T

he policy conflict over allegations of sexual assault on college campuses has become a battle over rights. Conservatives worry about the rights of the accused in the hands of overzealous administrators, while liberals fear that those same officials might ignore the rights of victims by focusing on due process and thereby adding to the trauma of those who say they were assaulted. In broad legal terms, conservatives say that evidence against the accused should be “clear and convincing,” while liberals favor a less demanding standard. This conflict is driven by radically different views of the rape problem itself. Conservatives are concerned that scarred and angry women might be leveling assault charges in the regretful wake of drunken hook-ups, while liberals insist that in such encounters, consent is rarely if ever actually granted.

While such disputes have dominated the discussion, comparatively little attention has been given to a more fundamental question: How can universities lessen the need for these painful trials by reducing sexual exploitation and drunken couplings? Even when reformers think about policies to address the deeper problem, they nearly always neglect social regulations, especially controls on alcohol and dorm life. The University of California, for example, established a task force on sexual assault that “specifically recommended separating the issue of sexual assault from alcohol-awareness education,” the Los Angeles Times reported. As for co-ed dorm life, the pundit Matthew Yglesias captured the prevailing wisdom when he criticized single-sex dorms, noting that a college student is an “adult, a person fully equipped to enjoy having sex—a fun, affordable, and ecologically sustainable pastime.”

Any successful approach to reducing sexual assault on campus will require the reconsideration of these very notions. We have completed a study of more than 1,300 colleges and universities and have found that far fewer accusations of sexual assault are levied at the schools that ban alcohol and prohibit opposite-sex overnight guests in residence halls. Our findings suggest that students are not so much victims of a “rape culture” as they are victims of faulty institutional policies that contribute to a higher risk of sexual assault.

The Campus Study

First, a brief description of our method. We collected data on every college in the U.S. News and World Report: 2015 rankings that maintains undergraduate dorms. For each college, we calculated the number of reported campus sexual assaults for every 1,000 female students living in its residence halls. We considered only on-campus assaults because our goal was to measure the effectiveness of campus social regulations in deterring them. In looking at the effect of alcohol policies, we gathered data from U.S. News and World Report as well as from individual college handbooks and websites. We classified campuses as “dry” if they prohibit all alcohol possession, consumption, and distribution on campus. Finally, using data from student handbooks and residence-life websites, we divided dorm-visitation policies into two types: Those that banned opposite-sex overnight guests outright, and those that permitted the practice, at least on some days or in some undergraduate dorms.

Here is what we found: In recent years, assault rates have been 3.1 to 4.4 times higher at the most permissive colleges and universities than at their more restrictive counterparts (see Table 1). That difference is substantial.

Consider two campuses—one permissive, the other restrictive—that both house 3,000 female undergraduates in their residence halls. The permissive campus is likely to receive somewhere between 65 and 100 more reports of sexual assault over a 10-year period.

One possible objection to our findings might be that these differences reflect underreporting of sexual assaults at conservative and religious colleges. But while it is true that such underreporting has been an issue on college campuses generally, there is no systematic evidence to suggest a plague of it at conservative or religious campuses in particular. In fact, our evidence shows that reports of sexual assault have been climbing at all types of colleges, including at regulated and religious ones (see Tables 1 and 2). The fact that reports have increased everywhere suggests that colleges of all stripes and the students enrolled in them are responding to the heightened national awareness about sexual assault.

Other studies support this conclusion. A 2009 study by the Center for Public Integrity profiled many permissive colleges with serious underreporting problems, including the University of Colorado, Eastern Michigan University, Florida State, West Virginia University, and Yale. And a 2002 Justice Department study on reporting profiled eight campuses, including two with restrictive social policies: West Virginia State University (dry) and Oklahoma State University (guest ban). While the Justice Department study praised these two colleges for the way they reported and adjudicated cases of sexual assault, it was critical of some of the permissive colleges it profiled. The study, for example, found that the reporting protocols at no-prohibitions UCLA “need[ed] to be tighter in terms of capturing all . . . cases of rape and sexual assault.”

Nonetheless, the same Justice Department suggested that dry policies might actually depress the number of assault reports. “If student victims know that they are in violation of a policy forbidding the use of drugs or alcohol,” the study read, “this might make them fearful to report a sexual assault.” Its own subsequent research undermined this conjecture. A 2007 DOJ study found that victims who had been incapacitated by drugs or alcohol were much less likely to report assaults, often because they did not consider their memory of the incident good enough to offer a credible report or because they were unsure about whether a crime occurred. Logically, then, the problem of unreported assaults may be especially acute at campuses that do a poor job of controlling alcohol consumption.

It is fair to wonder whether the apparent effects of social regulations are actually driven by religiosity. If they are, it would mean that religion both causes schools to implement these social regulations and depresses sexual violence. It is certainly true that religious colleges and universities are much more likely than secular ones to ban alcohol and overnight guests of the opposite sex. It is also true that assault rates are generally lower at religious colleges. But the profession or practice of religion itself does not appear to diminish violence very much. As Table 3 makes clear, permissive religious and secular colleges suffer from comparatively high levels of sexual assault. And when we assessed the unique influence of religion by including it in a regression analysis with variables for alcohol and guest bans, it had little influence on assault rates and was not statistically significant.

