Jan-Feb 1998 v35 n2 p356(7)
Realities and mythologies of rape.
Abstract: Advocacy research on sexual assault tends to inflate the scope of rape and emphasizes the perverse behavior of men. Statistics on rape are mostly exaggerated under the misguided belief that doing so would result in collective action against the assailants and, hence, mitigate the problem.
Full Text: COPYRIGHT 1998 Transaction Publishers
According to the alarming accounts of sexual assault by certain feminist groups, about one out of every two women will be a victim of rape or attempted rape an average of twice in her life, one-third will have been sexually abused as children, and many more will suffer other forms of sexual molestation. These claims are based on figures from several studies, among which the Ms. Magazine Campus Project on Sexual Assault, directed by Mary Koss, and Diana Russell's survey of sexual exploitation
are the most extensive, most widely disseminated, and most frequently cited.
Both studies were funded by the National Institute of Mental Health, giving them the imprimatur of endorsement by a respected federal agency. Often quoted in newspapers and journals, on television, and during the 1991 Senate hearings on sexual assault, the findings from these studies have gained a certain degree of authority by process of repetition. Most of the time, however, those who cite the research findings take them at face value without an understanding of where the numbers come from or
what they represent.
Prefaced by sophisticated discussions of the intricate research methods employed, the findings are presented in a blizzard of data, supported by a few convincing cases and numerous references to lesser known studies. But footnotes do not a scholar make, and the value of quantitative findings depends upon how accurately the research variables are measured, how well the sample is drawn, and the analysis of the data. Despite the respected funding source, frequent media acknowledgement, and an aura
of scientific respectability, a close examination of the two most prominent studies on rape reveals serious flaws that cast grave doubt on their credibility.
The Ms. study directed by Koss surveyed 6159 students at thirty-two colleges. As Koss operationally defines the problem, 27 percent of the female college students in her study had been victims of rape (15 percent) or attempted rape (12 percent) an average of two times between the ages of fourteen and twenty-one. Using the same survey questions, which she claims represent a strict legal description of the crime, Koss calculates that during a twelve-month period 16.6 percent of all college women
were victims of rape or attempted rape and that more than one-half of these victims were assaulted twice. If victimization continued at this annual rate over four years, one would expect well over half of all college women to suffer an incident of rape or attempted rape during that period, and more than one-quarter of them to be victimized twice.
There are several reasons for serious researchers to question the magnitude of sexual assault conveyed by the Ms. findings. To begin with, a notable discrepancy exists between Koss's definition of rape and the way most women she labeled as victims interpreted their experiences. When asked directly, 73 percent of the students whom Koss categorized as victims of rape did not think that they had been raped. This discrepancy is underscored by the subsequent behavior of a high proportion of
identified victims, forty-two percent of whom had sex again with the man who supposedly raped them. Of those categorized as victims of attempted rape, 35 percent later had sex with their purported offender.
Rape and attempted rape were operationally defined in the Ms. study by five questions, three of which referred to the threat or use of "some degree of physical force." The other two questions, however, asked: "Have you had a man attempt sexual intercourse (get on top of you, attempt to insert his penis) when you didn't want to by giving you alcohol or drugs, but intercourse did not occur? Have you had sexual intercourse when you didn't want to because a man gave you alcohol or drugs?" Forty-four
percent of all the women identified as victims of rape and attempted rape in the previous year were so labeled because they responded positively to these awkward and vaguely worded questions. What does having sex "because" a man gives you drugs or alcohol signify? A positive response does not indicate whether duress, intoxication, force, or the threat of force were present; whether the woman's judgment or control were substantially impaired; or whether the man purposely got the woman drunk to
prevent her from resisting his sexual advances. It could mean that a woman was trading sex for drugs or that a few drinks lowered the respondent's inhibitions and she consented to an act she later regretted. Koss assumes that a positive answer signifies the respondent engaged in sexual intercourse against her will because she was intoxicated to the point of being unable to deny consent (and that the man had administered the alcohol for this purpose). While the item could have been clearly worded
to denote "intentional incapacitation of the victim," as the question stands it would require a mind reader to detect whether an affirmative response corresponds to a legal definition of rape.
