The February 1988 issue of Social Problems published the first major effort actually to measure the protective value of firearms in America, by estimating the extent to which guns are used for protection, and what the result of attempted protective uses is. That study, by Florida State University criminologist Gary Kleck--and summarized in the July 1988 issue of American Rifleman--relied upon several national and state surveys to estimate that nearly one million adults each year use firearms for protection from criminals. The survey most relied upon by Prof. Kleck was conducted by Peter Hart Associates for an anti-gun organization, the National Alliance Against Violence (NAAV), since the Hart survey was, as of 1988, the most sophisticated at actually measuring protective uses of handguns, despite some limitations. For example, it asked only about protective use of handguns, so that long-gun estimates had to be made based upon various estimates on relative long-gun to handgun protective use.

Detailed evaluations of the Hart survey--for example, breakdowns by sex, age, ethnicity, and the like--were effectively prevented; published reports by the NAAV excluded all protective uses of handguns by non-owners, a significant exclusion. And, when the NAAV ceased to exist, it turned its files over to the National Coalition to Ban Handguns (now the Coalition to Stop Gun Violence), and the data have effectively vanished from the face of the earth, except for the summary results Kleck was able to obtain from Peter Hart Associates orally.

In addition to calculating how often guns were used for protection, Kleck used the National Crime Victimization Surveys (NCVS) to evaluate the utility of such protective use. Those data clearly indicated that using a gun for protection decreases the likelihood that a violent crime (particularly robbery and assault) will be completed or that the intended victim will be injured, compared to taking some other protective measures or taking no protective measure.

Kleck`s analysis was incorporated into his 1991 book, Point Blank: Guns and Violence in America, which won the 1993 Michael J. Hindelang Award from the American Society of Criminology--the nation`s preeminent criminological professional organization--given "for the book published in the past two to three years that makes the most outstanding contribution to criminology."

Anti-gun criminologists and public-health professionals have been denouncing Kleck`s research ever since and relying upon the NCVS surveys to estimate the number of times guns are used for protection.1 They`ve done this by attacking the single survey Kleck most relied upon and ignoring the fact that most of the surveys previously used by Kleck would have suggested about three-quarters of a million protective uses of guns (give or take a few hundred thousand). They thus acted as if it was the Hart survey versus the NCVS survey, and they preferred the latter, which would put the figure of protective gun uses in the 60-80,000 range.

The NCVS, however, has a number of serious flaws. Indeed, Kleck has noted that it is an outlier, yielding far lower estimates than any of at least 15 state or national surveys measuring protective gun use. This outlier status undermines its credibility. "The strongest evidence that a measurement is inaccurate is that it is inconsistent with many other independent measurements or observations of the same phenomenon . . . the gross inconsistency of the NCVS-based estimates with all other known estimates . . . would be sufficient to persuade any serious scholar that the NCVS estimates are unreliable."*

The NCVS has many flaws. It was developed to measure victimization--"getting people to report illegal things which other people did to them"--rather than "to get people to admit controversial and possibly illegal things which the Rs (respondents) themselves have done" to protect themselves.* Kleck notes, too, that the "NCVS is a nonanonymous national survey conducted by a branch of the federal government . . . on behalf of the U.S. Department of Justice, the law enforcement branch of the federal government."* With the ownership, carrying, and often the use of guns for protection possibly illegal, gun users would be discouraged from reporting too much such use to government personnel.2 In addition, while the NCVS may provide the best measure of criminal victimization, respondents are never directly asked whether they used a gun for protection. Respondents in the NCVS "are not even asked the general self-protection question unless they already independently indicated that they had been a victim of a crime."

