Integrity of Information
Professor John Fekete of Trent University wrote the book "Moral Panic" in which he develops the regression from politics called biopolitics. He writes: "Biopolitics has no time for humankind; nor does it care about individuals. In fact, even its concern for individual members of its own group depends on the extent to which they behave and express themselves in accordance with the stipulated essence of the group. Their concern is to promote the group, and to advance the groups cause against its enemies."
In the book, he also takes a close look at "Changing the Landscape", the Final Report of the Canadian Panel on Violence Against Women (1993 commissioned by the Canadian Ministry of Supply and Services). At one point the report concludes that " the tentative responses of the international community to the systematic rapes of women in Bosnia-Hercegovina were constant reminders to us that tolerance of violence against women knows no geographic boundaries." Fekete cites this as an extreme example of biopolitics and characterizes the report as being obsessively self-dramatizing, fixated in gender-thinking and bristling with hostility while it opportunistically pirates the terrible ordeal of an entire people to service its own political agenda.
He concludes that the $10 million report is a scandalous document which does a great disservice to those of the female population who suffer in Canada, as well as to suffering men and to the general well-being of all Canadian adults and children. He says: "Those of us who are surprised to find the Canadian government supporting, financing and issuing a document that accuses Canadian society and Canadian institutions of abuse of power and hatred of women will have to take stock of the extent to which organized biopolitical advocacy is becoming official policy in the Canadian Government."
Professor Fekete also writes about the genesis of the white ribbon campaign which became a symbol of mens collective guilt about violence against women. He says the demonstration needed an official scapegoat who could serve as the object of universal hatred and that was to be the mass murderer Marc Lepine.
Fekete writes: "December 6, 1989, when this obscure figure went supernova in public and made international headlines, has become unofficially commemorated in Canada as the day to reflect on the demon. Lepines own personal story has now completely disappeared into the myth of male evil. Marc Lepine, crazed with hostility, walked into a classroom at the Ecole Polytechnique in Montreal with a gun and slaughtered 14 female students. We all felt with various asymmetries the terror and the pity and the tragic necessity that every so often shatter the veneer of our remarkable systems of sublimation and control. But the panic that followed the mourning transformed individual deaths into biopolitical symbols. The 14 young female students became the essence of woman-victim; Lepine, the essence of male-predator. "Beasts in suits and ties," Craig McInniss article for The Globe and Mail, summarizes the lesson learned: "Never before has Canadian society had more insight into the nature of the beast and the certainty of its gender." One year earlier, in contrast, Laurie Dann, a Chicago woman, shot five elementary-school boys, poisoned food at two fraternities, burned down the Young Mens Jewish Council, burned two boys in their basement, shot her own son and justified her murder of an 8-year old boy by claiming he was a rapist. This story did not garner international headlines; nor did it come to symbolize a bloody and fundamental biological divide between women and men. In the aftermath of a single, male psychotics twisted internal narratives and outward violence, we continue to be robbed of the common humanity of a dark moment, which now survives only under glaring illumination as a sectarian cliché, and therefore a dangerous fraud about the nature of violence in the world."
No one can deny that some men do commit unjustifiable acts of violence against women and that some women also manifest terrible violence. Therefore, to seize the issue, the argument shifts to semantics and numbers. Some research suggests that the incidence of abuse of men is so insignificant, that social violence is a womens issue. In fact the sensationalized numbers are actually causing panic. However there is other research that finds that social violence is evenly split between men and women. The disagreements are rooted in defining what constitutes violence and how the data is collected and analyzed.
Marc Lepine, Paul Bernardo and O.J Simpson all contributed to the growing panic in the early 1990s. Coincidentally, three major Canadian surveys hit the public in 1993. In February 1993, the first national study of woman abuse in Canadian university and college dating relationships reported their astounding figure of female victimization at a level of 81% combined sexual, physical and psychological abuse based on a random sample of 1,835 women and 1,307 men. In July, 1993, the Canadian Panel on Violence Against Women claimed a global sexual violation rate of 98% of all Canadian women based only on a figure offered by two women calling themselves the Womens Safety Project. And finally in November, 1993, Statistics Canada published the results of its national telephone survey of a random sample of 12,300 women over the age of 18 on violence against women. It claimed that 51% of Canadian women have suffered from male violence at least once since the age of 16 and that 39% have been sexually assaulted, 5% in any given year, 16 to 20 times more than get reported to the police. The largest group, 25%, said they had been pushed, grabbed or shoved by their spouse at least once; 19% were threatened; 15% were slapped, 11% were the target of a flying object; 11% were kicked, bitten or hit; 9% experience severe physical or sexual assault; 7% were choked; and 6% were hit with an object. When StatsCan isolated the number of women who had been assaulted in the last 12 months, the figure was 10%. Each survey, reported in hundreds of articles, helps to build the panic but according to Prof. Fekete, each is fatally flawed in one or several respects.
