Chapter Five: Comparison of the Key Surveys Regarding Rape
Why Do Rape Statistics Vary So Widely? Introduction
- Table 1. Rate of Campus Rape and Sexual Assault
How Are Rape and Sexual Assault Defined?
- Table 2. Definitions of Rape and Sexual Assault
Brief Summary of Statistics
- NCVS: National Crime Victimization Survey (Dec 2014)
- NISVS: National Intimate Partner and Sexual Violence Survey, Centers for Disease Control (2011)
- CSAS: Campus Sexual Assault Study (December 2007), National Institute of Justice
- UCR: Uniform Crime Reporting Program, FBI (2013)
- DOE Department of Education “Clery Act” Report, (2013)
- SVCW: Sexual Victimization of College Women, National Institute of Justice (2000)
- NVAWS: National Violence Against Women Survey, National Institute of Justice (1998)
One thing that can help [the problem with misleading results] – though it’s by no means a catch-all solution – is to let people who don’t know the answer say so. A surprising amount of surveys…force people to choose between ‘agree’ or ‘disagree’. And that means that people who genuinely don’t know (or don’t care) have to misrepresent their opinions. Then again, ‘Majority of Americans noncommittal about DNA’ doesn’t make such a catchy headline, does it?–anonymous 
(Note: Much of this chapter is dry exposition and analysis of surveys with scant attempt to editorialize. Indeed, so much editorializing has occurred in other venues that a straight-forward presentation of their contents seems most appropriate. The word “study” and “survey” are used interchangeably.
I am deeply indebted to a statistician colleague who co-authored this chapter but wishes to remain anonymous due to the controversial nature of the book.)
The National Crime Victimization Survey (NCVS), which the Department of Justice (DOJ) conducts by telephone, has been called the gold standard of statistics on rape. Released in October 2012, the NCVS found that there had been 243,800 rapes or sexual assaults in the preceding year (2011). This was a decline. In Table 1 of “Criminal Victimization, 2001,” the DOJ reported on former NCVS data.  The survey cited the number of rapes or sexual assaults in 2002 as 349,810 and in 2010 as 258,570. (p.2). The rate declined by 30 percent between 2002-2011, and declined by 9 percent between 2010-2011. This would seem to be good news but the survey came under heavy criticism. Why?
Also focusing on the 2011 rate of rape, the Centers for Disease Control (CDC) issued a report that suggested “the NCVS is failing to capture a large portion of victims—as much as 88 percent.” The CDC concluded “almost one in five women (and 1.7 percent of men) have been raped in their lifetimes.” 
The New Republic offered plausible reasons  for the jarring discrepancy in statistics. For example, the questions differed. “The NCVS’s questions required the victims to identify the crime as a rape or sexual assault, asking questions like ‘How were you attacked?’ and providing answers like ‘raped’, ‘tried to rape’ and ‘sexual assault other than rape and attempted rape’.” By contrast, the CDC avoided legal terms and asked “questions like, ‘How many people have ever used physical force or threats to physically harm you to make you have vaginal sex?’” Victims who did not believe a sexual encounter was rape may have been willing to describe it in terms of “physical force.”
A second reason was a difference in the definitions of sexual assault. Significantly, the CDC considered females who had sex while under the influence to be victims of sexual assault because they were considered incapable of rendering consent. The NCVS made no such assumption.
The foregoing merely foreshadows the discrepancies between two key surveys from which rape stats are drawn. The rape data offered by standard reports have come to resemble chaos. Some confusion comes from honest incompatibilities between studies, such as a difference in the reporting period or in the age range of participants. Other problems are less pardonable. For example, the methodology of many surveys is so poor that their results should be discarded. (Chapter Four dwells upon the research habits and tactics that produce useless data…or worse.)
Yet other factors hinder the search for credible numbers. For example, political advocates and the media routinely misstate the basic findings of straight-forward reports. The previously mentioned New Republic article stated the NCVS findings for 2011 as 243,800 rapes or sexual assaults. An article in Time, which linked to the New Republic piece, declared, “While the CDC estimates that nearly 2 million adult American women were raped in 2011 and nearly 6.7 million suffered some other form of sexual violence, the NCVS estimate for that year was 238,000 rapes and sexual assaults.”  The difference is not great but it is disconcerting to see any variation in the reporting of fundamental findings from a single source.
This chapter explores the four most cited surveys in close detail. The chapter does not endorse any one report as the truth about the rate of rape or sexual assault. The four sources are:
- NCVS: The National Crime Victimization Survey, published by the Bureau of Justice Statistics in December 2014. 
- NISVS: The National Intimate Partner and Sexual Violence Survey, published by the Centers for Disease Control in 2011. 
- CSAS: The Campus Sexual Assault Study, published by the National Institute of Justice in December 2007. 
- UCR: The FBI Uniform Crime Reporting Program, 2013 data.  The UCR is not a survey, but rather an administrative compilation of crime reports.
The statistics are all over the map but the four preceding sources are the most significant ones and the most frequently consulted. Nevertheless, this chapter compares three additional surveys in a passing manner:
- SVCW: “Sexual Victimization of College Women,” a research report published by the National Institute of Justice in December 2000. 
- NVAWS: “Prevalence, Incidence, and Consequences of Violence Against Women: Findings of the National Violence Against Women Survey,” published by the National Institute of Justice in November 1998. 
- DOE: Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act and the Higher Education Opportunity Act, 2013 data. 
Unfortunately, much of the data cannot be directly compared. For example, some studies report incidence (how many assaults occurred); others report prevalence (how many women were assaulted).  Some render a one-year account while others gather lifetime totals or examine varied periods. There are surveys that include women of all ages, while others address only women of college age or they focus exclusively upon students. Some require an assault to be reported to an authority before counting it while others use self-reporting surveys. Many researchers randomize their sample population but some use self-selected samples. To add more confusion, completed and attempted assaults are often grouped into one statistic. And, as mentioned earlier, definitions differ widely.
But, with wide room for approximation, data comparison can be attempted. Table 1 is a summary chart of the rates of rape and sexual assault, both attempted and completed, of women on college campuses. These numbers are rates per thousand women per year. Numbers are prevalence (number of women affected), except for the DOE (Clery Act) ones which are incidence (number of reports). Boxes shaded in gray indicate data that is not available.
The table demonstrates how various studies classify and lump together different types of sexual violence. For example, the NCVS separates forcible rapes from sexual assault other than rape, but it does not include incapacitated rape (sex while unable to consent due to alcohol or drugs). The DOE provides a combined total for forcible rapes, sexual assault and incapacitated rapes.
Table 1. Rate of Campus Rape and Sexual Assault
per thousand women, per year
Table 1 Notes:
- Sexual coercion is generally defined as non-physical punishment, promise of reward or verbal pressure.
- Based on 6,069 incidents in the year 2013, over a female student population of 11,612,000.
- NCVS counts “coerced” sex as rape but does not define coercion. This chart includes “threat of rape or sexual assault” with sexual assault.
- Estimate is based on 0.8 per thousand reported for all ages, and assumes (per NCVS) that only 20% of rapes/attempted rapes of college students are reported to police. It also assumes the rate for the “college” age group is roughly 3x that for all ages. (See Note 11 below)
- Estimate is based on reported 16 per thousand of all rapes/attempted rapes, of which 10 per thousand are alcohol- or drug-facilitated. It assumes the rate for the “college” age group is roughly 3x that for all ages. (See Note 11 below).
- Estimate is based on reported 55 per thousand of all assaults, of which 20 per thousand are sexual coercion. It assumes the rate for the “college” age group is roughly 3x that for all ages. (See Note 11 below)
- CSAS four-year figures, divided by four. Categories overlap so the totals cannot be calculated by summing the columns. One-year rate for all sexual assault is 49 per thousand.
- Sum of prevalence figures for “attempted sexual contact” with force and without force; this almost certainly overstates the total prevalence due to victims responding in both categories. It does not include “attempted sexual coercion” at 13.5 per thousand.
- Sum of prevalence figures for “completed sexual contact” with force and without force; this almost certainly overstates the total prevalence due to multiple responses.
- Estimate is based on reported 3 per thousand rapes/attempted rapes the previous year, all ages. It assumes the rate for the “college” age group is roughly 3x that for all ages. (See Note 11 below).
- The NISVS reports 38.3% of rapes are first experienced at ages 18-24; for ease of calculation, this can be rounded upward to 40%. As a reasonable approximation, multiple rapes are ignored and it is assumed that 40% of rapes happen to women at ages 18-24. Per the 2010 Census, the college-age population is 10% of the total. If 40% of rapes happen to 10% of women, then the per-capita rate of rape for those women is 4 times the per-capita rate for all ages.
The NVAWS reports that 29.4% of rapes are first experienced at ages 18-24. If we round this up to 30 percent, the per-capita rate for college-age women can be calculated at 3 times the per-capita rate for all women.
The NCVS reports, “In 2013, college-age females had a similar rate of rape and sexual assault regardless of enrollment status (about 4.3 victimizations per 1,000), while the victimization rate for not college-age (ages 12 to 17 and 25 or older) females was 1.4 victimizations per 1,000.” The per-capita rate for college-age women can be calculated at 2.5 times the per-capita rate for all women. 
The discrepancy could be due to multiple rapes (e.g. women raped at 18-24 also being raped after age 24), or due to methodological differences in the surveys. As a reasonable “middle ground,” the chapter assumes the reported data indicates that the college-age rate is three times the all-age rate.
What lessons can be drawn from the preceding data?
