Date: Mon, 17 Jul 1995 13:03:10 -0800 To: cdn-firearms@sfn.saskatoon.sk.ca Subject: Re: Cdn-Firearms Digest V1 #299 You may wish to consider this in a future digest: To: CDN-Firearms From: Gary Mauser FYI, here is my reply to critical comments made by SFU Criminologists Neil Boyd in a recent article, "Bill C-68: simple problem, complex solutions," published in the Canadian Journal of Criminology, Volume 37 [2]: 1995, pp 214-219. In this article, Boyd's comments about me are the following: "Mauser [1993] has argued, to the contrary, that 'firearms are used in Canada to save lives.' He arrived at this conclusion through a telephone interview of 393 Canadians, six of whom stated, in response to one item on a questionnaire, that at some point in the preceding five years they or someone in their household had '... used a gun for self protection [or for protection] of property at home, at work, or elsewhere, even if it wasn't fired.' [For some reason, Boyd left out a few words in the question. The complete and correct wording is given below in my rebuttal]. "On the basis of the perceptions of these six respondents, Mauser concluded a reality of 'saved lives,' and warned that 'restrictive legislation .. may actually cost Canadian lives by rendering it difficult to obtain a firearm when one is needed.' "Needless to add, Mauser's unpublished study is best understood as a political intervention. Mauser is a former American gun collector, target shooter, and gun enthusiast who strongly endorses the right to bear arms as an important community initiative." The Secretary 17 July 1995 Canadian Journal of Criminology 304-383 Parkdale Avenue Ottawa, Ontario K1Y 4R4 fax: 613-725-3720 Dear Ms/Sir: It was recently brought to my attention that I and my research have been denigrated by Neil Boyd in an article published in the April 1995 issue of your journal. I am writing you to request an opportunity to rebut. Given the importance as well as the complexity of the issues in this matter, I feel it would be appropriate for you to allow me to respond in a brief article. I trust you will agree. The crux of the dispute is the proper method of estimating the frequency that Canadians use firearms in self protection. This is a rare but quite important event. Thus, it poses a difficult challenge for our discipline. Since many of the events of interest to criminologists occur relatively rarely [eg, crimes], I would think that an article examining the methodology of how rare events are measured would be of some interest to your readers. Given the high value that you and your editorial board place upon rebuttals, you may well be eager to permit me this opportunity. Mr. Boyd may wish to have an opportunity to reply as well. I have included a four-page rebuttal. Please let me know if you will accept this brief rebuttal for publication. As well, would you be interested as well in considering a full length article describing this study more fully? If it will be rejected out of hand, I will not submit it to your journal. I am looking forward to hearing from you at your earliest convenience. Respectfully yours, Gary A. Mauser, Ph D Professor Faculty of Business Administration and the Institute for Canadian Urban Research Studies Do Canadians use firearms in self protection? The article, "Bill C-68: simple problem, complex solutions," was exceptionally shallow (Boyd 1995). Boyd avoided dealing with the question of whether or not Canadians use firearms for self-protection by resorting to ad hominem arguments. I'm shocked that a Canadian criminology journal would publish blatant slurs against immigrants. The readers of the Canadian Journal of Criminology deserve to see a balanced account of the costs and benefits of firearms ownership in Canada. It is simply assumed in this article that firearms pose unacceptable risks for the public and that greater public safety could be achieved by reducing or eliminating private ownership of firearms. While such an opinion may be popular in some circles, this article fails to provide solid support for such a position. Boyd's argument consists of questionable statistics and a superficial discussion of my study of the use of firearms for self-protection. There is an important question at stake. If it could be shown that firearms actually have benefits as well as costs to Canadians, then there are significant implications for public policy. I believe that my study shows that the private ownership of firearms may actually save lives. I do not think it advances the discussion to insult people personally. Boyd cavalierly dismisses my research by saying I am a "former American." Am I disqualified from voicing my opinion on public policy issues because I wasn't born in Canada? Does Boyd similarly dismiss research done by immigrants from England, Jamaica, or China? Boyd also wants to disregard my study as "as a political intervention," because I'm a gun owner. This makes no sense. If gun owners' comments on firearms legislation should be ignored, then, logically, so should women on women's issues, Quebecers on Quebec independence, not to speak of drug users [such as more than a few academics] on drug laws. Any position must stand [or fall] upon the strength of the argument. The argument in Boyd's paper rests upon several unsupported claims. There isn't space available to thoroughly analyze them all, but a few are so egregious I feel compelled to briefly touch upon them. First, Boyd's knowledge of Canadian homicide statistics is faulty. The typical Canadian homicide victim is male, not female as he states. While it is true that the victim typically knows the accused, Canadian homicides are more likely to stem from an illegal business deal