This figure, often cited by the NRA, comes from a national telephone survey of 4,977 adults by Gary Kleck and Marc Gertz in 1995. About 1.1 percent of respondents reported using a gun for self defense an average of about 1.5 times each in the previous 12 months. Kleck and Gertz then extrapolated the sample rate to the total population in the United States and came up with an estimated 2.16 million self defensive gun uses per year. (Note: Over time, the Kleck and Gertz estimate has crept up to 2.5 million self defensive gun uses per year.)

So what’s wrong with this estimate? What’s wrong with this estimate is what epidemiologists call “false positives” and “false negatives.” In this case, a false positive would be a person who reports a self defensive gun use when there wasn’t one. A false negative would be a person who reports no self defensive gun use when in fact the person had used a gun in self defense.

The fact is people don’t always tell the truth when responding to self-reporting surveys, e.g., who they voted for in the last election; whether they were employed or unemployed in the last year; even their age, height, and weight.

Misreporting is a potential problem with all self-reporting surveys, but it is especially important in surveys that attempt to estimate the incidence of relatively rare events. For example, if the true rate of self defensive gun use were 1 percent, then in a survey of 5,000 individuals, 50 individuals would have used a gun in self defense in the last year and 4,950 individuals would not have used a gun in self defense. This means that a total of 4,950 respondents could report a false positive whereas only 50 respondents could report a false negative. Even if the misreporting rate for false positives and false negatives were the same, there would be 99 false positives for every false negative in the sample. This would greatly bias the results of the survey.

The bias could be even greater than 99 to 1. Regarding the Kleck and Gertz survey, do you consider it more likely that a person would exaggerate the use of a gun in self defense (a false positive) or downplay the use of a gun in self defense (false negative)?

Furthermore, it should be noted that even if the true incident rate of self defensive gun use were zero, if just 1 percent of the respondents in the Kleck and Gertz survey responded untruthfully, the number of false positives would be nearly equal to the number of actual respondents who claimed to have used a gun in self defense in the last year (1.1 percent). In dealing with estimates of rare events, just a small percentage bias in reporting can lead to huge overestimates of the true incident rate.

This fact is illustrated by a May 1994, ABC News/Washington Post survey of 1,500 adults in which 10 percent of respondents said that they had seen a spacecraft from another planet. Of these 150 respondents, 6 percent went on to say that they had also been in direct contact with aliens from other planets. Extrapolating to the total U.S. population using Kleck and Gertz’s methodology, one would conclude that up to 13 million Americans have seen an alien spaceship and 785 thousand Americans have been in direct contact with aliens.

The results of the Kleck and Gertz survey are absurd in other respects. For example, 8 percent of respondents who reported using a gun in self defense in the last year claimed to have killed or injured their attacker. If this were true, 2.5 million self defensive gun uses would lead to 200,000 gunshot deaths and injuries each year. Yet, this figure is greater than the total number of fatal and nonfatal shootings in the U.S. each year – by a factor of 2.

In the U.S., there are about 30,000 gun deaths and an estimated 75,000 nonfatal gun injuries each year, typically victims of criminal assaults, suicide attempts, and unintentional shootings. There is no evidence from police records or hospital records of 200,000 criminals being shot by guns each year. Evidently, none of these 200,000 shootings were reported to the police and none of the criminals who were shot went to the hospital for treatment of their wounds.

Furthermore, the FBI’s Uniform Crime Reports indicate that there are only about 200 justifiable gun homicides by private citizens each year. If we are to believe the Kleck and Gertz survey results, for every 1,000 criminals shot by private citizens in self defense, only 1 would be killed. It is simply not credible. (Or else the private citizens are pretty bad shots.)

It cannot be denied that civilians with guns do on occasion prevent crimes. But such use is greatly exaggerated by gun right’s advocates. The bottom line is that the risks of gun ownership far outweigh the benefits. For each justifiable gun homicide in the U.S., there are about 60 criminal gun homicides, 85 gun suicides, and 5 deaths due to unintentional shootings.

Sources: Hemenway, D. 2004. Private Guns: Public Health. The University of Michigan Press. Cook, P. J. and Ludwig, J. 2000. Gun Violence: The Real Costs. Oxford University Press. Sugarmann, J. 2001. Every Handgun Is Aimed at You. The New Press, NY.