Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

January 8, 2015

Research looks at drinking, sexual violence links

Drinking is related to sexual violence on college campuses, but where students drink and how often they do so may matter more than whether they drink heavily when it comes to a greater risk for victimization, or for victimizing others, said a Pitt public health researcher in a recent talk on campus.

“Not all sexual assaults occur with alcohol, but a large subset do,” said Christina Mair, faculty member in community and behavioral health sciences.

At what drinking levels are risks for sexual violence, victimization and perpetration greatest among college students who consume alcohol?

And how do the risks vary by where students are drinking? Is it riskier to drink at a Greek party than at a bar, dorm or tailgate party?

Researchers want to know because a better understanding of the contexts in which students’ drinking is likely to increase risky behavior could lead to more effective preventive interventions.

The issue of sexual assault on college campuses has been the focus of intense media attention in recent months. From an April 2014 White House task force report, to student complaints about how Columbia University handles assault claims to a controversial Rolling Stone article describing an alleged gang rape at a University of Virginia fraternity party, the issue is making the news.

“Of course this is not a new problem, but it is a problem that is receiving more scrutiny than it has, probably ever,” said Mair. “I think this is a really good opportunity for us in the public health field to become involved here.”

While much of the response has been focused on risk reduction or on intervention once a situation is occurring, “In public health it’s really interesting to think about how we could actually prevent these things,” she said as she shared some preliminary findings from her research, set to be published in the journal Addiction.

Commonly it’s thought that higher risk for alcohol-related problems goes hand in hand with higher levels of drinking. However, recent research is showing that it doesn’t necessarily take an excessive amount to raise some students’ risk, said Mair.

While four or more drinks for a woman and five or more for a man is considered “binge” drinking, “There’s no way to say five drinks is the magic number at which problems occur,” she said. “There’s a huge subset of individuals that we found who actually, at two drinks, three drinks, are at an increased risk level,” she said.

“Someone who, typically, when they go out has one drink may be at much greater risk for a set of alcohol-related problems when they have three or four drinks, which doesn’t necessarily sound like a huge volume of alcohol.”

In her research, Mair used a dataset that included nine years of data from students on 14 California college campuses who were surveyed about drinking and risky behaviors. The data included 37,000 drinkers who were asked about their drinking overall as well as within six common drinking contexts: Greek parties; dorm parties; on-campus events; off-campus parties; bars, and outdoor settings (such as tailgating).

Among other questions, the survey asked about victimization as well as perpetration: Whether the students felt pressured to have sex or whether they were sexually assaulted or raped, as well as whether they were concerned that they pressured someone to have sex or whether they may have sexually assaulted or raped someone.

Mair said 3.1 percent of respondents reported being pressured or forced to have sex (the equivalent of 19 events per 1,000 students in a 28-day period); 4.3 percent reported being taken advantage of sexually, with some students reporting both, for a total of nearly 6 percent reporting some sort of sexual victimization.

Fewer reported being perpetrators, with 1.2 percent responding that they’d pressured or forced someone to have sex. For every victim there’s not necessarily a perpetrator who sees himself/herself in that way, she said, adding that there were some repeat offenders in the survey.

Almost equal percentages of men and women reported being victimized, although men were twice as likely to report being perpetrators.

About half the students who reported victimization were freshmen in their first semester in college.

“Basically using alcohol — consuming one or more drinks — is  related  to  a  2  percent  increase in the population prevalence of reports of perpetration and 5 percent increase in prevalence of reports of victimizations. But these events are not generalized responses to differences in drinking levels,” she said, noting that the highest prevalence risk was when people went out and had two drinks.

“It’s not heavier drinking that’s causing this link between alcohol use and sexual violence. It seems like it doesn’t really matter how much you’re drinking, it’s being in these settings and drinking,” she said.

While severe outcomes were rare, preliminary analysis shows that contexts do matter. Mair said statistical significance was shown more often with regard to frequency of students’ alcohol use than to the quantity of alcohol use.

“Drinking more frequently at bars, outdoor events and off-campus parties was associated with increased risk of being forced or pressured to have sex,” she said. Quantity was significant at Greek parties. “Drinking more heavily at Greek parties led to increased probability of sexual violence or victimization,” she said.

For women, drinking more heavily on campus and at dorm parties was associated with a higher risk of sexual violence or victimization, while the frequency of drinking at Greek parties also was significant.

For men, drinking more at Greek parties was associated with higher levels of victimization while frequency of drinking was significant at bars, outdoor events and dorm parties.

As for perpetration, “We saw that drinking more frequently at all settings but dorm parties was associated with (respondents) being more likely to say they forced or pressured another person to have sex,” while none of the quantity effects was significant, Mair said.

Other factors remain to be analyzed, such as risks related to other people in each context: How many people were at the party? How many were drunk? What were the attitudes and expectations in that setting?

Understanding the context-specific risks has direct implications for prevention efforts, Mair said. “We can focus on specific locations and behaviors,” she said. “As a general rule, the more specific a preventive intervention program is, the more likely it is to be successful.”

Some solutions might include reducing access to high-risk contexts, eliminating drinking or reducing heavy drinking in high-risk contexts, although in these data, it doesn’t appear that heavy drinking is driving the associations with sexual violence, Mair said.

Given the recent spotlight on the issue, it’s an important time for research on sexual violence, Mair said.

“We have everyone’s attention for what’s been a sort of ignored problem: This is the time to do it.”

—Kimberly K. Barlow

Filed under: Feature,Volume 47 Issue 9