Kleck, too, recognized the flaws in the Hart survey and the other surveys. However, instead of relying upon the NCVS, he and his colleague at Florida State, Marc Gertz, developed a more refined survey questionnaire, with a national sample of nearly 5,000, compared to the 1,000-1,500 normally used in survey research. There was a legitimate question about what would happen if surveys became more refined. Anti-gun scholars such as Phil Cook of Duke University expected a more refined survey to reduce the estimates of protective gun use, but still expected hundreds of thousands of uses, rather than the NCVS`s 60-80,000.3

Refining the survey meant correcting for previous flaws. For one thing, some earlier surveys asked whether someone had ever used a gun, making annual estimates hard to come by, and missing forgotten incidents. So Kleck and Gertz provided two time frames, one year and five years, with the one-year period being better, since less is forgotten. In addition, prior surveys had sometimes asked about household use, rather than individual use. Kleck surmised, and his survey strongly supports the conclusion, that for a variety of reasons--ignorance by some respondents of what others have done, reluctance to talk about possibly unlawful protective use by other members of the family despite a willingness to talk about one`s own--household measures, followed by household projections, dramatically undercount protective gun uses. Some surveys have asked only gun owners about protective use, but many protective uses of guns are by persons who do not own a gun at the time of the survey. This was true in the Hart survey and in the more refined Kleck-Gertz survey. And the Kleck-Gertz survey, unlike any others, asked about how many protective uses had occurred, while previous surveys necessitated assuming just one for each affirmative respondent.

The new survey suggests 2.2-2.5 million protective uses of guns each year, of which 1.5-1.9 million incidents involve the use of handguns. While the refinements increased protective gun use dramatically, the reports are still similar to the various other surveys taken over the past two decades, unlike the NCVS, which reports about one-ninth of the protective uses recorded at the low end of the surveys attempting to measure protective gun use. That women account for 46% of the reported self-defense uses of guns suggests to Kleck and Gertz that some possibly less clearly justified protective uses by men are left out; that is, with female gun ownership relatively low and men more apt to be in criminal victimization situations, Kleck and Gertz suspect that there may actually be more male protective uses, but that those uses are in the area Cook and others have suspected might be "mutual combat" sorts of defensive use, where determining who was the initial aggressor could be difficult.

In addition to disproportionate protective use by women, relative to their rates of gun ownership, Kleck and Gertz found that a "disproportionate share of defenders are African-American or Hispanic compared to the general population and especially compared to gun owners. Additionally, defenders are disproportionately likely to reside in big cities compared to other people . . . ."The disproportionate need for protective gun use by inner-city minorities--who are often discriminated against in arrests and prosecutions--is important when considering current research suggesting the persons be disqualified from lawful gun ownership due to felny arrests or misdemeanor convictions. Such a policy could disarm upwards of half of young, inner-city adult males, persons with the least effective police protection and the greatest tendency to protective gun use.

Just under one-quarter of protective gun users said they fired a shot, with half believing they had struck the criminal. The other side tries to refute that by citing how many more persons that would leave injured by gunfire than are treated in hospitals--unconvincing since most minor injuries to criminals would not lead them to seek medical care--and then claiming that makes the whole survey indefensible. On the other hand, it would suggest marksmanship accuracy rate about three times what would be expected and is based on too small a portion of the sample to be deemed reliable. A substantial minority of those using guns for protection also reported perceiving the situation as serious enough so that a victim might have died if a gun had not been used for protection. As with the injury figure, the keys are the perception of the protective gun user, wishing to justify to himself and the questioner the use of deadly force, and the small numbers in that portion of the sample.

Frequency of Defensive Gun Use

While Kleck and Gertz recognize that telescoping--remembering as less than a year ago a protective gun use which actually occurred somewhat more--might reduce gun use to 2.1 million incidents annually, they note that their survey also missed some incidents because adolescents were not respondents, and, as a telephone survey, households without telephones would have been excluded. Those households are disproportionately low income persons who are more likely to be crime victims, and thus be in a position to use guns protectively, and rural Americans, who have higher levels of gun ownership and are more distant from the nearest police officer.

