<P> Another survey including DGU questions was the National Survey on Private Ownership and Use of Firearms, NSPOF, conducted in 1994 by the Chiltons polling firm for the Police Foundation on a research grant from the National Institute of Justice . in 1997 NSPOF projected 4.7 million DGU per year by 1.5 million individuals after weighting to eliminate false positives . Another estimate has estimated approximately 1 million DGU incidents in the United States . </P> <P> A commonly cited 1995 study by Kleck and Gertz estimated that between 2.1 and 2.5 million DGUs occur in the United States each year . After Kleck and Gertz accounted for telescoping, their estimate was reduced to 2.1 million DGU per year . Kleck and Gertz conducted this survey in 1992, and Kleck began publicizing the 2.5 million DGU per year estimate in 1993 . By 1997, the 2.5 million per year number from Kleck & Gertz' study had been cited as fact by news articles, editorial writers, and the Congressional Research Service . Besides the NSDS and NCVS surveys, ten national and three state surveys summarized by Kleck and Gertz gave 764 thousand to 3.6 million DGU per year . In the report "Guns in America: National Survey on Private Ownership and Use of Firearms" by Philip J. Cook and Jens Ludwig, projected 4.7 million DGU which Cook and Ludwig explained by pointing out all of the NSPOF sample were asked the DGU question . Cook and Ludwig also compared the U.S. crime rate to the number of DGU reported by Kleck and similar studies and said that their estimate of DGU is improbably high . </P> <P> Both Kleck and Gertz' and Lott's research are highly controversial within the academic community . Hemenway has asserted that Kleck and Gertz' methodology suffers from several biases leading them to overestimate the number of DGU, including telescoping, the social desirability bias, and the possibility that "some gun advocates will lie to help bias estimates upwards ." Hemenway contends the Kleck and Gertz study is unreliable and no conclusions can be drawn from it . He argues that there are too many "false positives" in the surveys, and finds the NCVS figures more reliable, yielding estimates of around 100,000 defensive gun uses per year . Applying different adjustments, other social scientists suggest that between 250,000 and 370,000 incidences per year . In 1996, Cook and Ludwig reported that based on their analysis of the National Survey of Private Ownership of Firearms, which "incorporated a sequence of DGU questions very similar to that used by Kleck and Gertz," they estimated that 4.7 million defensive gun uses occur in the United States per year . However, they questioned whether this estimate was credible because the same survey suggests that approximately 132,000 perpetrators were either wounded or killed at the hands of armed civilians in 1994 . They note that this number is about the same as the number of people hospitalized for gunshot injuries that year, but that "almost all of those are there as a result of criminal assault, suicide attempt, or accident ." </P> <P> Kleck asserts errors in his critics' statements that his survey's estimates of defensive gun uses linked with specific crime types, or that involved a wounding of the offender, are implausibly large compared to estimates of the total numbers of such crimes . The total number of nonfatal gunshot woundings, whether medically treated or not, is unknown, and no meaningful estimates can be derived from his survey regarding defensive gun uses linked with specific crime types, or that involved wounding the offender, because the sample sizes are too small . The fact that some crime - specific estimates derived from the Kleck survey are implausibly large is at least partly a reflection of the small samples on which they are based - no more than 196 cases . Kleck states that his estimate of total defensive gun uses was based on nearly 5,000 cases . Thus, he argues, the implausible character of some estimates of small subsets of defensive gun uses is not a valid criticism of whether estimates of the total number of defensive gun uses are implausible or too high . </P>

Fbi statistics on self defense with a gun