The evidence clearly suggests that secular institutions can control sexual assault without having to “find religion.” Indeed, many are doing just that. In our data, 171 secular colleges ban alcohol, 43 ban overnight guests, and 90 do both. The most restrictive secular colleges, moreover, have sexual-assault rates that are 1.7 to 2.8 times lower than those at the most permissive secular schools. Restrictive secular colleges tend to be located in states with conservative values, especially in the Midwest and the Deep South. But given the effectiveness of these regulations at controlling sexual assault, they might appeal to the citizens of deep-blue states as well, especially in this age of “helicopter” parents.

Other factors may influence the efficacy of social regulations. For example, bans on alcohol and overnight guests are unevenly enforced. But since we could only identify the policies themselves, we have no good measure of enforcement. In addition, all overnight bans are not equal. Some campuses, for example, ban opposite-sex guests after 11 p.m., while others allow them in the dorms until 3 a.m. Others still, especially conservative Protestant ones, forbid opposite-sex guests inside dorm rooms at any time of the day or night. Though we did not systematically study such campuses, sexual assault seems to occur rarely in their residence halls. We therefore believe that rates of sexual assault fall lower still when campuses embrace more stringent bans and enforce their own policies.

Why Social Regulations Work

Although our study offers the first quantifiable analysis of campus policies in relation to sexual assault, a large body of research supports its findings. We know, for example, that there is a strong link between sexual assault and alcohol, especially on college campuses. A 2004 study of college students in the Journal of Studies on Alcohol found that 72 percent of victims were intoxicated at the time of the assault. Research also shows that young men use alcohol to soften women’s reluctance to engage in sex. A 1985 study in the Archives of Sexual Behavior found that some 76 percent of college rapists admitted to using alcohol to weaken their victims.

It is not surprising that such men turn to alcohol. The sociologists Mark Regnerus and Jeremy Uecker say that regular alcohol consumption dramatically increases young women’s willingness to engage in casual sex. “It’s almost as if most students—especially but not only women—have a visceral aversion to casual sex that is only overcome with the help of alcohol,” Regnerus and Uecker concluded. This association is not lost on some young men. As one male freshman chillingly informed Regnerus and Uecker, sex under the influence “happens all the time . . . they’ll [the women] regret it, but it’s not like a tragedy.”

Even when alcohol is not used strategically, its consumption can lead women and men to misinterpret social cues. While alcohol depresses women’s ability to assess risk, it diminishes men’s ability to accurately perceive women’s interest in having sex with them. Alcohol also excites more aggressive and antisocial behavior in men. In a review of the literature, published in 2004 in Aggression and Violent Behavior, Antonia Abbey and her colleagues noted: “The cues that usually inhibit sexually aggressive behavior, such as concern about future consequences, sense of morality, or empathy for the victim are likely to be less salient than feelings of anger, frustration, sexual arousal, and entitlement.”

Less alcohol, naturally, is consumed on dry campuses. Evidence from a 1994 study of 140 campuses published in JAMA suggests that “individual binge drinking is less likely . . . if [the school] prohibits alcohol use for all persons (even those older than 21) on campus.” Are these differences caused by self-selection rather than by alcohol policies—or in other words, are dry campuses simply more appealing to students less likely to binge-drink? A 2001 study in the Journal of Studies on Alcohol suggests that self-selection only partly explains the lower rates of drinking on dry campuses. Drawing on a survey of more than 11,000 students at 19 dry and 76 regular colleges, it found that students who were heavy alcohol drinkers in high school tended to drink less heavily when they attended colleges with dry policies. The same study also undermined the theory that dry policies simply push drinking off campus. It found that on-campus residents at dry colleges were less likely to drink heavily at off-campus events, such as fraternity parties. In fact, the report concluded, “these findings do not support the assumption that displacement of heavy drinking off-campus and a heightened risk of drinking and driving and other drug use will occur at schools that ban alcohol.”

Our study found, nonetheless, that dry policies by themselves reduce the sexual-assault rate only modestly, though how much this finding is due to uneven enforcement is hard to say. The 2001 survey noted that while reports of unwanted sexual advances and assaults were less common on dry campuses, the differences were modest. Thus, dry policies seem to work best when they are combined with other regulations, especially bans on overnight guests.

Visitation policies are especially important for two likely reasons. First, they are almost certainly easier to enforce than alcohol bans (since it is much more difficult to conceal persons than, say, bottles). Second, visitation policies target the social settings in which so many campus assaults occur. A 2000 study by the Bureau of Justice Statistics found that approximately 90 percent of on-campus rapes occur in the room of either the victim or the assailant. It further found that half of all campus assaults happen after midnight, when students are more likely to be inebriated. Other studies find an even higher incidence of late-night assaults. A 2007 Department of Justice study of more than 5,000 undergraduate women found that 72 percent of assaults occurred after midnight, while some 90 percent of women who were incapacitated by drugs or alcohol reported late-night assaults. It seems, therefore, that visitation restrictions function to separate the sexes before they become too drunk to exhibit clear volition and good judgment.

