Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Ybarra ML, Mitchell KJ. Prevalence Rates of Male and Female Sexual Violence Perpetrators in a National Sample of Adolescents. JAMA Pediatr. 2013;167(12):1125–1134. doi:10.1001/jamapediatrics.2013.2629
Sexual violence can emerge in adolescence, yet little is known about youth perpetrators—especially those not involved with the criminal justice system.
To report national estimates of adolescent sexual violence perpetration and details of the perpetrator experience.
Design, Setting, and Participants
Data were collected online in 2010 (wave 4) and 2011 (wave 5) in the national Growing Up With Media study. Participants included 1058 youths aged 14 to 21 years who at baseline read English, lived in the household at least 50% of the time, and had used the Internet in the last 6 months. Recruitment was balanced on youths’ biological sex and age.
Main Outcomes and Measures
Forced sexual contact, coercive sex, attempted rape, and completed rape.
Nearly 1 in 10 youths (9%) reported some type of sexual violence perpetration in their lifetime; 4% (10 females and 39 males) reported attempted or completed rape. Sixteen years old was the mode age of first sexual perpetration (n = 18 [40%]). Perpetrators reported greater exposure to violent X-rated content. Almost all perpetrators (98%) who reported age at first perpetration to be 15 years or younger were male, with similar but attenuated results among those who began at ages 16 or 17 years (90%). It is not until ages 18 or 19 years that males (52%) and females (48%) are relatively equally represented as perpetrators. Perhaps related to age at first perpetration, females were more likely to perpetrate against older victims, and males were more likely to perpetrate against younger victims. Youths who started perpetrating earlier were more likely than older youths to get in trouble with caregivers; youths starting older were more likely to indicate that no one found out about the perpetration.
Conclusions and Relevance
Sexual violence perpetration appears to emerge earlier for males than females, perhaps suggesting different developmental trajectories. Links between perpetration and violent sexual media are apparent, suggesting a need to monitor adolescents’ consumption of this material. Victim blaming appears to be common, whereas experiencing consequences does not. There is therefore urgent need for school programs that encourage bystander intervention as well as implementation of policies that could enhance the likelihood that perpetrators are identified.
With more than 1 million victims and associated costs of almost $127 billion each year, sexual violence is a significant public health problem.1 In addition to societal costs, the impact on the individual can be high, including increased rates of posttraumatic stress disorder,2 physical health problems,3 and suicidal behavior.4
Sexual violence can emerge in adolescence,5-7 making this developmental time a critical period of inquiry. Nonetheless, nationwide estimates for adolescent perpetrators of sexual violence in community samples are lacking8 and state estimates vary significantly.1 Moreover, almost all of the sexual violence perpetration literature focuses on boys as the sexual aggressors and females as victims.9-12 In studies that include both adolescent males and females as perpetrators, females are less likely to engage in sexually violent behavior than males.13-15 Research examining specific event characteristics of male and female sexual offenders is even more limited and is conflicting as to whether differences exist.16,17 There is a clear need for a better understanding not just of prevalence rates but also how sexual violence may be different for older and younger adolescents as well as males and females. Given emerging interest in possible links between exposure to sexual material and sexual behavior and attitudes18-20 as well as violent pornography and violent sexual behavior,21 further research examining the associations between media and sexual violence also is critical.
Growing Up With Media is a longitudinal survey examining the associations between exposure to violent media and violent behavior. Wave 1 data were collected between August and September 2006 with 1586 youth-caregiver pairs. Extensive questions about sexual perpetration were added at wave 4. Therefore, data discussed in this article were collected in wave 4 (fielded between October 2010 and February 2011) and wave 5 (fielded between October 2011 and March 2012). The survey protocol was reviewed and approved by the Chesapeake Institutional Review Board, which is a private, paid Office for Human Research Protections–approved institutional review board. Caregivers provided informed consent for their participation and permission for their child’s participation; youths provided informed assent.
Fifty-seven caregivers and 8 youths actively declined to consent/assent at baseline. Because both caregiver permission and youth assent were required for youth participation, the household response rate was the same as the youth response rate, which was 31%. This is well within the expected range of well-conducted online surveys.22-24 Of the 1586 households who completed the baseline survey, 67% (n = 1062) completed either or both of the wave 4 (response rate = 56%, n = 887) and wave 5 (response rate = 59%, n = 939) surveys. Characteristics of nonresponders and responders were similar (Table 1).
