
Donna [71]: I call invalid straw man invocation. This is a post about a specific article published in a peer reviewed journal. Almost none of the comments above by anyone, by count, are about that article and what it says. Even some of the comments that seem to be are not really. So, what is this discussion about?
I think gun ownership as a broad category is a perfectly valid area for discussion on this thread. I think, though, that y semi-personal account of a pretty typical gun owner killing a pretty typical innocent teenager because he was acting exactly as pretty typical pro-gun people say one should act is making you a little uncomfortable.
I don't agree that both people were good guys. I think having a loaded shotgun next to your bed and finally, after years of waiting, you kill an innocent person makes you not a good guy. I should mention that this is a pretty run of the mill working class suburban neighborhood. I've seen the cops in my neighborhood "in action" twice over the last four months, and they were both DWI issues. Three years ago we had a string of violent muggings about a mile from here. A year before that a girl was run over by a train. Across the entire city I live in I'm sure there's more stuff going on, but that is what has been happening in the several blocks around where this shooting occurred.
The reason I know a lot of people who have killed someone is that for various reasons I happen to know a lot of US soldiers who were in combat, and a fair number of people outside the US who have been in active combat situations. With the exception of one person who comes to mind, all the people who have killed people that I know did so in combat or something like combat. So far as I know (one never really knows!) You came up with your one Marine, there are probably others you are not aware of. For some reason, people tend to tell me about these things. (Just last weekend, in fact, one such conversation happened!)
steel doesn't emit pheromones
Yeah, but gunoil and copper cladding have a certain ... je ne sai quoi
Otherwise, what Stephanie said....
You say this like your post isn't titled "Carry a gun = you get shot more often". Seriously? Some internal reflection, please? The tone of this thread is hardly just the "gun nuts'" fault.
Paul, thank you for telling me how to blog. Next time I write a blog post I'll check with you first on the title!
But seriously, suggest a better title, I'll change it.
It's also worth noting that this study's results are in line with prior studies that had been attacked on methodological grounds, like this one, which is fully available online. The current study was designed to address those methodological criticisms and found similar results.
Does anyone know of any studies that do show a direct protective effect of owning or carrying guns? Those are the studies you want to use for comparing methodology. Arguing about methodology isn't very useful if all the studies tell the same story.
Here's the new Title!
Carry a gun = you get shot more often in certain situations, but it might be flawed or limited, and you didn't even read it, and gangs are bad, and some cops shouldn't have guns, and you usually only get beat to death in England, have you ever been to Vermont? It's nice this time of year, and guns don't kill people, but yes they do kill people, and I want to carry a loaded gun into Babies R Us without people getting all up in my grill.
"Does anyone know of any studies that do show a direct protective effect of owning or carrying guns?"
There are many. I heard about them at the gun club. Can't lay my hands on them at this particular time, tho...
Ulys, I hear they exist too, but they're hard to compare without references. Most of what people bring up in arguments are the stats for a particular state or country before and after a change is made in the law, which isn't particularly direct.
José, +many internet points (certain restrictions may apply). I will point out, though, mostly for others, that someone who is comparing the researchers' conclusions to the press release has probably read the article.
I make no instruction on how you should blog, Greg. I was pointing out that it was disingenuous for you to talk about how "some" people (I assume you weren't talking about yourself with the distance you put in the phrasing) draw over-broad conclusions and then complain about them when you started the over-broad conclusions in the first place (and later admit you hadn't even read the paper).
For the record, I'm not pro-gun. I'd prefer the English model. The argument for keeping guns to hedge off government tyranny is the only one that ever made sense from a Constitutional basis, and now it's laughable to think you'll be able to prevent a government takeover with civilian arms. I simply found it somewhat dishonest and rude how you were framing the comment thread, which is the only reason I posted in the first place.
Paul: Of course I've read the paper. And, have you noticed the very disturbing 'us' vs 'them' reactionism that is happening here?
(Where the fuck did I "admit" that I've not read the paper? WTF?)
Ulys -
I have heard that snipe are really good eating, but fucking hard to catch. Heard it from several people on the kickass ranges by Hogsett lake's state game land. Fucked if I can find one though...
