Measuring Sexual Assault on Campus
“1 in 4 Women Experience Sex Assault on Campus,” shouted the headline of a New York Times article on September 21st, 2015. The lead paragraph went on to state that: “In four years of college, more than one-fourth of undergraduate women at a large group of leading universities said they had been sexually assaulted by force or when they were incapacitated, according to one of the largest studies of its kind, released Monday.” I found the statement especially alarming since my daughter was in college at the time.
During several years researching Yes Means Yes: A Novel, the question of how to measure the occurrence of sexual assault on college campuses was a hard one to answer. As I recalled my own days as an undergraduate, I didn’t remember meeting or hearing of anyone involved in sexual assault. And the more I read, the more I found it hard to quantify. Is sexual assault on campus a serious problem? Is it an epidemic requiring procedural changes in campus policy as recommended by Department of Education in its Dear Colleague letter? I wanted to know, so I explored two statistical measures of the same problem, in a sense book-ends to a range of estimates.
Jeanne Clery Act
In Yes Means Yes, I mentioned the history of the Jeanne Clery Act, which requires all colleges and universities that participate in federal financial aid programs to keep and disclose information about crime on and near their respective campuses. Since Yes Means Yes takes place in Boulder, I reviewed the University of Colorado 2016 Disclosure of Campus Security Policy and Campus Crime Statistics, a report published annually as required by the law. Here is a table summarizing crimes reported for 2015:
Note that the Clery report follows the state of Colorado legal code to define sexual assault and unlawful sexual contact, and the Violence Against Women Act for related offences. The sum of the reported offenses (16+10+20+4+15) is 65. The number of female students at CU in 2015 can be estimated as 44.7% of 18,201, or 8,132. Therefore, the rate of sexual assault reported crimes in 2015 was 0.8%. The report explains in detail the methodology used to arrive at its estimates, and I found this note most interesting:
For Clery Act purposes, CU Boulder is required to report crimes that occur at noncampus buildings or property that are owned or controlled by student organizations officially recognized by CU Boulder. Panhellenic sororities and Multicultural Greek organizations (listed at http://www.colorado.edu/greeks) are officially recognized by CU Boulder and own or control off-campus housing buildings. The Boulder Police Department has primary responsibility for responding to reports of crimes that occur at these noncampus locations in the City of Boulder. As such, the Boulder Police Department monitors and records criminal activity reported to have occurred at Boulder locations owned or controlled by the Panhellenic sororities and Multicultural Greek organizations. CU Boulder does not officially recognize Interfraternity Council member houses; therefore, crimes that occur on their properties are not counted in the annual Clery statistics.
It would be illuminating to know how many sexual assault crimes, originating on Greek properties, were reported to the Boulder Police in the same time frame.
Campus Climate Survey Validation Study Final Technical Report
A second example of measuring the rate of sexual assault on college campuses comes from the U.S. Department of Justice, Bureau of Justice Statistics, Campus Climate Survey Validation Study Final Technical Report. This detailed report presents the results of a nine-school pilot test that was conducted to develop a campus climate survey for collecting school-level data on sexual victimization of undergraduate students during the 2014-15 academic year. To achieve this goal, a web-based survey was developed to focus primarily on sexual assault victimization. A limited set of victimization questions were also asked about the broader reference periods of sexual assault experienced since beginning college and over the students’ lifetimes. In addition, the survey included items for capturing experiences with sexual harassment; coerced sexual contact; intimate partner violence; and perpetration of sexual harassment and sexual assault.
Fourteen schools were asked to participate in the survey, and they represented diversity of both public and private funding, and of various sizes, ranging from less than 5,000 students to more than 20,000. Nine schools agreed to participate, and surveys were completed by more than 23,000 undergraduate students (approximately 15,000 females and 8,000 males). The average response rates across all nine schools was 54% for females and 40% for males. Response rates for females ranged from 43% (School 4) to 71% (School 5).
As anyone who has taken an undergraduate statistics course learns, the quality of sample responses determines the true representative nature of the mathematically derived estimates across the total population. For example, even though the sample population was randomly selected at each school, were respondents who fully completed the survey self-selected? In other words, would someone who had experienced sexual assault be more likely to complete a survey that someone who hadn’t experienced sexual assault? The CCSVS final report addresses these issues in detail and adjusts for estimated bias across respondents. Also, most CCSVS sample members were offered a $25 gift card as an incentive for completing the survey.
