Unfortunately, disparate, decontextualized and sometimes illegitimate findings can be easily cited and are often employed to back up fallacious claims. Even cases of "scientific studies"…demonstrate serious inconsistencies and indicate the enigmatic nature of empirical research which is so often treated as "objective" (Hammer, 2002, p. 95).

On June 26, 2002 the large circulation Canadian national Globe and Mail’s web site announced : "Men as Likely to Suffer Spousal Abuse, Statscan Says" (Lawlor, 2002, p. 1). The story received top billing because of the high repute of Statistics Canada (Statscan), which means that any research they undertake on violent victimization is bound to be very influential (Denham & Gillespie, 1998). Briefly, the story reports on a study that uses a telephone interview technology to ask a national sample of Canadians slightly modified versions of items included in the Conflict Tactics Scales (CTS) (Straus, 1979 ; Straus & Gelles, 1986 ; Straus, Gelles, & Steinmetz, 1981 ; Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Known as the 1999 General Social Survey on Victimization (GSS), the study found that eight percent of 14,269 women and seven percent of 11,607 men reported at least one incident of intimate partner violence committed by a current or ex-spouse between 1994 and 1999 (Statistics Canada, 2002).

As was the case when Statistics Canada first released the 1999 GSS findings in 2000 (see Pottie Bunge, 2000), these results have been seized upon by some journalists and many fathers’ rights groups to support claims that women are as violent as men and that Canada is seeing a resurgence of what Steinmetz (1977-78) referred to as the "battered husband syndrome" (Jiwani, 2000). For example, Earl Silverman, Program Coordinator for the Calgary-based advocacy group Family of Men Support Society told the Globe and Mail that Statistics Canada’s findings show "that there has been a severe bias against men in the past not considering them as victims" and "[t]o try to deny the other side of the coin reduces the credibility of the first side" (Foss, 2002, p. 8).

Silverman is talking about a see-saw in the way that researchers, media and public policy figures have been dealing with investigations into intimate violence. Throughout North America and in many other countries, the late 1970s and early 1980s were marked by protests that not only was the criminal justice system stacked against battered women, but that researchers had completely ignored their plight. Through the 1980s study after study was released throughout the world documenting the tremendous emotional and physical harm to women that is caused by male intimates. Silverman and many others do not deny any more that women are battered. They argue instead that women are just as violent as men, and that researchers are biased because they mainly pay attention to the concerns of battered women.

These earlier studies, including Statistics Canada’s own 1993 national Violence Against Women Survey (VAWS)(1) and others like it (e.g., Smith, 1987) are now the basis behind the claim that researchers were biased against men in the past. The goal of this paper is not to argue that Statistics Canada purposely went out to pour disrepute on a generation of highly acclaimed surveys. Rather, the main objective is to show that Statistics Canada made a specific effort to return newer studies to the late 1970s and early 1980s, when researchers like Steinmetz (1977/78) used methods that portrayed an inaccurate picture of relationship violence. The first step in this story is to describe the broader political, economic, and social context in which Statistics Canada conducted the newest version of the GSS.

The Political, Economic, and Social Context of the 1999 GSS

Canada was, and still is, a country characterized by class, gender, and racial/ethnic inequality (2). Amazingly, in a country with important gender inequalities, there are a number of anti-feminist scholars (e.g., Fekete, 1994) and conservative men’s groups who argue to a receptive audience that the reverse is true ; that women often have the upper hand and that feminist politics are becoming deeply entrenched into the labor market, provincial and federal governments, and other institutions. Unfortunately, in Canada :

*Women make up only 5 percent of skilled trades, 10 percent of fire and police departments, and 21 percent of senior managers. * Thirty-five percent of women have not completed high school and 72 percent of these women had median after-tax incomes under C$13,786. * Women are underrepresented by almost a three-fold factor in the top 20 percent of Canadian wage earners. Only 11 percent of women get to the top 20 percent, while 29 percent of men access these incomes of C$32,367 and above. * Women aged 45-64 made only 51 percent of the wages of their male counterparts (Hadley, 2001, pp. 1-2).