Paul -
I was pointing out that it was disingenuous for you to talk about how "some" people (I assume you weren't talking about yourself with the distance you put in the phrasing) draw over-broad conclusions and then complain about them when you started the over-broad conclusions in the first place (and later admit you hadn't even read the paper).
Based on the press release, it was a very reasonable starting point - are you suggesting that he should have gone back and changed the starting point, because he discovered it was mistaken?
I simply found it somewhat dishonest and rude how you were framing the comment thread, which is the only reason I posted in the first place.
How exactly was he being dishonest and rude? I will admit that I didn't care for his characterization of my position, but we got that straightened out. Ultimately this was not an activity in going from one point and ending up at a completely different point abruptly - there has been a great deal of intervening discussion - not to mention several of us have been afforded the opportunity to actually read the paper, rather than relying on the press release.
I think you're overeacting just a tad...
DuWayne, I think you would not know snark from snipe if the snipe bit you in the as and the snark flew up your nose.
Based on the press release, it was a very reasonable starting point - are you suggesting that he should have gone back and changed the starting point, because he discovered it was mistaken?
First I'm being told that I admitted to having not read the paper, then I'm being told that I agree that the title of my post does not reflect what the paper said. Neither is true. (I believe some commenter up thread screamed at me that I must have not read the paper because I disagreed with him.)
The problem here, folks, is that YOU (whomever you are) are more or less failing to identify at all, even one iota with the YOU of the paper. My YOU (in the title) was meant to be inclusive of both those in the study and YOU'ALL reading this blog, including ME.
I found this paper to be an indication that the counterintuitive (from a pro-gun point of view) may be true. That is still the case. The paper is not perfect, but most of the criticisms I see in this comment thread are winged monkeys, miscomprhension about what an academic paper normally does, or a failure to recognize that although YOU (you who are reading this) and the study group in the paper are really not different species living on different planets. Yes, circumstances are very different, and maybe another study of YOU (your demographic/population/geography, whatever it is that makes you think this research does not apply to you) will show that carrying = you get shot less. Stephanie has already asked for examples of such papers. None are forthcoming. Maybe because it is not true. You would think with all the interest in this such a study would exist if the data were there to support it.
But anyway, DuWayne, thanks for the spirited defense of my blogosity.
In response to one of Donna's posts (I think it was Donna...) I *AM* an effete wuss whose afraid of big explosions.
I'm still not sure why thats in any way a bad thing...
And the idea of policemen with guns...just scares me. BADLY scares me. Properly trained emergancy squads sure, but the regular guy who walks the beat? That's just crazy...
Actually, Lab Rat, that was DuWayne riffing off something I said. There's nothing wrong with being afraid of explosions. They're dangerous. It's just that I've had gun nuts whom I can comfortably outshoot accuse me of being too scared to think about the issue because I don't see a compelling need for handguns in American life.
My son has a carry permit and does carry. A fellow tried to rob him at knifepoint. My son pulled his gun and advised the would-be thief to desist and disappear. The thief did so. A win-win situation. Are there studies relating to how carrying a gun makes you more or less at risk of being attacked with a knife, chain, club, etc?
DuWayne, I think you would not know snark from snipe if the snipe bit you in the as and the snark flew up your nose.
AHAHAHAHAHA!!!!!! Oh wait - I still haven't found that snipe and it would appear those mythical papers are nowhere to be found.
Jim: I'll state with confidence that there are no statistically valid comparisons or analyses of reliable data that would show that carrying a gun increased the overall outcome quality. The case of your son, and I'm glad he was not hurt, is an anecdote and may even have been an anomaly. There are no studies that show that carrying increases quality of outcome.
Before I tell you how I know what, let me say that I personally don't care if carrying increases quality of outcome, I still think there should be more reconstruction gun manufacture, wholesale, retail, distribution, ownership, and use than exist now. This is an opinion I am only recently coming to embrace. The reason I'm interjecting this should be obvious.
OK, how do I know there are no studies that demonstrate that "Carry a gun = you get shot LESS often" or bludgeoned or knifed or whatever? Because there is not one cited in this comment thread.
