Accidental v. Intentional, Fatal v. Non-Fatal, Gunshots, 2000-2007 in the USA

Spread the love

From 2000 through 2007, inclusively, approximately 780 thousand people in the United States took a bullet. Most of them were wounded by another person in an act of violence. A fairly large number were wounded by accident, killed by a bad guy, or killed themselves. A small number died in a shooting accident, tried to kill themselves but messed up, or were wounded or shot by a cop. Here’s the data culled form the CDC databases on injuries and deaths in the US, in crude rate per 100,000:


i-800dc0f1839d6c3ed20710acf0d3d1ae-US_shot_wounded_decade-thumb-500x430-64096.jpg

The numbers for the year 2000 are from a somewhat different set of data and I would be cautious about comparing them to later years at any fine scale. Having said that, we see very few trends in these data, other than a slight decrease in accidental shootings overall (both fatal and non-fatal).

Here’s something very interesting about these numbers. Different kinds of shooting events have very different rates of death (as opposed to non-fatal injury). Over this decade of data,

Accidental shootings are fatal 4% of the time.
If you are shot by a bad guy, you get killed 21% of the time.
Attempts at suicide by gun that are “successful” 83% of the time.
When the cops shoot you, they kill you: 27% of the time.

Have you read the breakthrough novel of the year? When you are done with that, try:

In Search of Sungudogo by Greg Laden, now in Kindle or Paperback
*Please note:
Links to books and other items on this page and elsewhere on Greg Ladens' blog may send you to Amazon, where I am a registered affiliate. As an Amazon Associate I earn from qualifying purchases, which helps to fund this site.

Spread the love

34 thoughts on “Accidental v. Intentional, Fatal v. Non-Fatal, Gunshots, 2000-2007 in the USA

  1. You seem to have missed the largest number – the number of bad guys killed or wounded by law-abiding citizens in acts of self-defense. I’ve always been assured that it’s a very statistically significant number.

    1. Statistically significant and large are two different concepts.

      Significant mean non-random. It means that there is enough data to demonstrate that the result is significant (i.e. real in some sense). 4% can be statistically significant if there is enough data to show that the correct value is within a narrow range.

      Large just means large and is hard to pin down. It might mean absolutely large, like 80% or 90%. It might mean more than half, i.e. 51%. It might mean the largest fraction of several, e.g. 32%, where the other fractions are 20%, 17%, 15%, 9%, 7%

      With that in mind, I strongly doubt that the number of bad guys killed or wounded by a law abiding citizen the largest number. Can you demonstrate with actual numbers what your claim is?

  2. And don’t forget good guys accidently killed by other good guys, cuz they thought they were bad guys…

    And to top it all off, we should have a Gun Nutz Killed By Other Gun Nutz, becasue they thought they weren’t Gun Nutzy Enough.

  3. LOL. It’s significant to whomever it happens to, I can assume, but since it hardly ever happens (except as anecdotal accounts by anonymous commenters on blog posts!).

  4. I have lived in the nyc area my entire life. I have never read about or seen a news story in which someone defended his home with a firearm. I am not saying it never happens, but I have never come across this story. It is possible that criminals are smart and avoid homes they know have firearms in them. I have a feeling it has more to do with home invasions actually being rare in the first place.

  5. There are about 1 million cases per year of where people in the U.S. defend themselves with a gun. Estimates of this number range from around 500,000 to 2.5 million.

    Less that 1% of all homicides and accidental deaths in the U.S. are due to firearms.

    When the District of Columbia banned handguns, the murder rate increased significantly (73%), while that in the U.S. decreased by 11%.

    During the Chicago hand-gun ban, murders committed with hand-guns were 40% higher than before the ban.

    Norway has just over a third the gun ownership rate of the U.S. (35%), but has a fractional rate of gun-related homicides (4.2%).

  6. Every year I travel from Canada to hunt deer in Michigan.

    In Canada, you are legally required to store and transport firearms locked or disassembled.

    This seems to make a lot of sense to me because a kid is going to find it harder to shoot themselves or someone else if they happen upon a locked or disassembled firearm.

    Yet I am amazed at the derisive comments I hear about trigger locks when I am down there.

    And lets face it: storing a loaded weapon is outright stupid. I’m not even sure the military lets you get away with that.

    How can you be opposed to safe storage of firearms? Isn’t safe storage basic parenting?

  7. How can you be opposed to safe storage of firearms?

    Because your dick extender doesn’t extend your dick nearly as effectively if it’s locked up and disassembled.

  8. Rob, the total number of violent crimes in the US in a given year is 6 or 7 million. Are you saying that in one in 7 (or more) cases a gun owner whipps out gun and solves theproblem?

    You need to provide sources for these statistics to be taken seriously. I happen to know the sources of some of them and they are highly disputed. But I’ll let you go ahead and dig up the references just in case I’ve got different sources.

