17

I came across this American Gun Facts web-site recently:

Every year, guns are used over 80x more often to protect a life than to take one!*

[...]

  • Based upon Kleck & Gertz estimates of 2.5 million defensive gun uses per year (see source #2). A similar study in 1994 under President Clinton (Source) found this number to be 1.5 million, which would result in guns being used over 47x more often to defend a life than to take one. According to the Bureau of Justice Statistics, about 436,000 crimes were committed with a firearm in 2008 (Source). This would mean guns are used 5.7 or 3.4 times (using Kleck or Clinton respectively) more often to defend against a crime than to commit one.

I wonder how truthful it really is? Even if the numbers presented are correct, it seems that they present it in a way that is not completely honest.

The principal claim (among many other related claims arguing that gun ownership is a net positive for american society) is that, for every incident where a gun is used in a suicide or assault, 80 or more incidents occur where a gun protects someone from harm (assault, rape, burglary...).

The claims being made do not seem entirely trustworthy as they frequently don't quote relevant context and mix and match unrelated statistics from non-comparable sources to produce a propaganda message rather than quoting comparable statistics that might enlighten a disinterested observer.

So is the principal claim even remotely verifiable with reliable and relevant statistics?

Community
  • 1
nctrnl
  • 181
  • 1
  • 5
  • What is the claim that you want examined? –  May 09 '13 at 16:41
  • 1
    Whether or not particular comparisons are fair is off-topic. This site analyzes particular claims for their correctness. Please see the [FAQ]. –  May 09 '13 at 16:43
  • I agree with Sancho. This appears to be looking for subjective discussion of the merits and flaws of the statistics selected, and not a request for verification of objective fact. – Beofett May 09 '13 at 16:52
  • 1
    Welcome to Skeptics. There are too many claims here, and you are asking for too much opinion about them. Select one claim, make this question about that, and ask more questions if you have more. – Oddthinking May 09 '13 at 17:14
  • @Oddthinking Tried a radical edit to see if focus on a single claim rescues question. I do think there is a notable and significant claim here, muddied by too many other (though related) claims. – matt_black May 09 '13 at 21:49
  • Further down the page (of the cite) there is a huge mass of statistical rubbish, much of which is debunked elsewhere on this cite. – DJClayworth May 10 '13 at 00:58
  • 5
    You might be interested in [this paper](http://www.stat.duke.edu/~dalene/chance/chanceweb/103.myth0.pdf) which debunks the 2.5 million self-defence incidents per year. – DJClayworth May 10 '13 at 01:02
  • 1
    Or [this](http://www.businessweek.com/articles/2012-12-27/how-often-do-we-use-guns-in-self-defense) – DJClayworth May 10 '13 at 01:05
  • 4
    If it's reopened, it turns out to be apples-to-oranges: the "80:1" ratio is from a (apparently controversial) 1995 paper that counts, e.g., the statement "get away -- I have a gun" as a "Defensive Gun Use" while the :1 counts only deaths and not, e.g., armed robbery. – Larry OBrien May 10 '13 at 01:06
  • @matt_black: I applaud the radical edit, but if I reopen it now, will it just be closed as a duplicate of [this one](http://skeptics.stackexchange.com/questions/14098/how-often-are-guns-used-by-private-citizens-in-self-defense-per-year-in-the-unit)? – Oddthinking May 10 '13 at 01:13
  • (Perhaps having the duplicate marker for later searchers is a positive improvement?) – Oddthinking May 10 '13 at 01:14
  • @Oddthinking It might, arguably, not be a duplicate on the grounds that it is a) asking about a very specific claim, and b) asking a more specific question (that is, the answer to the other question would feed the answer to this one but not be a complete answer). – matt_black May 10 '13 at 21:22
  • @matt: Okay, let's see... – Oddthinking May 11 '13 at 00:12
  • Armed bystanders don't seem to be a significant factor in resolving public shooting incidents, which is a much narrower scope than the question but should cause some thought about the reliability of the broader claim. See http://skeptics.stackexchange.com/questions/14123/what-is-the-value-of-armed-bystanders-in-preventing-gun-deaths?lq=1 – matt_black May 12 '16 at 17:07

1 Answers1

33

TL:DR Most studies find figures ten times lower than this one.

The "80 times more often" figure comes from a single study: "Kleck and Gertz:Armed resistance to crime: The prevalence and nature of self defense with a gun. Journal of Criminal Law and Criminology 1995 (Fall); 86:150–187", in which the authors estimated that there were 2.5 million occasions per year in which a gun was used in self-defense.

Here is a brief summary of the methodology:

Kleck and Gertz’s claim of 2.5 million defensive gun uses per year is derived from a telephone survey of 5,000 American adults conducted in 1992. Fifty-six respondents to this survey reported that they had used a gun in self-defense during the past year. Kleck and Gertz multiply the proportion of respondents in their survey who report a defensive gun use (X /5,000 = Y percent) by the number of adults in the U.S. (around 200 million) and the number of defensive gun uses equals 2.5 million per year.

