On the occasion of this year’s Grammy Awards, President Barack Obama released a public-service announcement targeting sexual violence, particularly against women and girls.
In his videotaped statement, Obama mentioned a statistic about the prevalence of rape -- with a frequency that was startling enough to provoke several PolitiFact readers to ask us to check it out.
"Right now," Obama said, "nearly one in five women in America has been a victim of rape or attempted rape."
Is that statistic correct? We took a closer look.
Vice President Joe Biden, for instance, said, "One in five of every one of those young women who is dropped off for that first day of (college), before they finish school, will be assaulted in her college years."
Biden and others were referring to college campuses -- unlike Obama’s most recent comment, which was referencing overall patterns in the population.
We’ll discuss the evidence for both "one in five" statistics in a moment, but first, let’s look at where Obama’s recent claim came from.
The CDC study
The number comes from the 2011 National Intimate Partner and Sexual Violence Survey, which was released in September 2014 by the federal Centers for Disease Control and Prevention. The survey was based on landline and cellphone interviews with more than 12,000 people who were 18 years or older, in all 50 states and the District of Columbia.
The survey’s summary table shows that 19.3 percent of women interviewed reported experiences that the authors categorized as rape or attempted rape. That’s "nearly one in five," so on the broad figure Obama cited, he correctly reported the CDC’s finding.
What qualified as rape in the survey? This is where it gets a bit complicated. The three subcategories listed are:
• "Completed forced penetration," 11.5 percent.
• "Attempted forced penetration," 6.4 percent.
• "Completed alcohol- or drug-facilitated penetration" without the ability to consent, 9.3 percent.
However, you can’t just add up all three subcategories -- you’d get 27.2 percent, which is significantly higher than the survey’s overall rate for rape of 19.3 percent.
CDC explained that the subcategories are derived from the overall rape estimate, and they overlap. Within each estimate, victims are counted only once, but due to the possibility of re-victimization, the same respondent may be represented in multiple categories, CDC spokeswoman Courtney Lenard said.
Comparing the CDC study with the campus study
As we noted, the campus study had some methodological problems. The biggest was that researchers surveyed undergraduates at two unnamed large public universities (one in the Midwest and one in the South) -- a sampling that may not be valid for the entire country. Large universities may not be representative of experiences at mid-size or small colleges. Further, the two colleges selected may not even be representative of large campuses.
In addition, the results for the campus survey were gleaned from 5,446 Web-based responses. Web-based polls yield a lower response rate than other methods, and researchers say that could skew results.
By contrast, the CDC study was based on the tried-and-true technique of telephone polling -- and used a far larger, and more random, sample.
It’s also worth noting that the studies had slightly different scopes. The campus survey was about "sexual assault," which included not only rape but also "forced touching of a sexual nature (forced kissing, touching of private parts, grabbing, fondling, rubbing up against you in a sexual way, even if it is over your clothes)." By contrast, the CDC figures cited by Obama focused exclusively on cases of penetration or attempted penetration.
For these reasons, we consider talking points based on the CDC study to be significantly more credible than those based on the campus study. Experts agree.
Last year, when we interviewed Mary Koss, a public health professor at the University of Arizona, she told us that the Campus Sexual Assault Study "is not the soundest data (the White House) could use."
For this fact-check, however, she said the CDC provides the best available data. Obama, she said, is "basically right. No survey is perfect, but this survey is considered very sound in the field."
We do see one legitimate reason for caution.
The CDC study takes a public-health approach in its methodology, asking respondents about their past experiences without looking at whether the activities may have qualified as crimes per se, as our friends at the Washington Post Fact Checker have noted. Surveys that take a criminal-justice approach, by contrast, focus on activities that were presumed to have been criminal.
Because of this methodological difference, criminal-justice surveys tend to report significantly lower rates of rape than public-health surveys do. For instance, the CDC study came up with 1.9 million cases of rape in 2012, whereas the National Crime Victimization Survey found far fewer rapes -- 347,000 -- over the same period.
So, given two survey methods to choose from, Obama picked the one that produced the higher number. Ordinarily, we take issue with claims that cherry-pick results. In this case, however, we aren’t as critical of his decision, since independent scientific experts have long questioned whether criminal justice surveys under-report instances of rape.
Notably, in 2014, the National Research Council concluded that under-reporting of rape in the National Crime Victimization Survey was "highly likely."
William D. Kalsbeek, an emeritus professor of biostatistics at the University of North Carolina and co-chair of the National Research Council panel, told PolitiFact that "the search continues for the best way to measure rape and sexual assault in population surveys. Our panel found that the broader, behavior-based definition of rape and sexual assault used by (the CDC study) is preferable to the more limiting, legal-based definition" used in the criminal-justice survey.
Obama said that "nearly one in five women in America has been a victim of rape or attempted rape."
A well-respected survey by the CDC found that 19.3 percent of women reported experiences that are considered to be rape or attempted rape under the survey’s guidelines. There are other surveys, using a different methodology, that show lower rates of rape, but researchers say the data in the CDC study is at least as credible, if not more so. The statement is accurate but needs clarification or additional information, so we rate it Mostly True.