Greg -
My bad. I actually looked half heartedly for where you had mentioned you didn't read the paper, but figured it was just buried.
I really don't see how this study can be extrapolated outside a very narrow population though. I am certainly not arguing that the conclusions aren't probably true to the general population - I just don't see this paper providing any evidence that that is the case. And not even this paper argued that it does.
Lab Rat -
I was most definitely not knocking anyone who is uncomfortable with gunfire and other loud explosion sounds.
And I am happy that you agree with me on cops. I thought the concept was absurd when a friend suggested it several years ago - then I started meeting cops and worse, one of my cousins almost became one. Those experiences rapidly changed my perspective...
OK, how do I know there are no studies that demonstrate that "Carry a gun = you get shot LESS often" or bludgeoned or knifed or whatever? Because there is not one cited in this comment thread.
Fuck this comment thread, try finding one on the NRA's website...
... where peer review and reloading clips are one and the same.
Greg, are you presuming every suicide should be prevented? Does that mean you are opposed the laws in states like Washington that legalize assisted suicide? There certainly are suicides we, as a society, should try to prevent, in particular, 1) those committed by young people motivated by problems that will seem much less given a few years greater perspective, and 2) those driven by organic depression or other mental illness where it can be treated successfully. But there are other suicides that seem to me well within an adult's appropriate sphere of liberty. In particular, I am quite leery of judging when anyone else has reached the end of their rope in dealing with chronic illness. Were I terminally ill and not wanting to suffer further pain or deterioration, a pistol would be my first choice as a means to an end. Because it is fast and certain. That anyone would try to prevent my choice in that situation is, for me, precisely a reason to want a guns legal.
Russell, when I was a sophomore in college, I happened to go to school with my ex-boyfriend's then girlfriend. I got a call from him one day asking me to meet her outside of the class she was sitting in and break some news to her (1) so she wouldn't hear it in a worse way and (2) so she'd know as soon as possible that he needed her.
The news was that my ex-boyfriend's younger brother had just figured out how to pull the trigger on a shotgun with the barrel in his mouth. He didn't do it well; his impulse-control and anger-management problems made him something of a fuck-up. He lived for about eight hours, some of which my ex said he was conscious.
Fast and certain and you're ticked at Greg? Kindly take your off-topic hyperbole and piss off.
Stephanie, where have I written in hyperbole? I'm sorry about your ex-boyfriend's brother, and explicitly pointed out the senselessness of young people committing suicide, for reasons that would soon resolve.
That is not the typical suicide. By ten year age-segment starting at 15, the suicide rate is lowest for ages 15-24 and highest for 75-84. (US Statistical Abstract.) People who have lived 75 years are not committing suicide from youthful lack of perspective. I suspect many of them are acting in the face of bleak circumstance that will not improve. The tragic example is the teenager with poor impulse control. More commonly, it is an old man not a young one.
Russell, did I say tht every suicide should be prevented?
So, are you saying that guns should be totally unregulated in case you get some terminal disease and want to kill yourself?
Russel -
I think Stephanie's point was that using a gun to commit suicide is not a very good idea. You would be amazed at how many people survive for hours, or survive altogether, after swallowing a gun. Using a gun is fucking stupid - it is messy for those who come after, even if you do succeed - not to mention cruel and it usually fails, at least for a few hours.
More to the point, no one - not one fucking person said a damned thing about committing suicide because one is old and/or infirm. But we don't need guns for that. There are drugs - can't get the script? There are other ways to procure them. There are also many other ways to go, that are considerably less cruel to those you leave behind and considerably more likely not to leave you writhing in agony or worse. Given my druthers, we would pass laws that allow physicians to proactively assist.
You are making a completely off topic and irrelevant point, that is completely and utterly pointless.
Given my druthers, we would pass laws that allow physicians to proactively assist.
But they'd have to ask first.
I'm a little surprised that this is the first of its kind since I read about 20 years ago that having a gun in the home made it 5 times more likely that someone in the family would get shot. I do not remember the source.
Incidentally, police are less likely to be injured on the job if they don't have a partner: they're less likely to charge into danger. But they want a partner for company and to help keep them awake.