There are many more statistics that reveal glaring examples of gender inequality. Still, Canadians, like U.S. citizens, keep hearing that the women’s fight for equality "has largely been won" (Faludi, 1991, p. ix). If, as some people contend, the battle has been won, then why is the female poverty rate so high compared to men, and why are at least 11 percent of Canadian wives/cohabiting partners beaten by their male partners at least once each year (DeKeseredy & MacLeod, 1997) ? The answer to these questions is that many Canadians, especially conservative upper class males, benefit from these and other forms of gender inequality. They have no interest in eliminating patriarchy, and in fact are determined to strengthen it (Naiman, 1997).

Nevertheless, Canadian feminists and other progressives periodically find reasons to be optimistic. Although Canada is a patriarchal country, every major social institution, such as the family, the workplace, and the military has been affected by laws and other means of eliminating sexism (Renzetti & Curran, 1995). Consider what happened in the late 1980s and early 1990s. Public awareness and outrage emerged over woman abuse (Denham & Gillespie, 1998), and due in large part to more than two decades of practical struggles waged by feminists to get politicians to define male-to-female violence as a major social problem (Levan, 1996), the mass murder of 14 women at Universite de Montreal’s Ecole Polytechnique, and alarming wife abuse survey data generated by sociologists (e.g., Smith, 1987), between 1988 and 1996 the Canadian federal government committed C$176 million to various initiatives on family violence, especially woman abuse in adult heterosexual relationships (Denham & Gillespie, 1998).

One outlet for this federal money was to fund two national representative sample surveys : Statistics Canada’s (1993) VAWS and the Canadian National Survey on Woman Abuse in University/College Dating (CNS) (DeKeseredy & Kelly, 1993 ; DeKeseredy & Schwartz, 1998a). As noted above, there was extensive concern in the literature about the limitations of earlier North American national representative sample surveys of family violence, such as those conducted by Straus et al. (1981) and Straus and Gelles (1986). The VAWS and CNS have been widely recognized by international scholars as "major achievements" (Denham & Gillespie, 1998) specifically because they overcame or minimized the limitations of these earlier surveys.

One of the limitations of earlier surveys is that they focused mainly on behaviors that were criminal, while many women saw their victimization in much wider terms than behaviors that would interest a police officer or a prosecutor. Thus, in addition to measuring male violence against female intimates, the VAWS paid attention to behaviors such as sexual harassment (Johnson, 1996). Another problem with earlier work is that it centered attention on raw counts of blows or times that someone was struck, without recognition that there are important differences in understanding such behavior based on the context in which the violence takes place, the meaning of the violence, and the motivation of the aggressor. An attempt to control your spouse by the use of violence can be very different than an attempt to protect oneself in the midst of an attack, but a question that just asks who hit whom, and how often, would miss these important factors. Thus, both the VAWS and the CNS used multiple measures of abuse, and the CNS added revised versions of Saunders’ (1986) context, meanings and motives of violence measures to the Conflict Tactics Scale measures. It should be noted in passing, too, that the CNS found that a substantial amount of women’s violence in dating was in self-defense or "fighting back" (DeKeseredy, Saunders, Schwartz, & Alvi, 1997).

The VAWS and CNS were also informed by gender-specific definitions of violence in intimate relationships, based on dozens of smaller studies through the 1980s that made it clear that the majority of persons killed or injured in intimate violence were women. Unfortunately, while these studies provided a more accurate picture of relationship abuse, they unintentionally also contributed to the anti-feminist backlash. In fact, of the 50 Canadian people who worked on the problem of woman abuse that were interviewed by Denham & Gillespie (1998, p. 46), many :

commented on how taken aback they were with the strength and destructiveness of the backlash movement, e.g., the appropriation of the feminist power analysis by men’s rights groups as a tool to attack women ; and the use by right-wing anti-feminist organizations of research into lesbian abuse to show how women are as abusive as men.