Finally, a vast disparity exists between the Ms. study findings and the rates of rape and attempted rape that come to the attention of various authorities on college campuses. The number of rapes formally reported to the police on major college campuses is remarkably low - two-to-five incidents a year in schools with thousands of women. It is generally agreed that many rape victims do not report their ordeal because of the embarrassment and frequently callous treatment at the hands of the
police. Over the last decade, however, rape crisis counselling and supportive services have been established on most major campuses. Highly sensitive to the social and psychological violations of rape, these services offer a sympathetic environment in which victims may obtain assistance without having to make an official report to the police. While these services usually minister to more victims than report to the local police, the numbers remain conspicuously low compared to the incidence of
rape and attempted rape on college campuses as Koss defines the problem.
Applying Koss's finding of an annual incidence rate of 166 in 1000 women (each victimized an average of 1.5 times) to the population of 14,000 female students at the University of California at Berkeley in 1990, for example, one would expect about 2000 women to have experienced 3000 rapes or attempted rapes in that year. On the Berkeley campus, two rapes were reported to the police in 1990, and between forty and eighty students sought assistance from the campus rape counselling service. While
this represents a serious problem, its dimensions (three to six cases in 1000) are a fraction of those (166 cases in 1000) claimed by the Ms. study.
What accounts for these discrepancies? Koss offers several explanations, some of which appear to derive from new data or additional analysis. Therefore it is important to distinguish between the data originally reported in 1987 and 1988 and later versions of the findings. The findings from the Ms. study were originally described in three articles, one by Koss and two co-authors in a 1987 issue of the Journal of Consulting and Clinical Psychology, the second (an expanded version of this article)
authored by Koss as a chapter in the 1988 book, Rape and Sexual Assault, (edited by Ann Burgess) and the third by Koss and three co-authors in a 1988 issue of the Psychology of Women Quarterly. Also published in 1988, was Robin Warshaw's book, I Never Called It Rape: The Ms. Report on Recognizing, Fighting, and Surviving Date and Acquaintance Rape, with an afterward by Koss describing the research methods used in the Ms. project on which the book was based.
Two articles reported that only 27 percent of the students whom Koss classified as rape victims believed they had been raped. The third article in the Psychology of Women Quarterly (1988) provided additional data on how all these supposed victims labeled their experience. The findings reported here indicate that: 1) eleven percent of the students said they "don't feel victimized;" 2) forty-nine percent labeled the experience "miscommunication;" 3) fourteen percent labeled it, "crime, but not
rape;" and 4) twenty-seven percent said it was "rape."
Although there was no indication that other data might have been available on this question, three years later a surprisingly different distribution of responses is put forth. In answer to questions raised about the fact that most victims did not think they had been. raped, Koss reported in the Los Angeles Daily Journal (July 17, 1991) that the students labeled as victims viewed the incident as follows: "One-quarter thought it was rape, one-quarter thought it was some kind of crime but did not
believe it qualified as rape, one-quarter thought it was sexual abuse but did not think it qualified as a crime, and one-quarter did not feel victimized."
In a later paper, "Rape on Campus: Facing the Facts," the gist of these new findings was revised, with Koss recounting: "One-quarter thought it was some kind of crime, but did not realize it qualified as rape; one-quarter thought it was serious sexual abuse, but did not know it qualified as a crime."
These inconsistencies in the reported findings aside, the additional data are difficult to interpret. If one-quarter thought their incidents involved a crime, but not rape, what kind of crime did they have in mind? Were they referring to illegal activity at the time such as drinking under-age or taking drugs? Despite Koss's elaboration on the data originally reported, at least one version of the findings reveal that 60 percent of the students either did not feel victimized or thought the
incident was a case of miscommunication. Although in the second version many more students assessed the sexual encounter in negative terms, the fact remains that 73 percent did not think they were raped.
Concerning the 42 percent of purported victims who had sex afterwards with their supposed assailants, again new data appear to have surfaced. Describing these findings in her chapter in Rape and Sexual Assault, Koss notes: "Surprisingly, 42 percent of the women indicated that they had sex again with the offender on a later occasion, but it is not known if this was forced or voluntary; most relationships (87%) did eventually break up subsequent to the victimization." Three years later, in a
letter to the Wall Street Journal (July 25, 1991), Koss is no longer surprised by this finding and evidently has new information revealing that when the students had sex again with the offenders on a later occasion they were raped a second time and that the relationship broke up not "eventually" (as do most college relationships), but immediately after the second rape.
Referring to this group's behavior, Koss explains: "Many victims reacted to the first rape with self-blame and thought that if they tried harder to be clear they could influence the man's behavior. Only after the second rape did they realize the problem was the man, not themselves. Afterwards, 87 percent of the women ended the relationship with the man who raped them." Koss also suggests that since many students were sexually inexperienced, they "lacked familiarity with what' consensual
intercourse should be like."