Monado, it's the first to use this methodology, not the first to find similar results.
Russell, what Greg (134) and DuWayne (135) said.
No, Greg, my point is simply that suicides shouldn't be lumped with murders.
DuWayne, it is undoubtedly a very personal choice that depends on medical circumstance, familial circumstance, personal knowledge, remaining ability, etc. Many people who attempt suicide by drugs also fail.
Speaking as a non-American about the efficacy of defensive gun uses
I have lived here in Sydney, Australia for almost all of 51 years. During this time, excluding television/the movies and the occasional sideways glance at a cop or security guard's holster, and one occasion when an acquaintance invited me to have a hold of his unloaded pistol -- I think it was a .22 but I'm not really sure now -- I have never seen anyone carrying a gun or imagined that anyone might be carrying one.
Nor have I ever been in a situation in which having a gun might have been a help (even allowing I knew how to handle it effectively and efficiently). Not once. And to the best of my knowledge, nobody I have ever met has ever been in a situation where a gun would have helped.
I hope it stays that way and have no reason to doubt that it will.
Occasionally, there has been some fuss, and I was glad nobody nearby was carrying a gun though I have on two occasions, persuaded someone to put down a knife they drew and a third time, an iron bar before they hurt someone. No harm and no foul.
I'm glad of these things because I am a school teacher, and my first duty is to keep the place safe.
If I take the US gun advocates at their word, that they can't feel safe without the right to carry a gun, or that they feel the need to carry guns to enable them to resist tyranny, then I think this is a dreadful commentary on US life and governance. Your society has failed you abjectly. I'm no patriot and still less a jingo, but on this reading, we Australians have a lot for which to be thankful.
Stephanie Z.:
"In fact, the more of these variables the researchers adjusted for, the greater the risk of being shot among those in possession of a gun."
The paper describes only two levels of lumped inclusion of multiple factors. For "all shootings", 13 out of 20 factors were noted as being significantly different between the case group and the control group. Maybe adding more stuff that is known to differ significantly between the groups and treating it in unknown ways makes your confidence go up, but I can't say that it has that effect on me.
The sentence quoted above, though, is imprecise, and makes it sound like factors were added individually with a monotonic change in relative risk resulting, which is not the case. The Branas et al. paper does not report on individual factor contributions to the "adjustment" of control group "gun possession" inclusion.
Procedurally, the draft manuscript prepared by "intelligent design" advocates William Dembski and Robert Marks had advantages over the current Branas et al. publication. The suspicious numbers reported in the Dembski and Marks manuscript came with not just a description of their methods, but with the actual code of their model. It was relatively simple with that in hand to demonstrate that their reported numbers were bogus, and determine with great precision why they were bogus.
With Branas et al. we are asked to accept, without any means of verification, that it is both reasonable and necessary that effectively they not merely exclude 40 out of the 49 instances of gun-possessing members of the control group in the "all shootings" condition, but instead classify them as if they were non-gun-possessing. Without this opaque and dramatic "adjustment" in the numbers, the relative risk actually showed a small benefit for the risk factor of "gun possession".
When people proffer a model as a methodology to accomplish something in science, I do expect them to publish the details needed to replicate it. I don't think that the expectation is at all "absurd". The classic artificial neural network of Hopfield and Tank was communicated efficiently via their short paper in Science, such that I could take up their Lyapunov functions and implement them myself. I know this because I have done so. The same goes for the back-propagation model of Werbos and the Adaptive Resonance Theory (1) model of Carpenter and Grossberg. That's not to say it was particularly easy, especially for ART-1, but the requisite information was provided via the publications. The same goes for several sound production models I investigated for application to the question of biosonar sound production in bottlenose dolphins. In my opinion, publishing replicable models is a distinguishing characteristic between science and commentary.
Quite true, Wesley, although I'm not sure why you're objecting to imprecise wording about how many variables were evaluated at once when you feel the need to put the entire "adjustment" in scare quotes. I also don't see that it affects my point in the least.
Additionally, a number of the citations in the paper are for methods, and your assertion that they are excluding participants is not in the paper, but based on your calculations surrounding a methodology you admit to not understanding.