Immediately after Statistics Canada (1993) and CNS researchers (1993) released their incidence and prevalence rates to the media, this backlash involved placing "well-placed lies" (3) about survey researchers in the media. Prominent fathers’ rights activists (e.g., Ferrel Christensen) sent petitions to the Canadian Sociology and Anthropology Association and to Health Canada stating that the CNS research team violated ethical principles (numerous human subjects boards had approved the research, and this claim was rejected), and men’s rights advocates placed harassing phone calls to researchers (DeKeseredy, 1999). The backlash, however, was not restricted to tactics such as these and quickly entered broader political and legal arenas. For example, drawing upon sexually symmetrical CTS data (men and women have equal violence rates) generated by researchers such as Straus and Gelles (1986), in 1998 Ferrel Christensen and others affiliated with the Movement for the Establishment of Real Gender Equality (MERGE) filed a complaint with the Alberta Human Rights and Citizen Commission (AHRCC) stating that the Edmonton Family Centre’s family violence brochure "discriminates against men" (Farrell, 2000) by "promoting the idea that only men are abusive in intimate relationships" (AHRCC, 2000, p. 1). In this complaint, MERGE also asserted that :

research shows men and women are equally likely to be physically and emotionally abusive toward their partners and children and therefore, on this basis the print material, in contravention of section 2(1) of the Human Rights, Citizenship and Multiculturalism Act. In addition to being in contravention of section 2(1)…the print material strongly suggests the Family Centre discriminates on the grounds of gender in the area of services, in contravention of section 3 of Human Rights, Citizenship and Multiculturalism Act (AHRCC, 2000, p.1).

In January 2000, the AHRCC ruled in favor of MERGE and the Family Centre could not afford to submit an appeal. Christensen called this decision a "major victory against the negative stereotyping of men" (cited in Farrell, 2000, p. 1), claiming that "Academic studies demonstrate men and women are responsible for family violence in equal proportions…. Those statistics have been overshadowed by studies which examine only violence against women. As a result, the "politically correct" view has pervaded Canadian society that family violence is a male vice…. (cited in Farrell, 2000, p. 2).

MERGE intends to use the legal system to target other agencies like the Family Centre that see women as the main targets of intimate violence, and it has the support of several prominent conservative federal politicians such as Members of Parliament Deborah Grey and Roger Gallaway (Farrell, 2000). Gallaway was a Joint Chair of the 1998 Special Joint Committee on Child Custody and Access (SJC) (Pearson & Gallaway, 1998), which devoted considerable attention to Ferrel Christensen’s (2000) submission claiming that feminist survey researchers engaged in "prostituted science and scholarship" and that violence in intimate relationships is sexually symmetrical. Gallaway has garnered "nationwide fame…for criticizing…anti-male bias in federal legislation" (Farrell, 2000, p. 2), and, like Christensen, announced to the media that, "in society at large women are as equally violent as men" (cited in Cobb, 1998, p. A3).

Obviously, Christensen and others with similar views influenced the SJC because it concluded that "because of the existence of violence against men, the Committee would not recommend that family law or divorce legislation employ a gender-specific definition of family violence" (Pearson & Gallaway, 1998, p. 81). Support for this statement was also provided by Murray Straus, one of the foremost proponents of the sexual symmetry of violence thesis. For example, the SJC quoted his following statement about the CTS used in Statistics Canada’s (1993) VAWS : "That is what the Canadian National Survey of Violence Against Women did. They used the techniques which I developed, the Conflict Tactics Scale. But they left out half of it which asks about violence by women, so they wouldn’t be left with politically embarrassing data" (cited in Pearson & Gallaway, 1998, p. 80).

