To: combjelly who wrote (868534 ) 6/29/2015 3:36:13 PM From: TimF Respond to of 1579334 ...In order for a survey respondent to report a typical DGU, she or he must be willing to report all three of the following elements of the event: (1) a crime victimization experience, (2) his or her possession of a gun, and (3) his or her own commission of a crime. The last element is relevant because most DGUs occur away from the user’s home, and only about 1 percent of the population in 1993, when we conducted our survey, had a permit that allowed them to legally carry a gun through public spaces. Thus, although survey-reported defensive gun uses themselves rarely involve criminal behavior (that is, the defender did not use the gun to commit a criminal assault or other offense), most (at least back in 1993) involved unlawful possession of a gun in a public place by the defender. So what does research on the flaws in surveys of crime-related behaviors tell us? It consistently indicates that survey respondents underreport (1) crime victimization experiences, (2) gun ownership and (3) their own illegal behavior. While it is true that a few respondents overstate their crime-related experiences, they are greatly outnumbered by those who understate them, i.e. those who falsely deny having the experience when in fact they did. In sum, research tells us that surveys underestimate the frequency of crime victimizations, gun possession and self-reported illegal behavior. Yet DeFilippis and Hughes somehow manage to conclude that defensive gun uses—incidents that always involve the first two of those elements, and usually the third as well—are overestimated in surveys. Like Hemenway, DeFilippis and Hughes fail to understand the most fundamental logical issue regarding whether surveys under or overestimate the frequency of defensive gun use. The point at issue is not whether there are “false positive” responses, i.e. respondents saying “yes, they used their gun defensively” when the correct answer was “no.” No one has ever disputed that there are some false positives in these surveys. But this by itself can tell us nothing about whether DGU estimates are too high or too low overall. Even if false positives were numerous, false negatives (when a respondent falsely denies a DGU that actually occurred) could be (and, according to extensive research, are) even more common. In that case, survey estimates of DGU frequency would be too low , not the enormous overestimate that DeFilippis and Hughes believe in. Since neither of those authors nor Hemenway—nor any other critics for that matter—have ever made the slightest effort to estimate the number of false negatives, they cannot possibly know whether false positives outnumber false negatives and therefore have no logical foundation whatsoever for their claims that erroneous responses to DGU questions result in an overestimate of DGU frequency...politico.com shorter - Kleck showed how this was especially evident in the work he did in the early 1990s where respondents who used a gun defensively not only had to admit using the gun but, in most cases, had to also admit a degree of criminal behavior inasmuch as they had the gun with them in a public place before concealed carry was as widespread as it is in 2015. In other words, a lawful use of a firearm was likely carried out by someone who possessed the firearm in a manner for which there was not legal sanction–therefore, even people who used a gun for self-defense rather than crime were likely to keep quiet about doing so.breitbart.com