These explanations are not entirely convincing. It is hard to imagine that many twenty-one year old college women, even if sexually inexperienced, are unable to judge if a sexual encounter is consensual. As for the victims blaming themselves and believing they might influence the man's behavior if they tried harder the second time, Koss offers no data from her survey to substantiate this reasoning. Although research indicates that victims of rape tend to blame themselves, there is no evidence
that this induces them to have sex again with their assailant. One might note that there are cases of battered wives who stay on with their husbands under insufferable circumstances. But it is not apparent that the battered-wife syndrome applies to a large proportion of female college students.
With regard to the operational definition of rape used in the Ms. study and described in the earlier reports, Koss continues to claim that the study measures the act of "rape legally defined as penetration against consent through the use of force, or when the victim was purposely incapacitated with alcohol or other drugs." No explanation is offered for how the researcher detects the "intentional incapacitation of the victim" from affirmative answers to questions such as: "Did you have unwanted
sex because a man gave you alcohol?" Although these responses account for about 40 percent of the incidents classified as rape and attempted rape, when describing the study to the Senate Judiciary Committee and in other writings, Koss's examples of typical items used to define rape do not include these questions.
Reviewing the research methodology for the Ms. survey in Rape and Sexual Assault (1988) and the Journal of Consulting and Clinical Psychology (1987), Koss explains that reliability and validity studies conducted in 1985 on the ten-item Sexual Experience Survey (SES) instrument showed that few of the female respondents misinterpreted the questions on rape. A serious question arises, however, whether the validity study cited by Koss was conducted on the version of the SES instrument that was
actually used in the Ms. survey or on the original version of this instrument, which differed significantly from the one the Ms. findings are based on. The Sexual Experience Survey instrument originally designed by Koss and Oros, and reported on in a 1982 issue of the Journal of Consulting and Clinical Psychology contained none of the questions dealing with rape or attempted rape "because a man gave you alcohol or drugs."
In 1985, Koss and Gidycz reported (again in Consulting and Clinical Psychology) on the assessment of this instrument's validity, which they said indicates that: "To explore the veracity of the self-reported sexual experiences, the Sexual Experiences Survey (original wording) was administered to approximately 4000 students." Although Koss cites this report as evidence of the Ms. study instrument's validity, if the SES as originally worded was used, it is not at all clear that the assessment of
validity included the vague items on "intentional incapacitation," which were absent from the original version of the SES instrument.
Finally, the vast discrepancy between Ms. study figures and the number of students who generally seek rape counselling or report incidents of rape to authorities on college campuses is accounted for by the assertion that most college women who are sexually violated by an acquaintance do not recognize themselves as victims of rape. According to Koss, "many people do not realize that legal definitions of rape make no distinctions about the relationship between victim and offender." Findings from
the Bureau of Justice Statistics suggest that the crime of being raped by an acquaintance may not be all that difficult to comprehend; in recent years 33 to 45 percent of the women who said they were raped identified their assailant as an acquaintance.
In support of the Ms. project findings, Koss invokes additional studies as sources of independent verification. Some of these use different definitions of forced sexual behavior (including verbal persuasion and psychological coercion) and involve samples too small or nonrepresentative for serious estimates of the size of the problem. Others are referred to without explanation or critical examination. For example, Koss cites Yegidis's findings in the Journal of Sex Education and Therapy (1986),
which show a prevalence rate of rape for college students in the range reported by the Ms. study, as supportive evidence. But Yegidis defined rape as forced oral sex or intercourse, where the use of "force" included verbal persuasion. As she explains: "This study showed that most of the sexual encounters were forced through verbal persuasion-protestations by the male to 'go further' because of sexual need, arousal, or love." According to this definition, the conventional script of nagging and
pleading "everyone does it," "if you really loved me, you'd do it," "I need it," "you will like it," is transformed into a version of rape.
Claiming that the Ms. survey's estimates of rape prevalence "are well-replicated in other studies," Koss refers to Craig's discerning review of the literature to confirm the consistency of prevalence data on college students. This is a curious citation, since Craig in fact is of a different opinion. Analyzing the problems of definition in Clinical Psychology Review (1990), she notes that they "vary from use of force to threat of force, to use of manipulative tactics such as falsely professing
love, threatening to leave the woman stranded, or attempting to intoxicate the woman." Even when studies use the same general definitions, their authors often develop idiosyncratic measures to operationalize the terms. All of this leads Craig to conclude "that this lack of consistency limits the comparability of studies and makes replication of results difficult."