To determine whether a respondent had experienced sexual assault, a number of definitions and survey questions were utilized. Here is the statement that defines unwanted sexual contact, and subsequent survey questions referenced this definition and asked participants about the type and number of unwanted sexual contact that they had experienced.
Based on completed surveys across all nine schools, and adjusted for bias and sample error, the prevalence rate for completed sexual assault experienced by undergraduate females during the 2014–2015 academic year, averaged across the nine schools, was 10.3%, and ranged from 4.2% at School 2 to 20.0% at School 1. The prevalence rate for completed sexual assault since entering college among the female sample ranged from 12% at School 4 to 38% at School 1, with a cross-school average rate of 21%. The percentage of undergraduate females who experienced sexual assault during their lifetime ranged from 26% at Schools 4 and 9 to 46% at School 1, with a rate of 34% for all nine schools combined.
Regarding the estimates for the incidence of sexual assault since entering college, the CCSVS notes several caveats:
As discussed in Section 5.2.2, no information about the number of such incidents, the type of unwanted/nonconsensual sexual contact that occurred, the tactic used, the month/year of the incident, or any other incident-level details were obtained about victimizations experienced prior to the 2014–2015 academic year. In addition, the longer reference period for these estimates of approximately 3.5 years for senior females might be more susceptible to measurement error in the form of recall bias or telescoping. Therefore, the “since entering college” estimates should be interpreted with caution.
Thinking back to the attention-grabbing headline of the New York Times article that “1 in 4 Women Experience Sex Assault on Campus,” the CCSVS study suggests a rate close to one in five during their undergraduate years, still a number that is too high. Another finding of the survey, one that lends support to efforts to provide a more inclusive environment on campus for victims of sexual assault, is that across the nine participating schools, only 4.3% of sexual battery incidents and 12.5% of rape incidents were reported by the victim to any official.
The University of Colorado Clery report put the rate of reported sexual assault on campus at less than 1.0%. The Campus Climate Survey Validation Study (CCSVS) found a rate of 10.3% across nine campuses. Why the difference? Clearly, reporting an incident to the police requires a willingness to engage with authorities, file a written report, and subject oneself to public scrutiny—a difficult decision for the victim if she feels she has no credible witnesses or evidence. A prosecutor in the King County Special Assault Unit told me that sex crimes are the hardest to prove because of the difficulty in establishing evidence. “At least with murder, there’s normally a corpse,” he added grimly.
What can We Infer?
What accounts for the over ten times higher rate reported by the CCSVS than the actual Clery crime statistics? One explanation could be that the definition used in the survey is broader than criminal sexual assault or rape, and includes unwanted sexual contact and touching. Another is that reporting a historical event on an anonymous web-based survey over twenty minutes presents much less personal investment than reporting an actual crime to the police. Regardless, the rate of reporting, which is low my any standards, is a concern. Surveys show that most college assaults are not reported for several reasons. When asked, victims have said they didn’t have proof that the incident occurred, were afraid of retaliation by the perpetrator, were scared of hostile treatment by the authorities, were uncertain the authorities would consider the incident serious enough, or wanted to prevent family and others from learning about it.
So how big is the problem? This past summer I visited a friend’s fifteen-year-old niece at a summer camp in the San Juan Islands. For four weeks, girls and boys age nine through sixteen live together in a natural setting and learn about teamwork, self-reliance, and spiritual awareness. My own daughter attended there and was a counselor during college summers. Outdoor skills are emphasized, like sailing, kayaking, and bicycling, over extended multi-day trips. As we spent the day visiting with the niece, enjoying an amazingly beautiful summer afternoon and strolling around the island, we inadvertently came across the camp director and a counselor engaged in serious conversation with a male camper. Once we were out of earshot, the niece explained that on one of the overnight trips, the male camper had touched a female camper in an inappropriate and unwanted manner while in a tent. Everyone in the camp was aware of the incident, and a decision was being made about whether to expel the camper. I asked her if that was something she ever worried about, of boys being too aggressive. Her response saddened me: “Yes, you can’t be too careful and it’s something I’m on-guard against all of the time.”
Our children should grow up confident that if someone wants to engage in a sexual relationship, it will be appropriate and it won’t happen without prior consent. Learning of how the niece, at age thirteen, was always on the lookout for predators, told me all I needed to know about the size of the problem.
Photo Credit: Lois Mueller firstname.lastname@example.org