Adult respondents were recruited at baseline through an e-mail sent to randomly identified adult Harris Poll Online panel members who reported having a child living in the household. Eligible adults were equally or more knowledgeable than other adult household members about the youth’s home media use. Youth participants were aged 10 to 15 years at recruitment (mean [SD], 12.7 [1.8] years; aged 14-21 years at waves 4 and 5), read English, lived in the household at least 50% of the time, and had used the Internet in the last 6 months. Recruitment was balanced on youths’ biological sex and age.
On average, caregivers took 15 minutes and youths took 32 minutes to complete their portion of the Growing Up With Media survey. Caregivers received $20 and youths received $25 as an incentive. To increase response rates at the end of the field period, an additional $10 bonus incentive was offered to nonresponders.
The term sexual violence broadly refers to acts ranging from sexual harassment to rape.25 Herein, we focus on coercive and forced sexual behavior. It should be noted that while related, sexual violence is different from teen dating violence, which by definition occurs within a dating relationship; sexual violence can be perpetrated by a romantic partner, someone else known by the victim, or someone completely unknown to the victim.
Sexual violence perpetration was queried using 4 items. Three items were modified from the Sexual Experiences Survey26,27 and are consistent with the Bureau of Justice Statistics definition of rape, which can include “psychological coercion as well as physical force.”28 Youths were asked about how often they had ever done the following: (1) “tried, but was not able, to make someone have sex with me when I knew they did not want to”; (2) “made someone have sex with me when I knew they did not want to”; and (3) “gotten someone to give in to sex with me when I knew they did not want to.” The language to convey lack of consent (“I knew they did not want to”) is developmentally appropriate for the age of survey participants and is similar to that used in a national survey of adults, which used the phrase “against their will.”29
Sexual assault, as defined by the Department of Justice, refers to “unwanted sexual contact between victim and offender… [which] may or may not involve force and include such things as grabbing or fondling… [as well as] verbal threats.”28 Our fourth item, which queried sexual assault (referred to as forced sexual contact in this study), has been included in the Growing Up With Media survey since wave 1. This item, created specifically for this study, read: “In the last 12 months, how often have you kissed, touched, or done anything sexual with another person when that person did not want you to?” To more closely align this measure with the time frame of the other 3 sexual violence perpetration items (ever), youths who reported forced sexual contact at any of the 5 waves were included as perpetrators.
Details about the perpetration were asked of youths who reported attempted or completed rape. First, youths were asked their age at first perpetration. Then, they were asked to think about the most recent perpetration event. Details about the most recent victim (eg, age, gender) and the event (eg, specific acts perpetrated, consequences) were asked. Youths also rated how responsible they thought they and the victim were for their most recent perpetration.
Harris Poll Online data are comparable to data that have been obtained from random telephone samples of adult populations once appropriate sample weights are applied.30-33 Data were weighted statistically at wave 1 to reflect the population of adults with children aged 10 to 15 years in the United States according to adult age, sex, race/ethnicity, region, education, household income, and child age and biological sex.34 Survey sampling weights also adjusted for adult respondents’ self-selection into the Harris Poll Online and accounted for differential participation over time.30-33
Missing data (ie, refused responses) were imputed using best set regression. To reduce the likelihood of imputing truly nonresponsive answers, participants were required to have valid data for at least 80% of the survey questions asked of all youths. Of the 1062 youths who responded in wave 4 or 5, 1058 met this criterion. Stata version 11 statistical software (StataCorp LP) was used to conduct all statistical analyses. As such, reported percentages are weighted, whereas sample sizes are not.
Nine percent of youths (n = 108) reported some type of sexual violence perpetration in their lifetime: 8% (n = 84) kissed, touched, or made someone else do something sexual when the youth knew the other person did not want to (ie, forced sexual contact); 3% (n = 33) got someone to give into sex when he or she knew the other person did not want to have sex (hereafter referred to as coercive sex); 3% (n = 43) attempted but were not able to force someone to have sex (ie, attempted rape); and 2% (n = 18) forced someone to have sex with him or her (ie, completed rape). Overlap was noted: among perpetrators, 12% reported 2 different behaviors, 11% reported 3, and 9% reported all 4 types.