Moving beyond the experiences of college students, Diana Russell's study is another major source often quoted as evidence that rape has reached, as she describes it, "epidemic proportions throughout society." Reported in several books, including Rape in Marriage (1982) and Sexual Exploitation (1984), Russell's findings indicate that 38 to 54 percent of the women sampled were sexually abused as children, 44 percent were victims of rape (26 percent) or attempted rape (18 percent) an average of
twice in their lives, and many other women suffered experiences in marriage that, if not rape, were very close to it.
As for the latter, if mutually desired intercourse and rape are placed at either end of a continuum, Russell explains, "our study suggests that a considerable amount of marital sex is probably closer to the rape end of the continuum." Indeed, beyond marital sex, Russell suggests that according to this view "much of what passes for normal heterosexual intercourse would be seen as close to rape." Although this analysis of sex relations is not as censorious as Andrea Dworkin's, for whom all
heterosexual sex is rape, it would seem to lean in that direction.
There are several fundamental problems with Russell's survey, which is based on interviews with a group of women in San Francisco. Although serious efforts were made to achieve a random sample of participants, the researchers were able to complete their interviews with only 930 people out of an original sample of 2000. Thirty-six percent of the people contacted refused outright to participate in the study.
As for the other nonparticipants, Russell offers two somewhat different accounts for their inaccessibility. In Rape in Marriage she explains: "Because of a high incidence of not-at-homes during the summer months when the interviews were conducted, and because of an unexpectedly large number of households in which no eligible women resided, the original sample of two thousand drawn by the methods described proved insufficient for obtaining one thousand completed interviews." Later, it appears
Russell really did not know the number of households ("unexpectedly large") in which no eligible women resided. As she describes the sampling difficulties in Sexual Exploitation: "Many of the households that were inaccessible or where no one was at home might have been households without eligible women - there are a large number of all-male households in San Francisco."
In any event, for whatever reasons, more than 50 percent of the sample did not participate in the interview survey. The standard textbook criterion recommends that properly executed interview surveys should achieve a completion rate of 80 to 85 percent, which is the range usually required of surveys by federal agencies. It is highly doubtful that the 930 participants who agreed to be interviewed for Russell's study should be considered a representative random sample of the women in San
A more basic problem, however, is that starting with a questionable sample, Russell goes on to claim that her respondents' sexual experiences reflect not only those of all women in San Francisco, but are representative of the entire female population of the United States. A brief disclaimer to the effect that generalizing from the San Francisco sample "would be highly speculative," is quickly forgotten as, after adjusting her findings for age-specific probabilities, Russell concludes: "It is
indeed shocking that 46 percent of American women are likely to be victims of attempted rape or completed rape sometimes in their lives."
This conclusion she continues to share with the media. She also notes that these victims are likely to be attacked an average of two times. The fact that only 31 percent of the women in Russell's sample were married compared to a 63 percent marital rate nationally, is one of many reasons why this national estimate drawn from the sexual experiences of the San Francisco sample is not only highly speculative, but scientifically groundless. Russell offers a more detailed analysis using the San
Francisco findings to extrapolate, not the lifetime probability, but the national incidence of rape and attempted rape in 1978. As she explains in Sexual Exploitation:
The incidence figure for rape in the Russell survey cited above was 35 per 1000 females. (This includes cases of rape and attempted rape that occurred to residents of San Francisco, both inside and outside the city.) This is 24 times higher than the 1.71 per 1000 females reported by the Uniform Crime Reports (emphasis in original).
Based on this calculation, the 1978 Uniform Crime Report's figure of 67,131 for all cases of rape and attempted rape in the United States is then multiplied by twenty-four, which yields Russell's national estimate of 1.6 million incidents for that year.