Statistics Canada is highly respected and generally has a very good reputation because of the diversity and complexity of its reporting. Prof. Fekete, however, took a close look at this study and apparently has not only shattered the myth but put into serious doubt the credibility of StatsCan. In their November report, StatsCan concludes: "The threat of violence is considered so real that fully 60% of women in Canada who walk alone in their area after dark feel worried doing so." According to Prof. Fekete, StatsCan interviewers asked a leading question, bonding with the interviewee, woman to woman, signalling what is the rational position that they can share. The telephone interviewer starts the "Fear of Violence" section of the interview by reading the following from the questionnaire: "I would like to start by asking you some questions about your personal safety. I am going to ask you about some everyday situations, and I would like to ask you how you feel about each of them. First, when walking alone in your area after dark, do you feel: Very worried? Somewhat worried? Not at all worried about your personal safety?"
The StatsCan Daily reports "Six in ten Canadian women who walk alone in their own area after dark feel "very" or "somewhat" worried in doing so." Given the leading question and one in three possible responses, the 60% seems plausible. However the detailed data shows that only 41% of respondents fell into the "very" and "somewhat worried" categories and 27% in the not worried at all. But 33% of the women said they are never in that situation at all. Effectively StatsCan just excluded those women and recalculated the 41% as a percent of 68% giving them 60%. Thats the same answer you get if you assume the 33% split in the same ratios. This is flawed logic and constitutes fabricating information. The actual result was that only 6% of women surveyed were very worried when walking alone in their areas after dark. The headline using 6% would be true and doesnt create panic but it doesnt give the desired spin. According to Prof. Fekete : "Statistics Canada is making up the numbers and making up the interpretations that give the numbers meaning. The ends are expected to justify the means." Since over 10% of the population suffers from some sort of anxiety disorder, the 6% who are very worried theoretically could have no relation whatsoever to violence.
In my experience, particularly in the area of family research, Statistics Canada has from time to time provided editorial commentary which has not been a fair reflection of the data collected. They have structured the answers to questions such that there are more options for the positive response than the negative response and then they group the responses in a way that produces high enough numbers to support a concensus view. Ostensibly, the answers can be configured in a way that the conclusion you want is virtually assured. Since the data gathering organization is the highly regarded Statistics Canada, who is going to question the validity or perceived bias either in the questions or in the options for a response.
In his book, Moral Panic, Professor John Fekete also shares that view. He talked about the Statistics Canada survey "Violence Against Women" which he characterized as being against basic scientific integrity and fairness. He said the survey uses standard CTS questions to measure spousal violence. CTS or "Conflict Tactic Scales" were developed by Dr. Murray Straus and Dr. Richard Gelles and have been used extensively in their research, and internationally, including all three Canadian research studies published in 1993. In fact, it has become the standard in the field of violence research. It turns out that numbers look more alarming if research subjects tick off micro-actions from a list of event descriptions, rather than describing their interpretation of their own experience.
The CTS is a quantitative instrument designed to measure how couples settle their differences and conflicts. It presents the interviewees with a choice of 18 micro-acts, 10 non-violent and 8 violent, including the degrees of minor, severe and very severe. The idea is to measure the use of tactics of reasoning, verbal aggression and physical aggression ranging from "discussed the issue calmly," to "refused to talk about it" or "did or said something to spite the other." It also uses a list of physical acts in order of escalating possibility of injury like grabbing, slapping, hitting, beating up and knifing.