One lesson: The truth is in there. But where? In 2011, the National Research Council asked much the same question when it convened a panel of experts to review the methodology of the NCVS survey. In their report, the panel compared the NCVS to older surveys , (pp.3-4),
The panel found that a comparison across these sources of estimates of rape was particularly problematic because of the differences in the populations targeted, the definitions used, the data collection methodology, and the survey timing. The panel determined that it could not scientifically conclude which source was overall better, and it does not recommend any source as the best or as a standard. However, in reviewing all of this material, the panel judges that it is likely that the NCVS is under-counting rape and sexual assault victimization…
It is important to note that the panel did not perform the same in-depth examination of the error structure of the other surveys for measuring rape and sexual assault because of limitations of time and resources. Presenting findings focused on the NCVS does not imply that the panel believes that the other surveys have fewer errors: the panel did not examine them carefully and so cannot draw overall conclusions about their error structures.
There is no statistically meaningful way to average such disparate reports. Only inferences can be drawn. Assuming from the CSAS that there are 2.5 incapacitated rapes for every 1 forcible rape on campus, then the NCVS would seem to overlook 5 incapacitated rapes per thousand women per year. The discrepancy is massive.
Perhaps the most informative approach to the differing surveys is to view them as a way to set an upper and lower bound on the real numbers. If so, it is safe to conclude that the rate of forcible rape on campus is more than 2.0 per thousand women per year (NCVS), and less than 16.6 per thousand per year (SVCW). But the surveys offer more information than merely setting a range.
To decide how much weight to give the data from each survey, however, it is necessary to compare their methodology and definitions. Start with the most basic question: How are rape and sexual assault defined?
How are Rape and Sexual Assault Defined?
Compilations of data use different definitions for rape and sexual assault. As mentioned earlier, the NCVS does not consider sex while incapacitated by alcohol or drugs to be rape. The FBI’s UCReport has only counted such sexual acts as rape since 2013 when its definition was updated.
Table 2 below presents a summary of acts that may or may not be considered rape or sexual assault. It indicates how they are classified by the NCVS, NISVS, CSAS, UCR, SVCW and NVAWS — or, at least, they are categorized based upon a reasonable reading of the varying definitions. The authors, researchers. survey subjects and law enforcement officers may classify borderline cases differently. For example, no survey defines “regret the next morning” as rape. But if it were reported as non-consensual, then it could be counted as rape. These studies are unable to provide such distinctions.
A more subtle difference was noted by the National Research Council :
There are two quite different perspectives for the measurement of rape and sexual assault—the criminal justice perspective and the public health perspective. These different perspectives have led to methodological differences in designing and implementing surveys, which, in turn, have resulted in different estimates of the incidence rates. The NCVS reflects the criminal justice perspective, and its purpose is to measure criminal victimizations: “point-in-time” events that are judged to be criminal. In contrast, surveys that reflect the public health perspective look at victimization as a condition that endures over a period of time, and may not necessarily be criminal.
To distinguish between rape and sexual assault, Table 2 follows the updated FBI definition, which may be the simplest and least ambiguous of the various definitions of rape: “penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim.” All other unwanted sexual contact is classified here as “sexual assault.”
Table 2. Definitions of Rape and Sexual Assault
Table 2 Notes
A question mark “?” indicates that it is not clear whether the act is deemed to be rape or sexual assault.
- Included in FBI statistics since their 2013 redefinition of rape.
- Attempted rape includes verbal threats of rape.
- Attempted rape is counted only for force or threats of harm (rows 1-3 above).
- Yes, if by physical force or when incapacitated or drunk.
- Counted as sexual coercion by the SVCW
The NCVS has the narrowest definition of rape, followed by the UCR, and then the CSAS. The NISVS has the most expansive definition of rape by far; for example, it is the only survey that counts “rape by deception,” which means you have sex with someone because he or she lies to you. It is no surprise that, in Table 1, the NCVS reports the lowest rate of rape, followed by the UCR and CSAS; the NISVS reports the highest.
The NISVS also has the broadest definition of sexual assault, which includes sexually suggestive remarks and being forced to view sexually explicit media. (Note: The UCR does not count sexual assaults, only rapes.) Again, these definitions are subject to interpretation by the victim. For example, one victim may regard a dirty joke as a “sexually suggestive remark” but another may not.
The following sections offer an in-depth look at these four touchstone sources of rape statistics.
National Crime Victimization Survey, Bureau of Justice Statistics (NCVS, December 2014)
In December 2014, the BJS issued a Special Report based on NCVS data, which was entitled“Rape and Sexual Assault Victimization Among College-Age Females, 1995–2013 ” . (Unless otherwise noted, the page numbers below refer to this report.) The BJS findings caused a sensation because its reported rate of rape and sexual assault among college students was 6.1 per thousand per year, a factor of ten lower than some headlines.
According to the report:
The National Crime Victimization Survey (NCVS) is an annual data collection conducted by the U.S. Census Bureau for the Bureau of Justice Statistics (BJS). The NCVS is a self-report survey in which interviewed persons are asked about the number and characteristics of victimizations experienced during the prior 6 months. The NCVS collects information on nonfatal personal crimes (rape or sexual assault, robbery, aggravated and simple assault, and personal larceny) and household property crimes (burglary, motor vehicle theft, and other theft) both reported and not reported to police… (p.11)
All first interviews are conducted in person with subsequent interviews conducted either in person or by phone. New households rotate into the sample on an ongoing basis to replace outgoing households that have been in the sample for the 3-year period… (p.11)
In 2013, 90,630 households and 160,040 persons age 12 or older were interviewed for the NCVS. Each household was interviewed twice during the year. The response rate was 84% for households and 88% for eligible persons. (p.12)
Because the NCVS is a household survey and not based on police reports, it offers a glimpse into “dark” numbers such as how many sexual assaults are not reported to police. Because it is conducted annually with a consistent methodology, it also captures trends. Another advantage of the NCVS: As a randomized survey, it is less prone to selection bias. But because it surveys all crimes, and not just sexual assaults, the amount of detail on sexual crimes is limited.
The NCVS collects information for two sexual crimes – “rape” and “sexual assault.” These are defined as follows:
Rape is the unlawful penetration of a person against her or his will by the use or threat of force, as well as the attempt to commit such an act. Rape includes psychological coercion and physical force. Forced sexual intercourse means vaginal, anal, or oral penetration by the offender. Rape also includes incidents where penetration involves a foreign object such as a bottle. Attempted rape includes verbal threats of rape.
Sexual assault is defined to include a wide range of acts that are separate from rape or attempted rape. They include attacks or attempted attacks, which usually involving unwanted sexual contact between a victim and an aggressor. They may or may not involve force but they include grabbing or fondling. (p.11)
Actual NCVS questions include:
–(Other than any incidents already mentioned), has anyone attacked or threatened you in any of these ways:… (e) any rape, attempted rape, or other type of sexual attack; …
–Incidents involving forced or unwanted sexual acts are often difficult to talk about. (Other than any incidents already mentioned), have you been forced or coerced to engage in unwanted sexual activity by (a) someone you didn’t know before, (b) a casual acquaintance? OR (c) someone you know well? (p.15) [Note: Unfortunately, the question does not specify what constitutes “coerced,” so this is left up to the interpretation of the survey respondent.]
Also, “the survey does not specifically ask about incidents in which the victim was unable to provide consent because of drug or alcohol consumption.” (p.14) As Libby Nelson observed  at Vox:
One thing the survey doesn’t ask about specifically is incapacitated rape — sex when one person was too drunk or drugged to legally consent. Studies of campus sexual assault have consistently found that this kind of rape is more common than rape under a threat of force. That’s one big difference between the National Crime Victimization Survey and other studies about campus sexual assault that have found campus rape to be much more prevalent.
The most frequently cited NCVS result is “6.1 per 1,000” per year for the prevalence of rape and sexual assault (combined) of female college students. This is subdivided (p.4) into: 2.0 per 1,000 completed rape; 1.5 per 1,000 attempted rape; 1.9 per 1,000 sexual assault; and 0.7 per 1,000 threat of rape or sexual assault. (In Table 1 above, threats are counted in with sexual assaults.)
Several other NCVS results are interesting:
The rate is decreasing. The 6.1 per 1000 figure is the average for the years 1995-2013. Figure 2 on page 3 of the report shows the trend is decreasing. In 2013, for example, the rate for students was 4.4 per 1,000. (p.17)
Non-students are more likely to be victims. Contrary to the popular belief that college is especially dangerous for women, the NCVS statistics reveal that college-age (18 to 24) women who are non-students have a higher rate of sexual victimization. For the 1995-2013 period, the rate is 7.6 per thousand for non-students, versus 6.1 per thousand for students. (p.1)
Students are less likely to report to the police. “Among student victims, 20% of rape and sexual assault victimizations were reported to police, compared to 32% reported among nonstudent victims ages 18 to 24.” (p.1) The most common reasons given for not reporting, by both students and non-students, are “fear of reprisal,” “personal matter,” and “other reason.” (p.9)
College-age is the most vulnerable age group. In 2013, college-age females (students and non-students) were about three times more likely to be victimized than other age brackets: 4.3 per 1,000 for ages 18 to 24, v. 1.4 per thousand combined for ages 12 to 17 and 25 or older. (p.3) (This report does not break out the statistics for 17 and under v. 25 and older.)
Men are victimized but less often. The rate of rape and sexual assault for male students is roughly 1/5 the rate for female students. For the period 1995-2013, 1.4 per 1,000 male students were victimized, compared to 6.1 per 1,000 female students. Male non-students of college age report a rate of 0.3 per 1,000. (p.5)
Gang rape is rare. 95 percent of student rapes and sexual assaults involved only a single offender (v. “two or more” or “unknown number”). (p.8)
As noted earlier, the NCVS does not count incidents of sex while incapacitated due to alcohol or drugs or unconsciousness.
