The Sex Study That Could Alter Our Understanding of Campus Assault

Columbia University researchers are “spying” on undergraduates, the headline of a New York Post story asserted. The article, which appeared beside a page-length photo of Beyoncé cradling a chicken, criticized the methods of a new study of students’ sex lives. How the story came to be highlights the ethical and communications challenges facing the university’s up-close research project — and why that project might change our understanding of campus sexual violence.

The story starts with Erin Mizraki, who in 2015 was a Barnard College freshman covering sexual-assault activism for the Columbia Daily Spectator. Columbia had been under a national spotlight the year before, when Emma Sulkowicz protested how the university had dealt with a sexual-assault complaint by lugging around a 50-pound mattress. So it was significant when Mizraki and her editors learned that Columbia hoped to improve assault prevention by investigating the social and sex lives of its undergraduates.

As part of the project, called the Sexual Health Initiative to Foster Transformation, or Shift, ethnographers had started to observe students at locations such as sports events, club meetings, dorm parties, and neighborhood bars. A sophomore spotted Shamus Khan, a sociologist known for his popular introductory course, taking notes one Wednesday night in a booth at a bar called 1020. How, Mizraki began asking, did students feel about being watched?

That October, Mizraki published their responses in a Spectator article describing how some students worried about being observed drinking underage, or about being watched by a researcher with whom they might have to interact later in class. Her story balanced such concerns with a discussion of the numerous steps researchers were taking to safeguard students’ privacy and confidentiality. For example, the ethnographers refrained from recording any students’ names in their field notes.

Two days later, the Post published a rewrite of her article, nuances buried, under the headline, “‘Spying’ Prof on Campus: Columbia Party Kids Balk.” All this watching students, the Post story began, was “creeping them out.” At a time when skewed news can easily go viral, especially in a media capital like New York, it was the kind of publicity that might harm a sensitive research project.

“When we were writing the story about Shift, it made a lot of sense to write it from that angle,” Mizraki says. “But, I think upon reflection — and now that they’ve actually done the study — I don’t know if I would have done that again. Because there were just so many measures taken to make it an ethical study.”

Two years later, the Columbia project is once again generating buzz, as researchers embark on a tightly managed rollout of their findings.

In November, the Shift team published the first of some 10 initial papers, reporting the results of a survey on the prevalence of sexual assault and risk factors among Barnard and Columbia students. Then, in December, news emerged that Khan and a colleague, Jennifer S. Hirsch, had sold a book about the project, described as “the most rigorous study of sexual assault ever conducted,” to Norton. The social scientists would examine “the often hidden forces of campus ecosystems that determine how and when assault happens, and why so few assaults are ever reported,” according to Publishers Lunch, a book-industry newsletter.

After the 27-campus study done in 2015 by the Association of American Universities, the White House report produced in 2014 by the Obama administration, and the saturation coverage in articles and books, you might wonder what remains to be learned about college sexual assault. Answer: a lot.

“It’s almost shocking that we know so little,” says Justin R. Garcia, a sex researcher and evolutionary biologist at Indiana University’s Kinsey Institute. He added, “We recognize there is a problem. Everyone is scrambling to do something. There’s not even close to enough data or clearly demonstrated interventions that work.”

At a time of national reckoning with sexual violence, could Columbia’s study help change that?

Many of the Columbia team’s findings are still under wraps. But in December, the two lead researchers — Hirsch, a medical anthropologist, and Claude A. Mellins, a clinical psychologist — gave The Chronicle a broad overview of their study. Alexander Wamboldt, an ethnographer hired to focus on male student groups, like fraternities, also spoke with The Chronicle via Skype. Perhaps not surprisingly, given the “spying” professor episode, a Columbia communications official monitored these conversations, which took place in a 15th-floor conference room at the Mailman School of Public Health.

Shift originated in the fall of 2014, when Hirsch approached senior administrators with an idea to fill a gap in the national discussion about sexual assault. Much research involved surveying the prevalence of assault. Prevention efforts, meanwhile, emphasized changing individuals’ beliefs or expanding their knowledge. In a 2014 review of such programs, the Centers for Disease Control and Prevention found scant evidence of their effectiveness in reducing sexual violence.

Hirsch had spent her career studying the social forces that shape sexual behaviors, such as men’s engagement in extramarital sex. She proposed a small ethnographic study in that vein focused on the organization of students’ social and sex lives: where they live and hang out, how they spend their time, how institutional policies affect them.

“What are the social factors that shape students’ regular, normal, desired, consensual sexual behavior?” Hirsch hoped to understand. “And what are the modifiable social factors that create the conditions under which sexual assault is likely to occur?”

The project came at a charged moment. That September, Sulkowicz, who now uses the pronoun “they,” began carrying the mattress across campus. Sulkowicz pledged to keep carrying it until the student they said had raped them, Paul Nungesser, departed the university. Columbia, which had cleared Nungesser, found itself engulfed in fliers, meetings, litigation, open letters, political speeches, federal complaints, and student demonstrations. On the facade of a campus building, one student group projected the phrase, “Columbia protects rapists.”