gone wrong than from an angry spouse [Fedorowycz 1994]. Over two-thirds of the accused were known to have a previous criminal record, as did almost half of the victims [Wright 1992]. In the typical homicide, police report that one-half of all accused [and 42% of victims] had "consumed alcohol, drugs or both at the time of the offence. In the case of homicides involving a firearm, Boyd should know that the police data are too unreliable to determine if the accused was legally in possession of the firearm [Stenning 1994]. Second, a study by Kellerman [1993] is cited as showing that keeping guns at home is likely to lead to a homicide. Kellerman's study shows nothing of the kind. Kellerman only selected homes where there had been a homicide with a firearm, and he ignored homes with firearms where no one was killed. Moreover, Kellerman's results are not generalizable to Canadian gun owners. His study was limited to only three American cities. Sample size should not be confused with representativity. Third, another paper, published in a coffee-table magazine, is cited as support of the "correlation" between national rates of firearms deaths and firearms ownership [Killias 1993]. Beyond the obvious point that correlations should not be confused with causation, the methodology used in this study simply cannot stand rigorous examination. Killias subdivides countries that support this correlation into 'regions' [eg, Great Britain is split into Wales, England, and Scotland], and he ignores other countries that do not support his claim [eg, Israel]. Moreover, Killias changes his definitions whenever it suits him. Turning to my study, Boyd criticizes it because of the small sample size. He neglects to mention that this sample was representative of the Canadian population 18 years of age or older. According to basic statistical sampling theory, the representativity of a sample depends upon the sampling methodology, not the size of the sample [Deming 1950]. This sample was collected as part of a comparative study of the United States and Canada [Mauser 1992]. It was funded by the International Council for Canadian Studies, which is affiliated with the Canadian Embassy in the United States. Stratified random sampling methods were used to ensure representative samples drawn. More precisely, unlisted telephone numbers were generated using random digit dialing and professional interviewers conducted the telephone interviews in all 10 Canadian provinces [including Quebec which were conducted in French] in April 1990. The small sample size does mean that the projections have a relatively large sampling error -- plus/minus 5 percentage points. In this study, respondents were asked a series of questions to investigate their use of firearms in self protection: "Aside from military service or police work, in the past five years, have you yourself, or a member of your household, ever used a gun for self-protection or for protection of property at home, at work, or elsewhere, even if it wasn't fired?" If the respondent answered, "yes," s/he was then asked, "Was this to protect against an animal threat or a person [or both]?" This question has been praised as one of the best wording in the literature [Kleck 1991]. Approximately 3.1% of the Canadian adult population report that someone in their household used a firearm, at least once, in self protection in the past 5 years. Over half of these incidents [1.8%] involved protection against animals -- most likely bears; but almost as many [1.6%] involved the use of firearms to defend against other people. Thus, 6 people reported that they had used a firearm to defend against human threats, while 7 people said they had used a firearms to defend against animal threats. [One person reported that they had used a firearm against both human and animal threats]. Very few Canadians report using a firearm for protection. Nevertheless, since there were over 10 million households in Canada in 1990, these percentages imply that there were approximately 312,463 Canadian households where at least one person reported having used a firearm to protect themselves or their family between 1985 and 1990. Calculating the typical 5% confidence limits for a sample of this size gives an estimate ranging from 296,840 to 328,086. Assuming that these incidents were equally distributed over the entire 5-year period, firearms were reported used approximately 62,493 times each year during this time period. Roughly half of these incidents involved a defense against human threats. That is, firearms were reported used 32,255 times each year between 1985 and 1990 in Canada to defend against a human threat, and 36,286 time against animal threats. Confidence in these results is increased because they agree with the results of two other representative surveys. First, an earlier study conducted of British Columbia found a similar but slightly higher percentage of respondents who reported using firearms in self protection [Mauser 1990]. This study, using a similar question, found a slightly higher reported usage against animal threats and a similar rate against human threats. A second, and somewhat larger, national representative survey [N = 1,500], conducted earlier this year also found approximately the same percentage of Canadians reporting they had used a firearm to protect themselves from human or animal threats [Mauser 1995]. Additional confirmation of these findings is found in their consistency with other studies. First, the 1993 General Social Survey found that [2%] of the Canadian population 15 years of age or older reported they had ever "obtained a gun" to protect themselves or their property from crime. This is strong corroboration of