How Many Times A Year Do Americans Defend Themselves With Guns?
- Thread starter Biker
- Start date
- Status
"With the avalanche of horrific news stories about guns over the years, it's no wonder people find it hard to believe that, according to surveys (one I conducted for 2002 for my book, "The Bias Against Guns," and three earlier academic surveys by different researchers published in such journals as the Journal of Criminal Justice) there are about two million defensive gun uses each year; guns are used defensively four times more frequently than they are to commit crimes."
http://www.foxnews.com/story/0,2933,107274,00.html
According to the National Self Defense Survey conducted by Florida State University criminologists in 1994, the rate of Defensive Gun Uses can be projected nationwide to approximately 2.5 million per year -- one Defensive Gun Use every 13 seconds.
http://www.pulpless.com/gunclock/noframedex.html
http://www.foxnews.com/story/0,2933,107274,00.html
According to the National Self Defense Survey conducted by Florida State University criminologists in 1994, the rate of Defensive Gun Uses can be projected nationwide to approximately 2.5 million per year -- one Defensive Gun Use every 13 seconds.
http://www.pulpless.com/gunclock/noframedex.html
Monkeyleg
Member.
The US Department of Justice puts the number of self-defense uses of guns at about 500,000 a year.
A 1994 study commissioned by the US Bureau of Justice Statistics put the figure at 1.5 million.
And, as mentioned, Gary Kleck estimates the number could be as high as 2 million or more.
So, it's probably somewhere in between.
A 1994 study commissioned by the US Bureau of Justice Statistics put the figure at 1.5 million.
And, as mentioned, Gary Kleck estimates the number could be as high as 2 million or more.
So, it's probably somewhere in between.
TallPine
Member
My gun is used (carried) every day. Just because I don't have to shoot something doesn't mean it isn't being used defensively.
Just like seat belts are not only "used" when you have a wreck, but you use (wear) them every time you drive and buckle up.
I know that's not what you're looking for, but I guess I'm tired of having to justify safety precautions. And if I go outside at night to see what the dogs are barking at, I sure feel better having some means to defend myself and my family/property.
Just like seat belts are not only "used" when you have a wreck, but you use (wear) them every time you drive and buckle up.
I know that's not what you're looking for, but I guess I'm tired of having to justify safety precautions. And if I go outside at night to see what the dogs are barking at, I sure feel better having some means to defend myself and my family/property.
RNB65
Member
I have no idea.
I do know that the US is far safer than the media would have you believe. Most Americans will go their entire lives without ever once having to defend themselves or a loved one with a firearm or any other type of weapon.
But it's an awfully nice option to have if you ever need it.
I do know that the US is far safer than the media would have you believe. Most Americans will go their entire lives without ever once having to defend themselves or a loved one with a firearm or any other type of weapon.
But it's an awfully nice option to have if you ever need it.
PlayboyPenguin
member
That is a high number. Is that 2,000,000 different idividuals or just 2,000,000 instances. I doubt it would make much change in the numbers either way.
When you take into account the US population is around 300,000,000 that means you have a 1 in 150 chance of needing a gun every time you leave your home.
That makes this Walther hanging on my belt look alot less paranoid now (but still every bit as stylish).
PS: That number doesn't include law enforcement does it?
When you take into account the US population is around 300,000,000 that means you have a 1 in 150 chance of needing a gun every time you leave your home.
That makes this Walther hanging on my belt look alot less paranoid now (but still every bit as stylish).
PS: That number doesn't include law enforcement does it?
Art Eatman
Moderator In Memoriam
The 2,000,000 number comes from Prof. Gary Kleck's work. He's a statistician at FSU and a card-carrying ACLU member.
Some eight-ish years back, he did the largest telephone survey ever made in the U.S. I disremember, but it was somewhere around 8,700 or so. It was a statistical scatter of ZIP Codes across the country, in all states.
Anyhow, the use of a firearm, whether a lie in the case of, "I have a gun!" or a display, a warning shot or an actual shot at a Bad Guy, prevented at least 800,000 to as many as 2.4 million crimes per year, nationwide.
At the time, per federal statistics, some 600,000 crimes involving firearms were reported.
Art
Some eight-ish years back, he did the largest telephone survey ever made in the U.S. I disremember, but it was somewhere around 8,700 or so. It was a statistical scatter of ZIP Codes across the country, in all states.
Anyhow, the use of a firearm, whether a lie in the case of, "I have a gun!" or a display, a warning shot or an actual shot at a Bad Guy, prevented at least 800,000 to as many as 2.4 million crimes per year, nationwide.
At the time, per federal statistics, some 600,000 crimes involving firearms were reported.
Art
SoCalShooter
Member
Wow I am actually surprised at the actual number of defensive gun uses. the anti gun lobby sure would not want these stats around.
Librarian
Member
They deny they could be possible.