Although published in a reputable journal, the figure has been widely criticised by other researchers on many grounds: statistical, methodological, and by comparison with other known figures.

Here is a typical criticism::

The primary problem is that, even if the Kleck and Gertz’s estimates were accurate, defensive gun use is a relatively rare occurrence in that only 1% of respondents reported a defensive gun use during the previous 12 months. As David Hemenway of Harvard University has pointed out, inaccurate reporting of these events by a relatively small number of respondents could lead to population projections that are orders of magnitude different from the true incidence.14 For example, if one-half of one percent of the survey respondents incorrectly reported that they had used a gun to defend themselves against a criminal attack during the past year, the estimated number of defensive gun uses would be twice as high the true number.

Other researchers find that attempts to cross-reference these figures with other better-established ones lead to contradictions:

[According to Kleck and Gletz] Guns were reportedly used by defenders for self-defense in approximately 845,000 burglaries. We know that there were 6 million burglaries in the year of the survey and in only 22% of these cases was someone certainly at home.Since only 42% of American households own guns, and since the victims in 2/3 of the occupied dwellings were asleep, the 2.5 million figure requires us to believe that burglary victims use their guns in self-defence more than 100% of the time.

Other researchers conducted a similar survey (obtining compatible figures) and then checked on the credibility of the reports:

Our closer examination of the [Defensive Gun Use] reports in the [out survey] suggests that almost half of the incidents appear to contain some internal inconsistency, or other- wise do not make sense. We are persuaded that surveys of this sort generate estimates that grossly exaggerate the true number of DGUs

This overview cites the National Crime Victimization Survey which places the number of defensive gun uses at 100,000, and other estimates come it at about 250,000 - ten times less than the figure quoted.

In short: while there is much disagreement over the true figure, the "80 times" figures is very much at the top end of the range of estimates. Most estimates are ten times lower.

DJClayworth
  • 57,419
  • 26
  • 209
  • 195
  • 40
    Additionally, the phrase "more often to protect life than to take it," is comparing apples to oranges. To achieve the claimed ratio, they are comparing *all* "defensive gun uses" as defined in Kleck & Gertz 1995 versus firearm *mortalities* (suicide, murder, accident). In other words, a gun used to *prevent* an armed robbery would count as a use to "protect life" but a gun used *for* an armed robbery would not be used in the calculation. – Larry OBrien May 11 '13 at 03:03
  • 4
    Good debunking of the specific claims (+1). I wonder, though, whether also including some other omitted information would also help put the claims in context. For example, the known number of accidental gun injuries or frequency of use in household arguments. Any fair risk analysis of gun ownership needs to look at *all* incidents of gun use: intended or unintended, good or bad. – matt_black May 11 '13 at 13:53
  • @matt_black I don't disagree with looking at the whole picture, but if we start to bring in other arguments we'll end up having the whole gun debate over again. – DJClayworth May 13 '13 at 19:30
  • 3
    @matt_black: That's the first thing that occurred to me. If an idiot lady leaves a loaded .22 around the house, and a 5-year-old kills his 2-year-old sister with it, would the "researchers" just say "Oh, well, that's the price of safety." – Mike Dunlavey May 13 '13 at 21:30
  • 13
    The other issue is the attempt to corrolate "Defensive gun usage" with "lives saved" is inherrently flawed comparison. We don't know that the presense of the gun in these situations lead to a life saved. For instance, using a gun to prevent a robbery counts as a DGU, but the robber likely would not have killd the home owners even if the buglary was successful. The robber may also have been scared off when he realized a home owner was present and willing to defend their home even if the owner didn't have a gun. Presence of a gun in such a situation doesn't gaurentee lives saved. – dsollen May 13 '13 at 21:43
  • 8
    It's really a case of omitting or misusing information. E.g. gun-ownership in Germany and UK is under 10%, yet homicide rate is 0.7 - 1, less than 1/4th of the US, conveniently omitted information. Also the violent crime statistic, in the US only serious offenses are considered a "violent crime", in the UK every crime against a person is a violent crime. By the US definition of violent crime, the UK would be somewhere between 300-400 in the infograph, a far cry from the stated 2000. A case of misusing the information and omitting the true definitions to further your agenda. – Dulkan May 11 '16 at 13:45
  • 1
    @Dulkan We're only focussing on the defensive use of guns question here. – DJClayworth May 11 '16 at 14:17
  • And I was referring to the entirety of the infograph linked in the question, how it misuses and omits information in several cases, the question itself being one of them. Thus, a comment with additional information. – Dulkan May 11 '16 at 14:33
  • 1
    I understand. I'm pointing out that we focus here only on the question being asked, not on other issues being raised by the same sources. – DJClayworth May 11 '16 at 14:35
  • 2
    Yet another problem with this survey: A and B get into an argument in a bar over whose football team is better. A reaches towards his hip. B reaches towards his hip. The Bartender reaches under the counter. That night, A, B, and the Bartender all record a Defensive Gun Use in their diaries. – Andrew Lazarus Feb 28 '18 at 00:43