Faulty logic and poor arithmetic invalidate this analysis. First, the initial calculation (ironically italicized) is incorrect. The San Francisco rate of 35 per 1000 is 20.5 times (not 24 times) higher than the Uniform Crime Report's 1.71 per 1000. Ignoring, for a moment, this arithmetical error, the logic of the second calculation assumes a correspondence between the Uniform Crime Report rates for San Francisco and the nation at large. If Russell's sample had an incidence rate twenty-four times
higher than the Uniform Crime Report's rate for San Francisco, in that case one need only multiply the Uniform Crime Report's national rate by twenty-four to project the local difference on a national scale. However, those who study this subject know very well that the reported rates of rape are considerably higher for metropolitan areas than for the national population. Indeed, the Uniform Crime Report's total of 67,131 cases of rape and attempted rape in 1978 amounted to a national rate of .6
per 1000 females, which was about one-third the rate of 1.7 per 1000 females it showed for San Francisco. Thus, if we include the initial arithmetic error, Russell's national estimate of the incidence of sexual assault exaggerates by about 350 percent the figure that would result from simply an accurate reading of her own facts.
The Ms. study by Koss and Russell's survey of sexual exploitation are highly sophisticated examples of advocacy research. Elaborate research methods are employed under the guise of social science, to persuade the public and policy-makers that a problem is vastly larger than commonly recognized. This is done in several ways: 1) by measuring a problem so broadly that it forms a vessel into which almost any human difficulty can be poured; 2) by measuring a group highly impacted with the problem and
then projecting the findings to society-at-large; 3) by asserting that a variety of smaller studies and reports with different problem definitions, methodologies of diverse quality, and varying results, form a cumulative block of evidence in support of current findings; and 4) by a combination of the above.
Advocacy research is a phenomenon not unique to feminist studies of rape. It is practiced in a wide variety of substantive problem areas and supported by groups that, as Peter Rossi suggests, share an "ideological imperative," which maintains that findings politically acceptable to the advocacy community are more important than the quality of research from which they are derived. Playing fast and loose with the facts is justifiable in the service of a noble cause, just as is condemning or
ignoring data and sentiments that challenge conventional wisdom. Denounced for expressing objectionable sentiments, for example, folk singer Holly Dunn's hit, "Maybe I Mean Yes - When I say no" was clearly out of tune with the feminist mantra, "no means no." The controversy over these lyrics ignored Muehlenhard and Hollabaugh's inconvenient findings that 39 percent of the 610 college women they surveyed admitted to having said no to sexual advances when they really meant yes and fully intended
to have their way.
Although advocacy studies do little to elevate the standards of social science research, they sometimes serve a useful purpose in bringing grave problems to public attention. No matter how it is measured, rape is a serious problem that creates an immense amount of human suffering. One might say that even if the rape research magnifies this problem in order to raise public consciousness, it is being done for a good cause, and in any case the difference is only a matter of degree. So why make an
issue of the numbers?
The issue is not that advocacy studies simply overstate the incidence of legally defined rape, but the extent to which this occurs and what it means. After all, the difference between boiling and freezing is "only a matter of degree." The tremendous gap between estimates of rape and attempted rape that emerge from data collected annually by the Bureau of Justice Statistics (BJS) and the figures reported in advocacy studies have a critical bearing on our understanding of the issue at stake.
The BJS surveys, actually conducted by the Census Bureau, interview a random sample of about 62,000 households every six months. The confidentiality of responses is protected by federal law and response rates amount to 96 percent of eligible units. The interview schedule asks a series of screening questions such as: Did anyone threaten to beat you up or threaten you with a knife gun or some other weapon? Did anyone try to attack you in some other way? Did you call the police to report something
that happened to you that you thought was a crime? Did anything happen to you which you thought was a crime, but you did not report to the police?
A positive response to any of these screening items is followed up with questions like: What actually happened? How were you threatened? How did the offender attack you? What injuries did you suffer? When, where did it happen, what did you do, and so forth.
As a guide to trends in sexual assault, the BJS data show that rates of rape and attempted rape declined by about 30 percent between 1978 and 1988. As for recent experience, BJS findings reveal that 1.2 women in 1000 over twelve years of age were victims of rape or attempted rape. This amounted to approximately 135,000 female victims in 1989. No trivial number, this annual figure translates into a lifetime prevalence rate of roughly 5 to 7 percent, which suggests that one woman out of fourteen
is likely to experience rape or attempted rape sometime in her life. As do other victimization surveys, the BJS studies have problems of subject recall, definition, and measurement, which, as Koss and others have pointed out, lead to underestimation of the amount of sexual assault.
Assuming that the BJS survey underestimated the problem by 50 percent - that is, that it missed one out of every two cases of rape or attempted rape in the sample - the lifetime prevalence rate would rise to approximately 10 to 14 percent. Although an enormous level of sexual assault, at that rate the BJS estimates would still be dwarfed by the findings of Koss and Russell's studies, which suggest that one in two women will be victimized an average of twice in their life.