Stephanie Z.:
I'm not sure why you are so assiduously trying to be personally offensive and tendentious. Case-control methodology and calculation of the odds ratio or estimated relative risk is perfectly straightforward, and it is via this part of their methodology that one can go some way toward analyzing the characteristics of the black-box portion of their work. And characterizing input and output conditions of such black boxes is also a pretty straightforward process. We know the summary initial conditions from the Branas et al. Table 1, and we know the relevant calculated outcome in the form of the reported final odds ratio, so assessing the effect of the intervening black box is not a matter of guesswork, as you seem to wish to impute. It's a matter of quite simple math.
What is revealed about the workings of the black box are broad statements about certain of the components used, but this fails to deliver the information needed to allow a third party to replicate the actual function of their black box such that the effect, the one equivalent to transferring 40 out of 49 control cases in the "gun possession" category to the "non-gun-possessing" category, can be understood in any rational sense, or to note possible problems in the details with which the models in the black box are constructed. Certainly the citations you mention from the "Statistical analyses" section may illuminate why they wish to keep inflation factors to less than 10, for example, but they are not informative of what values actually obtained in their regressions and models in this study.
I'm sure that there is a reason that Branas et al. do not state the effect of their black box in terms of what the equivalent effect on the degree of change in their raw numbers would apply. You yourself have appealed to them having a finite amount of space as explanatory of other lacunae in their exposition. Others could be speculated upon. However, that changes not in the least my point, which is that their reported results can be understood in those terms and are a straightforward application of the math behind case-control studies.
Forgive me if I'm repeating some things, but this is what I sent to the study's authors:
As a long-time CCW permit holder and one who carries a firearm on a daily basis, I read your recent study with extreme interest. I wanted to share my thoughts with you and see what you think. My quick first impression is that the study claims that "guns did not protect" those who were involved in a criminal attack from being shot, yet this has never generally been the objective. If you're attacked by an armed criminal, the assumption is that you'll be shot. Statistics show that even complying with the criminal is no guarantee that you won't be shot. (Those stats were supplied to me during my CCW class, so I can't recall what the source was.)
The idea of being armed is simply to have a fighting chance. If protection is what you want, buy a bullet proof vest. You'll still get shot, but you'll have a better chance of not being injured. If you have a gun, you'll have the ability to fight back.
At the very beginning of the study, the abstract states: "Objectives. We investigated the possible relationship between being shot in an assault and possession of a gun at the time." So the hypothesis presumed a relationship. They conclude, given a statistical analysys of the very limited sample (less than a thousand crime victims), that you're just under 5% more prone to being shot if you have a gun than if you don't, yet a significant number of those without a gun were also shot! (Wasn't it 100%?) Further, if the victim with a gun tried to resist, it clicked the percentage up almost exactly one point. This low percentage is both insufficient to cause alarm and might be consistent with other studies that show that once a person involved in an attack is known by the criminal to have a gun, they then become the criminal's focus. This is why armed guards or police openly wearing guns get shot at first. Which leads to the next bone of contention...
The study also doesn't say whether or not these victims were licensed to carry, nor what their level of training might be. In short, they may have interviewed drug dealers or gang bangers and not even known it! Undercover police officers, trained citizens and security personnel carrying concealed are not distinguished from the victim pool as a whole. From a purely objective, scientific standpoint, the study provides a tantalizing avenue for further investigations, but the data are hardly a good basis for drawing any hard conclusions, let alone anything to base public policy on.
Of course I only had the overviews / articles that were printed, and can't seem to find your actual data. (I only made a cursory search -- please don't take this as any sort of accusation!) I'd love to see the whole study. I'd also like to know if you plan to expand on this one and possibly take into account some of the variables I mentioned. Thank you for your time and attention, and for the fascinating work!
NOTE: I'm assuming that these researchers also have no other ulterior motives. :)
Personally offensive and tendentious, Wesley? For noting imprecise language on your part? For noting the use of scare quotes? For noting an argument from ignorance? For not acting as though multivariate analysis were something new and strange and dubious?
Tell me, Wesley, just what I've done that's so wrong, aside from argue with you and do it well.