Questions were not as specific and guided as NISVS and CSA. For example, it is not clear if forced kissing, fondling, etc. qualifies as “unwanted sexual acts.”
The NCVS takes the “criminal justice perspective” rather than the “public health perspective” in the design of its survey:
In comparison, the NISVS focused on sexual violence, stalking, and intimate partner violence and was presented as a survey collecting data on a range of behaviors that impact public health. This public health perspective may encourage respondents to recall and report on experiences that they may not typically think of as criminal victimization. It also may result in the collection of incidents that may not be considered criminal behavior. Similarly, the CSA study focused specifically on rape and sexual assault, also from a public health and safety perspective. (p.14)
The NCVS report acknowledges that this may lead to under-reporting:
The NCVS is an omnibus survey designed to collect information on experiences with a broad range of crimes. It is likewise presented to respondents as a survey about criminal victimization. Because victims of rape or sexual assault may not consider their victimization a crime, this context could discourage or suppress recall and reporting of those incidents. Additionally, because the NCVS covers a wide range of criminal victimization, the number of screening questions related to rape and sexual assault are limited. (p.14)
Also, because of the limited number of screening questions, respondents receive less guidance. The NISVS and CSAS specifically ask about forced kissing and fondling; the NCVS has no such explicit question. It is up to the respondent to decide if a forced kiss is an “unwanted sexual act.”
Comparison to Other Studies
Because this Special Report was issued after the NISVS and the CSAS (called CSA in the Special Report), it was able to include comparisons to those prior studies:
The NCVS, NISVS, and CSA target different types of events. The NCVS definition is shaped from a criminal justice perspective and includes threatened, attempted, and completed rape and sexual assault against males and females (see Methodology). The NISVS uses a broader definition of sexual violence….The CSA definition of rape and sexual assault includes unwanted sexual contact due to force and due to incapacitation, but excludes unwanted sexual contact due to verbal or emotional coercion. (p.2)
Returning to the issue of specific questions and focus:
Unlike the NCVS, which uses terms like rape and unwanted sexual activity to identify victims of rape and sexual assault, the NISVS and CSA use behaviorally specific questions to ascertain whether the respondent experienced rape or sexual assault. These surveys ask about an exhaustive list of explicit types of unwanted sexual contact a victim may have experienced, such as being made to perform or receive anal or oral sex. (p.2)
A strength of the NCVS is its high response rate compared to other studies:
Surveys with low response rates have an increased potential for nonresponse bias compared to surveys with higher response rates. Nonresponse bias means that those who participated in the survey may differ in important ways from those who did not participate, which could in turn impact the survey findings. In 2013, the NCVS had an 88% response rate for eligible persons and a combined persons and household response rate of 74%, while the 2011 NISVS had an overall response rate of 33.1%, and the CSA response rate was between 33% and 43% for males and females at the two schools. (p.16)
A second significant and frequently quoted source of data is the NISVS.
- NISVS: National Intimate Partner and Sexual Violence Survey, Centers for Disease Control (2011)
The NISVS was conducted in the year 2011 by the CDC. In September 2014, the CDC published a Surveillance Summary, “Prevalence and characteristics of sexual violence, stalking, and intimate partner violence victimization — National Intimate Partner and Sexual Violence Survey, United States, 2011” . (Unless otherwise noted, page numbers below refer to this report.)
The NISVS survey is a random-sample telephone interview:
NISVS is a national random-digit–dial telephone survey of the noninstitutionalized English- and Spanish-speaking U.S. population aged ≥18 years. NISVS gathers data on experiences of sexual violence, stalking, and intimate partner violence among adult women and men in the United States by using a dual-frame sampling strategy that includes both landline and cellular telephones. The survey was conducted in 50 states and the District of Columbia; in 2011, the second year of NISVS data collection, 12,727 interviews were completed, and 1,428 interviews were partially completed. (p.1)
A total of 6,879 women and 5,848 men completed the survey. The estimates presented in this report are based on completed interviews. An interview is defined as having been completed if the respondent completed the demographic and general health questions as well as all of the violence victimization questions. (p.3)
The survey collected information about many types of “sexual violence”:
The specific types of sexual violence assessed included rape (completed or attempted forced penetration or alcohol- or drug-facilitated penetration) and sexual violence other than rape, including being made to penetrate a perpetrator, sexual coercion (nonphysically pressured unwanted penetration), unwanted sexual contact (e.g., kissing or fondling), and noncontact unwanted sexual experiences (e.g., being flashed or forced to view sexually explicit media). (p.3)
The survey’s definition of the activities that might constitute “forced sex” — vaginal sex, anal sex, oral-genital contact, oral-anal contact, inserting fingers or an object into vagina or anus — is fairly close to the 2013-revised definition now used by the FBI for the Uniform Crime Reporting program.
The questions about these activities are :
When you were drunk, high, drugged, or passed out and unable to consent, how many people have ever …
How many people have ever used physical force or threats to physically harm you to make you…
For the question about physical force or threats there is a further option, “sex did not happen”, for vaginal/oral/anal sex. This appears to be counted as “attempted forced penetration” in the report’s Table 1. (p.5)
There is also a “nonphysical pressure” question :
How many people have you had vaginal, oral, or anal sex with after they pressured you by…
- doing things like telling you lies, making promises about the future they knew were untrue, threatening to end your relationship, or threatening to spread rumors about you?
- wearing you down by repeatedly asking for sex, or showing they were unhappy?
- using their influence or authority over you, for example, your boss or your teacher?
This appears to be counted as “sexual coercion” in Table 1.
The most frequently cited results from the NISVS are “16 per 1,000” (per year), for the prevalence of rape, and “55 per 1,000” for sexual assault, for all women. From the report’s abstract (p.1):
In the United States, an estimated 19.3% of women and 1.7% of men have been raped during their lifetimes; an estimated 1.6% of women reported that they were raped in the 12 months preceding the survey. The case count for men reporting rape in the preceding 12 months was too small to produce a statistically reliable prevalence estimate. An estimated 43.9% of women and 23.4% of men experienced other forms of sexual violence during their lifetimes, including being made to penetrate, sexual coercion, unwanted sexual contact, and noncontact unwanted sexual experiences. The percentages of women and men who experienced these other forms of sexual violence victimization in the 12 months preceding the survey were an estimated 5.5% and 5.1%, respectively…
With respect to sexual violence and stalking, female victims reported predominantly male perpetrators, whereas for male victims, the sex of the perpetrator varied by the specific form of violence examined. Male rape victims predominantly had male perpetrators, but other forms of sexual violence experienced by men were either perpetrated predominantly by women (i.e., being made to penetrate and sexual coercion) or split more evenly among male and female perpetrators (i.e., unwanted sexual contact and noncontact unwanted sexual experiences)…
The lifetime and 12-month prevalences of rape by an intimate partner for women were an estimated 8.8% and 0.8%, respectively; …
The NISVS also reported statistics for stalking but, as none of the other key surveys include stalking, those figures are omitted here.
Also from the abstract (p.1):
Results suggest that these forms of violence frequently are experienced at an early age because a majority of victims experienced their first victimization before age 25 years, with a substantial proportion experiencing victimization in childhood or adolescence.
This suggests that rapes/assaults may be skewed towards younger (college-age) women, which is a finding consistent with the NCVS.
The survey asked respondents to report on acts of sexual violence during a “respondent’s lifetime and during the 12 months before interview” (p.3). The latter were used to estimate 12-month prevalences. In some categories, there were too few cases to establish a statistically significant result. Table 1 reports the following 12-month prevalences (p.5):
- Completed forced penetration – no result
- Attempted forced penetration – no result
- Completed alcohol or drug-facilitated penetration – 1.0%
Other sexual violence: 5.5%
- Made to penetrate – no result
- Sexual coercion – 2.0%
- Unwanted sexual contact – 2.2%
- Noncontact unwanted sexual experiences – 3.4%
Note: There is overlap in the four subcategories of “other sexual violence,” which means these numbers cannot be simply added together to compute total risk. Judging from the lifetime prevalence figures, there is likewise an overlap in the three subcategories of rape.
This study has, by far, the most expansive definition of sexual violence. Any of the following might be considered sexual assault: Being kissed against your will, being “touched…in a way that made you feel unsafe,” being verbally harassed in a public place in a way that made you feel unsafe, or being made to look at sexual photos. Sexual coercion might include a boyfriend saying that if you didn’t want to have sex with him, he would end the relationship; “wearing you down by repeatedly asking for sex”; “showing [he is] unhappy”; and “telling lies.” All are reported as sexual violence. This may be because reporting a 6 percent chance (per Table 1 above) that a male will lie to a female student in order to have sex is less alarming than a 6 percent chance she will be sexually coerced.
Perhaps due to the low case counts for the 12-month period, the report focuses more heavily on lifetime prevalence rather than the risk per year.
There is no breakdown by age groups. The closest it is possible to come to estimating college-age risk is to look at “age at the time of first victimization” for completed rape. According to the report (pp.11-12):
Among female victims of completed rape (completed forced penetration and completed alcohol- or drug-facilitated penetration), this form of sexual violence was first experienced by an estimated 78.7% before age 25 years, by an estimated 40.4% before age 18 years (28.3% at ages 11–17 years and 12.1% at age ≤10 years), and by an estimated 38.3% at age 18–24 years (Figure 3). In addition, among female victims of completed rape, an estimated 15.2% first experienced this at age 25–34 years, an estimated 4.6% at age 35–44 years, and an estimated 1.5% at age ≥45 years.