Another point of contention between research methodologists and backlash politicians has been the very method of defining abuse. Many feminist researchers have argued that by keeping definitions narrow, and tied too closely to the criminal statutes, surveys do not uncover large portions of woman abuse. This notion of what Stanko (1990) calls a "continuum of unsafety" in women’s lives is key to virtually all research in the area, including the CTS research widely cited by backlash critics. Yet, Gallaway claims that studies like the CNS "are distorted even further by a broadening of the concept of ’abuse’" (cited in Farrell, 2002, p. 2). Of course, it can be argued that members of MERGE, Roger Gallaway, and other supporters of the anti-feminist backlash are simply "extremists" or "right-wing ideologues" who have little, if any influence on the general public and government policy. However, if this was the case, why, then, are gender-neutral terms like "spousal violence," "family violence," and "domestic violence" frequently featured prominently in recent government documents like Statistics Canada’s fifth annual Family Violence in Canada report (Trainor, 2002a), which features 1999 GSS data ? Moreover, why did Statistics Canada move from taking a feminist approach to studying violence in the early 1990s to using gender-neutral survey methods in 1999 ? And, why did Statistics Canada not add contexts, meanings, and motives of violence measures to the CTS employed in its 1999 GSS ? We will return again to these questions.

Of course we have no way of knowing just why it is that Statistics Canada stopped being guided by work done by feminist survey researchers like Smith (1987, 1994). Regardless of what influenced this agency return to the older gender-neutral approach resembling that used in the first and second U.S. National Family Violence Surveys (Straus & Gelles, 1986 ; Straus et al., 1981), Statistics Canada has contributed to making terms like spousal violence a central "part of our everyday popular lexicon" and its recent survey "has been appropriated and exploited by a variety of antifeminist pundits and organizations" (Hammer, 2002, p. 111). It is to precisely how Statistics Canada’s scientific techniques have taken intimate violence research "back to the future" that we now turn.

Statistics Canada’ s Return to "Family Violence Research"

"Family violence" researchers (4), such as Straus and Gelles (1986) and Steinmetz (1977/78) hold beliefs similar to those of MERGE and Member of Parliament Roger Gallaway. For example, they question the validity of data showing that women are the primary targets of abuse in marriage/cohabitation and dating. They also use gender-neutral terms like spousal violence, administer versions of the CTS that exclude measures of the contexts, meanings and motives of violence, and some of them claim that women who slap male "cads" provoke men to hit them back harder. According to Straus and Gelles (1990, p. 105), the :

danger to a woman of such behavior is that it sets the stage for the husband to assault her. Sometimes this is immediate and severe retaliation. But regardless of whether that occurs, the fact that she is being obstinate, "bitchy," or "not listening to reason" as he sees it. Unless women also forsake violence in their relationships with male partners and children, they cannot expect to be free of assault. Women must insist as much on non-violence by their sisters as they rightfully insist on it by men (emphasis in original).

Although family violence researchers see male and female intimates as equally violent, they have always argued that male violence causes more injuries than female violence (Straus, 1993). Still, they contend that this is unimportant because what influences whether a person is injured or not "are typically random phenomena such as aim or luck" (Straus et al., 1981, p. 22), and the fact that on average men are bigger. This assertion, their gender-neutral definitions, and their methodology have, since the late 1970s, been subject to extensive feminist critiques (5) and several attempts have been made to overcome family violence researchers’ methodological limitations. Again, some social scientists (e.g., Saunders, 1986 ; DeKeseredy et al., 1997) have added contexts, meanings, and motives of violence measures to the CTS.

Statistics Canada’s (1993) VAWS is one prime example of a recent survey that addressed some of the key shortcomings of earlier studies done by family violence researchers. In fact, due in part to advice provided by Michael D. Smith and other leading feminist experts in the field (e.g., Rebecca and Russell Dobash), the VAWS research team specifically designed their study to be distinct from those conducted by Straus et al. (1981) and Straus and Gelles (1986). For example, although it used 10 modified versions of items used in the CTS to measure violence by male spouses, some major differences were important. The CTS is set up with an introduction explaining that the researchers are interested in various tactics employed to resolve couple conflict. However, according to Statistics Canada researcher Holly Johnson (1996, p. 59) :