Another reputable research tool is the Sexual Experiences Survey (SES) developed by Mary Koss which she used in a study of sexual aggression and victimization in a national sample of higher education students. Like CTS, the SES contains a list of 10 acts in descending order of sexual coercion and force, from being talked into a kiss to forced intercourse and other sexual acts. The CTS assumes a continuum of conflict tactics in the relationships; similarly, the SES assumes a continuum of sexual aggression and victimization.
Both the CTS and SES have high degrees of credibility in research on violence provided the methodology is not modified. This appears to be the case according to Fekete who says that it is scandalous that the Statistics Canada study which purports to provide pioneering national information about relationship interaction in Canada, is a single-sex survey. He suggests that the decision to ask only women about acts of violence perpetrated against them is highly partisan and biased.
To quote Professor Fekete directly, he writes: "In other words, the reputation of Canada's number one number cruncher, the gold standard of the truth about Canadian life, is hostage in all this to the tyranny of pain, grievance, fear, and resentment. Still, it boggles the mind that Statistics Canada would have opted for the inferior, biopolitical option, the panic option of going through the motions of a fake survey. The one-sex survey of Canadian women is a completely uncorroborated, worthless waste of money and public trust." This is serious charge against StatsCan which should not go unanswered.
He also notes that the survey follows up on the transformation of Canadian sex laws on consent and sexual assault. The change from rape laws to sexual assault laws has meant that casual fondling gets thrown in with shattering transgression, with the result that although the statistics in the sex crime category get much bigger, they tell us much less about what's going on. Pressing ahead with legislative changes, which give greater recognition to victims' rights, the government has also created a new definition of consent and removed the evidentiary requirement of corroboration from the sexual assault law. Fekete concludes that: "Statistical science now puts itself at the service of this fictional Canadian woman, whose uncorroborated testimony about all the unwanted incidents in her life raises her to a level of mythic participation in a sacrificial ritual beyond her death, where her chances at real equality are sacrificed to the creation of Woman-as-Victim."
The most significant claim in the survey is that 51% of Canadian women have experienced at least one incident of violence since the age of 16. However by defining it as a strictly male-to-female victimization survey, based only on complaints, they gave up any serious interest in the nature of interactions between men and women. This means that they made no effort to discover if men and women might have different perceptions of the questions they asked. They also made no effort to verify the accuracy of memory recall. Many of the questions, as is characteristic of CTS and SES type measures, include poorly conceived categories of violence often combining threatened, attempted and actual violence that are not mutually exclusive. As an example, the question from which estimates were arrived at on sexual attack by non-spouses was: "It is important to hear from women themselves if we are to understand the very serious problem of male violence against women. Has a male stranger, date or boyfriend, or other man known to you ever forced you to or attempted to force you into any sexual activity by threatening you, holding you down or hurting you in some way? Yes or no.
Professor Fekete also notes that the question that was one of the two main staples for generating the surveys physical assault data is a yes or no question: "Now I am going to ask you some questions about physical attacks you may have had since the age of 16. By this I mean any use of force such as being hit, slapped, kicked, or grabbed to being beaten, knifed or shot. Has a male stranger, date or boyfriend, or other man known to you ever physically attacked you?" Later in the marital violence section, this question is simply broken into the usual CTS range, from pushing, grabbing and shoving to being knifed or shot.
Professor Fekete also highlights further misleading information in the Statistics Canada survey. They reported that: "For one-in-three (34%) victims of wife assault, the abuse or threats of abuse were so serious that they feared for their lives. While this percentage was higher in the case of past marriages (45%), it is important to know that 13% of women reporting violence in a current marriage had at some point felt their lives were in danger (130,000 or 2% of all currently married women)." In fact Statistics Canada misreported its own data since the figures actually referred to perpetrators and not to victims.
It was not that 34% of women fear for their lives; nor that 45% of women feared for their lives in the past marriage. It is that 34% of those partners who were complained about made the woman fear for their lives, including 45% of the allegedly violent partners in past marriages. Fekete points out that this makes a difference because Statistics Canada is double counting in its tables. Women who have experienced violence in both the current and a previous partnership are counted as two partnerships. If the issue is what percentage of ever-married women have ever feared for their lives from their partner, at some time in their lives, it does not matter how many partners frightened them. One woman, one life. What Statistics Canada has presented is perpetrator data, presented as though it were victim data though we are never told how many partners these ever-married women had altogether. We learn only how many partners they complain about.