As an approximation, then, assume that 38.3% of the rapes occurring in any given year (estimated in Table 1 as 1,929,000 women) are occurring in the group aged 18-24. Table 1 in the report (p.5) suggests a total female population of 121,000,000. Per the 2010 Census , about 10% of the population is age 18-24. So, 738,800 rapes in a population of 12,100,000 is a rate of 6.1%, or 61 per thousand (per year), for college-age victims.
While mentioning the limitations of the report (p.17), the authors speculate (without offering evidence) that the NISVS might be under-counting the prevalence of sexual violence:
The findings of this report are subject to at least five limitations. First, the overall response rate for the 2011 NISVS survey was relatively low (33.1%). However, the cooperation rate was high (83.5%), and multiple efforts were made to reduce the likelihood of nonresponse and noncoverage bias. These included a nonresponse follow-up in which randomly selected nonresponders were contacted again and offered an increased incentive for participation as well as the inclusion of a cellular telephone sample. Second, although NISVS captures a broad range of self-reported victimization experiences, it is likely that the estimates presented underestimate the prevalence of sexual violence, stalking, and intimate partner violence. Victims who are involved in violent relationships or who have recently experienced severe forms of violence might be less likely to participate in surveys or might not be willing to disclose their experiences because of unresolved emotional trauma or concern for their safety, among other reasons. Third, a telephone survey might be less likely to capture some populations that could be at higher risk for victimization (e.g., persons living in nursing homes, military bases, prisons, or shelters, or those who are homeless). Fourth, self-reported data are vulnerable to recall bias because respondents might believe that events occurred closer in time than they did in actuality (i.e., telescoping) , and this type of bias might particularly affect 12-month prevalence estimates. Finally, follow-up questions were designed to reflect the victim’s experience with each perpetrator across the victim’s lifetime and there were limitations associated with how these questions were asked. Respondents were asked about the impact from any of the violence inflicted by each perpetrator. Therefore, the impact of specific intimate partner violence behaviors cannot be assessed. Also, because victims’ reports of the age and relationship at the time any violence began with each perpetrator were used, it was not always possible to assess the age or relationship at the time specific types of intimate partner violence occurred.
It is equally plausible to suggest that victims who have experienced sexual violence could be more motivated to participate in the survey, rather than less. This remains unknown.
A third significant and frequently quoted source is the Campus Assault Study.
- CSAS: Campus Sexual Assault Study (December 2007), National Institute of Justice
CSAS – also widely known as CSA – surveyed undergraduate women and men on their experiences of sexual assault both before and after entering university. Conducted during the winter of 2005-2006, 5,466 undergraduate women and 1,375 undergraduate men participated.
The web-based survey was conducted by RTI International (RTI) at two “large, public universities” under a grant from the National Institute of Justice (NIJ). The NIJ is the research, development and evaluation agency of the Department of Justice. “The Campus Sexual Assault (CSA) Study Final Report”  was published in December 2007. (Unless otherwise noted, page numbers below refer to this report.)
CSAS drew from random samples of full-time students aged 18-25. Volunteers were offered a $10 Amazon gift certificate and an iTunes song download as compensation. Nevertheless, the response rate was “relatively low” (p.3-7), 42.2 and 42.8 percent at the two universities. (p.x). The study stated, “about 84% of students who completed the CSA survey followed through to obtain their incentives .” (p.3-6) It is not known if the 16% who did not follow through were the result of multiple completions of the study; nor is it known whether only questionnaires with completion codes were counted.
CSAS explained its methodology (p.x):
We drew random samples of students aged 18-25 and enrolled at least three-quarters’ time at each university to participate in the CSA Study. Sampled students were sent an initial recruitment e-mail that described the study, provided a unique CSA Study ID#, and included a hyperlink to the CSA Study Web site…
The survey was administered anonymously. (students did not enter their CSA Study ID # to take the survey)
A follow-up survey of 2,000 non-respondents elicited 296 replies (p.3-9):
Fourteen sample members stated that they never received the original recruitment e-mails, and 126 students (45%) were not sure whether they had received these e-mails…..Among the sample members who either did receive the e-mails and chose to not participate or said that they would not have participated if they had received the e-mails, the most commonly reported reasons for nonparticipation were that they did not have time (reported by about two-thirds of the sample members) or that they never participate in Web-based surveys (reported by just over 20% of the sample members). An extremely small number of the respondents to the nonrespondent survey indicated that they did not participate because they had never experienced sexual assault (n=9, 15.0%), or because they did not want to discuss their experiences with sexual assault victimization (n=4, 6.7%).
The foregoing raises a question. Did the non-responsive students simply have nothing to report or no interest? Or had they experienced sexual violence about which they did not wish to speak? Either could bias the results.
According to the report (p.xi, emphasis in original):
Sexual assault included forced touching of a sexual nature, oral sex, sexual intercourse, anal sex, and/or sexual penetration with a finger or object [sic]. For both physically forced and incapacitated sexual assault, information was collected on completed and attempted assaults experienced before entering college and since entering college. For completed sexual assaults, a series of follow-up questions enabled us to define the assault as sexual battery (i.e., sexual assault that entailed sexual touching only) and/or rape (i.e., sexual assault that entailed oral, vaginal, or anal penetration).
Appendix A of the report (p.A-1) provided the survey questions:
This section of the interview asks about nonconsensual or unwanted sexual contact you may have experienced. When you are asked about whether something happened since you began college, please think about what has happened since you entered any college or university. The person with whom you had the unwanted sexual contact could have been a stranger or someone you know, such as a family member or someone you were dating or going out with.
The questions ask about five types of unwanted sexual contact:
- 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)
- oral sex (someone’s mouth or tongue making contact with your genitals or your mouth or tongue making contact with someone else’s genitals)
- sexual intercourse (someone’s penis being put in your vagina)
- anal sex (someone’s penis being put in your anus)
- sexual penetration with a finger or object (someone putting their finger or an object like a bottle or a candle in your vagina or anus).
In a subsequent article in Time magazine, two of CSAS’s authors clarified what constituted “rape” and “sexual assault” in evaluating responses :
Among other items, the students, after being told they were going to be asked about their experiences with unwanted sexual contact, were asked these two key questions:
Since you began college, has anyone had sexual contact with you by using physical force or threatening to physically harm you?
Since you began college, has someone had sexual contact with you when you were unable to provide consent or stop what was happening because you were passed out, drugged, drunk, incapacitated, or asleep? This question asks about incidents that you are certain happened.
To be counted as a victim of sexual assault or rape and included in the 1-in-5 statistic (19.8%), a woman would have to be a senior and answer “Yes” to one or both of those questions.
In our reports, sexual-assault victims who selected only “Forced touching of a sexual nature” in a follow-up question asking about the type of contact that happened were classified as victims of sexual battery only, whereas victims who selected any of the other response options (oral sex, sexual intercourse, anal sex, or sexual penetration with a finger or object) were classified as victims of rape.
“Forced” and “incapacitated” were spelled out in the survey questions (pp.A-1,A-2, emphasis in original):
Force could include someone holding you down with his or her body weight, pinning your arms, hitting or kicking you, or using or threatening to use a weapon against you.
Has anyone had sexual contact with you by using physical force or threatening to physically harm you?
Has anyone attempted but not succeeded in having sexual contact with you by using or threatening to use physical force against you?
The next set of questions ask about your experiences with unwanted sexual contact while you were unable to provide consent or stop what was happening because you were passed out, drugged, drunk, incapacitated, or asleep. These situations might include times that you voluntarily consumed alcohol or drugs and times that you were given drugs without your knowledge or consent.
Has someone had sexual contact with you when you were unable to provide consent or stop what was happening because you were passed out, drugged, drunk, incapacitated, or asleep? This question asks about incidents that you are certain happened.
Have you suspected that someone has had sexual contact with you when you were unable to provide consent or stop what was happening because you were passed out, drugged, drunk, incapacitated, or asleep? This question asks about events that you think (but are not certain) happened.
What does CSAS actually say?
An interesting result is the number of women who experienced sexual assault before college (pp.xii-xiii, emphasis in original):
Nearly 16% of the 5,446 women experienced attempted or completed sexual assault before entering college….Nineteen percent of the women reported experiencing completed or attempted sexual assault since entering college….
This is the source of the frequently-heard claim that 1-in-5 women will be sexually assaulted in college. To be precise, 1-in-5 will experience a sexual assault or an attempt during a four-year stay on campus. In CSAS, 1,073 women out of the 5,446 (19.7%) reported experiencing an attempted or completed sexual assault since entering college. In Figure 1 (p.xiii) and Exhibit 5-1 (p.5-2) these reports are categorized for assaults since entering college:
- Attempted or completed sexual assault since entering college, n=1073, 19%
- Attempted, n=682, 12.6%
- Completed, n=782, 13.7%
- Physically forced, n=256, 4.7%
- Sexual battery only, n=75, 1.4%
- Rape, n=181, 3.4%
- Incapacitated, n=651, 11.1%
- Sexual battery only, n=144, 2.6%
- Rape, n=507, 8.5%
- Alcohol and/or other drug (AOD)-enabled SA, n=466, 7.8%
- Certain drug-facilitated SA, n=31, 0.6%
- Physically forced, n=256, 4.7%
iii. Suspected drug-facilitated SA, n=103, 1.7%
- Other incapacitated SA (e.g. asleep), n=48, 1.0%
It is not clear from the report how subcategories 2.a/b and 2.i/ii/iii/iv overlap; that is, it is not clear how many rapes had drug involvement.
Subcategories B)1 and B)2 do overlap. 782 women reported completed assaults; 256 women reported being physically forced, and 651 reported being incapacitated. This suggests that 125 reported both physically forced assaults and incapacitated assaults (an overlap of 16% of the total number).
