The Battered Statistic Syndrome

By now, everyone knows about the murder of Nicole Brown Simpson and Ronald Goldman. But there's a third victim of these tragic killings: the truth about the prevalence of domestic violence and female victimization, a truth that is daily being maimed almost beyond recognition by the irresponsible use of statistics. 

Consider, for example, the wildly varying statements being issued on all sides regarding the number of women who are supposedly beaten by men in the United States. The National Coalition Against Domestic Violence, for example, estimates that more than half of married women (over 27 million) will experience violence during their marriage, and that over one third (over 18 million) are battered repeatedly every year. Shocked by these statistics--both of which are frequently quoted in the media--I called the NCADV and asked where they came from. Rita Smith, the group's coordinator, told me these figures were only ""estimates.'' From where? " "Based on what we hear out there.'' Out where? Battered women's shelters and other advocacy groups. 

Common sense should tell you that asking women at a shelter whether they've been hit would be like asking patrons at McDonald's whether they ever eat fast food. It would be irresponsible and intellectually dishonest to apply those answers to the country as a whole. But when there's a sensational story to run, common sense and intellectual honesty are rarely taken into consideration. 

Even those who have a public responsibility to be accurate on these issues sometimes falter. According to Donna Shalala, Secretary of Health and Human Services, for example, 4 million women are battered each year by their male partners. But where did Shalala get her figure? From a 1993 Harris poll commissioned by the Commonwealth Fund. Two percent of the 2,500 women interviewed said they had been "kicked, bit, hit with a fist or some other object.'' Apply that to the approximately 55 million women married or living with a man and you get a total of 1.1 million. So where did the other 2.9 million come from? They were women who said they had been ""pushed, grabbed, shoved, or slapped.'' That's a form of abuse, to be sure, but is it what most people would call battering? 

By far the worst distortion of the numbers of battered women comes from Miami talk show host Pat Stevens, who appeared on a segment of CNN's Crossfire show called ""OJ on the Air'' in June. Stevens estimated that when adjusted for underreporting, the true number of battered women is 60 million. No one bothered to tell Stevens--or Crossfire's millions of viewers--that 60 million is more than 100% of all the women in this entire country who are currently in relationships with a man. Instead, Stevens' ""estimate'' and the other ""facts'' on battered women all serve to fuel the claims that there's an ""epidemic of domestic violence'' and a ""war against women.'' 

How many battered women are there? ""Because many feminist activists and researchers have so great a stake in exaggerating the problem and so little compunction about doing so, objective information on battery is very hard to come by,'' writes Christina Hoff Sommers, author of Who Stole Feminism: How Women Have Betrayed Women (Simon & Schuster, 1994). But Murray A. Straus, head of the Family Research Laboratory at the University of New Hampshire, and Richard A. Gelles, a sociologist at the University of Rhode Island, who have been tracking spousal abuse for over 20 years, have come up with what are widely believed to be the most accurate estimates available--the National Family Violence Survey (NFVS). 

Their Survey, sponsored by the National Institute of Mental Health, found that 84% of American families are not violent. In the 16% of families that do experience violence, the vast majority of that violence takes the form of slapping, shoving, and grabbing. Only 3-4% of all families (a total of about 1.8 million) engage in ""severe'' violence: kicking, punching, or using a weapon. 

Moreover, a recent study published in the Archives of Internal Medicine found that 44% of ""severe violence'' to wives did not cause any injury, and 31% caused only a slight bruise. Still, Straus and Gelles estimate that about 188,000 women are injured severely enough to require medical attention. That's a horrifying number of victims, but it's a far cry from 4 million, or 18 million, or 60 million. 