This brings us to the crux of the issue, that is, the huge differences between federal estimates and advocacy research findings have implications that go beyond matters of degree in measuring the size of the problem. If almost half of all women will suffer an average of two incidents of rape or attempted rape sometime in their lives, one is ineluctably driven to conclude that most men are rapists. "The truth that must be faced," according to Russell, "is that this culture's notion of masculinity
- particularly as it is applied to male sexuality - predisposes men to violence, to rape, to sexually harass, and to sexually abuse children."
In a similar vein, Koss claims that her findings support the view that sexual violence against women "rests squarely in the middle of what our culture defines as 'normal' interaction between men and women." Catherine MacKinnon, one of the leading feminists in the rape crisis movement, offers a vivid rendition of the theme that rape is a social disease afflicting most men. Writing in the New York Times (December 15, 1991), she advises that when men charged with the crime of rape come to trial,
the court should ask "did this member of a group sexually trained to woman-hating aggression commit this particular act of woman-hating sexual aggression?"
Advocacy research not only promulgates the idea that most men are rapists, it provides a form of "scientific" legitimacy for promoting social programs and individual behaviors that act on this idea. When asked if college women should view every man they see as a potential rapist, a spokeswoman for the student health services at the University of California, Berkeley told the Oakland Tribune (May 30, 1991), "I'm not sure that would be a negative thing." This echoes the instruction supplied in one
of the most popular college guidebooks on how to prevent acquaintance rape. "Since you can't tell who has the potential for rape by simply looking," the manual warns, "be on your guard with every man."
These experts on date rape advise college women to take their own cars on dates or to have a back-up network of friends ready to pick them up, to stay sober, to go only to public places, to be assertive, to inform the man in advance what the sexual limits will be that evening, and to prepare for the worst by taking a course in self-defense beforehand. Separately, some of the instructions, such as staying sober, are certainly well advised. Collectively, however, this bundle of cautions transmits
the unspoken message that dating men is a very dangerous undertaking.
Beyond seeking courses in self-defense, the implications drawn from advocacy research sometimes recommend more extreme measures. Last year, at a public lecture on "The Epidemic of Sexual Violence Against Women," Diana Russell was asked by a member of her largely feminist audience whether, in light of the ever-present danger, women should start carrying guns to protect themselves against men. Stating that personal armament was a good idea, but that women should probably take lessons to learn how
to hit their target, Russell's response was greeted with loud applause.
Not all feminists, or members of the rape crisis movement, agree with the view that all men are predisposed to be rapists. Gillian Greensite, founder of the Rape Prevention Education program at the University of California, Santa Cruz, writes that the seriousness of this crime "is being undermined by the growing tendency of some feminists to label all heterosexual miscommunication and insensitivity as acquaintance rape." (One is reminded that 50 percent of the students whom Koss defined as
victims of rape labelled their experience as "miscommunication.") This tendency, Greensite observes, "is already creating a climate of fear on campuses, straining relations between males and females."
Heightened confusion and strained relations between men and women are not the only dysfunctional consequences of advocacy research that inflates the incidence of rape to a level that indicts most men. According to Koss's data, rape is an act that most educated women do not recognize as such when it has happened to them, and after which almost half of the victims go back for more. To characterize this type of sexual encounter as rape trivializes the trauma and pain suffered by the many women who
are true victims of this crime, and may ultimately make. it more difficult to convict their assailants. By exaggerating the statistics on rape, advocacy research conveys an interpretation of the problem that advances neither mutual respect between the sexes nor reasonable dialogue about assaultive sexual behavior.
It is difficult to criticize advocacy research without giving the impression of caring less about the problem than those engaged in magnifying its size. But one may be deeply concerned about the problem of rape and still wish to see a fair and objective analysis of its dimensions. Advocacy studies have, in their fashion, rung the alarm. Before the rush to arms, a more precise reading of the data is required to draw an accurate bead on this problem and attack it successfully.
Neil Gilbert is Chernin Professor of Social Welfare and Social Services at the School of Social Welfare of the University of California at Berkeley. He is author of Capitalism and the Welfare State; Clients and Constituents; and co-author of The Enabling State: Modern Welfare Capitalism in America; and Protecting Young Children from Sexual Abuse: Does Preschool Training Work?
[Current affiliation: University of California, Berkeley School of Social Welfare]