Similarly, it is reasonable to deduce that 391 women reported both attempted (A) and completed (B) assaults.
Since only senior women were surveyed, the report is presumed to be over four years. (In Table 1 above, the CSAS figures are divided by four to give a per-year risk.)
A particularly interesting result is that the authors frequently considered the women to have been raped, when the women themselves did not (p.5-20):
When subsetting to victims who were raped, 64.6% of physically forced rape victims and 37.8% of incapacitated rape victims considered the incident to be rape.
Exhibit 5-8 (p.5-22) describes whether students reported the assault. 12.9% of the victims of forced sexual assault reported the assault to law enforcement as compared to only 2.1% of the victims of incapacitated sexual assault. Unfortunately, this table does not distinguish between “sexual battery” versus rape.
Because CSAS is the primary source of the 1-in-5 statistic that has been wielded to such political advantage, the limitations of the study deserve in-depth analysis.
A recurring problem is that CSAS does not generally break out rape figures; it considers all sexual assaults together. Moreover, there is some confusion because the authors use the term “sexual battery” for acts that others commonly refer to as “sexual assault.” By contrast, CSAS broadly defines “sexual assault” and includes rape within the definition. But the term also includes forced kissing, grabbing, and “rubbing up against you in a sexual way.” (p.A-1) This has implications for the reporting statistics. For example, how many women would call the police if a breast had been groped through clothing?
The study also does not generally count attempted rapes. It seems likely, however, that respondents would report an attempted and uncompleted rape as a sexual assault.
Cathy Young pinpointed another limitation in a Minding the Campus article, “The White House Overreaches on Campus Rape.” 
The vast majority of the incidents counted as assault involved what the study termed “incapacitation” by alcohol (or, rarely, drugs). But ‘incapacitation’ is a misleading term, since the question used in the study also measured far lower degrees of intoxication: ‘Has someone had sexual contact with you when you were unable to provide consent or stop what was happening because you were passed out, drugged, drunk, incapacitated, or asleep?’ This wording does not differentiate between someone who is unconscious or barely conscious and someone who is just drunk enough to go along with something he or she wouldn’t do when sober. The questions related to sexual assault by physical force–particularly attempted sexual assault–are also worded so ambiguously that they could refer to a clumsy attempt to initiate sex, even if the ‘attacker’ stops at once when rebuffed.
Three quarters of the female students who were classified as victims of sexual assault by incapacitation did not believe they had been raped; even when only incidents involving penetration were counted, nearly two-thirds did not call it rape. Two-thirds did not report the incident to the authorities because they didn’t think it was serious enough.
Libby Nelson at Vox commented on the discrepancy between the researchers’ view of rape and that of the the victims . One interpretation is that the victims were lacking awareness; another is that the authors over-counted rapes:
Studies of college women find that a large proportion don’t define what happened to them as a crime. In Krebs’ survey of women at two large public universities, 56 percent of the victims of forced sexual assault who didn’t report their assaults to the police, and 67 percent of victims of incapacitated sexual assault who did not do so, said it was at least partly because it was not ‘serious enough to report’. Slightly more than one-third of women in both categories said it was ‘unclear if a crime or harm was intended’.
Nelson referred to an earlier study:
A national survey from the Medical University of South Carolina  found that 15 percent of college women who were victims of forcible rape would describe what happened to them as ‘unpleasant, but not a crime’. An additional 32 percent called it a crime, but not rape. For incapacitated rape, the proportions were even higher: 31 percent called it an unpleasant experience, and 40 percent a crime other than rape.
The authors of CSAS themselves openly noted another limitation, which has been resoundingly ignored ever since. In the aforementioned Time magazine article, researchers Christopher Krebs and Christine Lindquist stated ,
First and foremost, the 1-in-5 statistic is not a nationally representative estimate of the prevalence of sexual assault, and we have never presented it as being representative of anything other than the population of senior undergraduate women at the two universities where data were collected—two large public universities, one in the South and one in the Midwest.
Second, the 1-in-5 statistic includes victims of both rape and other forms of sexual assault, such as forced kissing or unwanted groping of sexual body parts—acts that can legally constitute sexual battery and are crimes. To limit the statistic to include rape only, meaning unwanted sexual penetration, the prevalence for senior undergraduate women drops to 14.3%, or 1 in 7 (again, limited to the two universities we studied).
Third, despite what has been said in some media reports, the 1-in-5 statistic does not include victims who experienced only sexual-assault incidents that were attempted but not completed. The survey does attempt to measure attempted sexual assaults, but only victims of completed incidents are included in the 1-in-5 statistic.
Fourth, another limitation of our study—inherent to web-based surveys—is that the response rate was relatively low (42%). We conducted an analysis of this nonresponse rate and found that respondents were not significantly different from nonrespondents in terms of age, race/ethnicity or year of study. Even so, it is possible that nonresponse bias had an impact on our prevalence estimates, positive or negative. We simply have no way of knowing whether sexual-assault victims were more or less likely to participate in our study.
The authors suggested that the low response rate (42%) could bias the study toward either higher or lower rates. There is reason to believe the bias would tend toward higher. If a $10 gift card did not motivate 58% of the solicited population to participate, then it seems plausible that those with additional motivation – for example, those who were victims of sexual assault or politically invested in victims’ rights – would participate in disproportionate numbers. The authors stated, “the anonymity and privacy we afforded respondents may have made women comfortable with responding honestly” — that is, more likely to report. Respondents who experienced violence or were acutely sensitive to the rape culture may well have been disproportionately represented.
Yet another criticism leveled against CSAS is that only two universities were surveyed, which means the results cannot be generalized. In 2015, however, the Washington Post and the Kaiser Family Foundation conducted a nationwide poll to duplicate the CSAS survey, and found nearly identical results :
Twenty percent of young women who attended college during the past four years say they were sexually assaulted, according to a Washington Post-Kaiser Family Foundation poll. But the circle of victims on the nation’s campuses is probably even larger.
Many others endured attempted attacks, the poll found, or suspect that someone violated them while they were unable to consent. Some say they were coerced into sex through verbal threats or promises.
In all, the poll found, 25 percent of young women and 7 percent of young men say they suffered unwanted sexual incidents in college….
Most notably, two-thirds of victims say they had been drinking alcohol just before the incidents.
Many victims were not clear on whether to categorize the incident as “sexual assault”:
Forty-six percent said it’s unclear whether sexual activity when both people have not given clear agreement is sexual assault. Forty-seven percent called that scenario sexual assault.
While the methodology was more random than CSAS, the definitions remained essentially the same:
Conducted by telephone from January through March , the poll surveyed a random national sample of 1,053 women and men ages 17 to 26 who were undergraduates at a four-year college — living on campus or nearby — or had been at some point since 2011. They attended more than 500 colleges and universities, public and private, large and small, elite and obscure, located in every state and the District of Columbia.
The Post-Kaiser poll used questions and definitions similar to those in the 2007 [CSAS] study.
The Washington Post did not report any statistics for rape, just “sexual assault” (including rape).
In conclusion: The biggest problem with CSAS is an overbroad definition of “sexual assault,” which produces the inflated statistic that 1-in-5 women will be sexually assaulted during four years of college. A major problem is that there is no adequate distinction between rape and sexual assault. If a distinction is attempted, then the most aggressive interpretation of the data is that the odds of being raped during college are 1-in-7.9. A more likely result would be 1-in-9.2. And, depending on the significance of the non-responses, the results could be register as as low as 1-in-21. In addition, almost 1/3 of “forcible rapes” and almost 2/3 of “incapacitated rapes” were not considered “rape” by their victims.
A fourth significant and frequently-quoted source of rape statistics is the Uniform Crime Reporting Program.
- 4. UCR: Uniform Crime Reporting Program, FBI (2013)
The Federal Bureau of Investigation’s Uniform Crime Reporting Program 
is a nationwide, cooperative statistical effort of more than 18,000 city, university and college, county, state, tribal, and federal law enforcement agencies voluntarily reporting data on crimes brought to their attention. Since 1930, the FBI has administered the UCR Program and has continued to assess and monitor the nature and type of crime in the nation.
A UCR is issued every year, with a break-down on different kinds of crime. Information presented here is from the online 2013 UCR “Rape” report. 
The FBI receives reports of violent crime and property crime from most of the law enforcement agencies in the U.S. :
In 2013, law enforcement agencies active in the UCR Program represented more than 309 million United States inhabitants (98.0 percent of the total population). The coverage amounted to 98.8 percent of the population in Metropolitan Statistical Areas, 92.9 percent of the population in cities outside metropolitan areas, and 93.5 percent of the population in nonmetropolitan counties.
This represents a larger sample size than any randomized or focused survey. However, by the nature of its collection, the UCR can only count those assaults which are reported to law enforcement.
The FBI has recently revised its definition of rape :
In 2013, the FBI UCR Program initiated collection of rape data under a revised definition within the Summary Reporting System. Previously, offense data for forcible rape was collected under the legacy UCR definition: the carnal knowledge of a female forcibly and against her will. Beginning with the 2013 data year, the term “forcible” was removed from the offense title, and the definition was changed. The revised UCR definition of rape is: Penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim. Attempts or assaults to commit rape are also included; however, statutory rape and incest are excluded.
The UCR Program counts one offense for each victim of a rape, attempted rape, or assault with intent to rape, regardless of the victim’s age. Sexual relations without the victim’s consent which involves a familial offender is counted as a rape and not an act of incest. All other crimes of a sexual nature are considered to be Part II offenses; as such, the UCR Program collects only arrest data for those crimes. The offense of statutory rape, in which no force is used but the female victim is under the age of consent, is included in the arrest total for the sex offenses category.
