Victimization Surveys

  1. Methodology
  2. Data
  3. Table - Selected NCVS Statistics
  4. Critical Assessment


Victimization surveys attempts to bypass the underreporting problem by going directly to the victims. The National Crime Victimization Survey (NCVS) is conducted by the U.S. Bureau of the Census in cooperation with the Bureau of Justice Statistics and the U.S. Department of Justice. The NCVS polls over 50,000 households, totaling over 100,000 individuals, in the United States annually using a multistage sample of housing units. Individuals over 12 years old in selected households are interviewed every six months for about three years. The first interview is conducted face-to-face and only used to bound the responses. Further interviews are conducted over the telephone.

Rape is the most severe personal crime measured by the NCVS (US Bureau of Justice Statistics 1991). The definition of rape used by the NCVS is "Carnal knowledge through the use of force or threat of force, including attempts. Statutory rape (without force) is excluded. Both heterosexual and homosexual rape are included" (US Bureau of Justice Statistics 1991, p.141). Rapes and sexual assaults are ascertained through three questions in the NCVS-1 Basic Screen Questionnaire. Question 41a asks, "(Other than any incident already mentioned,) has anyone attacked or threatened you in any of these ways (Exclude telephone threats) … (e) Any rape, attempted rape or other type of sexual attack." Question 42a asks, "People often don’t think of incidents committed by someone they know. (Other than any incidents already mentioned,) did you have something stolen from you OR were you attacked or threatened by (Exclude telephone threats) – (a) Someone at work or at school – (b) A neighbor or friend – (c) A relative or family member – OR (d) Any other person you’ve met or known?" Question 43a asks, "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?" Followed by question 43b, "Did any incidents of this type happen to you?" and question 43c, "How many times?" In the NCVS-2 Crime Incident Report, Question 29 asks, "How were you attacked? Any other way?" Possible responses to question 29 include raped, tried to rape, and sexual assault other than rape or attempted rape. Question 31 asks, "What were the injuries you suffered, if any? Anything else?" Possible responses to question 31 include raped, attempted rape, and sexual assault other than rape or attempted rape. If the response to question 29 is raped, or the response to question 31 is raped and the response to question 29 is not raped, the field representative is instructed to ask, "Do you mean forced or coerced sexual intercourse?" Similarly, if the response to question 29 is tried to rape, or the response to question 31 is attempted rape and the response to question 29 is not tried to rape, the field representative is instructed to ask, "Do you mean attempted forced or coerced sexual intercourse?" If no is the response to either of the two preceding questions, they are to ask, "What do you mean?"



Table - Selected National Crime Victimization Survey Statistics

Year Statistic
1994 Rapes and sexual assaults in the US 433,000
1994 UCR/NCVS rapes reported .31
1994 Reported to law enforcement .32
1994 Female rape rate .004
1994 Male rape rate .0002
1994 Female victims .91
1994 Male single-victim perpetrators .99
1993 Occurring between 6 PM and 6 AM .66
1993 Own home or non-stranger home .60
1993 Within 1 mile of victim’s home > .50
1993 Single offender .91
1993 Firearm present .06
1993 No weapon .84
1993 Stranger perpetrator .24
1993 Perpetrator <21 .25
1993 Self-protective measures taken .72
1993 Self-protective measures helpful >.50
1993 Self-protective measures harmful .20
1993 Rape victims/Victims of violence .04
1993 Rapes requiring medical attention/
incidents requiring medical attention
Source: Greenfeld 1997.

The redesign of the NCVS in 1992 produced sexual assault rate estimates about 4 times higher than were previously reported for rape by the NCVS (Greenfeld 1997). Table 7 summarizes more recent selected sexual assault statistics from the National Crime Victimization Survey. In 1994, the NCVS found that 433,000 rapes were experienced in the United States (Greenfeld 1997). The rape rates were 1 rape for every 270 female and 1 rape for every 5,000 male residents 12 years or older. Rape rates were highest among those 16 to 19 years of age, low-income residents, and urban residents. Rape rates did not differ significantly among racial groups. Ninety-one percent of rape victims were female. Ninety-nine percent of single-victim incidents involved male perpetrators. Approximately 32% were reported to a law enforcement agency.

In 1993, two-thirds of rapes occurred between 6 PM and 6 AM. Almost 60% of the incidents took place in the home of the victim or the home of a friend, relative, or neighbor. Over 50% occurred within one mile of the victim’s home. Ninety-one percent of rapes involved single offenders. Only 24% of incidents involved strangers. Victims reported that approximately 25% of perpetrators were under 21 years of age and 40% were 30 years of age or older.

Seventy-two percent of victims reported taking self-protective measures during the rape. Nineteen percent resisted or captured the offender. Eleven percent scared or warned the offender. Eleven percent persuaded or appeased the offender. Other self-protective measures include running away or hiding, attacking the offender, screaming, and getting help or giving alarm. A majority of those who took self-protective measures felt they helped the situation. However, about 20% felt they made the situation worse.

In 1993, rape or sexual assault victims accounted for 4% of victims of violence. Rapes accounted for 6% of the incidents in which medical assistance was obtained. Less than 10% of rape victims reported suffering an economic loss. The average reported economic loss was approximately $200. Seven percent of victims reported losing time from work.

In 1994, the ratio of people reporting rapes to the NCVS to the number of rapes known to law enforcement, .31, was approximately equal to the NCVS proportion of rape victims revealing that they reported the incident to law enforcement, .32. This indicates that individuals willing to report sexual assault to a government agency are approximately 3 times more likely to report sexual assault to NCVS interviewers than to law enforcement.


Critical Assessment

Victimization surveys have the potential for being the most accurate source of data on the incidence of sexual assault. The major advantage of victimization surveys is that they go out into the population and ask for information, not waiting for incidents to be reported to an agency. However, there is no guarantee that individuals will be any more willing to report sexual assaults to census workers than to the police. In addition, the quality and quantity of information obtained by a survey is very sensitive to how questions are asked.

Although the NCVS is more effective than the UCR program at uncovering the incidence of rape at the national level, it has several disadvantages. The sampling scheme of the NCVS does not allow reliable estimation of rape rates at the state or local level. Therefore, it cannot be used as a source of sexual assault incidence and service planning data within the State of Arizona. The NCVS provides a more valid and reliable definition of sexual assault by including males and females as victims and perpetrators and measuring a broader range of sexual behavior beyond sexual intercourse. However, despite recent revisions, the NCVS does not ask about rape and sexual assault explicitly. Unless the respondent asks what is meant by sexual assault, the definition of rape is solely in the mind of the respondent. In addition, the context of the interview is uncertain (Koss 1996). The interviews are conducted over the telephone and other household members may be present. The privacy needed to maximize reporting is not ensured by the interview protocol of the NCVS. Finally, questions about rape are put into a violent crime context that is not associated with the typical rape. For example, the National Women’s Study found that 70% sustained no physical injuries during rape (National Victim Center and Crime Victim Research and Treatment Center 1992).

The major disadvantage of victimization surveys is that they are extremely expensive. Rape has a relatively low incidence rate. A survey of the population is not financially feasible because a large sample is needed to reliably project statewide victim rates.


Arizona Rape and Sexual Assault Surveillance Project
University of Arizona, University of Arizona Prevention Center
All contents copyright © 1998. All rights reserved.