On a technical point, Fekete demonstrates the kind of manipulation that severely distorts the presentation of data. He points out that what is more significant is how little purpose these data have apart from validating and promoting fear. The survey showed that about one million current male marital partners (1 in 6) stand accused of some form of abuse; of those, 13% are accused of being the cause of mortal fear which means that 2.6% of the six million currently married Canadian women have experienced mortal fear from their partner. Notwithstanding, Statistics Canada still reported that 34% of women fear for their lives. It doesnt seem to be a plausible figure since the actual number of women killed by their spouses was 13 for every 1 million couples or 76 female homocides in the past year.
The Statistics Canada survey was a relationship survey, but they did not ask any men any questions at all. They did not ask any woman to report on her own acts of violence towards the man in a relationship. As such it is a victimization survey of violence against women. Even as a victimization survey, they could have asked whether any woman has ever hurt her, such as a mother or sister. They could have asked whether a sister had said any unkind things about her body or whether a daughter pushed or shoved her around in her older years. And they could have asked whether a female superior has taken advantage of her sexually.
Prof. Fekete concludes that Statistics Canada has sold its soul to the dark powers of demonization and has traded in science for voodoo. There was no excuse for a one-sex survey. It shows that there is some reason to believe that StatsCan has politically embarrassing data on women abuse against men. If it is true, but not being reported, then StatsCan has lost its objectivity and may, directly or indirectly, be under the control or duress of those in whose interest it is to supress the information.
The criticism of the StatsCan survey was also written up in the Alberta Report of December 6, 1993 in an article titled "Is It StatsCan or Propcan? - Another lopsided report ignores women's violence, and sticks to the feminist model." In the article they wrote that feminist groups tried to stop StatsCan from doing the study. They feared it would produce too low a number. But now that StatsCan has given them a percentage they can work with, feminist groups are using the report to lobby the federal government for more money. With the National Action Committee on the Status of Women leading the charge, they are demanding a change to government funding structures which will guarantee them a core funding base each year.
The story goes on to say that there is no unanimity among experts that women are victimized by men because they are women, as the feminist faction believes. Dr. Reena Sommer, a family violence researcher at the University of Manitoba, is glad that the study on violence has finally come out which is broadly representative of the Canadian female population. Earlier studies were skewed, she says because they amounted to theoretical extrapolations using data collected from women's shelters and victims groups, where naturally, nearly 100% of the respondents were victims.
But while Dr. Sommer is grateful for a more accurate study on violence against women, her applause of the StatsCan report ends there. She is extremely disappointed that StatsCan spent $1.9 million to produce what amounts to a lopsided view of what is really happening, especially in homes. StatsCan should also have asked questions relating to female perpetrated violence, since most family violence is mutual. It would have cost more money to ask more questions, she concedes, but it would have produced a far more valuable result.
Dr. Sommer also faults StatsCan for insufficient analysis of the limited information it did get. She considers it almost irresponsible to publish findings like the 51% number, without offering any indication of how prevalent the violence occurs. Paul Brantingham, Professor of Criminology at Simon Fraser University, agrees that the 51% figure is misleading if it isn't also revealed how often the violence is a present an ongoing occurrence for women. He adds that StatsCan should have asked the women why they did not report the assaults to the police; in other surveys women have frequently indicated that it was too trivial to bother with.
Professor Brantingham also adds that StatsCan failed to distinguish between different types of violence in some categories. For instance, sexual assault included both rape and gropings. One is unambiguously violence against women, but the other is one of the more unfortunate experiences in teen dating. He asked why the two were lumped together and the question should be answered.
Dr. Sommer's own survey, which was part of a larger study funded by Health and Welfare Canada, found that although women are more often injured when men hit them, women are more likely to perpetrate household violence against men. Though feminists and most news media have ignored her findings, they confirm the findings of many similarly ignored Canadian and American studies. In a sample of 452 married or cohabiting females randomly selected in Winnipeg, Dr. Sommer found that 39% of female subjects participated in some form of spousal abuse: 38% had committed acts of minor violence, while 16% admitted to seriously abusing their partner. The rate of male-perpetrated violence in marriage, according to the Sommer study, is 26%, very close to StatsCans 29%.
