Evaluating the One-in-Five Statistic: Women’s Risk of Sexual Assault While in College

Abstract

In 2014, U.S. president Barack Obama announced a White House Task Force to Protect Students From Sexual Assault, noting that “1 in 5 women on college campuses has been sexually assaulted during their time there.” Since then, this one-in-five statistic has permeated public discourse. It is frequently reported, but some commentators have criticized it as exaggerated. Here, we address the question, “What percentage of women are sexually assaulted while in college?” After discussing definitions of sexual assault, we systematically review available data, focusing on studies that used large, representative samples of female undergraduates and multiple behaviorally specific questions. We conclude that one in five is a reasonably accurate average across women and campuses. We also review studies that are inappropriately cited as either supporting or debunking the one-in-five statistic; we explain why they do not adequately address this question. We identify and evaluate several assumptions implicit in the public discourse (e.g., the assumption that college students are at greater risk than nonstudents). Given the empirical support for the one-in-five statistic, we suggest that the controversy occurs because of misunderstandings about studies’ methods and results and because this topic has implications for gender relations, power, and sexuality; this controversy is ultimately about values.

Notes

^1. We use the word college broadly to refer to any institution of higher education for undergraduates—that is, for students who have completed high school but who have not yet earned a bachelor’s degree.

^2. We use a dagger (^†) to identify articles from the popular media.

^3. Some of these sources might more accurately be described as data compilations rather than studies.

^4. Even acknowledging that someone could be sexually assaulted more than once, a formula assuming uniform risk across students and years would overestimate the prevalence: If prob_SA = the probability of being sexually assaulted during one year, and prob_NoSA = the probability of not being sexually assaulted during one year, then prob_NoSA = 1 − prob_SA. The probability of not being sexually assaulted over four years = prob_NoSA⁴ (i.e., prob_NoSA to the 4th power, which = prob_NoSA × prob_NoSA × prob_NoSA × prob_NoSA), and the probability of being sexually assaulted over four years = 1 − prob_NoSA.⁴ The AAU Survey (Cantor et al., 2015) found a risk of 13.2% for the current school year; assuming uniform risk would have resulted in a four-year risk estimate of 1 − (1 − 13.2%)⁴ = 1 − (0.868)⁴ = 1 − 0.568 = 43.2%, which is much higher than the 27.2% of seniors who actually reported this experience since entering college. The Campus Sexual Assault Study (Krebs et al., 2009) found an annual risk of 7.5% for completed sexual assault; assuming uniform risk would have resulted in a four-year risk estimate of 1 − (1 − 7.5%)⁴ = 1 − (0.925)⁴ = 1 − 0.732 = 26.8%, which is much higher than the 19.8% of seniors who actually reported this experience since entering college.

^5. Most of the studies that we review assessed prevalence, “the number of unique persons in the population who experienced one or more victimizations in a given period”; in contrast, the NCVS assesses incidence, “the number of victimizations experienced by persons in the population during a given period” (Sinozich & Langton, 2014, p. 2).