Demographic and media usage characteristics of youths reporting sexual violence perpetration vs youths not reporting perpetration are shown in Table 2 and Table 3. Males were significantly more likely than females to report coercive sex or attempted rape, with similar but nonsignificant results observed for completed rape. White youths were more likely than nonwhite youths to report perpetrating coercive sex, and Hispanic youths were less likely than non-Hispanic youths to report perpetrating coercive sex. Similar, nonsignificant patterns were noted for youths reporting completed rape. Youths living in low-income households were less likely than youths living in higher-income households to report attempted rape. Similar, nonsignificant patterns were observed for youths reporting coercive sex perpetration.
Consumption of X-rated material significantly differed for perpetrators and nonperpetrators of all types of sexual violence. Differences were almost entirely explained by whether the material was violent in nature (ie, depicted one person physically hurting another person while doing something sexual). Although not always statistically significant, perpetrators also tended to report more frequent exposure to media that depicted sexual situations (ie, kissing, fondling, and sex that young people saw in television, heard in music, saw in games, and saw online) and media that depicted violent situations (ie, physical fighting, shooting, and killing in television, music, games, and online) compared with nonperpetrators.
Youths who reported attempted or completed forced sex were asked follow-up questions about the most recent event. Characteristics for all youths as well as comparisons between male and female perpetrators are provided in Table 4. Comparisons between perpetrators based on age at first perpetration are shown in Table 5. Sixteen years of age was by far the most common age at first perpetration (n = 18 [40%]). Youths whose first sexual violence perpetration was at age 15 years or younger were overwhelmingly (98%) male; similarly high rates (90%) were also noted among perpetrators who began at ages 16 or 17 years. By ages 18 or 19 years, the split of male to female perpetrators was nearly equivalent. More females reported older victims, and more males reported younger victims.
Thinking about the most recent time they tried to force or were able to force someone to have sex, youths reported using coercive tactics, including arguing or pressuring the person (32%) or getting angry or making the person feel guilty (63%), more commonly than using threats (5%) or physical force (8%). Alcohol was used in a notable minority of situations (15%). Tactics were generally similar by perpetrators’ sex and age at first perpetration.
Three in 4 victims (73%) were a romantic partner. Sixty-six percent of perpetrators reported that no one found out about the perpetration. Contact with the justice system was uncommon: 1% of perpetrators reported police contact and 1% an arrest. Consequences were similar for males and females, although youths reporting a younger age at first perpetration were more likely than youths first perpetrating at an older age to get in trouble with their parents. Youths first perpetrating when older were more likely to report that no one found out about the perpetration.
Fifty percent of perpetrators said that the victim was completely responsible; one-third (35%) said that they, the perpetrator, were completely responsible for the incident. Again, differences by perpetrators’ sex or age at first perpetration were not noted.
Vaginal sex was the most common type of sexual act reported among youths who attempted or completed a rape (60%), followed by oral sex (48%). Males were significantly more likely than females to report oral sex and the use of a finger or sex toy. Youths who started perpetrating younger were significantly more likely to report forcing oral sex, whereas youths perpetrating first when older more frequently reported forcing vaginal sex.
This study reports important findings about the extent of sexual violence perpetration among adolescents. A substantial number—almost 1 in 10 adolescents in the national Growing Up With Media study—reports perpetrating sexual violence, including coercive sex, forced sexual contact, attempted rape, and completed rape. Importantly, different developmental trajectories for male and female perpetrators are suggested. Youths who begin perpetrating at younger ages are overwhelmingly male. It is not until age 18 years that female perpetrators emerge at notable rates. For some reason, young female perpetrators are very unusual. Future research should focus on better understanding these sex and age differences to inform the ways in which prevention programs can be tailored and timed.
Perhaps because females tend to start perpetrating later, they are more likely than males to report aggressing on victims older than themselves. Consistent with previous research on youths who had police contact,36 females also appear to be more likely than males to engage in perpetration as part of a team or group: 2 of the 10 female perpetrators in this study engaged in group sexual assault compared with 1 of the 39 male perpetrators. Whether these differences are related to suggested variation in male and female developmental trajectories is an important area of future research.
A notable spike in first-perpetration rates appears at age 16 years. This age also marks the beginning of the late adolescent period of development, a time of rapid and intense growth physically, mentally, and emotionally.37,38 As youths continue to assert their independence, they also face many new challenges and temptations. Additionally, this period of adolescent development is when teens begin to develop more serious romantic relationships and a clear sexual identity.39,40 It is critical that adolescent health care professionals, parents, and others interacting with youths model and educate young people about how to develop and maintain healthy sexual relationships well before this pivotal age.
