Sexual Contact from Victimization to Crime

face of a prisoner

Crime victimization surveys attempt to measure sexual victimization that is criminal. Sexual victimization can be a deeply personal feeling. Sex crimes, in contrast, are defined and judged under public criminal law. Measuring sexual victimization isn’t a straight-forward, objective statistical exercise of measuring unreported crimes. Crime victimization surveys both reshape victims’ perceptions of sexual victimization and independently judge sex crimes.

Understanding what sexual victimization surveys report requires detailed understanding of survey officials’ crime codes and judging procedures. Crime victimization surveys establish crime definitions and conventions for coding incidents that respondents recall in response to a variety of probing questions. Survey officials judge crime based on the incident report that field interviewers submit for the victim’s responses. What counts as particular crimes depends on survey officials’ criminal judgments. Those judgments can differ greatly from how the victim would judge her or his victimization, or how an actual judge would judge the case if presiding over a trial for the incident in question.

Survey officials’ judgments of sex crimes have great quantitative importance for rape victimization statistics. In a survey of sexual victimization of college women in 1997, the surveyed subject did not regard as rape 49% on the incidents that survey officials categorized as completed rape.^ The criminal justice system likewise declares far fewer convictions for rape than the number of rape victimizations that crime victimization surveys report. In 1992, state courts made 21,655 rape convictions. A major crime victimization survey reported 607,000 rape/sexual assault victimizations. Reported trends in rape from victimization surveys don’t even parallel trends in rape convictions and rapes known to the police. From 1992 to 2006, rape convictions, rapes known to the police, and rapes in victimization surveys increased 53%, -13%, and -69%, respectively.

Obscure aspects of survey design and administrative procedures greatly affect rape statistics that crime victimization surveys report. Consider the U.S. National Crime Victimization Survey (NCVS). It’s a large, ongoing victimization survey that the U.S. Department of Justice manages. NCVS categorizes “sex-related crimes” into four categories: rape, attempted rape, sexual assault, and unwanted sexual contact. Each of these categories are associated with survey questions asking about acts, attempted acts, and threats to act.^ The NCVS Field Manual defines rape as follows:

For the NCVS, rape is defined as forced sexual intercourse including both psychological coercion, as well as physical force. Forced sexual intercourse means vaginal, anal, or oral penetration by the offender(s). This category also includes incidents where the penetration is from a foreign object, such as a bottle. A rape victim can be either female or male and the rape can be heterosexual or homosexual.^

In NCVS, forced sexual intercourse doesn’t include men being made to penetrate. That’s a quantitatively significant form of rape. According to a U.S. official, national victimization survey, 1.3 million men per year are made to penetrate another sexually.^ Being made to penetrate another sexually is reasonably regarded as real rape. The NCVC survey instrument differentiates sexual assault from rape and attempted rape:

Enter Precode (13) {sexual assault} when the respondent was sexually assaulted in some way other than rape or attempted rape; that is, the sexual assault did not involve forced or coerced sexual intercourse or attempted sexual intercourse (for example, fondling the respondent’s breasts against her will).^

The NCVS Field Manual defines unwanted sexual contact without force thus:

Enter Precode (16), Unwanted sexual contact without force (grabbing, fondling, etc.), if the offender made some type of unwanted sexual contact without the use of force. For example, during the incident, the offender may have embraced, fondled, or touched the respondent against his/her will, but did not grab, push, or restrain the respondent in any way.^

Identifying rapes, attempted rapes, and sexual assaults requires classifying threats, including psychological coercion. Threats can constitute attempted rape or sexual assault. The NCVS requires field workers to interpret victims’ responses using a narrow definition of threats:

Only accept threats that are verbal, face-to-face threats to physically harm the respondent. Do not accept threats that an offender makes over the telephone, in a letter, FAX, or electronic message. Also unacceptable are threats or warnings delivered by another person for the offender, as well as a respondent saying that he/she just felt threatened.^

An NCVS screening question that probes for sex-related crime encompasses “unwanted sex”:

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 —

– Someone you didn’t know -
– A casual acquaintance -
OR
– Someone you know well?

Unwanted sexual activity does not necessarily imply a sex crime. For example, a man may have sex with his wife when he doesn’t want to have sex, but she does. NCVS officials identify within unwanted sexual activities those incidents that NCVS declares to be sex crimes. Both field workers and NCVS officials processing field workers’ reports make those judgments. Emphasizing the importance of NCVS officials’ crime judgments, the NCVS Field Manual advises field workers:

Since sex-related crimes are rare compared to other types of crimes, include as many details as the respondent is willing to provide. This is important so that we can classify any sex-related crimes into the correct category– rape, attempted rape, sexual assault, or unwanted sexual contact.

Avoid using phrases like “made sexually explicit comments,” “unwanted sexual contact,” or “unwanted sexual advances.” These phrases do not provide us with enough information to determine what actually happened. We need to know what was actually said, what parts of the body were touched, whether or not force was used, and so on.^

NCVS classifies as violent crimes victimizations that NCVS categorizes as attempted or completed rape, sexual assault, or verbal threats of rape or sexual assault.^ In its annual report Criminal Victimization, the NCVS reports under the heading “violent crime” the category “rape/sexual assault.” The category “rape/sexual assault” thus groups together a wide range of behaviors. Survey officials judge across that wide range of sexual behavior what behavior is unwanted sexual contact, sexual assault, attempted rape, or rape. While sex is common behavior, most persons have no clear understanding of the difference between the crimes of unwanted sexual contact, sexual assault, attempted rape, and rape.