  9. Azkyroth @7: I’m sure you know this, but a gun safe doesn’t mean shortening or disassembling your dick extender; it just means locking it away to keep children from getting their hands on it.

    What this suggests about the motivations of dick-extender owners who let children handle their dick extenders, I leave as an exercise to the reader.

    BTW, my own dick extenders are all on the short side, mostly because I don’t need that much extension in the first place.*

    Greg – on the plus side, the numbers don’t seem to be showing a huge, godawful increase, at least. Actually, they don’t appear to be fluctuating much at all.

    ==

    * Good heavens, a whole new subset of Freudianism has just opened itself up to me: My phallic symbol is smaller than your phallic symbol!

  10. Of course not Greg. That is not what I said. There are upwards of 500,000 incidents where people defend themselves with a gun. In the vast majority of cases, no violent crime is committed. You totally made up the 1/7 statistic by conflating two completely unrelated numbers.

    http://www.ncbi.nlm.nih.gov/pubmed/9591354

    By defend, I mean arm themselves when they believe they are threatened. I hope the CDC is a reliable source.

  11. I did read the abstract, Rob. It requests particular attention to the definitions and methodology, which I don’t currently have access to. I’ll happily read it if someone wants to send it to me.

    You’ve read the whole thing before arguing based on it, right, Rob? Pass it along to Greg. He’ll forward it to me.

    I am also aware that violent crime reporting includes attempted crime and includes robbery. That means that Greg’s point about the 1 in 7 stands. It also means that if you’re going to accuse Greg of totally making something up because you don’t understand crime statistics, it gets much more tempting to swat you when you say something stupid.

    The point about the tinfoil hats is perfectly serious, by the way. As far as I can tell, gun nuts are scared of every little thing. That’s why the definition of “intruder” is critical to your argument.

  12. Rob: “There are about 1 million cases per year of where people in the U.S. defend themselves with a gun. Estimates of this number range from around 500,000 to 2.5 million.”

    Greg: “Rob, the total number of violent crimes in the US in a given year is 6 or 7 million. Are you saying that in one in 7 (or more) cases a gun owner whipps out gun and solves the problem?”

    Rob: “Of course not Greg. That is not what I said. There are upwards of 500,000 incidents where people defend themselves with a gun.”

    FOS, Rob.

    Anyway, I love this study. Its been sitting in my pile of stuff to blog about forever.

    Every year, almost two million people, in their homes, whip out their gun for no demonstrable reason. Half a million times they whip out their gun and later claim to have seen an intruder. About the same number of times, based on asking a gun owner if they ever actually scared someone off, an intruder, it is claimed, was scared off.

    But of course the numbers are bullshit. They interviewed 5,238 people of whom a mere 1,678 claimed to have a gun in the home. (That’s about right, by the way: About one third of homes have firearms in them, thus the severe risk to children across the country [ http://tinyurl.com/3jmnojs ]) Then, from this sample the SELF REPORTED data are used to extrapolate. So, a couple of hundred people reported doing something witha gun, and this is extrapolated to several hundred thousand in the population. Funny how people will claim that “polls are always wrong” but then swallow extrapolations that are convenient.

    Anyway, I’m sure the numerical estimates are reasonable (but need to be taken with a grain of salt) but the reason we are talking about a study from 1994 is because since then sel-reported data as a reliable source of information has been largely discredited.

    Still, taking the results at face value, they clearly prove my point. We need to assume the claims are exaggerated, possibly by a factor of two or four. For every actual case of an intruder scared off, there is a very large number of cases of a home owner waving their gun around. and, there is no way that one can be certain much of the time that the gun is what scared off the intruder (see this: http://tinyurl.com/3nvedd9 and http://tinyurl.com/3jg4b9c and http://tinyurl.com/y8tjbwc )

  13. It would be interesting to know what facts would change Greg, Stephanie Z’s and Rob’s minds. So far, Rob is the only one to put some facts on the table. For example, how many defensive uses of a firearm would make change Greg or Stephanie’s attitude toward gun ownership? And from what sources? How high would the firearm death rate have to increase to change Rob’s mind? If you can’t say what would change your opinion, you can’t engage in dialogue.

  14. It would be interesting to know what facts would change Greg, Stephanie Z’s and Rob’s minds.

    Change my mind about what?

    So far, Rob is the only one to put some facts on the table.

    Are you on drugs?

    For example, how many defensive uses of a firearm would make change Greg or Stephanie’s attitude toward gun ownership?

    So, you are asking, what is the number of defensive uses of a firearm that would make me stop thinking that US citizens have the right to bear arms?

  15. (I’m thinking of the shotgun. Adds a whole new dimension to deer hunting. Duluth, BTW, is having an extended “city deer” hunt this year. Could be a good fit.)

  16. Greg,

    The authors of the article from CDC state 95% confidence levels. I am going to have to assume that they did their statistics correctly. I also assume that the sociological journal in which the article was published was peer reviewed. Not being either a statistician or a sociologist, I would necessarily need to assume the authors are making reasonable argument, just as they would have to assume the same for me, as they are not chemists.