The Violence Against Women Survey departs from the conflict tactics approach in the extensive lead-up it has to questions about spousal violence, through detailed questions about fear of violence in public places and precautions taken to protect oneself, sexual harassment, and sexual and physical violence by strangers, dates and boyfriends, and other known men. The VAWS does not use the "verbal reasoning" scale or the "verbal aggression" scale to ease respondents into questions about violence. Moreover, the introduction to the section inquiring about wife assault states very directly that "We are particularly interested in learning more about women’s experiences of violence in their homes. I’d like to ask you to tell me if your husband/partner has ever done any of the following to you…" This survey is concerned not with ways of settling differences but with violence against women, and this context is established at this point (emphasis in original).

On pages 56 to 59 of Dangerous Domains : Violence Against Women in Canada, Johnson devotes considerable attention to the major pitfalls of the CTS, including the fact that "it ignores the gendered power imbalances that exist within marriage and society and excludes crucial details about motives, intentions, and consequences…." (1996, p. 57). She also argues that "[t]his scale rests on the assumption that a level playing field exists in families and society whereby men and women enjoy equal power, authority, and resources" (1996, p. 57).

Now, however, it seems that close to a decade after conducting the feminist-informed VAWS, Statistics Canada has revised its position in partial congruence with that held by family violence researchers. Consider the following preamble to 10 modified CTS items used in the 1999 GSS. This introduction is based on the assumption that male and female intimates have equal power :

It is important to hear from people themselves if we are to understand the serious problem of violence in the home. I’m going to ask ten short questions and I’d like you to tell me whether, in the past 5 years, your spouse/partner has done any of the following to you. Your responses are important whether or not you have had any of these experiences. Remember that all information provided is strictly confidential (Pottie Bunge, 2000, p. 13).

Like the two national U.S. family violence surveys (Straus & Gelles, 1986 ; Straus et al., 1981), the above preamble when combined with other methodological pitfalls, resulted in the 1999 GSS containing CTS-style data - raw counts of violent acts committed — that are sexually symmetrical. The 1999 GSS misses the fact that much of male and female violence is committed for different reasons (Jiwani, 2000). Again, a common cause of women’s violence is self-defense, while men typically use violence to control their partners (DeKeseredy & Schwartz, 1998b ; Ellis & Stuckless, 1996). Statistics Canada researchers (e.g., Johnson, 1996), as stated above, are well aware of these issues and other pitfalls of the CTS, yet elected to use methods like those used by non-feminists close to 20 years ago.

Another point to consider is that like U.S. family researchers, Statistics Canada researchers, although aware of the problems associated with misinterpreting crude counts of behavior (see Johnson, 1996), claim that "[m]en and women are exposed to" spousal violence "in almost equal proportions (Trainor, 2002b, p. 1). So, not only are the methods used similar those developed by family researchers, the interpretation of CTS data is the same. A basic difference, however, between family violence researchers and those who conducted the 1999 GSS is that the latter (e.g., Pottie Bunge, 2000) do not state that their research provides evidence of a "battered husband syndrome" (Steinmetz, 1977/1978). Nevertheless, as York University sociologist Desmond Ellis recently asserted in his critique of the 1999 GSS, "Ignoring context, meaning and motive is misinforming…And not separating different types of violence is misleading" (cited in Foss, 2002, p. 2).

Still, Pottie Bunge (2000) presents data showing that the consequences of spousal violence are more serious for Canadian women. Between 1994 and 1999, female victims (15%) were five times more likely than their male counterparts (3%) to have received medical attention. Of course, family violence researchers have not necessarily accepted the importance of such figures even when they themselves come up with the results. Stets and Straus (1990, p. 193), for example, question "whether women’s greater need for medical care and time off from work is a direct result of abuse." Similarly, Canadian men’s rights advocates have often found it easy to just dismiss or challenge findings they do not wish to acknowledge. Note Ferrel Christensen’s (2000, p. 7) statement to the SJC :

The great bulk of assaults between partners are the actions of fairly ordinary people, with fairly ordinary levels of conflict and emotional dysfunction and substance abuse. And in general, those ordinary assaults cause little or no physical harm to either sex - hence, in the large majority of case, the injury levels are about equal for the two sexes.