Russell: I don't think murder and suicide should be lumped but when it comes to gun control you can't ignore the carnage that happens in the overlap between suicide and gun availability in the great Venn Diagram of life.
Stu, I suggest reading the comments on this thread, as much of what you say has already been covered. Also, if you look at the names on the study and the name on the blog, you'll note that the investigators in the study didn't blog this.
More guns = Less crime
#98 - Of course none of the victims of domestic violence I worked with were dead. Most were beaten repeatedly and many left only after their children were also in danger.
As for domestic violence (where I live) the ones that make the news where someone dies as a result are most often the ones where the abuser has turned on a child and it's likely a boyfriend, not the father.
A substantial number of cases involve non-married couples breaking up and one or the other (it's not always the male) going after revenge, usually for having a new boyfriend/girlfriend.
Again, these are my personal observations of the community I live in. But now that you've got me thinking about it, I'll probably have to actually do some research. Gee... thanks a lot! :-)
Donna: Read The Truth about Cinderella: A Darwinian View of Parental Love (Darwinism Today series)
#109 - Greg, I still think that old man and that teenager were good people who both acted stupidly creating a tragedy. Maybe if I knew the old man, I'd feel differently. But your characterization of him "waiting to kill someone" and finally getting the chance... is that fair?
I know of only one person who has "slept" with a gun and it was an M-16 because that's what she was most familiar with. While her husband was in Iraq, their house was robbed (not invaded) twice. It took a few weeks to get an alarm system installed, and then she put the gun away.
As for knowing people who have killed someone, I was thinking of those I know now. There are several who are now dead who killed people in WWII and Korea. You must have a sympathetic ear.
#123 - Lab Rat, I don't mean it's a bad thing to not like explosions or be afraid of guns. I simply do not understand it. I think they are a lot of fun to shoot.
It's a good thing you don't live near an Army installation where artillery training takes place :-)
Donna 109: your characterization of him "waiting to kill someone" and finally getting the chance... is that fair?
Only in the sense that it is what he did.
At some level everyone is a good person, of course.
Thanks for the recommendation, Greg. I just ordered the book.
I responded to your little racism tirade here.
Greg #155, I don't think that's what he did at all. You make it sound like he was staying up nights practicing his SWAT moves when, in reality, it seems that he was an old man living alone that was all to aware of how vulnerable he was. The elderly are frequently targeted as victims for robberies and home invasions because they are considered weak and less able to resist.
Furthermore, there is evidence (http://www.jstor.org/stable/2095123?seq=1) to support the idea that people that buy guns for protection are not part of a violent subculture that is eager to kill any and all that they percieve as a threat.
What happened was a terrible misfortune resulting from stupidity on the boy's part and fear-based overreaction on the old man's part. But that doesn't make him a bloodthirsty killer. Many defensive gun owners, including myself, dread the thought of killing anyone, even the criminals that you seem to think we dehumanize. If my apartment was being robbed, of course I wouldn't go out to try to kill the robber, because anyone's life, crims included, is worth more than any amount of property (assuming that I can live without it. An interesting moral situation crops up when someone is trying to steal something that you need to live. In that case, does that property equal life? But that's a digression).
On the suicide front, research (http://content.nejm.org/cgi/content/abstract/322/6/369) supports the so-called "substitution" hypothesis, which states that people who want to kill themselves will regardless of the means. In studies that compared the suicide stats pre- and post-gun control implementation, the gun suicides dropped but the overall suicide rate was unchanged. This suggests that suicidal people kill themselves by the easiest means available, and casts doubt on the hypothesis that having a gun readily available increases the risk of suicide.
Stephanie Z.:
Come now, Stephanie. Certainly you should be able to appreciate that continuous attempts to cast your interlocutor as incompetent are personally offensive, and for tendentious you have not been shy in stating precisely where your advocacy interests lie on this topic. Pointing out truth, as I did there, is only to be expected.
We've already had exchanges on the topics you bring up again, and I fear that your self-assessment of how your contributions have fared is somewhat off.