Another commonly accepted ""truth'' about domestic violence is that 95% of the time, women are the victims and men the perpetrators. Nothing could be further from the truth. The Family Violence Survey--as well as numerous other studies--have found that men are just as likely to be the victims of domestic violence as women. But aren't these women just defending themselves against their more violent partners? Straus and Gelles found that among couples reporting violence, the man struck the first blow in 27% of cases; the woman in 24%. The rest of the time, the violence was mutual, with both partners brawling. The results were the same even when the most severe episodes of violence were analyzed. They were also the same when only the woman's version of the events was considered. 

Even more interesting are Straus' findings, released earlier this month, that men's violence against women--even as reported by women—has dropped 43% between 1985 and 1992. Over this same period, in contrast, assaults by women against men increased by about 28%. Straus concludes that ""part of the reason may be that there has been no effort to condemn assault by wives parallel to the effort to condemn assaults by husbands.'' 

So where did the claim that 95% of domestic violence is initiated by men come from? From the U.S. Department of Justice, which collects data on the number of reports of domestic violence. But as women's rights groups rightfully claim, reports are not always an accurate measure of the severity of the problem. Certainly, some female victims of domestic violence fail to call the police, fearing retaliation by their abusers. But other Justice Department studies have shown that men, too, are reluctant to ask for help, reporting all kinds of violent victimization 32% less frequently than women. 

Confessing to being beaten up by another man, however, is a piece of cake compared to admitting being victimized by a woman. After all, men are socialized to ""take it like a man.'' As a result, men tend to report only the most extreme abuse. ""They wouldn't dream of reporting the kind of minor abuse--such as slapping or kicking--that women routinely report,'' says Suzanne Steinmetz, director of the Family Research Institute at Indiana University / Purdue. 

Another example of how data on female victimization is distorted, is the claim that ""domestic violence is the most common cause of injury to women.'' The source for this claim is a 1991 study of extremely poor, inner-city African-American women in Philadelphia--which doesn't even find that domestic violence was the leading cause of injury. ""And even if it did,'' says Dr. Jeane Ann Grisso, one of the lead researchers of the study, ""I'd never apply that conclusion to the total population of American women.'' Nevertheless, Grisso's study has been widely cited as proof that there's an epidemic of violence against women. 

Some advocates have taken Grisso's study one step further, claiming that as many as 50% of women's hospital emergency-room admissions are the result of ongoing abuse. At the source of this so called fact are several studies done in the 1970's by Evan Stark and Anne Flitcraft, co-directors of the Domestic Violence Training Project at the University of Connecticut. They compiled their data by going through old medical records in urban hospitals and estimating how many women were battered by using what they called an ""index of suspicion.'' Christina Hoff

Sommers has analyzed Stark and Flitcraft's methods and writes: ""if a woman was assaulted but the records do not say who hit her, Stark and Flitcraft classify this as a case of "probable' domestic abuse; if she has injuries to her face and torso that are inadequately explained, they classify it as "suggestive of abuse.''' Apparently no one considered the possibility that someone other than a husband or boyfriend might have been responsible for the woman's injuries. 

Compare Stark and Flitcraft's results to those reached in a 1992 survey of 397 emergency rooms in California. Nurses were asked to estimate the number of patients per month who have been diagnosed with injuries caused by domestic violence. Estimates ranged from two per month for small hospitals to eight per month for large ones. The California study concluded that the number of perceived domestic violence victims was so low because many health professionals are poorly trained in recognizing domestic violence. That may be correct, but it's doubtful that it would account for the enormous difference between a handful of domestic violence cases a month and the claim that such cases account for 50% of all women's emergency room admissions. 

There's no question that many women who have been severely battered are afraid to leave their batterers--either because they are economically dependent, or because they fear further abuse. In one of their ""fact sheets,'' the National Coalition Against Domestic Violence tells us that women who leave their batterers ""increase by 75% their chances of getting killed.'' When I asked her to explain that figure, the NCADV's Rita Smith admitted that that statistic isn't true at all, and that the Coalition has no concrete evidence of the effect--if any--leaving a violent partner will have on a woman. I then asked Ms. Smith whether it bothered her that her organization was responsible for spreading an imaginary statistic. ""Not really,'' she said. ""We think the chance of getting killed goes up and we're just trying to make a point here.'' 