    You cite an article that states: “As identified by police and medical examiners, they randomly selected 677 cases of Philadelphia residents who were shot in an assault from 2003 to 2006. Six percent of these cases were in possession of a gun (such as in a holster, pocket, waistband, or vehicle) when they were shot.” So here we have 40 cases where gun-carrying people got shot, whereas the other 637 people did not have a gun. What does this tell us? From this, they conclude that people who have a gun are 4.5 times more likely to be shot in an assault than those without a gun, compared to a control population that did not get shot or assaulted, but presumably had the same rate of carrying a gun. What exactly does this prove? Six percent of the total number of the sample population who were shot in assaults also carried a gun. Ninety-four percent did not have a gun. (BTW, the control population was self-reporting.)

    My point in the OP is that firearms are responsible for a tiny fraction of accidental deaths: just 0.5% (613 of 123,706) of all unintentional fatal accidents, again in a data obtained from the CDC.

  17. Greg, it would go in a lock-box, as it would be one of the handguns. Ben and I had a discussion about the desirability about having a gun on one’s hip in the unlikely event of laying the bike down. None of the eventualities, which mostly involved losing the gun, were appealing.

    I’ll have to look at the Duluth hunt. “City” usually means archery. Fewer worries about what will stop the projectile before it hits someone.

  18. Rob, I’ll take that as a “No, I didn’t read the study before I decided to cite it, despite the abstract going to the unusual step of cautioning about definitions and methodology.” And why are you lecturing Greg about the statistical portion of the study, when that’s the part he already said was likely sound?

  19. Yes, there is a very good chance that, based on the fact that they were self-reported and having to do with an emotional topic, the numbers are bullshit. You can do all the excellent statistical manipulation you want on bullshit numbers. That doesn’t make them any more valid.

    But Greg explained all that already.

  20. The authors of the article from CDC state 95% confidence levels. I am going to have to assume that they did their statistics correctly.

    I don’t know what statistics they used, so I won’t judge or assume. However, really, self-reporting data is highly unreliable and there are reasons to believe that these data in particular would be biased in one direction.

    Not being either a statistician or a sociologist, I would necessarily need to assume the authors are making reasonable argument, just as they would have to assume the same for me, as they are not chemists.

    I’m not a statistician but I’m trained in this area, have taught it at the graduate level,and have a PhD in Anthropology, so if we are going for qualifications, I don’t know as much about chemistry as you do but I know as much about this sort of study as the authors do. So if you are going to take their argument based on Authority, give me the same benefit. Although I admit my PhD is from a small and insignificant school out east.

    The study I cite is a reasonable study that produces a conter-intuitive result if you believe, in advance, in the unmitigated good of packing heat. That is what I think it is useful for.

    My point in the OP is that firearms are responsible for a tiny fraction of accidental deaths

    That does not matter at all. Not even a little bit. Lung cancer is responsible for only a small fraction of cancer deaths. Therefore, don’t do lung cancer research? Therefore, no one with lung cancer is really that sick? Therefore, causes of lung cancer (i.e. smoking) are not relevant? The argument you make is utterly irrelevant and kind of embarrassingly dumb. Don’t feel bad, a lot of people make that sort of argument. But it really is not relevant at all.

    Had the uncle in the post that started this all said, of his niece being shot in the head by her dad unjamming a pistol across the living room from her “well, she had a higher chance of getting in a car wreck so this does not matter” … what would that have meant?

  21. Cool, Greg.

    I agree that self-reported data are unreliable, but a large number of sociological studies rely on such data, and this is one of them. I originally posted as a response to the first couple of comments, and someone who called self-defense woundings/deaths anecdotal.

    The remainder of the numbers I cited are because I get sick of people giving more significance to accidental gun deaths than they deserve. Some are preventable, some are not, given that guns exist, but they account for a VERY small fraction of accidental deaths.

    I personally wouldn’t call 1/4 a “small fraction,” especially when the total contains 10-20 categories, and that 1/4 class is the most highly populated. That would be embarrassingly dumb, and would be like calling cancer deaths a “small fraction” of all deaths. Half a percent is a small fraction, but not 25%.

  22. On raw numbers alone:

    – the number of deaths was on the decline pre-Bush and pre-FOX propaganda

    – the number of deaths increased after 9/11

    – the numbers topped 100,000 after Hurricane Katrina (remember the stories of NON-accidental shootings in Louisiana?)

    Correlation does not prove causation, but this does not surprise. I would not be the least bit surprised to see the numbers continue to increase after 2008 both because of Obama being president and because gun companies have been “exporting” guns to Mexico (making sales easier, actually) and the blowback is hitting the US, not just Mexicans.

    http://wonkroom.thinkprogress.org/2011/04/22/mexico-gun-makers/

Leave a Reply to Stephanie Z Cancel reply

Your email address will not be published. Required fields are marked *