Junk Science and Campus Rape

A new inquiry casts serious doubt on the most influential study on collegiate sexual assault.

Alex Milan Tracy/Sipa USA/Newscom

David Lisak is hardly a household name. But over the last decade, he has become the single most important expert on a topic of increasing national concern—sexual violence on college campuses. Lisak's authority on the subject is well-established: The White House cites him in briefing papers, anti-rape activists promote his work in movies and books, and university administrators invite him to give lectures and sit on panels. Even those who are skeptical about the existence of a massive new campus rape crisis have largely declined to dispute Lisak's most significant finding: that the majority of campus rapists are serial offenders who commit routine violence until and unless they are apprehended.

Lisak's views have dominated the conversation about campus rape ever since the release of his 2002 study, "Repeat Rape and Multiple Offending Among Undetected Rapists." He believes campus rape adjudicators should view each and every accusation "as an opportunity to identify a serial rapist," since the accused student is likely to have raped multiple women. He has encouraged colleges to do more to root out serial rapists and banish them from campus. Efforts to reduce campus rape that focus on education and awareness are unlikely to succeed on their own, according to his line of thinking. These men can't be taught not to rape—they are undetected career criminals, and their very existence justifies the federal government's meddlesome intervention into students' sex lives.

But unquestioned deference to Lisak may have been a serious tactical and intellectual mistake. Why? An investigation into Lisak's signature work casts serious doubt on the reliability of his serial predator theory. The 2002 study routinely cited as foundational evidence in collegiate sexual assault discussions isn't even about campus rape—and that's just one of its many flaws.

For years, Lisak has exaggerated the scientific support for his theory while selling himself and his policy solutions to advocates, administrators, and politicians. Given that his science is much less convincing than it seems, perhaps the policies based on it also deserve a more skeptical look.

Dubious Data

Public outrage over a purported epidemic of sexual violence on American college campuses may have reached a high water mark on January 22, 2014. That's when the Obama administration released a 34-page report establishing a White House Task Force to Protect Students from Sexual Assault.

The report included several of the erroneous statistics that have inspired mass hysteria over the prevalence of rape on campuses. "College students are particularly vulnerable: 1 in 5 women has been sexually assaulted while in college," it asserts on page 1.

These are astonishing claims. Also astonishing is how quickly they collapse when scrutinized by fact checkers. Women attending college aren't "particularly vulnerable." The truth is precisely the opposite: Women attending college are less likely to be raped, according to the Bureau of Justice Statistics—an unsurprising finding, since wealthier, better-educated people typically experience less violence than the socioeconomically disadvantaged. The 1-in-5 statistic holds up only slightly better; the study that produced that number had serious limitations. When surveyed, large percentages of college students will admit they've endured nonconsensual touching, but they do not see themselves as victims of rape.

That these claims overstate the extent of the problem is something that a growing and disparate chorus of voices now seems to recognize. In February 2015, the education reporter for the left-leaning politics/policy website Vox admitted that the 1-in-5 statistic was "probably inaccurate"; just five months prior, Vox Editor in Chief Ezra Klein had argued that the 20 percent figure was why he was supporting what he called a "terrible law" mandating affirmative consent for collegiate couples daring to touch each other. Nearly a year after the release of the White House memo, Inside Higher Ed ran a story skeptical about the statistic, and The Washington Post's fact-checking operation asserted flatly that the White House claims had gone too far.

Yet left unscrutinized until now was an equally disturbing "fact" cited later in the White House memo. "Notably, campus assailants are often serial offenders," it stated. "One study found that of the men who admitted to committing rape or attempted rape, some 63% said they committed an average of six rapes each."

That study was authored by Lisak, a former professor of psychology at the University of Massachusetts Boston. Lisak knew from his previous work on sexual assault that incarcerated rapists were often repeat offenders, and wondered whether "undetected rapists"—i.e., people who were never caught—would share this characteristic. His study interviewed 1,882 subjects at UMass Boston, 120 of whom he determined to be rapists who had never been caught. Seventy-six of those 120 were deemed to have committed multiple assaults, which led Lisak to conclude that a majority of undetected predators, just like their incarcerated brethren, were serial rapists.

"Those serial offenders were prolific," Lisak said in an October 28, 2013, interview with Al Jazeera. "The average number of rapes for each one of those serial offenders was six."

But were they in fact "campus assailants," as the White House claimed and based policy recommendations on? The researchers didn't even ask the participants—who ranged in age between 18 and 71, averaging 26.5, with more than 20 percent older than 30—whether they were students. They just set up booths at a commuter college and asked willing men to answer a long questionnaire for some pocket change (between $3 and $5). And none of the questions—not one—asked participants specifically about violence they had committed against other students, or on campuses.

Occasionally, Lisak has acknowledged his study's limitations, albeit without emphasizing that the participants were never asked about their enrollment status. "Students are a little bit older," he told an audience at Emory University during an April 2013 presentation. "They are working mainly, half-time, full-time some of them. Kind of a cross section of working-class Boston, but still, young people, in college."

But in the Al Jazeera interview, Lisak also asserted that the repeat offenders have "perfected ways of identifying who on campus, for example, are most vulnerable," as if he had gleaned personal insight into the minds of serial campus rapists in the course of extensive back-and-forths with the subjects. That impression, it turns out, is highly misleading.

Linda LeFauve, associate vice president for planning and institutional research at Davidson College, dug into the study for a July article at reason.com and made a startling discovery: The underlying research wasn't even conducted by Lisak. Rather, he had appropriated the data from four different surveys administered by his graduate students during the 1990s. The surveys, which are not specified in the 2002 study, were derived from his students' dissertations and master's theses. During a phone interview with LeFauve, Lisak was unable to recall which surveys had wound up in the study. Part-time Harvard University instructor Dr. James Hopper, a former student of Lisak's whose research was used in the study, was able to reconstruct the sources of much of the data.