Statistics Canada took another step back to the work of family violence researchers by ignoring many highly injurious male behaviors such as sexual harassment (Jiwani, 2000). There are other now standard methods of greatly enhancing the quality of survey data on abuse in intimate relationships. For example, supplementary open- and closed-ended questions about male violence that were developed by Smith (1987, 1994) elicited reports of harms that were not reported at the start of a self-report questionnaire or telephone interview. Multiple measures of violence have long been recommended as a method of enhancing the reliability and validity of survey data (DeKeseredy, 1995 ; Schwartz, 2000, Smith, 1994), but for some reason, Statistics Canada did not use them. In sum, unlike when they developed the VAWS, Statistics Canada has now gone back in time to follow family violence researchers.

Ironically, even family violence researchers might have some trouble understanding the lack of results found in the CTS-based data reported by Pottie Bunge (2000) and other Statistics Canada employees (e.g., Trainor, Lambert, & Dauvergne, 2002). For example, regardless of the sample size, virtually every survey that uses the CTS has found that at least 10 percent of respondents, regardless of their sex, report being victimized by at least one form of violence in a one-year time period. Sometimes the figure is much higher. Yet, the 1999 GSS found that three percent of women and two percent of men with a current partner or ex-spouse were victimized in the year before the survey (Trainor et al., 2002). Further, even the male (7%) and female (8%) victimization rates for a five-year period are markedly lower than one-year CTS figures uncovered by previous Canadian large-scale representative sample surveys (e.g., Kennedy & Dutton, 1989 ; Smith, 1987), which found that at least 11 percent of women reported having been victimized by a male spouse or cohabiting partner (DeKeseredy, 2000).

As pointed out by Straus (1998), Schwartz (2000) and many others, low estimates like those uncovered by Statistics Canada are functions of presenting a study to respondents as a crime victimization study. For example, the definition of spousal violence used in the 1999 GSS was informed by the Canadian Criminal Code (Jiwani, 2000 ; Pottie Bunge, 2000 ; Trainor, 2002a). Crime surveys create a set of "demand characteristics" and unless respondents clearly label acts as criminal in their own mind, they tend not to report them (Koss, 1996 ; Schwartz, 2000 ; Straus, 1998). If people do not think of their spouses’ violence as "criminal," they may not report it in such a survey. In fact, close to 83 percent of marital violence incidents are not reported in contexts where the research emphasis is on criminal assault and victimization (Mihalic & Elliot, 1997).

A narrow, legalistic definition of violence also informed Statistics Canada’s (1993) VAWS, and not surprisingly, the one year rate of female victimization generated by this study is three percent. Clearly, surveys driven by broader definitions of violence typically elicit higher rates of violence regardless of the sample size and composition (DeKeseredy, 2000). Statistics Canada researchers are aware of this and one recently stated that, "Obviously, more all-encompassing definitions of family violence willproduce higherestimatesof the problem" (Trainor, 2002c, p. 4).

Feminist scholars(e.g., Kelly,1988 ;Radford,1987)stronglyadvocate using broad definitions of violence, while many scholars (e.g., Fekete, 1994 ; Gelles & Cornell, 1985 ; Gordon, 2000) and some anti-feminist politicians are sharply opposed to them. Recall what Gallaway said about the "distorted" nature of studies that use broad definitions of abuse. It is unclear whether politically powerful opponents to broad definitions heavily influenced Statistics Canada. Still, regardless of what motivated this agency to maintain a narrow focus, a large literature shows that just because the law does not define an abusive incident as serious does not mean that these narrow definitions coincide with women’s real life feelings and experiences (DeKeseredy, 2000 ; Smith, 1994). For example, Jiwani (2000, p. 2) asserts that isolation should have been included in Statistics Canada’s list of physical violence items "based on the reality that abusers will begin by isolating their victims from family, friends and acquaintances, and through isolation, make their victims more vulnerable to violence."

