Is a Gun an Effective Means of Self-Defense?

Summary Contrary to myth that using a gun in self-defense is more likely to result in injury or death to the victim or innocent bystanders and fail to successfully thwart the crime rather than the criminal, the evidence, as opposed to selective anecdotes, suggests the opposite. (Of course this doesn't mean that all people should have a gun, or a gun should be used in all life-threatening situations.)

Discussion Florida State University criminologist, Gary Kleck, analyzed data from the Department of Justice's National Crime Victimization Survey (1992-1998). Describing his findings on defensive gun use, in Armed: New Perspectives on Gun Control, New York:Prometheus Books (2001), Kleck writes:

"In general, self-protection measures of all types are effective, in the sense of reducing the risk of property loss in robberies and confrontational burglaries, compared to doing nothing or cooperating with the offender. The most effective form of self-protection is use of a gun. For robbery the self-protection meaures with the lowest loss rates were among victims attacking the offender with a gun, and victims threatenting the offender with a gun. For confrontational burglarly, attacking with a gun had the second lowest loss rate of sixteen self-protection measures, bested only by another mode of armed self-protection, threatening the offender with a nongun weapon." (p. 291) "[W]hile defensive gun use is generally safe, it does not appear to be uniquely safe among self-protection methods as data from earlier NCVS data suggested. Nevertheless, there does not appear to be any increase in injury risk due to defensive gun use that counterbalances its greater effectiveness in avoiding property loss." (p. 292)

Kleck summarizes the effectiveness and risks of victim self-protection measures, gleaned from NCVS data, in this table. 

Are the lower rates of injury and property loss with defensive gun use simply due to the victims having a more favorable set of circumstances than non-gun victims? Perhaps the criminals failed to surprise gun defending victims giving them time to ready their weapons. Kleck responds to this speculation by writing:

"These data indicate that victims who use guns for self-protection actually face less favorable circumstances than other victims, and that the post-self-protection injury rates for defensive gun use, low though they are, may still be misleadingly high compared to tother self-protection measures because victims who used guns faced tougher crime circumstances. More dangerous situations apparently prompt victims to adopt more dangerous self-protection measures. Two pieces of information available in the NCVS support this view. First, victims who used guns were substantially more likely than victims in general or victims using other self-protection measures to face offenders armed with guns — 32.7 percent of victims who attacked the offender with a gun, and 21.8 percent of those who threatened the offender with a gun, and 21.8 percent of those who threatened the offender with a gun, faced offenders with guns, compared to only 6.8 percent of all victims who used self-protection measures, and 2.2 percent of all victims. Second, victims who used guns were more likely to face multiple offenders — 33.2 percent of victims who attacked offenders with a gun and 34.5 percent of those who threatened with a gun confronted multiple adversaries, compared to 20.6 percent of all those who used self-protection measures, and 6.2 percent of all victims. These findings are consistent with the view that crime circumstances likely to appear more dangerous to victims are more likely to push victims into using guns. They are contrary to the speculation that crime outcomes are better for gun-wielding victims merely because other circumstances of the crime made successful outcomes more likely." (pp. 291-92)

Further supporting his contention Kleck writes:

"The simple percentage table results concerning robbery completion and injury rates are, however, supported by more sophisticated multivariate analysis of NCVS robbery incidents. In a logistic regression analysis, Kleck and Miriam Delone ("Victim Resistance and Offender Weapon Effects in Robbery," Journal of Quantitative Criminology 9 [1993]: 55-82) found that robbery victims who used guns in self-protection were significantly less likely to either be injured or lose their property than victims who used any other form of self protection or who did nothing to resist. This was true even when controlling for other characteristics of the robbery situation that could influence the effectiveness of defensive actions, such as the number of robbers, the number of victims, whether the robbery occurred in a private place, whether it occurred when it was dark, whether the robbers were armed, the age and gender of victims, and so on. Thus, there is no support for the speculation that gun defenders do well merely because of other advantageous crime circumstances associated with defensive gun use." (pp. 293-94)

Appearing to contradict Kleck's assertions are the pre-self-protection injury rates. 27% percent of victims were injured prior to taking any self-protection measures, but only 5% of gun-defenders were injured prior to taking their actions. However, "injuries are less common in gun robberies than in nongun robberies... Analysts typically attribute the lower injury rate among gun robbery victims to their lower rates of resistance. Although this is part of the explanation, gun robbers are also less likely to attack or injure their victims, even controlling for resistance. Further, since resistance often follows injury, it is not clear that the resistance-injury association indicates that resistance provokes robber attack. To the extent that injury precedes resistance, one cannot entirely explain the lower injury rates of gun robberies by less frequent victim resistance... Murder of the victim is more likely in gun robberies than non-gun robberies." (Gary Kleck, Targeting Guns: Firearms and Their Control, Walter de Gruyter, Inc., New York, 1997. pp 238-39. [numerous citations omitted])

Regardless of whether gun-defenders faced easier circumstances or not, as Kleck concludes, defensive gun use is often an effective form of self-defense.
How often do gun owners accidently shoot a family member in the course of defensive gun use? Again in Targeting Guns (p. 310), after reviewing studies, Kleck claims this kind of accidental shooting is rare. He estimates that less than 2% of fatal gun accidents occur during defensive gun use. With approximately 1,000 fatal gun accidents annually, that would imply 20 per year.
If defensive gun use is common then many criminals should certainly have encountered armed resistance. Professors James D. Wright and Peter Rossi surveyed 2,000 felons incarcerated in state prisons across the United States. Wright and Rossi reported that 34% of the felons said they personally had been "scared off, shot at, wounded, or captured by an armed victim"; 69% said that they knew at least one other criminal who had also; 34% said that when thinking about committing a crime they either "often" or "regularly" worried that they "[m]ight get shot at by the victim"; and 57% agreed with the statement, "Most criminals are more worried about meeting an armed victim than they are about running into the police." (James D. Wright & Peter H. Rossi, Armed and Considered Dangerous: A Survey of Felons and Their Firearms [1986]. See Guns and Public Health: Epidemic of Violence or Pandemic of Propaganda? by Don B. Kates, et. al. Originally published as 61 Tenn. L. Rev. 513-596 [1994]).

---