Part II offenses are lesser offenses, including :
Other assaults (simple) ― Assaults and attempted assaults where no weapon was used or no serious or aggravated injury resulted to the victim. Stalking, intimidation, coercion, and hazing are included.
Sex offenses (except forcible rape, prostitution, and commercialized vice)— Offenses against chastity, common decency, morals, and the like. Incest, indecent exposure, and statutory rape are included. Attempts are included.
Since the Part II definitions include many crimes that are not considered sexual assault, the UCR statistics are of limited value. Only the UCR statistics for attempted and completed rape are considered here.
The Rape Overview states :
There were an estimated 79,770 rapes (older definition) reported to law enforcement in 2013. This estimate was 6.3 percent lower than the 2012 estimate, and 10.6 percent and 16.1 percent lower than the 2009 and 2004 estimates, respectively. (See Tables 1 and 1A.)
The rate of rapes (older definition) in 2013 was estimated at 25.2 per 100,000 females.
Some confusion exists, however. Table 1  gives the rate of rape (older definition) as 25.2 per 100,000 inhabitants or 79,770 rapes over a population of 316,128,839. This renders the rate of rape at approximately 25.2 per 50,000 females, assuming that the majority of rapes reported to the FBI involve female victims. (The UCR does not seem to collect data on prison rape.)
No summary report is available for the new rape definition. Per Table 16 , for “All Agencies,” a total of 53,621 rapes were reported using the new definition, over a population of 134,788,319, for a rate of 39.8 per 100,000 inhabitants. A total of 36,209 rapes were reported from agencies using the older definition, over a population of 156,887,921, for a rate of 23.1 per 100,000 persons. (It is not clear how to reconcile this number with that reported in Table 1.) The higher rate may result from the changes to the definition of rape. For example, the crime now includes penetration with an object and counts male victims.
Assuming, again, that rapes reported to law enforcement overwhelmingly involve female victims, and that 50% of the population is female, the “new-definition rapes” are occurring at a rate of 79.6 per 100,000 females, or 0.8 per thousand.
From reports that used the the new definition, 4.7% were for attempted rape or assaults to commit rape, and 95.1% were for “rape by force.” In short, FBI statistics overwhelmingly deal with completed rapes.
The “Clearances” report  indicates that “40.6 percent of rape offenses (new definition), 40.0 percent of rape offenses (old definition), … were cleared.”
The three conditions [for clearance] are that at least one person has been:
- Charged with the commission of the offense.
- Turned over to the court for prosecution (whether following arrest, court summons, or police notice).
In its clearance calculations, the UCR Program counts the number of offenses that are cleared, not the number of persons arrested. The arrest of one person may clear several crimes, and the arrest of many persons may clear only one offense.
40 percent of reported rapes are prosecuted. The UCR does not collect information on how many prosecutions returned a guilty verdict.
UCR counts only crimes reported to law enforcement and so underestimates the total number of rapes. In short, unreported rapes are not counted. Also, sexual assaults are not included in FBI statistics.
Attempted rapes are included, though not separated from forcible rape or (under the new definition) incapacitated rape. The earlier 2011 report did state 
The rate of forcible rapes in 2011 was estimated at 52.7 per 100,000 female inhabitants. Rapes by force comprised 93.0 percent of reported rape offenses in 2011, and attempts or assaults to commit rape accounted for 7.0 percent of reported rapes.
Unfortunately, the UCR offers no breakdown by age groups.
The UCR by itself cannot estimate the rate of rape among college-age women. From other surveys, for example, NCVS, it is possible to form a rough idea of how many rapes are not reported to police, and how many rapes happen to college-age women v. women of all ages. A broad estimate is offered in Table 1, presented earlier.
- DOE Department of Education “Clery Act” Report, (2013)
The Department of Education, Office of Postsecondary Education collects data on crimes committed on or near campuses as required by the “Higher Education Opportunity Act” and the “Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act” (commonly known as the Clery Act). 
The Clery Act requires the following compilation of statistics for all universities that participate in federal financial aid programs. The data is to be published by by October 1 of each year. The crime statistics must be sent to the Department of Education each year.
Crimes are categorized as occurring :
(1) on campus, (2) on public property within or immediately adjacent to the campus, and (3) in or on noncampus buildings or property that your institution owns or controls.
Note that the “on campus student housing” report is a subset of “on campus.”
The Data Analysis website  further cautions:
The crime statistics found on this website represent alleged criminal offenses reported to campus security authorities and/or local law enforcement agencies. Therefore, the data collected do not necessarily reflect prosecutions or convictions for crimes. Because some statistics are provided by non-police authorities, the data are not directly comparable to data from the FBI’s Uniform Crime Reporting System which only collects statistics from police authorities.
According to the Clery Center ,
“When not in conflict with the Clery Act, the standards of the FBI’s Uniform Crime Reporting program are to be used.”
The U.S. Department of Education “Handbook for Campus Safety and Security Reporting” is more specific :
The FBI’s National Incident-Based Reporting System (NIBRS) edition of the UCR defines a sex offense as any sexual act directed against another person, forcibly and/or against that person’s will; or not forcibly or against the person’s will where the victim is incapable of giving consent.
Per the Clery Act, you must classify crimes based on the Federal Bureau of Investigation’s (FBI’s) Uniform Crime Reporting Handbook (UCR). For sex offenses only, use definitions from the FBI’s National Incident-Based Reporting System (NIBRS) edition of the UCR.
And in greater detail ,
- Sex offenses. Sex offenses are separated into two categories: forcible and non-forcible. Include attempted sex offenses, but do not include in your Clery statistical disclosures any sex offenses other than the four types of Forcible Sex Offenses and the two types of Non-forcible Sex Offenses described in this chapter.
- a) Sex Offenses—Forcible is defined as any sexual act directed against another person, forcibly and/or against that person’s will; or not forcibly or against the person’s will where the victim is incapable of giving consent. Count one offense per victim. In cases where several offenders commit a Forcible Sex Offense against one person, count one Forcible Sex Offense. Do not count the number of offenders.
There are four types of Forcible Sex Offenses:
- Forcible Rape is the carnal knowledge of a person, forcibly and/or against that person’s will; or not forcibly or against the person’s will where the victim is incapable of giving consent because of his/her temporary or permanent mental or physical incapacity (or because of his/her youth). This offense includes the forcible rape of both males and females. Count one offense per victim.
If force was used or threatened, classify the crime as forcible rape regardless of the age of the victim. If no force or threat of force was used and the victim was under the statutory age of consent, classify the crime as statutory rape. The ability of the victim to give consent must be a professional determination by a law enforcement agency.
- Forcible Sodomy is oral or anal sexual intercourse with another person, forcibly and/or against that person’s will; or not forcibly or against the person’s will where the victim is incapable of giving consent because of his/her youth or because of his/her temporary or permanent mental or physical incapacity. Count one offense per victim.
- Sexual Assault With an Object is the use of an object or instrument to unlawfully penetrate, however slightly, the genital or anal opening of the body of another person, forcibly and/or against that person’s will; or not forcibly or against the person’s will where the victim is incapable of giving consent because of his/her youth or because of his/her temporary or permanent mental or physical incapacity. An object or instrument is anything used by the offender other than the offender’s genitalia. Examples are a finger, bottle, handgun, stick, etc. Count one offense per victim.
- Forcible Fondling is the touching of the private body parts of another person for the purpose of sexual gratification, forcibly and/or against that person’s will; or, not forcibly or against the person’s will where the victim is incapable of giving consent because of his/her youth or because of his/her temporary or permanent mental incapacity. Count one offense per victim.
- b) Sex Offenses—Non-forcible is defined as unlawful, non-forcible sexual intercourse.
There are two types of Non-forcible Sex Offenses:
- Incest is non-forcible sexual intercourse between persons who are related to each other within the degrees wherein marriage is prohibited by law. Count one offense per victim.
- Statutory Rape is non-forcible sexual intercourse with a person who is under the statutory age of consent. Count one offense per victim.
In summary: under the Clery Act, “forcible” sexual assault includes the use of force, threat of force, or incapacity. Note that e.g. in the case of incapacity, the perpetrator’s intent (sexual gratification) must be established. 
For the 2013 reporting year, the totals reported for all campuses are :
Sex offenses – forcible: 5,050
Sex offenses – non-forcible: 46
Sex offenses – forcible: 588
Sex offenses – non-forcible: 9
Sex offenses – forcible: 377
Sex offenses – non-forcible: 2
Reported by Local & State Police
Sex offenses – forcible: 0
Sex offenses – non-forcible: 0
Total all offenses 6,072
To explain the “Reported by Local & State Police” category :
In addition to collecting crime reports from campus security authorities, Clery requires that every institution make a “reasonable, good-faith effort” to obtain Clery crime statistics from local law enforcement agencies that have jurisdiction over the school’s Clery geography. Local law enforcement agencies do not include your campus police or security department (if you have one). Those are campus security authorities.
All “Reported by Local and State Police” numbers for all crimes – not just sex offenses – are suspiciously low. Either students are reporting crimes only to campus police, or the schools aren’t diligent about getting statistics from local police. For example, in 2013 there were a reported 24 instances of murder/manslaughter on campus, but 0 reported to local police. How is that possible?
The rate of sexual assault is also suspiciously small. Assume all female victims. The projected enrollment  for the year 2013 is 11,612,000 female college students (and 8,985,000 male). 6,069 victims in 11,612,000 is a rate of 0.52 per 1,000 per year. This is a total for forcible rape, incapacitated rape, and sexual assault other than rape.
Universities are accused of deliberately under-reporting crimes, since this information must be published and could alienate prospective students. In September 2014, the Columbus Dispatch reported ,
The crime statistics being released by colleges nationwide on Wednesday are so misleading that they give students and parents a false sense of security.