When you accused me of imprecision because of placing quotes around a word, you were offering a tu quoque to attempt to divert from the rather more substantial issue of your own claim that embroidered and exaggerated results in the Branas et al. paper: "In fact, the more of these variables the researchers adjusted for, the greater the risk of being shot among those in possession of a gun." It was an enthusiastic and emphatic claim that, unfortunately, appears to be empirically not well founded, and which may well warrant the Churchillian description of "terminological inexactitude". The fact was that there was no report of monotonic change in response with addition of individual factors to the model, as your emphasized claim implied. The fact was circumscribed to one more inclusive and one less inclusive level of modeling, each of which incorporated multiple factors. Nor were the particular factors you listed just above your quoted bit all definitively assigned to one or the other level of model. The fact that 13 out of 20 available factors were known to be significantly different between case and control groups virtually guarantees that the more inclusive model had to lump in more of those mis-matched factors.
Hopefully my mentioning that tu quoque is an argumentative fallacy is simply review material.
Then you claim to have noted an argument from ignorance. That you did not do so explicitly can be confirmed by word search on the comment thread above. So what, exactly, was that supposed to refer to? I'm guessing that it refers to this claim of yours: "Additionally, a number of the citations in the paper are for methods, and your assertion that they are excluding participants is not in the paper, but based on your calculations surrounding a methodology you admit to not understanding." I responded to that one, too, showing that my basis for my argument was directly taken from the description of the methodology expressed in the paper, and that the mathematics were quite simple. Perhaps you overlooked that, as that would be the easiest explanation for your curiously canted perception of the outcome of that exchange. What I have "admitted" to was the same thing that any reader of the paper would have to admit to, that it would be impossible to independently replicate the results of the study based upon the provided description of the methods since the particulars of the models underlying adjustment of the raw data were not disclosed. Casting that enforced and universally applicable degree of ignorance (S.Z. partakes of it just as much as I do) as a lack of understanding is a remarkably tendentious bit of personal attack.
Then there is this bit of yours: "For not acting as though multivariate analysis were something new and strange and dubious?" I should think that it is obvious to any reader that you were imputing something about my background knowledge, and doing it in a personally offensive way. The tendentious component of it lies in the knowledge that if you can dismiss any and all critics of the paper's content as being less then technically competent, then it is easier to avoid cognitive dissonance by denying that a critic has a point. But you obviously do not have knowledge of my background, even though my previous comments here clearly indicated my professional involvement in the fields of epidemiology, artificial neural networks, and bioacoustic modeling. I do have decades of experience with complex modeling, plus a co-edited technical volume on neural networks to my credit, and I have an appreciation for the pitfalls that can beset the use of models. This may not be something that you have experienced professionally, so perhaps that accounts for your persistent refusal to allow that others may rationally find a vaguely described model unconvincing, especially when it causes a radical re-interpretation of the raw data set. Nor is the statement as given supported by anything in our exchanges. I certainly claimed no such thing as that the methods, insofar as they were actually revealed, were novel, strange, or dubious in and of themselves. But just as not everything that can be assembled comprising an engine, transmission, and set of wheels functions as an automobile, not every complex model comprised of several component parts functions as even those constructing it suppose it ought to, as my example of the Dembski and Marks manuscript affair was intended to illuminate. I have emphasized throughout that my issue is about the lack of transparency concerning this critical component of the research, and that is certainly not due to any lack of understanding on my part.
That brings us to your closing bit: "Tell me, Wesley, just what I've done that's so wrong, aside from argue with you and do it well." In each of the instances you've held up as demonstrating the disinterested and civil treatment you give this topic, I have demonstrated instead that in each you have partaken of the precise flaws I mentioned at the outset, and been incorrect to boot. So you did not "do it well", and you have been an uncivil and remarkably partisan commenter. I certainly never expected my arguments to be universally adopted, and I am content to remain in respectful disagreement with people of goodwill on many an issue. I do hope that you recognize that insofar as I have had to gainsay your peculiar miscomprehension of the record of our exchanges, I did so in my own defense and in the hope, however optimistic, that you might come to a better understanding of your own actions.