Against the backdrop of that controversy, every bit of it chronicled in the national media, Hirsch’s small proposal grew into a sweeping project. The faculty-run study would eventually comprise a team of ethnographers immersed in campus life for more than a year, plus in-depth student interviews and surveys — all of it backed by more than $2 million from the president’s office.

Announcing Shift in 2015, Columbia’s president, Lee C. Bollinger, said that the project would “help remedy what is a national deficit in evidence-based information relevant to creating the most effective prevention programs and policies.”

The Columbia project is more far-reaching than previous studies of this kind, says Elizabeth A. Armstrong, a sociologist at the University of Michigan at Ann Arbor who has done ethnographic research on undergraduates’ social and sexual lives. “To my knowledge,” she says, “there has not been a single university that has actually commissioned its own faculty to engage in such a comprehensive self study of its undergraduate sexual culture.”

Researchers, of course, can’t observe sexual assault. But a challenge of getting near the problem, and understanding its causes, is that ethnographers face what Armstrong describes as a “quagmire” of ethical dilemmas. The heart of the issue is how lenient universities are willing to be in allowing researchers to temporarily step outside of their faculty roles to pursue important knowledge.

For example, during Armstrong’s study of assault, conducted at a large research university that she did not identify, the institutional review board that vetted her work prohibited her from observing anything unlawful. That meant no parties, she says, since “there is no such thing as an undergraduate party that doesn’t involve illegal behavior.”

“We just haven’t been able to get close enough,” she adds, to “really get at it.”

Columbia’s ethnographers could attend parties, following a consent protocol. One of the party’s hosts had to invite them. People could ask them to leave. Hirsch and Khan, the senior professors who ran the ethnography, would not enter students’ private spaces.

Wamboldt, one of the junior ethnographers who did enter those private spaces, is a densely bearded anthropologist finishing up his doctorate at Princeton University. He did not blend in at dorm parties. Most students, he says, tried to figure out why he was there.

With each new hand he shook, Wamboldt offered the same awkward introduction: “I’m a researcher at the Sexual Health Initiative to Foster Transformation. We’re looking at sexual assault and sexual health issues on campus. You don’t need to talk to me if you don’t want to. But if you do, it’s completely confidential.”

At first, he says, students tended to be wary, concerned that he might report on drinking or other violations of the code of conduct.

“Once they figured out that that was not our role, students in general — I would say across the board — viewed the topic as a really important thing to learn more about,” he says. “Many of them had things in place, either formally or informally, for themselves that they were trying to do to minimize the risks of sexual assault on campus.”

Still, Khan, who declined to speak with The Chronicle, did acknowledge in a 2015 interview with The Blue and White, an undergraduate magazine, that his research in student bars might come off as “a bit creepy.” One Barnard student, Sara Taffel, agreed. Being observed while out at a bar would be “pretty weird and uncomfortable,” she told the Spectator, “especially if it’s a professor you’ve had in the past.”

Her comment underscores the challenge of drawing neat divisions between scholars’ research and teaching roles in this kind of campus self-study — a separation that Columbia students who participated in Shift sometimes had difficulty understanding.

“Those roles are complex,” Armstrong says. “They can move back and forth really quickly — in the sense of, well, what happens if a student that was in a research project shows up in your class?”

For his part, Wamboldt says the field research could be “stressful” and “fast paced.” In previous projects, he had not dealt with as many people who were using substances, especially alcohol. Situations arose where Wamboldt worried about “rapid changes in students’ health.”

“We had an outstanding policy while doing work that we were always going to be human beings first and researchers second,” he says. “So if we saw anything that we were worried was placing someone in immediate risk to themselves or others, we would intervene in that.”

Ethnographers also conducted private interviews with students about their backgrounds and sexual practices. To hear candid stories, they worked with the general counsel’s office at Columbia to get an exemption from the federal requirement that most university employees must report an assault if a student tells them about it. In Mellins’s experience, most victims never report assault. The exemption allowed researchers to promise confidentiality. But it also necessitated them to refer students in need of help to the appropriate services.

Wamboldt estimates that Columbia students disclosed just over 80 assaults during interviews. Only five of those had been previously reported.

During those interviews, researchers tried to discover students’ shared assumptions regarding sexual consent — knowledge that is key for shaping prevention efforts. They asked students to divulge at least one sexual encounter in detail. Then they asked students to retell the story, focusing on when consent was or was not given.

In many instances, students described encounters that they themselves did not label as an assault, but the researchers did.

Those stories typically involved students who had gotten so intoxicated that they didn’t remember the encounter, or weren’t conscious as it happened, while their partner had been far less inebriated. Some students said all of their sexual activity had taken place under such circumstances. They viewed it as how they have sex. Or, in any case, how they have sex at this point in their lives.