Even the U.S. Department of Education official who oversees compliance with a federal law requiring that the statistics be posted on Oct. 1 each year admits that they are inaccurate. Jim Moore said that a vast majority of schools comply with the law but some purposely under-report crimes to protect their images; others have made honest mistakes in attempting to comply.
In addition, weaknesses in the law allow for thousands of off-campus crimes involving students to go unreported, and the Education Department does little to monitor or enforce compliance with the law — even when colleges report numbers that seem questionable.
Phrases, such as “for the purpose of sexual gratification” in defining sexual assault, may tempt campus authorities to reclassify some sexual assaults as “ordinary” assaults if the intent is uncertain. But additional “loopholes” also allow under-reporting :
One is geographic. If a reported sexual assault took place in an off-campus building that isn’t owned by an officially recognized student group, a school doesn’t have to publish it. So, off-campus parties, like those thrown by fraternities, tend to be left out of the published statistics entirely. And fraternity brothers are more likely to rape than men who do not join a frat — three times as likely in their first year in college.
Of the 12 reported sexual assaults in Yale University’s Clery report, for example, none took place off campus. How many are they missing? There’s finally a clue. Starting in 2012, Yale became the only college to voluntarily publish data on all its reports of sexual assault. In 2013, that number was 19, or 58 percent more than its Clery figure.
Another Clery booby trap is where schools collect their numbers. Under the law, a university must gather its stats from campus and local police, as well as school officials with “significant responsibility for student and campus activities,” such as deans, coaches and academic advisers.
In 2003, Princeton University decided to go above and beyond the Clery law, including sex offense reports made to confidential counselors – inflating its own numbers in the interest of accuracy….
If Princeton had included confidential counselor reports again this past year, its total would have been 23, according to Mbugua. But the university’s Clery report posted only six.
The bottom line: Clery Act reports are probably best viewed as a “lower bound” for the real numbers.
- SVCW: Sexual Victimization of College Women, National Institute of Justice (2000)
Two older reports deserve brief mention. The more recent is “Sexual Victimization of College Women,” published in December 2000 by the National Institute of Justice . This is the same organization that commissioned the December 2007 Campus Sexual Assault Study, described above. (Page numbers below refer to the SVCW.)
SVCW was a telephone survey of 4,446 randomly selected college women, conducted at schools with a minimum of 1,000 students. Participants were contacted between February and May 1997, and asked about assaults during the previous school year (starting in the fall of 1996). The response rate was 85.6%.
“Rape” was defined as (p.8):
…penetration by force or the threat of force. Penetration includes: penile-vaginal, mouth on your genitals, mouth on someone else’s genitals, penile-anal, digital-vaginal, digital-anal, object-vaginal, and object-anal.
“Sexual coercion” included (p.6):
[someone made or tried to make you have] sexual contact when you did not want to by making threats of nonphysical punishment, such as lowering a grade, being demoted or fired from a job, damaging your reputation, or being excluded from a group…
…by making promises of rewards, such as raising a grade, being hired or promoted, being given a ride or class notes, or getting help with coursework…
…by simply being overwhelmed by someone’s continual pestering and verbal pressure.
“Sexual contact” included (p.8):
…touching; grabbing or fondling of breasts, buttocks, or genitals, either under or over your clothes; kissing; licking or sucking; or some other form of unwanted sexual contact.
Exhibit 3 (p.11) reports a prevalence of completed rape of 16.6 per 1,000 women for the preceding year. The prevalence of attempted rape was 11.0 per 1,000.
Exhibit 5 (p.16) reports the prevalence of sexual coercion was 16.6 per 1,000 completed sexual coercion, and 13.5 per thousand attempted.
The prevalence of completed sexual contact was 19.1 per thousand with force or threat of force, and 18.0 per thousand without force. Attempted sexual contact was 20.0 per 1,000 with force or threat of force, and 29.0 per thousand without force. It is not clear how much overlap there is between these groups.
The survey uses broad definitions of sexual contact and sexual coercion. It does not seem to record intoxicated/incapacitated rape.
- NVAWS: National Violence Against Women Survey, National Institute of Justice (1998)
The NIJ, in conjunction with the CDC, conducted an all-ages study from November 1995 to May 1996, which was reported in November 1998.  (Page numbers below refer to the report.)
A telephone survey of 8,000 women and 8,005 men, 18 years and older, was conducted by the Center for Policy Research.
The questions (p.13) employed a fairly standard definition of rape: vaginal, oral, or anal sex, by force, threat of force, or threatening to harm someone close to the victim. There were no specific questions about sexual assault other than rape.
Exhibit 1 (p.3) reports that 17.6% of women surveyed experienced an attempted or completed rape during her lifetime. Exhibit 2 (p.4) reports that 0.3% of women surveyed (3 per 1,000) experienced an attempted or completed rape during the previous year. The statistics reported for physical assault refer to non-sexual assault, and so are not relevant here.
Exhibit 6 (p.6) indicates that 54.0% of female rape victims were under 18 years old when first raped; 29.4% were age 18-24. (The age distribution of the respondents was not mentioned in this report; the omission may skew these statistics. As an extreme example, if half the respondents were 24 years old or younger, none of those respondents will report a first rape after the age of 25.)
As noted, this survey measured the prevalence and incidence only of rape and attempted rape. Other kinds of sexual assault were not included.
The data is a onslaught of confusion. The best statistics probably come from the NCVS, which has sound methodology and is repeated consistently on an annual basis. NCVS provides information that other studies do not, such as the rate of sexual assault for students v. non-students, and the rate of non-reporting. It is perhaps unique in that the Bureau of Justice Statistics convened a panel of experts to evaluate and improve its methods. Its main weaknesses: Being an all-crime survey, it has only a few questions about rape and sexual assault, and it does not address being assaulted while incapacitated. Other studies point to the latter as the most frequent form of sexual assault.
It would be incredibly helpful if researchers converged on a common set of definitions. As it is, some studies are very liberal in their definitions and others are conservative, which strips value and precision from comparisons.
It is clear, however, that statements like the one by Vice President Biden  are misleading and unnecessarily alarmist. He declared, “We know the numbers: one in five of every one of those young women who is dropped off for that first day of school, before they finish school, will be assaulted, will be assaulted in her college years.”
More accurate and useful advice might sound like this:
In any given year, there is something like a one percent chance that you will be raped. That’s true whether or not you go to college – in fact, college is slightly safer for you. If you are raped, the most likely scenario will be that you are incapacitated by alcohol. So be careful around men who want to ply you with liquor. It is more likely – maybe a 5 percent chance per year – that someone will try to fondle you or kiss you against your will or pester you into having sex. Plan right now on how to handle these situations.
On the bright side, in any given year, 95 percent of you are statistically likely to encounter only men who – even if they cannot be described as well-behaved — are men who want your willing and active consent to sexual activity.
 As quoted in “Beware the bad survey: Science literacy isn’t as bad as the statistics make it look,” PLOS blog, May 12, 2015. http://blogs.plos.org/scicomm/2015/05/12/beware-the-bad-survey-science-literacy-isnt-as-bad-as-the-statistics-make-it-look/ Retrieved Sept. 27, 2015.
 Jennifer L. Truman and Michael Planty, Criminal Victimization, 2011, Oct. 2012, NCJ 239437. http://www.bjs.gov/content/pub/pdf/cv11.pdf Retrieved Sept. 27, 2015.
 “Prevalence and Characteristics of Sexual Violence, Stalking, and Intimate Partner Violence Victimization — National Intimate Partner and Sexual Violence Survey, United States, 2011,” Centers for Disease Control.
http://www.cdc.gov/mmwr/preview/mmwrhtml/ss6308a1.htm?s_cid=ss6308a1_e Retrieved Sept. 27, 2015.
 Claire Groden, “Why is it so Hard to Determine Exactly How Many Women are Raped Each Year?” New Republic, Sept. 8, 2014. http://www.newrepublic.com/article/119364/cdcs-report-one-five-women-raped-other-statistics-disagree Retrieved Sept. 27, 2015.
 Cathy Young, “The CDC’s Rape Numbers Are Misleading,” Time, Sept. 17, 2014. http://time.com/3393442/cdc-rape-numbers/ Retrieved Sept. 27, 2015.
 Sofi Sinozich and Lynn Langton, “Rape and Sexual Assault Victimization Among College-Age Females, 1995–2013 ”, U.S. Department of Justice, Bureau of Justice Statistics, Dec. 2014. http://www.bjs.gov/content/pub/pdf/rsavcaf9513.pdf Retrieved Sept. 27, 2015.
 M.J. Breiding, S.G. Smith, K.C. Basile, M.L. Walters, J. Chen and M.T. Merrick, “Prevalence and characteristics of sexual violence, stalking, and intimate partner violence victimization — National Intimate Partner and Sexual Violence Survey, United States, 2011. Morbidity and Mortality Weekly Report, Surveillance Summaries, 63(8), 1–18. http://www.cdc.gov/mmwr/preview/mmwrhtml/ss6308a1.htm Retrieved Sept. 27, 2015.
 C.P. Krebs, C.H. Lindquist, T.D. Warner, B.S. Fisher and S.L. Martin, S.L. (2007), “The Campus Sexual Assault (CSA) study. Final report to the National Institute of Justice, grant number 2004-WG-BX-0010, document number 221153.” https://www.ncjrs.gov/pdffiles1/nij/grants/221153.pdf Retrieved Sept. 27, 2015.
 “Crime in the United States, 2013,” U.S. Department of Justice, Federal Bureau of Investigation, Uniform Crime Reporting (UCR) Program, Rape. http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/violent-crime/rape Retrieved Sept. 27, 2015.
