How an Influential Campus Rape Study Skewed the Debate
Widely cited study relies on surveys that don't actually have anything to do with on-campus sexual assaults.
The public debate over the extent and causes of the campus sexual assault crisis is fraught with misleading information. The previously acclaimed work of psychologist David Lisak deserves that distinction as well.
The federal government, universities, and members of Congress have all used Lisak's theories to justify rape adjudication policies that are biased against accused students. They should reconsider those policies in light of new discoveries about the inapplicability of Lisak's work.
Lisak has cultivated a reputation as one of the nation's foremost authorities on sexual assault, and his thinking undergirds the most vexingly anti-due process policies currently mandated by the Education Department's Office for Civil Rights. His authority on the subject is so uncontested that even critics of draconian anti-rape policies feel obligated to grapple with his assertions, according to Slate's Emily Yoffe, who described Lisak's work as foundational "in the movement to curb campus sexual assault."
President Obama's January 2014 memo announcing the creation of a White House task force to address campus sexual assault repeatedly cites Lisak. His research provides evidence of the notion that "campus rapists are often serial predators" who perpetrate a "cycle of violence" unless stopped, according to the memo.
The 2002 Lisak study that supposedly makes that case—"Repeat Rape and Multiple Offending Among Undetected Rapists"—is fundamental to the activist campaign to reduce campus rape. But despite the study's prominence, its assertions about the serial nature of campus rapists are dubiously sourced, according to a thorough investigation conducted by Reason contributor Linda LeFauve.
The study pooled data from four separate surveys of interpersonal violence that were conducted at the University of Massachusetts-Boston during the '90s, at which time Lisak was employed as an associate professor. Lisak's study had a total sample size of 1,882 men, 120 of whom gave responses in the surveys indicating that they were predators. Of the 120 rapists, 76 were judged to be repeat offenders, leading to the oft-cited claim that the majority of campus sexual assault is the work of serial predators who remain "undetected," i.e., are never convicted for their crimes.
The claim suffers when scrutinized. For reasons left unclear, the four surveys that contributed data are never actually identified in the study. In fact, Lisak struggled to recall which ones he used when asked about them during the course of a telephone interview with LeFauve. When LeFauve suggested to him that the data in question came from his doctoral students' dissertations and masters' theses, he agreed that this was "probably" the case.
I spoke with James Hopper, one of Lisak's former students at UMass-Boston, who confirmed that the survey data he conducted for his own dissertation was included in the 2002 study. He also identified several other students as near-certain contributors via their masters' theses and dissertations.
What's remarkable about these surveys is that they don't actually have anything to do with campus sexual assault (aside from the location where they were conducted).
Researchers set up tables at different areas of campus and handed out questionnaire packets to men who passed by them; participants who returned the questionnaires received a few dollars. The surveys made no attempt to prevent non-students from participating. The researchers had no reason to do so, since their questions weren't aimed at on-campus attacks and did not specifically ask about violence committed by or against students. And the average respondent was 26.5 years old—several years older than the typical college student—reflecting the fact that UMass-Boston is a commuter school with a significant number of older, non-traditional students.
This is quite the revelation: The canonical text of the campus sexual assault crisis is filled with data repurposed from academic papers that never intended to survey campus violence in the first place.
Lisak admitted as much during his conversation with LeFauve, agreeing with her contention that the surveys weren't designed primarily to study campus sexual assault and admitting that "a number of these cases were domestic violence situations." During a presentation at Emory University in 2013, he also said that UMass-Boston was "demographically different than a traditional four-year college."
In an interview with Reason, Hopper described the survey respondents as "working-class, first-generation college students" who didn't live on campus.
"This is not a typical college sample," he said.
The issues don't end there. In an August 20, 2012 interview with the Star-Telegram, Lisak asserted that he subsequently interviewed the sexual predators in the study. "We started with questionnaires, then interviewed men who responded to the questionnaires, we did screening interviews and to verify and confirm that everything they had reported was accurate," he said.
Several other interviews and news articles about Lisak imply that he extensively interrogated the subjects of his 2002 study. He also told LeFauve during his conversation with her that he had interviewed "most of them." And yet when LeFauve asked him to explain how this was possible—given that most of the surveys he relied on were anonymous—he hung up the phone.
It's true that Lisak interviewed 15 students who had committed sexual assault during his graduate studies at Duke University in the '80s. But the idea that he conducted similarly comprehensive interviews with a majority of the subjects of the 2002 study—and thus gained some special insight into the minds' of serial campus rapists that would allow him to make claims beyond the scope of the study—has no basis in fact.
Lisak did not answer my repeated phone calls and emails asking for clarification. Paul Miller, a former student of Lisak's who coordinated at least one of the four surveys and is co-author of the study, also did not return calls. The UMass-Boston psychology department did not respond to a request for comment.