If separating the sexes lowers assault rates, we should also expect fewer reports at single-sex colleges. And that is precisely what we found (see Table 4). Average assault rates are many times lower at single-sex colleges than in co-ed colleges. Nonetheless, social regulations still seem to make a big difference at single-sex colleges.

Why the Alternatives Fail Us

Despite the possible benefits of bans on alcohol and overnight guests, those who have dedicated their careers to cracking down on sexual assault prefer other remedies. That is why many universities are shifting the burden of proof from victims to perpetrators by lowering evidentiary standards. Making punishment a somewhat more likely consequence of assault may lower its incidence. But given the choice between limiting a student’s social freedom or depriving a student his due-process rights, the former is preferable. Those who are examining and policing assault on campus also argue for more education on sexual misconduct. While such efforts might be worthwhile, they are far less likely to protect young women and men as effectively as bans on alcohol and overnight guests.

Why? Because such sobering lessons, even well received, can get lost in the haze and confusion of intoxication, and because every college population is laced with antisocial men who never pay much heed to lessons in empathy. As a large body of research shows, rapists tend to be impulsive and exhibit high levels of narcissism. Such characteristics are found in not only the hardened and poorly educated class of rapists in prisons—but also in the well-educated offenders in our universities. A 1997 study published in Aggression and Violent Behavior, for example, found that college rapists possessed much higher levels of psychopathic traits than their nonviolent student peers. Another study, published in 1994 in the Journal of Social and Clinical Psychology, asked college men to listen to an audiotape of a date rape. Students who had committed sexual assault were much more aroused by the tape than those with no history of violence.

Such men need to be controlled, and it cannot be done through moral suasion. Even the best education campaign cannot inoculate a whole population from crime. Instead, we need to change a social context that is currently tailored to the preferences of those with psychopathic tendencies.

The emphasis on “rape culture” cannot account for the lower incidence of sexual assault on the most conservative and religious campuses where feminist sensibilities are weakest.  Such campuses are better at controlling crime, not because they are any more likely to condemn rape and certainly not because they are more progressive, but primarily because they are more regulated social environments.

Progressives who worry about the coercive nature of these regulations should remember that they still grant considerable freedom. Limiting visiting hours, for example, allows plenty of opportunities for sexual intimacy. Established couples can arrange their lives to take advantage of those ample opportunities, just as they are already accustomed to doing. And given how little time the average student spends studying outside the classroom, there is no shortage of hours in the college day for such encounters. Bans on overnight guests are more likely to frustrate drunken hook-ups at 3 a.m. than stifle the sex lives of caring couples.

The deeper problem with social regulations has to do with their limits, not their coerciveness. Most assaults, after all, do happen off campus, beyond the easy reach of college social regulations. It is possible, however, that even the off-campus problem could be partially mitigated by on-campus regulations in at least two respects. First, more social order might help students cultivate temperate habits while they are still adjusting to their new freedoms as freshman and sophomores. The 2007 Department of Justice study found that “women who are victimized during college are most likely to be victimized early on in their college tenure,” before they have had much experience navigating their new freedoms. Ordered campuses give them time to do so.

Second, colleges could also experiment with curfews in their residence halls, which would help prevent new students from venturing into the late-night party scene off campus. But even if the off-campus problem proves intractable, social regulations can make a significant difference at many schools, especially residential liberal-arts colleges where practically all students live on campus. If these colleges solve their on-campus problem, they solve the problem entirely. On the other hand, student resistance to new social regulations may undermine their effectiveness. If so, new rules may need to be phased in and accompanied by a public-education campaign.

Some conservatives, meanwhile, make a different objection to new regulations: They say that much of what we call assault is mostly the result of drunken hook-ups. In this view, the campaign against sexual assault is a moral panic, fueled by an ideological faith in a mythical problem. But even if conservatives are correct, social regulations should limit these soul-crushing hook-ups, not to mention the false charges and painful trials they can create. As conservatives often emphasize, universities should show more concern for the souls of their students.

Unfortunately, our best universities and colleges have shown little interest in dampening the sexual revolution, at least not until someone yells “rape.” As Heather Mac Donald reported recently, Brown University’s student services works to help students integrate sex toys into their relationships, including whips and restraints. Tufts University’s sex fair, meanwhile, includes dental-dam slingshots and “dildo ring toss,” while NYU offers orgasm workshops.

These liberationist campaigns forget that students are still moving into adulthood and have not gotten there yet. Colleges should recognize this transitional period by providing a more structured and less sexualized space for young people to grow into their new freedoms.

Even if our study’s findings are confirmed by further research, most campuses may reasonably decide to strike some balance between sexual liberty and security. Few want to follow the lead of fundamentalist colleges by banning any opposite-sex visitation in campus dorm rooms, even though doing so would greatly reduce the risk of assaults in their residence halls. But rejecting the most draconian policies does not mean that no balance could or should be struck. Campuses with comparatively high rates of sexual assault must decide whether they are willing to place some modest restrictions on their students, or whether their many rape charges are the necessary price of the sexual revolution.