Rape involves deep conceptual difficulties in surveying criminal victimization. Both the relevant case law concerning rape and perceptions of rape victimization vary widely across relatively common behavior. Consider the following statement/threat:

If you don’t have sex with me, I’ll stop going out with you.

That doesn’t count as a threat under the NCVS definition of threats. But persons receiving that threat and acquiescing to sex might well feel victimized. They might feel that they were forced to have unwanted sex via psychological coercion. Such incidents appear to meet the NCVS definition of rape. A leading work on unwanted sex that considers in detail the legal meaning of consent argues that such coercion should not imply the crime of rape.^ However, if the NCVS ignores such incidents, it may ignore deeply felt feelings of sexual victimization.

Recognizing emotional harm to be often the primary harm in sexual victimization implies recognizing broad possibilities for unwanted sex forced via psychological coercion. Suppose Abby’s intimate would feel deeply hurt if Abby had sex with Bill. Abby’s intimate also thinks that Abby having sex with Bill would be emotionally harmful for Abby herself. Suppose Abby said to her intimate:

If you don’t have sex with me, I’ll have sex with Bill.

Abby’s intimate may feel forced to have unwanted sex with Abby. If Abby’s intimate doesn’t have sex with Abby, Abby may feel forced to have unwanted sex with Bill. Do these circumstances imply that either Abby’s intimate will be raped, or Abby will be raped. Could Abbey’s statement/threat cause both Abbey and her intimate to be raped? Drawing force from both self-concern and other-regarding concern, Abby’s statement/threat is highly potent psychological coercion. Emotional harm is difficult to judge objectively. Perceptions of such harm can vary widely across intimately related persons. Statements interpreted as including a threat of emotional harm to multiple parties are likely to be relatively common means of sexual coercion.

Survey design, survey crime codes, and survey officials’ judgments appear to be much more significant than sampling error in rape victimization surveys. Beginning in 2008, NCVS reported standard errors for its annual rape estimates. The standard errors from 2008 to 2010 were about 20% of the estimated figures. NCVS annual rape estimates are based on a small number of incidents (36 to 57 from 2008 to 2010). The standard error accounts for sample size. The standard error doesn’t account for ambiguous categorization. Particular administrative decisions in judging the sexual crime for just ten incidents would cover roughly a quarter of the incidents that determine the NCVS annual national estimate of rape/sexual assault. Redesigns of NCVS have produced sevenfold increases in rape figures. Survey officials’ choices in defining and judging rape are much more important than standard errors in evaluating NCVS rape estimates. Those crucial crime definitions and crime judgments are publicly opaque.

NCVS has obscured in technical expertise fundamental challenges in surveying sexual victimization. The annual NCVS report for 1992, Criminal Victimization 1992, estimated 141,000 rape victimizations in 1992. Criminal Victimization 1993 estimated 607,000 rape/sexual assault victimizations in 1992, based on 1992 data from the redesigned NCVS. Criminal Victimization 1993 didn’t publish simple comparative figures showing the effect of the redesign on total estimated sexual crimes.^ It described the redesign as a matter of technical expertise:

the NCVS underwent a thorough, decade-long redesign to improve the survey’s ability to measure victimization in general and certain difficult-to-measure crimes, such as rape, sexual assault, and domestic violence, in particular.

A consortium of experts in criminology, survey design, and statistics performed extensive study and testing to update the questionnaire as well as survey procedures. Among the changes was the addition of sexual assault to expand the types of sexual crimes counted. Direct questions about these crimes were added to encourage victims to report to interviewers incidents that may have been committed by someone known to them.^

The large increases in rape reported by redesign in 1992 and 2010 aren’t merely a technical matter of better measurement. NCVS rape figures involve significant administrative judgments about sexual crimes. If NCVS doesn’t make those judgments transparently and reasonably, NCVS rape (or rape/sexual assault) figures aren’t objective, credible data.

A key survey administrative decision is how to treat men being made to penetrate sexually another person. Victimization surveys commonly haven’t even asked men if they have been sexually victimized by being made to penetrate. Major U.S. national surveys that have asked about sexual victimization in a gender-neutral way have found the prevalence of sexual victimization of men to be similarly to the prevalence of sexual victimization of women.^ If rape victimization is defined as nonconsensual sex, then being made to penetrate is real rape.

The public judgment of crime in a law-governed society differs from personally perceived victimization. The National Crime Victimization Survey tends to collapses the distinction between publicly defined crime and personally perceived victimization. Victimization of women attracts highly disproportionate public concern. The criminal justice system highly disproportionately incarcerates men. Because of its emotional potency, mis-information about sexual victimization is particularly pernicious in public discourse. Public discussion of sexual victimization, like public discussion of domestic violence, is deeply connected to disproportionate incarceration of men and mass incarceration.

Leave a comment (will be included in public domain license)

Your email address will not be published. Required fields are marked *