Extensive digging, as well as interviews with Hopper (who relayed some of my questions to Lisak) revealed that it was possible for Lisak's team of researchers to conduct interviews with some, though by no means a majority, of questionnaire respondents. One of the four surveys asked subjects to provide their contact information for subsequent interviews, and a minority of those subjects did so. According to Hopper, Lisak said his team may have subsequently interviewed approximately 22 of the 120 participants who gave responses indicating that they were rapists. These interviews were validation exercises designed to ensure that respondents had given truthful answers during the initial survey. Lisak did not recall precisely how many were conducted—whether it was greater or fewer than 22—and was unwilling to look it up for me, according to Hopper.
Lisak did not respond to me directly and only answered questions about his methodology through Hopper. The fact that he avoided answering such simple questions—questions that should have been answered in the text of the study itself—suggests an obscuring (either deliberate or coincidental) of the study's actual purpose.
Unlike Lisak, Hopper agreed to participate in a wide-ranging conversation about methodology. He vigorously defended Lisak, their fellow researchers, and the work that was done.
"This went through institutional review and this was people of integrity conducting this research," he said.
Hopper also believes Lisak's conclusions about the serial nature of campus rapists are well justified.
"They are entitled!" he insisted during our conversation after I suggested that students might be less likely to commit violence than the general populace. "They think they can get away with it. If anything, they might be at higher risk for committing sexual assault."
Nevertheless, in the years since "Repeat Rape and Multiple Offending" was published, the subject and scope of Lisak's research has been vastly misinterpreted. I'll leave it to others to debate the extent to which Lisak himself is culpable for encouraging this confusion, but at the very least, he allowed it to continue. He has repeatedly claimed in interviews that the overwhelming majority of campus sexual assaults—more than 90 percent—were premeditated by serial perpetrators.
"The vast majority of sexual assaults on campuses, in fact over 90 percent, are being perpetrated by serial offenders," Lisak told Al Jazeera. That article cites the 2002 study—a study that had nothing to do with campus sexual violence, remember—as evidence of this.
This curious interpretation of Lisak's work has spawned a glut of federally-mandated anti-rape policies. To understand why, it's necessary to first understand what the theory of serial predation implies. Prior to the widespread adoption of Lisak's views, campus rape was often considered to fall into the supposedly less serious category of "date rape." Students who committed rape were assumed to be one-off offenders motivated by alcohol and circumstance into crossing blurry lines. But the 2002 study turned this thinking on its head by revisiting campus rapists as sociopaths inclined to commit violence over and over again. Abuse was in their nature, and reforming them was difficult. They were not motivated by circumstance: In fact, they planned out their crimes in advance, much like the sociopathic rapists who allegedly attacked Jackie in the infamous, false Rolling Stone story, "A Rape on Campus."
Since most campus rapists* are serial perpetrators rather than date rapists, according to Lisak, universities should feel obligated to take stronger corrective measures. Lisak himself has said that every rape accusation "should be viewed and treated as an opportunity to identify a serial rapist." This logic makes some sense, but only if one accepts this interpretation of the research. Such thinking makes it much easier for administrators to justify the abridgment of due process rights for accused students, and to operate from the presumption that accused students are guilty—of a great number of rapes, no less.
But Lisak's 2002 study falls well short of proving that this approach is justified. His surveyed perpetrators weren't traditional college students. It's possible that some of them weren't students at all, since the surveys had no mechanism for ensuring this. What the study did find was a small proportion of the UMass-Boston community—perhaps but not necessarily students—had a history of violence. This violence may or may not have happened in proximity to campus. It may or may not have happened to students. It may or may not have happened to children, spouses, or the elderly.
Given that the 2002 study is much less than what it seems, it should come as little surprise that more recent research contradicts its findings. On July 13, Georgia State University Psychology Professor Kevin Swartout and his colleagues published a study, "Trajectory Analysis of the Campus Serial Rapist Assumption," that seriously undercuts Lisak's ideas. Swartout's study, which did specifically set out to gauge student-on-student violence, found that most college rapists could not accurately be described as serial predators.
"Although some men perpetrate rape across multiple college years, these men are not at high risk entering college and account for a small percentage of campus perpetrators—at least 4 of 5 men on campus who have committed rape will be missed by focusing solely on these men," wrote the study's authors.
Policymakers would be wise to let go of the notion that the average American campus is home to a dedicated contingent of rampaging sociopaths that must be stopped at all costs. Such thinking has provided cover for federal bureaucrats to endlessly expand their efforts to root out imaginary monsters—to the detriment of due process and academic freedom. It has also duped the media into uncritically accepting the lies of people like Duke University's Crystal Mangum and UVA's Jackie, whose nightmarish tails of ritualistic, premeditated violence destroyed the reputations of dozens of innocent people.
None of which is to say that the campus rape crisis is made up. Women are assaulted on college campuses all the time. But if we want to do anything to stop this, we might look for solutions outside the lens of a study that was never about campus violence in the first place.
(For more on this story, please see Linda M. LeFauve's piece on David Lisak's work, "Campus Rape Expert Can't Answer Basic Questions About His Sources.")
*This sentence has been corrected: most rapists are serial offenders, and 90 percent of rapes are committed by serial offenders, according to Lisak.
Hopper's quote referencing a sense of "entitlement" was expanded, and contexted provided, to improve clarity.