In a very small number of tragic cases, abusive men do kill their partners. But women aren't the only ones killed in domestic disputes. A Justice Department study released earlier this month showed that 41 percent of spousal murder victims were male. Battered women's advocates claim that those women who kill their husbands do so only out of self-defense. But in an extensive study of women imprisoned for murder, Coramae Richey Mann, a researcher at the Department of Criminal Justice, Indiana University/Bloomington found that only 59% claimed self-defense and that 30% had previously been arrested for violent crimes. 

As for the perception that women who murder their husbands are treated harshly by the justice system, Dr. Mann found that few female domestic homicide offenders receive prison sentences, and that those who do rarely serve more than four or five years. These findings were are confirmed by a recent Los Angeles Times article. The article, which quoted Justice Department sources, reported that women who kill their husbands were acquitted in 12.9% of the cases, while husbands who kill their wives were acquitted only 1.4% of the time. In addition, women convicted of killing their husbands receive an average sentence of only six years, while male spousal killers got 17 years. 

Why are these statistics being battered? ""The higher your figures for abuse, the more likely you'll reap rewards, regardless of your methodology,'' says Dr. Sommers. Those who create and disseminate inflated statistics are often invited to testify before Congress, they're written about in the New York Times, and some even get to be interviewed on Oprah. 

Not everyone who manipulates data does so for personal gain. Some are simply trying to get people to sit up and pay attention to the plight of battered women--a truly important goal. But to do so, they've created a false epidemic. If advocates confined themselves to the truth—that 3-4% of women are battered each year--domestic violence might still be regarded as the unfortunate behavior of a few crazy men. But if enough people are led to believe that 19 or 50 or 100 percent of women are ""brutalized,'' the only logical conclusion can be that all men are dangerous and all women need to be protected. 

Is it OK to lie shamelessly if your cause is a noble one? Is half a solution better than no solution at all? On the one hand, lying about the extent of the problem of domestic violence has had some very positive effects, opening the public's eyes as well as their wallets. Battered women are now the hottest story in town and Congress is about to pass the $1.8 billion Violence Against Women Act which, among other things, will fund toll-free hotlines, battered women's shelters, and education and training programs. It's certainly possible that none of this would be happening if advocacy groups stuck strictly to facts. 

On the other hand, even supposedly harmless ""puffing'' can have been some extremely negative consequences. Inaccurate discussions about domestic violence, for example, can quickly turn into smear campaigns in which almost every man who hasn't exhibited his natural vicious and misogynist tendencies yet, is expected to do so at any moment. Members of Congress, seeing a golden opportunity to appease a large block of voters, have chosen a quick solution rather than attempting to correct their constituents' misapprehensions. The Violence Against Women Act, for example, doesn't devote a nickel to the same kind of special protection for men, even though males make up 75% of all murder victims and 61% of the victims of all violent crime. 

Women, too, are being hurt by the lies. Having fought so hard to be taken seriously and treated as equals, women are again finding themselves portrayed as weak and helpless--exactly the stereotypes that have been traditionally used to justify discriminating against them. As the author and feminist critic Katherine Dunn writes in the current issue of The New Republic, ""The denial of female aggression is a destructive myth. It robs an entire gender of a significant spectrum of power, leaving women less than equal with men and effectively keeping them "in their place' and under control.'' 

Worst of all, the inflation of domestic violence statistics produces a kind of ratchet effect. The same people who complain that no one listens if they don't exaggerate only find it that much more difficult to get people's attention the next time around--which in turn seems to justify another round of exaggeration. Eventually, the public either stops listening altogether, or finds the statistics too absurd to believe. And when we're trying to alleviate the tragedy of domestic violence, the last thing you want anyone to do is laugh. 

Armin A. Brott