“There’s a lot of drunk sex, and it’s actually kind of intentional,” Hirsch says. “It’s not like people are like, ‘Oh my goodness, I got drunk and then I hooked up.’ It’s like, drunk hooking up is part of the fun.”

And, while that finding may hardly come as a shock, it points to a tension between the requirements of law and the realities of student life. New York, like California and several other states, has an “affirmative consent” statute that compels partners to obtain the equivalent of, “Yes, I want to have sex.” Columbia provides explicit video tutorials on good and bad ways to get and give that consent. Top of the bad list: You can’t give consent when you’re drunk.

The lesson, for Hirsch, is that a university should go beyond the drunks-can’t-consent message. It should also adopt the public-health approach of harm reduction. That means admitting that people are doing something dangerous and figuring out ways to make it safer.

For example: alternative spaces whose availability might steer drunk kids away from ending up on a dorm bed. At Columbia, many lounges had been converted into study areas, Hirsch says, so there were not a lot of couches to sprawl on at 4 a.m. Now the university’s dining officials, who have been speaking with Shift’s research team, are keeping a campus eatery open all night.

This emphasis on the physical environment extends to details as minute as the furniture in dorm rooms. When students described their sexual encounters, one thing they emphasized repeatedly — to the dismay of the ethnographers interviewing them — was the belief that the moment they had consented to have sex with someone was when they had entered their room or sat on their bed.

Interviewers would ask whether, realistically, there had been any other place to sit. Often, there hadn’t. “On the one hand, we have this bed symbolizing consent,” Wamboldt says. “And, on the other hand, the built environment might not be giving people that many choices.”

What Shift’s findings will add up to, ultimately, is one broad theme: disaggregating the different experiences lumped under the phrase “sexual assault.” Universities, the thinking goes, could apply that fine-grained knowledge toward more targeted prevention efforts.

You see that potential in some of the surprising results emerging in the early published work, such as the prevalence of male sexual-assault victims. It’s clear that women are much more likely than men to be victimized. In Columbia’s research, 28 percent of female survey respondents had experienced a sexual assault. But in surveying and interviewing students, the Shift team also discovered that a nontrivial percentage of men — nearly one in eight — had also suffered an assault.

Consider, Hirsch says, the example of a man who is plied with drinks all night by a woman, which leads to sex that he hadn’t wanted. When he subsequently describes that situation to a female friend, she informs him that he has been sexually assaulted. This particular student shrugs it off. But other men, Hirsch says, find such experiences “deeply disturbing,” even more so because their stories don’t fit the standard narrative. Sometimes, their friends treat them as a joke.

Or consider verbal coercion. Incapacitation through drugs or alcohol is the most-common way people are assaulted, the Columbia scholars found. But they also identified a less-discussed pattern of students getting verbally coerced into unwanted sex. This might involve lying to someone or threatening to divulge their sexuality or end the relationship.

It matters because many prevention efforts emphasize the intervention of bystanders. “That implies there are people around,” Mellins says. “But if verbal coercion is going on, and no one is around, that requires a whole other level of intervention. And just focusing on whether the person said yes isn’t going to help.”

Will any of this help put a dent in the problem?

Armstrong, the Michigan sexual-assault researcher, believes Shift has the potential to produce changes her own work could not.

In 2006, Armstrong published a paper on sexual assault based on a year that she and her colleagues had spent observing and interviewing the student residents of one dorm floor at the unnamed university. Her study, like Columbia’s, probed the interplay of attitudes, interactions, and institutional arrangements. An organizational sociologist, she came to view universities as environments that systematically and predictably produce assault.

As she saw it, this resulted, in part, from the homogeneity of dorms, with their concentrations of single, white, affluent, 18-year-old, party-oriented kids. This culture “intensified social anxiety,” she wrote, “heightening the importance of partying for making friends.” University policies that seemed gender neutral exacerbated risk for these women. Armstrong found, for example, that policing of alcohol in dorms pushed freshmen to party in male-controlled fraternities.

Men preyed on women by getting them drunk, obstructing doors, and controlling transportation. Women were expected to play the part of sexual “gatekeeper,” she wrote, so men saw sex obtained by diminishing their capacity to resist as “consensual.” Their behavior was legitimated by the cultural belief that men were “naturally” sexually aggressive.

Armstrong speculated on interventions, like more consistent policing of alcohol (avoiding contrasts like heavy policing of residence halls and light policing of fraternities) and new and more diverse housing arrangements. Many people have taught her article in academic classes and used it to train staff at sexual-assault awareness and prevention centers. But the article appeared at a cultural moment when the academic discussion of campus sexual assault was small and public interest minimal.

The Columbia project, with its institutional support, its close-up view of student life, and its emergence amid a new awakening about sexual violence, is different. It might help colleges change student culture, she says, if they muster the resources and political will.

Marc Parry is a senior reporter who writes about ideas, focusing on research in the humanities and social sciences. Email him at marc.parry@chronicle.com, or follow him on Twitter @marcparry.