my findings since the 1993 GSS involved random digit dialing and telephone interviews with approximately 10,000 Canadians, 15 or over [Sacco 1995]. However, this question differs importantly from the question that I asked. The GSS asked if the R "obtained a gun," while my question concerned "using a gun." Also, my question asked about both animal as well as human threats; the GSS question was limited to human threats. Furthermore, the GSS question did not include a specific time frame, while my question focused upon the past five years. The GSS also confirmed that Canadians are taking a wide variety of measures to protect themselves or their property from crime. Among the measures taken for self protection, 12% of the Canadian population 15 years and over report that they "routinely carry something to defend [themselves] or to alert other people" [Sacco 1995]. In the light of these results, it should not be too surprizing that 3% of the adult population report having actually used a firearm for self protection during the past five years. Finally, the use of a firearm for self protection is consistent with other facts about Canada. Estimates vary, but there are at least 3 million Canadian firearms owners, and there may be as many as 7 million. Whether or not the fear of animal attacks is exaggerated, animal attacks actually do pose a real problem, at least in Western Canada. In BC alone, 2 people are killed annually by bears. There are about 7,000 complaints about problem bears, and about 1,000 bears are destroyed or relocated annually. Cougars pose less of a problem: there are hundreds of problem cougars reported each year, but it rare for humans to be killed by cougars [BC Wildlife Branch 1994]. Firearms may be useful for self protection even if they are not fired. Clearly this would be so if a woman deterred a rapist simply by showing him her firearm and her resolve to use it if necessary. Equally clearly, a hunter need not shoot [or kill] the bear who menaces him, but having his rifle loaded and ready to fire would be a legitimate instance of self protection. Such incidents would probably not be brought to the attention of the police. In the case of the woman who used a firearm to deter a potential rapist, she has nothing to gain by going to the police, and given the official attitude towards the defensive use of firearm, much to lose by doing do. In the case of the hunter who did not have to shoot a dangerous bear, there is nothing to report. How much confidence can we have in these verbal reports? Would respondents be more likely to "telescope" time, reporting events that happened long ago, or even to exaggerate the number of times that they used a firearm in self protection, or would they be more likely to refuse to admit [or to forget] actual instances? Errors of both types may occur. However, the use of a relatively long recall period [5 years] suggests that memory loss problems would predominate over telescoping [Sudman and Bradburn 1973]. More importantly, since the use of a firearm in self-defensive exposes the user in Canada to potential legal sanctions, many people would probably prefer to conceal actual instances of self protection use of firearms rather than to report falsely "heroic" events. In conclusion, my study should not be dismissed. Despite its limitations, my study remains the best available estimate of the frequency with which Canadians use firearms for self protection. If firearms are actually used over 60,000 times a year in Canada to defend against either human or animal threats, then conceivably firearms ownership may contribute to public safety as well as pose a danger. It is unknown how many lives were actually saved, but if a life were saved in only 5% of these incidents, then the private ownership of firearms would save more than 3,000 lives annually in Canada. Thus it is possible that more lives are saved each year, than lost, in Canada due to the private ownership of firearms. References BC Wildlife Branch 1994 Activity Report. BC Ministry of Environment. Boyd, Neil 1995 Bill C-68: simple problem, complex solutions. Canadian Journal of Criminology, Volume 37 [2]: 214-219. Deming, William E. 1950 Some Theory of Sampling. Dover. New York. Fedorowycz, Orest 1994 Homicide in Canada - 1993, Juristat, Vol 14, No 15, August. Kellerman, A. L., et al. 1993 Gun ownership as a risk factor for homicide in the home. New England Journal of Medicine 329: 1084-1092. Killias, M. 1993 International correlations between gun ownership and rates of homicide and suicide. Canadian Medical Association Journal 148: 1721-1725. Kleck, Gary 1991 Point Blank. Aldine de Gruyter, New York. p 107 Mauser, Gary 1995 Support for registration is a mile wide but only an inch deep, unpublished paper, Simon Fraser University. Mauser, Gary 1993 Firearms and Self Defense: the Canadian Case, unpublished paper, presented to the American Society of Criminology. Phoenix, AZ. Mauser, Gary and Michael Margolis 1992 The Politics of Gun Control: Comparing Canadian and American Patterns, Government and Policy, Vol 10, pp 189-209. Mauser, Gary 1990 A Comparison of Canadian and American Attitudes Towards Firearms. Canadian Journal of Criminology Vol 32(4): 573-589. Sacco, Vincent F. 1994 Fear and Personal Safety, Juristat Vol 15, No 9, March. Stenning, Philip 1994 Gun Control: A Critique of Current Policy. Policy Options, October, 13-17. Sudman, Seymour, and Norman M. Bradburn 1973 Effects of time and memory factors on response in surveys. Journal of the American Statistical Association 68: 808-815, Wright, Christine 1992 Homicide in Canada - 1991, Juristat Vol 12, No 18, October. Gun registration: The Home Shopping Network for burglars. Gary