David [158]: Absolutely not. I describe the situation exactly as it was, and I did not describe someone practicing swat moves. That is something in your head. And please, spare me the blaming of the victim. Not only do you call the boy stupid, as if that is true, or stands out in relation to all teenagers, but you go on to imply that he was robbing the house, and he was not. Seriously, man, you owe him and his family an apology.
Regarding the suicide study you cite, I invite people to look at it and note that "We conclude that restricting access to handguns might be expected to reduce the suicide rate in persons 15 to 24 years old" ... that part of the conclusion relates to hand gun availability being greater where the suicide rate is five times greater. They further note that where the suicide rate is reduced because of lack of hand guns, that lack of handguns is PARTLY (no where near wholly) off set by suicide by other means.
Wesley, you can put away your argument from authority. Your math is just fine. Surely even I can see that. (Since when is condescension a civil virtue?) It's the inferences you draw from it that are the problem.
You implied that the researchers were doing something wrong by not including their detailed methods directly so you could follow along in the paper. You went so far as to as to delegitimize these methods you couldn't see by placing them in scare quotes. I pointed out that their paper was standard for the journal and incorporated additional methods by citation.
You said the researchers effectively excluded participants. I clarified for lay readers and those without access to the paper that the researchers did not, in fact, actually exclude anyone in the population you pointed to.
You wanted to disregard the paper completely based on the fact that you did not understand it. I put it in the context of other work in order to determine whether it was making some sort of extraordinary claim and asked for additional research on the topic for comparison.
At no point in this discussion have I advocated for anything. At no point have I questioned your credentials. At no point have I been less than civil. I accepted a technical correction on a statement I made. I assumed you could recognize a logical fallacy when described in the wild without it having to be named or a lecture on logical fallacies given.
If there is something about my arguing with your conclusions that makes you feel personally attacked, it doesn't originate with me. Go figure out what it is and get back to us.
And just to underline your point Greg, this just in ...
Sigh ... this is a dreadful tragedy for the community and of course the kids and one hopes they will recover from this but it's just so damned foreseeable.
Today's local news:
"Remember that lancaster woman who was arrested for bringing her gun to her kids soccer game? Well, today her funeral proceedings were started. She was shot by her husband, who then shot himself".
Anecdotes != data. But sad.
Something new to the internet is a url called "Google." You can search information about just about anything. You can search 'Charles C. Branas, PhD, Joyce Foundation'; (a totally anti-gun of any type foundation that spends millions each year funding anti-gun articles, law review, medical papers and anti-gun groups and causes that our own Pres. BHO was on the board of directors of for 8 years when they gave over $10 million to anti-gun causes,) and you will find out that, SURPRISE!!!, that the Dr. has had papers funded by the Joyce Foundation in the past, (Big Cities and Small Towns Bear Similar Risks of Gun Death) and guess what, they are all anti-gun.
Follow this link to see what is wrong with the study: http://volokh.com/2009/10/05/guns-did-not-protect-those-who-possessed-t…
Read the abstract, (it cost $20 to get a copy of the report,) and you will see that they did not control for people who were in legal possession of a gun and those who owned and used a gun for illegal purposes so that if a drug dealer shot someone who was selling in their turf and that person shot back and wounded the original person that counts as two people who had guns and were shot anyway. What if someone shot someone who was breaking into their house and the criminal had a gun. That still gives you one person who had a gun and was still shot.
Note that the report has no margin of error. How is that possible? I am sure that all of those people who they called up were honest. Even the gang members.
Typical of the anti-gun foundations and the press who does not do any work to discover if one of these fake foundation papers is true or not or has some major problems. That is unless the paper is done by any pro-civil rights group or person.
George, can you back up even one of your assertions with a valid reference, including possibly reading the actual paper rather than drawing all of your conclusions from the abstract?
From what I can gather after reading the authors' press commentary and the abstract, this study probably failed to control for the (apparently accurately) perceived environmental risk factor that motivated the carrying of firearms in the first place. An obvious point that would have had to be addressed in any serious press communication. Otherwise, it seems rather like arguing that life-preservers cause drowning, because vehicles with life-preservers cause drowning. Those vehicles being boats, naturally.
Babaganoosh here and this was such a treat, boost out another one asap