There were also two Alberta studies, one at the University of Calgary, the other at the University of Alberta, which also found high rates of female-perpetrated violence. In the former study of 562 couples, 38%, admitted to spousal abuse ranging from mild threats to using a weapon; 80 cases involved mutual abuse; 75 were wife-to-husband abuse; and 58 were husband-to-wife abuse.
Holly Johnson, project manager for the Canadian Center for Justice Statistics, an arm of StatsCan, together with feminist groups, criticize the findings which show that women are as physically aggressive as men. They suggest the method of inquiry most often used in the studies, the Conflict Tactics Scale, CTS, is inaccurate. Indeed, argues Ms. Johnson, asking women if they ever perpetrated violence is utterly inappropriate. She claims women tend to overstate their own violence because they feel guilty while they accurately report male-perpetrated violence. However, she cannot cite any clinical evidence to support this hypothesis.
Both Dr. Sommers and Professor Straus say that although every methodology has its flaws, the CTS is the most widely used tool in North American family violence research. They note that when the CTS found higher rates of male-perpetrated violence in the U.S., feminist groups used its findings to lobby the government for funding. Only when the same scale reported violence by women did they get critical.
Ms. Sommers considers these objections mere double-talk since the questions asked in the StatsCan survey are actually modeled after the CTS. Professor Sraus writes in a recent report on female-perpetrated violence that "attacks on the CTS are examples of blaming the messenger for the bad news. Moreover, no matter what that one thinks of the CTS, at least four studies that didnt use the CTS also found roughly equal rates of violence by women."
Professor Straus takes the modern view that all violence is bad and that all disagreements can and must be pursued in other ways. He maintains that even minor assaults by wives are a serious social problem. The idea persists that a woman is justified in slapping a man who makes an insulting or outrageous statement. Even if these acts rarely produce injury, contends Professor Straus, they increase the chances of violent retaliation. In fact, he cites the 1985 National Family Violence survey of U.S. women which found that in 53.1% of mutually violent exchanges between husbands and wives, the female struck the first blow.
Prof. Fekete notes that Canadian national data on family violence tends to corroborate trends established in the United States. All revealed high rates of wife to husband violence. Even the renowned researcher Murray Straus is quoted as saying: "It is painful to have to recognize the high rate of domestic assaults by women. Moreover, the statistics are likely to be used by mysogynists and apologists for male violence."
There are also examples where survey data did not get published for some reason or other. Straus notes that a survey conducted for the Kentucky Commission of Women in 1979 showed that 38% of attacks in violent couples were attacks initiated by women. This data on female perpetrated assaults was intentionally suppressed. Straus suggests that this brings out a troublesome question of scientific ethics.
The problem does not however seem to be unique to Kentucky. In Canada, Eugene Lupri, in Alberta, has conducted a random national survey on conjugal violence, based on 652 women and 471 men. Although he has not published his results in English, his study fully confirms Strauss work and all the other random two-sex surveys showing symmetry of violence up and down the severity scale. Lupri used the self-report approach to gather offender information and found that in every category including hitting, kicking, beating up, threatening, or violating with a weapon, women reported that they perpetrated more incidents of violence against their partners than men did against women. Before and since, Lupri has published work on male perpetrators but the report on female perpetrators has so far been withheld.
Lupris story is not untypical of today's family violence research. Many male feminists in the field of family violence research are deeply embarrassed by their own findings of sexual symmetry in violence perpetrated. On the 1993 Canadian study on university dating abuse, one of the researchers, Dr. Kelly writes that the study was designed for a specific purpose: to survey women's experiences with abuse. It would appear from all they published, as well as from the way they show their questions and analyze the data, that they asked women for self-reports only on victimization and men for self- reports only on perpetration. The presentation of the results in several articles and to the media leaves a distinct impression of women as victims in assaultive relationships.
The range of examples raises some serious flags about the credibility and ethics of research and reporting. Legislators and a host of others depend on scientific research and studies to make judgements and decisions. We assume that the work is comprehensive and is in accordance with generally accepted principles and practices. Apparently, at least in regard to domestic violence, this may not be the case.