As Kleck and Gertz note, whatever the precise figure, protective gun use is far more common than misuse of guns in victimization, particularly since the NCVS reports on the latter exaggerate such use. "[T]he NCVS estimate of `gun crimes` overstates the number of crimes in which the offender actually used the gun," since the respondents are reporting whether they thought a gun was present, and the "victims are not asked why they thought the offender possessed a gun or if they saw a gun."*

Because Kleck and Gertz used a large sample, their analysis is based on 213 respondents reporting actual gun use for protection.4 Although the 213 is a large enough sample for projecting annual protective gun use, further breakdowns--fractions of the 213--are more problematic, because the small numbers make details less reliable. Curiously, some critics of the Kleck/Gertz survey have used the admittedly less reliable breakdowns as a basis for rejecting the overall estimate of about 2.5 million protective uses of guns. This would be like rejecting an overall survey on the number of gun-owning households nationally because analyses from the same survey about the nature of low-income, educated, rural unmarried gun-owning households presented some odd possible conclusions.

Guns were most commonly used for protection against burglary, assault, and robbery. As was true with the NCVS surveys, using guns for protection is rarely associated with loss of property or injury, and in the few instances where injury occurred, it preceded rather than followed protective gun use. Thus there is no indication that protective gun use provokes criminals to further violence. Contrary to hypotheses by anti-gunners that guns are used protectively more in "easy" circumstances, gun users "were more likely than other victims to face gun-armed criminals and multiple offenders."* And protective gun use--similar to criminal misuse --involves handguns about 80% of the time.

It should be noted that Kleck and Gertz are willing to endorse some restrictive gun laws, if carefully aimed at criminals. "(P)rohibitionist measures," they write, "whether aimed at all guns or just handguns . . . (would) discourage and presumably decrease the frequency of DGU (defensive gun use) among non-criminal crime victims because even minimally effective gun bans would disarm at least some noncriminals. The same would be true of laws which ban gun carrying. In sum, measures that effectively reduce gun availability among the noncriminal majority also would reduce DGUs that otherwise would have saved lives, prevented injuries, thwarted rape attempts, driven off burglars, and helped victims retain their property."* As with Kleck`s earlier studies, then, the conclusion remains that general efforts to restrict gun availability, inside or outside the home, are likely to be counterproductive in terms of ensuring the safety of the law-abiding citizenry.

Anti-gunners understandably are aghast at the Kleck-Gertz survey, and the ever-increasing evidence that guns are effectively used for protection much more than they are criminally misused. Thus, when HCI`s Center to Prevent Handgun Violence decided to show that its disdain for the Second Amendment could almost be matched by its disdain for the First, and it asked the Federal Trade Commission to prohibit ads suggesting handguns were an effective means of protection, they said that Kleck`s findings had been denounced in the scientific community. That ignored, of course, the award Kleck had won from the American Society of Criminology. The more denunciatory of the two sources it cited was to a book co-authored by an employee of HCI`s Center, with a second co-author a close associate of a second employee of HCI`s Center.

More significantly, the Journal of Criminal Law and Criminology invited Marvin Wolfgang to submit comments on the Kleck-Gertz study.5 The late Prof. Wolfgang was one of the most prominent criminologists in the world. His "round criticism" speaks for itself:

"I am as strong a gun-control advocate as can be found among the criminologists in this country. If I were Mustapha Mond of Brave New World, I would eliminate all guns from the civilian population and maybe even from the police. I hate guns--ugly, nasty instruments designed to kill people . . . . What troubles me is the article by Gary Kleck and Marc Gertz. The reason I am troubled is that they have provided an almost clear-cut case of methodologically sound research in support of something I have theoretically opposed for years, namely, the use of a gun in defense against a criminal perpetrator . . . . I have to admit my admiration for the care and caution expressed in this article and this research.

"Can it be true that about two million instances occur each year in which a gun was used as a defensive measure against crime? It is hard to believe. Yet, it is hard to challenge the data collected. We do not have contrary evidence. The National Crime Victim Survey does not contravene this latest research . . . .