Kleck published summaries of earlier results in Point Blank: Guns and Violence in America, New York, Aldine de Gruyter, 1991. He published his National Self Defense Survey in 1995 - "Armed Resistance to Crime", Journal of Criminal Law and Criminology, Vol 86 Number 1, Fall 1995. page 150. The actual survey was done Feb - Apr 1993.
Phillip Cook and Jens Ludwig did a similar survey (NSPOF - National Survey of Private Ownership of Firearms) in 1996, and essentially confirmed Kleck's results on a different dataset. Cook and Ludwig are antis; they were unhappy about that result.
See also Kleck, Gary, Targeting Guns : Firearms and Their Control, New York, Aldine de Gruyter, 1997.
PlayboyPenguin
member
Wait, are these numbers based on surveys or actual crime data?
CNYCacher
Member
PlayboyPenguin said:
The 2M stat is per year, so it would be 1/150 each year, not each time you leave the home.
Then again, since non-gun-owners couldn't possibly use a gun in self-defense, the odds are more like 2M/(total gun owners) not 2M/(US Population)
Then again, most gun owners don't have a gun available 24/7, it's really more like:
2Million/(Gunowners in the us)(% of time the average gun owner has a gun available)
Lets take the widely accepted number of 80 million gun owners in the US. I am willing to bet that a very small percentage of them have a gun available 24/7. Let's say 10% of them CCW nonstop, and 80% have at least a gun in the home accessible, where they are maybe 12 hours a day, on average, and 10% have everything in a safe, or in storage, or buried in boxes in th attic. We come out to a VERY GENEROUS 50% of the time, the average gun owner has a gun available to them.
Now, we are talking 2M/80M(50%) or 1 in 40 years the average gun owner will need a gun for self defence. My friends, that is once or twice in a lifetime, which is definately odds that warrant full-time CCW. Chances are, you will need a gun to defend yourself at least once, maybe twice in your lifetime.
Since gun owners do not have a sign around their neck that makes them MORE likely to be attacked, you have to assume that these start are uniform across the US population. So you have to feel sorry for the other 220M non-owners and 40-million owners who will be caught without a gun at crunch-time.
Disclaimer: It's Quarter past 12 on the east coast and I have been hacking PHP all day.
Librarian
Member
Surveys, generally. That's why the numbers are controversial. The usual criticisms are
- if there are all those defensive shootings, where are the bodies?
- why don't the cops know about these?
- why don't we see these reported in the papers?
- one doesn't have to shoot someone to defend with a gun
- what would you tell the cops - a crime DIDN'T just occur? (Besides, not every crime gets reported; things like murders get noticed, things like robberies/muggings not so much.)
- if I don't tell the cops, how would the papers find out? (And, some of them are reported in the papers, you just have to look for them.)
PlayboyPenguin
member
Depends...if it was a home or property invasion I would definately call the cops. If I was in public, such as walking down the street, and had to show my firearm to stop a possible assailant then I would probably not involve the cops. They would probably find some way to charge me with brandishing if I reported it.
I am not writing off these numbers. I will take them for what they are worth and use them as an indicator...but I cannot hold then too highly since it is not a true scientific method. I would never argue tham as "fact" but I can always use them as a "polls have shown" type argument.
NineseveN
member
Wel, the work that Kleck and Gertz and then also Ludwig and Cook did was based on a scientific statistical analysis and survey. The results aren't definitive, but this wasn't just some wild-eyed guess work, all the data and methods were published for review. Not hard facts per se, but factual evidence (if you include your cite in your argument).
Wiley
Member
I can't give cites as it was a number of years ago but: The lowest figure I ever saw was from an anti survey using crime/police reports. It was 760,000 DGU's/year.
So even at the lowest reported number, the number of DGU's is 25 times the number of firearm deaths, even including suicides. Exclude suicides and the number is close to 51 times.
So even at the lowest reported number, the number of DGU's is 25 times the number of firearm deaths, even including suicides. Exclude suicides and the number is close to 51 times.
Zundfolge
Member
One stop shopping for the stats: www.gunfacts.info
Police call the person to call the "victim" and the guy who didn't call the "perp".
Call anyway ... if the bad guy gets to the phone first and calls then the police come looking for you and charge you with assault with a deadly weapon and it comes down to your word against his (and since you didn't call they won't believe you).
Police call the person to call the "victim" and the guy who didn't call the "perp".
Aside from what CNYCacher added, you're assuming that none of these DGUs occured at home?
Librarian
Member
It's as scientific as carefully designed telephone surveys can be; there are arguments about that, too. A text summary of Cook/Ludwig can be found here; you can estimate the scientific aspect for yourself.
See also the Google result for searching on "NSPOF Cook Ludwig" - lots of links to critical responses.
Probably the wisest position - but it's a bit better than most 'polls'; 'polls' have the connotation of less rigor than 'a survey conducted by a PhD criminologist at the University of X'. Either one, in reality, could be useful or just smoke.