Significantly, the surveys were not specifically about campus rape; they were about more general acts of violence that men did or did not commit over the course of their lives. Participants who answered "yes" to certain questions (such as: "Have you ever had oral sex with an adult when they didn't want to because you used or threatened to use physical force if they didn't cooperate?") were labeled rapists for the purposes of the study, and participants who had committed multiple instances of rape under this definition were labeled serial predators.

Lisak acknowledged during the course of a telephone interview with LeFauve that the surveys weren't primarily about campus violence. They "may have been about child abuse history or relationships with parents," he said. He also supposed that many examples of repeat rape in the study were domestic violence, i.e., not the kind of serial predation one would normally expect to find on a college campus.

Lisak defended his prior claims that he had interviewed most of the rapists in the 2002 study. When LeFauve pressed him to explain how this was possible, given that the studies were anonymous and had been coordinated by his students, he ended the call. He has not responded to subsequent requests for interviews.

Hopper explained that it would have been possible for researchers to interview some, though by no means a majority, of the participants, since one of the four surveys provided for respondents to be interviewed at a later date. But these subsequent interviews were validation exercises designed to ensure that respondents had given truthful answers during the initial survey, not probing examinations of the inner minds of serial predators.

Date Rapists or Serial Rapists?

Prior to Lisak, campus rapists were often branded "date rapists." The thinking was that students who committed sexual violence against other students were typically single-instance offenders who had gotten carried away due to a variety of circumstances: heavy drinking, lack of understanding about sexual consent definitions, and so on. Date rape is no less serious than serial predator rape, but the tools needed to address a crisis consisting predominantly of the first kind of assault are different than the tools needed to prevent the second kind. Teaching students to drink responsibly, for instance, might mitigate date rape, but is an inadequate tactic for confronting a fringe element of career sociopaths.

In the 13 years since "Repeat Rape and Multiple Offending Among Undetected Rapists" was published, Lisak has become a minor celebrity in the world of anti-rape activism. His serial predator theory is the dominant lens through which the issue is examined, and his policy preferences are given center stage.

But according to feminist professor Mary Koss, who specializes in sexual assault at the University of Arizona's College of Public Health, the reliance on Lisak's theory has been a grave mistake. "It's one of the most egregious examples of a policy with an inadequate scientific basis that lives on because it offers a simplistic solution," Koss tells reason.

Koss believes that Lisak himself is directly responsible for elevating his faulty science to the level of popular acceptance and governmental influence. "It didn't happen by accident," she says. "He played more than a role. He described his information as being relevant to serial rape from day one."

Lisak's serial predator theory has been directly contradicted by a new paper Koss co-authored, titled "Trajectory Analysis of the Campus Serial Rapist Assumption." Based on two surveys of the adolescent sexual histories of a combined 1,645 male college students, the authors found that 177 (11 percent) had committed at least one rape between age 14 and the end of their college careers, of which a whopping 75 percent "only perpetrated rape during 1 academic year."

"Although a small group of men perpetrated rape across multiple college years," the paper concludes, "they constituted a significant minority of those who committed college rape and did not compose the group at highest risk of perpetrating rape when entering college."

If Lisak is wrong, and most campus rapists are not serial predators, what should be done about collegiate sexual assault? Koss has some ideas. Many of the young men who commit rape in college can be taught better behavior, she says. They also might be well-suited to a different, less Orwellian campus adjudication process than what has become increasingly common across the country.

Koss would like to see more rape disputes settled via "restorative justice" involving discussion between the two sides. The goal is not necessarily to punish a rapist, but to allow both the victim and the accused to achieve closure. Universities—which possess significant resources to assist students who are dealing with trauma but are often unequal to the task of fairly adjudicating sex crimes—are an ideal vehicle for restorative justice in cases where victims are looking for validation and assistance instead of punitive justice, she says.

"In my work with the restorative processes, there are a couple of preconditions," said Koss. "One precondition is that due process is observed. A second precondition is that victims' rights are observed. A third thing is that participation is voluntary by both people."

Those guarantees might not fully allay due process concerns, given how poorly university administrators have handled sexual assault allegations in the past. But voluntary university-sponsored dispute-resolution would still be a vast improvement over the current regime, which requires that universities play judge, prose­cutor, and jury at mandatory, farcical extrajudicial proceedings.

There is a big obstacle standing in the way of this approach, however: the federal government. In the last few years, the Department of Education's Office for Civil Rights has consistently told universities that federal law—Title IX of the Higher Education Act, specifically—requires them to adjudicate rape disputes. Koss' solution would also no doubt be opposed by many victims' advocates, who see expulsion as the only just outcome when a student is even accused of rape.

And that's why it matters so much whether Lisak's theory is accurate or misleading. If most student rapists are irredeemable villains, it's easy to justify draconian efforts to hunt them down and expel them from campus. But if most student rapists are one-off offenders—ignorant about relationships, capable of learning from their mistakes—school might actually be the right place for them.

"These are educational institutions, so at our foundation we have some faith that people can change over time," Koss said.

The validity of Lisak's work and influence is very much an open question within the community of experts who have studied sexual assault. But it remains unscrutinized by policy makers who have been all too eager to implement Lisak's suggestions and media figures who have cheered them for doing so. The blunt fact remains that the canonical study on campus sexual violence comprised four 1990s surveys of men between the ages of 18 and 71—at a commuter school with no campus housing—who were never asked specifically about student-on-student rape and who may not even have been students themselves.

The ensuing serial predator theory casts a long shadow over the campus rape issue, but neither victims nor the accused are well-served by the sacrosanct status it has achieved.