"The Kleck and Gertz study impresses me for the caution the authors exercise and the elaborate nuances they examine methodologically. I do not like their conclusions that having a gun can be useful, but I cannot fault their methodology. They have tried earnestly to meet all objections in advance and have done exceedingly well."

Less willing to accept Wolfgang`s endorsement, the Clinton/Reno Justice Department instead funded to NSPOF, originally to be conducted by the anti-gun Police Foundation with the aid of Kleck. Kleck was eased out by the Justice Department, and anti-gun criminologist Phil Cook substituted. When the results nonetheless confirmed the Kleck/Gertz survey, the Police Foundation and the Justice Department essentially renounced the finding of their own study. Professor Wolfgang will be missed.

Footnotes:

1. David McDowall and Brian Wiersema, "The Incidence of Defensive Firearm Use by US Crime Victims, 1987 through 1990," American Journal of Public Health 84 (12):1982-1984 (December 1994). They also prefer studies that look exclusively at self-defense killings compared to other gun-related deaths. But, as Kleck and Gertz note, the large number of protective gun uses is "too serious a matter to base conclusions on silly statistics comparing the number of lives taken with guns with the number of criminals killed by victims. Killing a criminal is not a benefit to the victim, but rather a nightmare to be suffered for years afterward." Since only about one-thousandth of protective gun uses involve killing a criminal, "The number of justifiable homicides cannot serve as even a rough index of life-saving gun uses . . . (and) can shed no light on the benefits and costs of keeping guns in the home for protection."

2. Still worse, of course, would be an attempt to measure protective use of guns by relying upon reports to police. Where gun use prevents a crime from being completed, the crime itself is often unreported--indeed, nationally, only about half of victimizations reported to NCVS indicates there was a police report as well. And police rarely ask about, and never systematically record, protective measures taken by victims reporting crimes. Thus a Centers for Disease Control and Prevention-funded study by Arthur Kellermann which reviewed some home invasion crimes reported to the Atlanta police was not worth the paper the Journal of the American Medical Association printed it on.

3. When Cook himself helped analyze another more refined survey, aimed at testing the accuracy of the Kleck/Gertz survey, which reached precisely the same conclusions, he concluded that surveys of relatively rare events were worthless, and that respondents were apparently lying to survey researchers--;apparently about the same portion lying in about 15 different surveys--;either to pull their leg, or because under the influence of drugs or alcohol, or to make gun ownership look beneficial. He voiced none of these objections to using surveys to measure rare events like protective or accidental gun use prior to results from the so-called National Survey of Private Ownership of Firearms (NSPOF) coming in. Philip J. Cook and Jens Ludwig, Guns in America: Results of a Comprehensive National Survey on Firearms Ownership and Use: Summary Report. Washington, D.C.: Police Foundation, 1996.

4. Interestingly, among those most critical of Kleck`s analysis, as being based on too small a sample, is Douglas Weil, of Handgun Control`s Center to Prevent Handgun Violence. Weil once based an entire article on how NRA members feel about the NRA on a survey including only 102 such persons. Apparently, to HCI, 213 is too few, but 102 is plenty, even if only 30 persons in a survey of that size (about 600 respondents) should have been NRA members, if the sample were honest and randomly selected.

5. Marvin E. Wolfgang, "A Tribute To A View I Have Opposed," Journal of Criminal Law and Criminology; 86(1): 188-192 (Fall 1995).

*Gary Kleck and Marc Gertz, "Armed Resistance to Crime: The Prevalence and Nature of Self-Defense with a Gun," Journal of Criminal Law and Criminology 86(1):150-187 (Fall 1995). The survey is also reported in chapter 5 of Gary Kleck, Targeting Guns: Firearms and Their Control (N.Y.: Aldine de Gruyter, 1997), a revision of his award-winning Point Blank.