 Bonnie S. Fisher, Francis T. Cullen, Michael G. Turner, (Dec. 2000). “Sexual Victimization of College Women”. National Institute of Justice. http://www.ncjrs.gov/pdffiles1/nij/182369.pdf Retrieved Sept. 27, 2015.
 “Prevalence, Incidence, and Consequences of Violence Against Women: Findings of the National Violence Against Women Survey”. National Institute of Justice. November 1998. https://www.ncjrs.gov/pdffiles/172837.pdf Retrieved Sept. 27, 2015.
 The U.S. Department of Education, Office of Postsecondary Education collects data required by the “Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act” (a.k.a. the Clery act) and the “Higher Education Opportunity Act”. http://ope.ed.gov/security/ Retrieved Sept. 27, 2015.
 For example, if in one year a woman was assaulted once, and another woman was assaulted twice, then the incidence for that year would be 3 (assaults) but the prevalence would be 2 (victims).
 Calculated as follows: let P be the population of women of all ages. Then the number of college-age women is 0.1P, and the number of other women is 0.9P. If the college-age rate of rape is 0.0043 (4.3 per thousand), and the non-college rate 0.0014 (1.4 per thousand), then the total number of rapes for all women is R = (0.1P)(0.0043) + (0.9P)(0.0014). The per-capita rate of rape for all women, r, is calculated as R/P; dividing R by P gives r = R/P = (0.1)(0.0043) + (0.9)(0.0014) = 0.00169. That is a rate of 1.69 per thousand. The college rate, 4.3 per thousand, is about 2.5 times the all-women rate.
 Candace Kruttschnitt, William D. Kalsbeek, and Carol C. House, eds., “Estimating the Incidence of Rape and Sexual Assault,” Panel on Measuring Rape and Sexual Assault in Bureau of Justice Statistics Household Surveys, National Research Council, 2014. Committee on National Statistics, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press, pp. 3-4. http://www.nap.edu/catalog.php?record_id=18605 Retrieved Sept. 27, 2015.
 National Research Council, op. cit., p.2.
 “Rape and Sexual Assault Victimization Among College-Age Females, 1995–2013,” U.S. Department of Justice, Bureau of Justice Statistics, Dec. 2014. http://www.bjs.gov/content/pub/pdf/rsavcaf9513.pdf Retrieved Sept. 28, 2015.
 Libby Nelson, “Why some studies make campus rape look like an epidemic while others say it’s rare,” Vox, Dec. 11, 2014. http://www.vox.com/2014/12/11/7378271/why-some-studies-make-campus-rape-look-like-an-epidemic-while-others Retrieved Sept. 28, 2015.
 Op.cit, Breiding, “Prevalence and characteristics of sexual violence.”
 “National Intimate Partner and Sexual Violence Survey (NISVS) 2011 victimization questions,” National Center for Injury Prevention and Control (U.S.), Division of Violence Prevention. pp. 3-4. http://stacks.cdc.gov/view/cdc/24726 Retrieved Sept. 28, 2015.
 Ibid., p.4.
 “Age and Sex Composition: 2010,” U.S. Department of Commerce , Economics and Statistics Administration, U.S. Census Bureau. http://www.census.gov/prod/cen2010/briefs/c2010br-03.pdf Retrieved Sept. 28, 2015.
 “Telescoping” refers to the tendency to include events that happened more than a year ago, when asked about events during the past year. If this is happening, it would tend to inflate the numbers.
 Op.cit., Krebs, “The Campus Sexual Assault (CSA) study.”
 Christopher Krebs and Christine Lindquist, “Setting the Record Straight on ‘1 in 5’,” Time, Dec. 15, 2014. http://time.com/3633903/campus-rape-1-in-5-sexual-assault-setting-record-straight/ Retrieved Sept. 28, 2015.
 Cathy Young, “The White House Overreaches on Campus Rape,” Minding the Campus, Jan. 23, 2014.
http://www.mindingthecampus.com/2014/01/the_white_house_overreaches_on/ Retrieved Sept. 28, 2015.
 Op.cit., Libby Nelson.
 Dean G. Kilpatrick, Heidi S. Resnick, Kenneth J. Ruggiero, Lauren M. Conoscenti, and Jenna McCauley, “Drug-facilitated, Incapacitated, and
Forcible Rape: A National Study ,” Medical University of South Carolina, Feb. 1, 2007. https://www.ncjrs.gov/pdffiles1/nij/grants/219181.pdf Retrieved Sept. 28, 2015.
 Op.cit., Krebs, Time.
 Nick Anderson and Scott Clement, Scott, “1 in 5 college women say they were violated,” Washington Post, June 12, 2015. http://www.washingtonpost.com/sf/local/2015/06/12/1-in-5-women-say-they-were-violated/ Retrieved Sept. 28, 2015.
 “Crime in the United States, 2013: Summary of the Uniform Crime Reporting Program,” U.S. Department of Justice, Federal Bureau of Investigation.
 “Crime in the United States 2013: Rape,” U.S. Department of Justice, Federal Bureau of Investigation.
http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/violent-crime/rape Retrieved Sept. 28, 2015.
 Op.cit., “Crime in the United States 2013: Summary of the Uniform Crime Reporting Program.”
 Op.cit., “Crime in the United States 2013: Rape.”
 “Crime in the United States 2011: Offense Definitions,” U.S. Department of Justice, Federal Bureau of Investigation. https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/11offensedefinitions_final.pdf Retrieved Sept. 28, 2015.
 Op.cit., “Crime in the United States 2013: Rape.”
 “Crime in the United States 2013: Table 1,” U.S. Department of Justice, Federal Bureau of Investigation. https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/tables/1tabledatadecoverviewpdf/table_1_crime_in_the_united_states_by_volume_and_rate_per_100000_inhabitants_1994-2013.xls Retrieved Sept. 28, 2015.
 “Crime in the United States 2013: Table 16,” U.S. Department of Justice, Federal Bureau of Investigation. https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/tables/table-16/table_16_rate_by_population_group_2013.xls Retrieved Sept. 28, 2015.
 “Crime in the United States 2013: Clearances,” U.S. Department of Justice, Federal Bureau of Investigation. http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/offenses-known-to-law-enforcement/clearances/clearancetopic_final Retrieved Sept. 28, 2015.
 “Crime in the United States 2011: Forcible Rape,” U.S. Department of Justice, Federal Bureau of Investigation. https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/violent-crime/forcible-rape Retrieved Sept. 28, 2015.
 “The Campus Safety and Security Data Analysis Cutting Tool,” U.S. Department of Education, Office of Postsecondary Education. http://ope.ed.gov/security/ Retrieved Sept. 28, 2015.
 “Handbook for Campus Safety and Security Reporting,” U.S. Department of Education, p.11.
http://rems.ed.gov/docs/ED_CampusSafetyAndSecurityReportingHandbook.pdf Retrieved Sept. 28, 2015.
 Op.cit., “The Campus Safety and Security Data Analysis Cutting Tool,” U.S. Department of Education, Office of Postsecondary Education.
 Clery Act Compliance Resources, Clery Center. http://clerycenter.org/clery-act-compliance-resources Retrieved Sept. 28, 2015.
 “Handbook for Campus Safety and Security Reporting,” U.S. Department of Education, p.141.
 Ibid., p. 34.
 Ibid., pp.37-39.
 Ibid., p. 39.
 “The Campus Safety and Security Data Analysis Cutting Tool: Aggregated Data,” U.S. Department of Education, Office of Postsecondary Education. (Using the link “Get aggregated data for a group of campuses”. ) http://ope.ed.gov/security/GetAggregatedData.aspx http://ope.ed.gov/security/ Retrieved Sept. 28, 2015.
 “Handbook for Campus Safety and Security Reporting,” U.S. Department of Education, p.82.
 “Digest of Education Statistics, 2013, Table 303.10, National Center for Education Statistics. “Total fall enrollment in degree-granting postsecondary institutions, by attendance status, sex of student, and control of institution: Selected years, 1947 through 2023.” http://nces.ed.gov/programs/digest/d13/tables/dt13_303.10.asp See also http://nces.ed.gov/fastfacts/display.asp?id=98 Both retrieved Sept. 28, 2015.
 Collin Binkley, Jill Riepenhoff, Mike Wagner and Sara Gregory, “Reports on college crime are deceptively inaccurate,” The Columbus Dispatch, Sept. 30, 2014. http://www.dispatch.com/content/stories/local/2014/09/30/campus-insecurity.html Retrieved Sept. 28, 2015.
 Clair Gordon, “Sexual assault reports jump 61 percent at top colleges in two years,” Al-Jazeera America, Oct. 7, 2014. http://america.aljazeera.com/watch/shows/america-tonight/articles/2014/10/7/colleges-clery-sexualassault1.html Retrieved Sept. 28, 2015.
 “Sexual Victimization of College Women,” National Institute of Justice. http://www.ncjrs.gov/pdffiles1/nij/182369.pdf Retrieved Sept. 28, 2015.
 “Prevalence, Incidence, and Consequences of Violence Against Women: Findings of the National Violence Against Women Survey,” National Institute of Justice, Nov. 1998. https://www.ncjrs.gov/pdffiles/172837.pdf Retrieved Sept. 28, 2015.
 Joseph Biden as quoted by Glenn Kessler, “One in five women in college sexually assaulted: the source of this statistic,” Washington Post, May 1, 2014. http://www.washingtonpost.com/blogs/fact-checker/wp/2014/05/01/one-in-five-women-in-college-sexually-assaulted-the-source-of-this-statistic/ Retrieved Sept. 28, 2015.