Heywood Case
Member
A few thoughts from a survey researcher:
There's no magic dividing line between a "scientific survey" and an entertainment poll. As far as I can tell the surveys being cited are reasonably serious efforts but there are a few reasons to be careful with the "2.5 million" figure.
1. A survey of 8,700 is impressive but by no means the largest ever done in the US, even at the time it was done. Not even close.
2. The number of interviews doesn't tell you anything about the quality of the sample or the confidence intervals you can place around the estimates. When you are dealing with low base rates, funny things can happen to your "margins of error". Even with 8,700 interviews, how many defensive gun uses were reported? That is the number you are using to project to the entire U.S. population. In Cook and Ludwig's survey, that number was 45 out of 2,568 interviews. Generalizing to a population of 250 million from 45 cases is not something you can do with a high degree of precision.
3. Cook and Ludwig's survey only involved up to 6 callback attempts per number. This is nowhere near enough for an acceptable response rate. Cook and Ludwig reported a response rate between 44 and 59 percent, which I can guarantee you is not possible with only 6 callbacks (even in 1997). Something is way off here. They also employed quotas for "racial minorities" and gun owners, which is not a good way to get a random sample. The weights you have to apply after the fact also tend to inflate the margins of error by a substantial factor (which they didn't correct for).
4. All the above doesn't mean the population estimates from surveys like the one by Cook and Ludwig are wrong, but rather that it is hard to place a lot of confidence them. The survey-based reports I've seen by Lott and Kleck have similar problems. As do many of the ones you'll see cited by "anti" folks as well, not to mention the polls you see in the media. Surveys don't have to be perfect to produce useful information, but you can't make use of the information without a realistic assessment of precision.
5. The methods of the FBI's National Crime Victimization Survey (http://www.icpsr.umich.edu/NACJD/NCVS/) are far more defensible in terms of scientific rigor. The estimates of annual defensive gun use numbers from that survey are more on the order of 800,000 per year if I'm remembering right. There are probably good arguments as to why that number might be too low, but I feel more confident saying "at least 800,000" than "maybe as many as 2.5 million".
John
There's no magic dividing line between a "scientific survey" and an entertainment poll. As far as I can tell the surveys being cited are reasonably serious efforts but there are a few reasons to be careful with the "2.5 million" figure.
1. A survey of 8,700 is impressive but by no means the largest ever done in the US, even at the time it was done. Not even close.
2. The number of interviews doesn't tell you anything about the quality of the sample or the confidence intervals you can place around the estimates. When you are dealing with low base rates, funny things can happen to your "margins of error". Even with 8,700 interviews, how many defensive gun uses were reported? That is the number you are using to project to the entire U.S. population. In Cook and Ludwig's survey, that number was 45 out of 2,568 interviews. Generalizing to a population of 250 million from 45 cases is not something you can do with a high degree of precision.
3. Cook and Ludwig's survey only involved up to 6 callback attempts per number. This is nowhere near enough for an acceptable response rate. Cook and Ludwig reported a response rate between 44 and 59 percent, which I can guarantee you is not possible with only 6 callbacks (even in 1997). Something is way off here. They also employed quotas for "racial minorities" and gun owners, which is not a good way to get a random sample. The weights you have to apply after the fact also tend to inflate the margins of error by a substantial factor (which they didn't correct for).
4. All the above doesn't mean the population estimates from surveys like the one by Cook and Ludwig are wrong, but rather that it is hard to place a lot of confidence them. The survey-based reports I've seen by Lott and Kleck have similar problems. As do many of the ones you'll see cited by "anti" folks as well, not to mention the polls you see in the media. Surveys don't have to be perfect to produce useful information, but you can't make use of the information without a realistic assessment of precision.
5. The methods of the FBI's National Crime Victimization Survey (http://www.icpsr.umich.edu/NACJD/NCVS/) are far more defensible in terms of scientific rigor. The estimates of annual defensive gun use numbers from that survey are more on the order of 800,000 per year if I'm remembering right. There are probably good arguments as to why that number might be too low, but I feel more confident saying "at least 800,000" than "maybe as many as 2.5 million".
John
Librarian
Member
I'm not an NCVS participant, but having read the survey questions, the facial problem is that before they ask about how one resisted a crime, they ask if the respondent has been a victim of a crime; it's reasonable (but hair-splitting) to believe that some people, having successfully defended themselves in some manner, do not believe they are victims. Thus, they answer 'no' to the screening question, and the information question on methods never gets asked. I have no idea whether the interviewers actually ask the questions in such a way that the first 'no' precludes the followup discussion, but it seems the right/standardized way to do the interview.
Larger dataset, established relationship between researchers and respondents, followup every 6 months for 4 years, annual data compilation - it all feels more solid than a 1-time telephone survey.
Besides, coming from the Gubmint, it must be true. None of those nasty biased things from outside academics...
- Status