There are many other problems with the 1999 GSS that could easily be described here ; however, it is beyond the scope of this paper to do so (see, e.g., Jiwani, 2000). Rather, the most important point to consider here is that Statistics Canada researchers were aware of the problems with previous family violence surveys conducted in the U.S. (e.g., Straus et al., 1981), minimized and/or overcome some of them in the VAWS, and recently conducted a survey that excluded some of the most important methodological developments in the field. Unfortunately, Statistics Canada’s latest survey is more than one step backward and is likely to lead many people to discount the results of the 1993 VAWS and similar studies (Jiwani, 2000).

Conclusions

The retreat to the past approach taken by those who developed and conducted the 1999 GSS has been hailed as a major victory for the men’s rights movement and those with similar beliefs. However, others see it as a failure for several reasons. For example, we suggest that the GSS spousal violence data are the products of flawed and outdated research methods. Workers in the field take a stronger position that it is a resource for those seeking to curb women’s struggles for safety and equity while providing policy makers with a rationale for further reducing scarce resources given to women’s groups, shelters, batterers’ programs and other social support services designed to curb male-to-female violence (Jiwani, 2000). Thus the backlash politics involved here have a major whiplash effect on both workers and survivors, making it more difficult to garner public support and resources. Women who have been victimized in abusive relationships not only find themselves without concrete help, but are belittled and trivialized in the public arena by men’s rights groups in the process. To have social science researchers as conscious or unconscious fellow travelers in this process is a stain on the profession.

What is to be done ? Here, we could yet again call for feminist scholars and activists to engage in "newsmaking criminology" (Barak, 1988 ; Schwartz & DeKeseredy, 1993). This involves disseminating critiques such as this one to the dominant or mainstream mass media (e.g., Globe and Mail). Of course, the widely read or watched media will occasionally broadcast or publish such commentaries. However, the "frequency is…low" (Caringella-MacDonald & Humphries, 1998, p. 7), and as Faludi (1996, p. 10) reminds us, typically, when it comes to reporting stories about intimate violence, "The media continue to pass along without question bogus findings promoted by right-wing foundations while going on high alert whenever a feminist writer fumbles a statistical footnote."

This is not to say that feminists and other progressives should not disseminate their concerns, opinions, or research to journalists. Nevertheless, at best this approach when done alone will only result in minor chips at the patriarchal forces that contribute to the development of studies such as the 1999 GSS. In conjunction, then, with newsmaking criminology, pro-feminist men and feminist women must join together to engage in several other struggles to preclude research and policies from returning to the past. Repeated contacts with politicians to discover what they are doing to promote gender equality, innovative research, and strategies aimed at alleviating violence against women are all important steps that do not cost a substantial amount of money (Morris, 2002). Other strategies worth considering are (some of which involve spending money) :

* Promoting high quality action research that generates data required to respond in an informed manner to the deluge of popular and scientific material that promotes men’s rights and sexual symmetry in the analysis of and response to violence in intimate relationships (Denham & Gillespie, 1998, p. 48). * An increasingly important site for reaching people are web sites, especially attractive ones that provide extensive information in easy to read and digest formats. The high quality research that we generate should be distributed as widely as we can arrange. * Sharing empirical, theoretical, and political information among different progressive communities and organizations that struggle to end violence against women. This might include press conferences, of course, but also brochures and handouts available in public buildings, health facilities, and workplaces. * Organizing "town hall meetings" to discuss research and policies that harm the movement to end violence against women, as well as to generate strategies that curb this social problem.

COPYRIGHT STATEMENT

All articles appearing in the OJJS are the copyright of the Journal.

The online edition of OJJS is free to individuals and institutions. Copies of the individual articles may be distributed for research or educational purposes free of charge and without permission. However, commercial use of the OJJS or the articles contained herein is expressly prohibited without the written consent of the publisher.