Building on previous findings in the same data set suggesting links between violent X-rated material and sexual violence,21 our study shows clear associations between the consumption of violent X-rated material and the perpetration of a broad range of sexually violent behaviors. Similar, although not always statistically significant, patterns are noted for violent media of a nonsexual nature as well as media depicting sexual situations. Although the data do not denote causality, it seems appropriate to suggest that frequent consumption of sexual and violent material and especially sexually violent material should be a marker for concern for adolescent health care professionals.
Many incidents occurred within the context of romantic relationships, yet 1 in 4 victims was not a boyfriend or girlfriend. All perpetrators in this study had some form of a relationship with their victims, however. This is consistent with the greater sexual violence literature, which suggests that adolescent victims are overwhelmingly more likely to be victimized by someone they know rather than a stranger.41,42 Prevention programs need to provide youths with safety skills that can be used in both nondating and dating scenarios.
Few perpetrators experience consequences: only 2 reported being arrested. More effort needs to be made to enhance methods of detecting and investigating sexual violence cases.43 This finding also highlights the limitations of juvenile justice data. Like sexual victimization,44 the overwhelming majority of sexual violence perpetration does not come to the attention of law enforcement. Arrest and conviction data characterize youths coming into contact with the justice system for sexual violence offenses and are helpful in designing secondary prevention programs among perpetrators, but they do not provide a full picture of sexual aggression among youths. Population-based studies are needed to inform the development of population-based prevention efforts occurring before arrest.
It is not uncommon to believe that a man cannot be raped by a woman.45,46 Gender stereotypes can make it difficult to imagine a dominant woman coercing or forcing an unwilling man to have sex.47 Accordingly, male victims of female perpetrators are judged more harshly than male victims of male perpetrators.48 Moreover, the same behaviors perceived to be sexually aggressive when committed by a male can be perceived as romantic or promiscuous when committed by a female.47,49 Nonetheless, physiological data suggest that men can be raped; an erection does not necessarily mean sexual arousal50 and can be reflexogenic.51 Adolescent health care professionals need to assess the potential for their own gender biases in this area so that they can be more effective in identifying and treating female perpetrators and male victims when they present.
Few differences by race, ethnicity, and household income were noted; when they were, disadvantaged youths were less likely to report perpetration. Adolescent health care professionals also need to assess and perhaps challenge our assumptions about sexual perpetration as an ill conscripted solely to underprivileged populations.
Some may argue that the definitions of rape and sexual assault in our investigation are too broad. Indeed, this may be why the perpetration rate among females is higher than might be posited. Rape includes acts beyond those in which the victim is physically overpowered, however.28 Restrictive definitions have potentially led to undercounting of sexual assault experiences. For example, in the National Violence Against Women Survey, respondents were asked whether anyone had ever made them engage in a sexual activity “by using force or threat of force.”52 Psychological coercion was not clearly specified even though there are multiple coercive strategies other than physical force that can be used in a rape.53,54 To ensure that comprehensive rates of sexual assault and rape are identified as well as to begin building the research base on female perpetrators, research needs to include a fuller spectrum of rape scenarios.
Perpetrator characteristics are examined by age at first perpetration, whereas the assault characteristics relate to their most recent perpetration, which may or may not be their first perpetration. Thus, it is possible that the perpetration data reflect behavior that one engaged in at age 18 years even though the youth first perpetrated at age 16 years, for example. Data should be interpreted as reflective of patterns for youths who start perpetrating earlier vs later rather than patterns of younger vs older youth per se.
Additionally, details about each type of sexual violence were not available in many cases. It is possible, for example, that age at first perpetration for completed rape is older than age at first attempted rape. Also, given the sensitivity of the subject, observed rates may be underestimates of the true prevalence of sexually violent behavior. Nonetheless, rates are much higher than the lifetime rate of 0.15% yielded in a national study of adults that was conducted face to face.29 Finally, because this was a community-based sample, the number of youths reporting sexual violence perpetration was small and yielded limited statistical power.
One in 7 perpetrators said they were not at all responsible for what happened. Accordingly, more than 4 in 5 perpetrators said the victim was at least somewhat responsible for what happened. Research about victim blaming has been studied among young adult and adult victims of rape, typically by examining attributions of blame toward the perpetrator and victim in hypothetical scenarios.55-57 Generally, individuals have a tendency to perceive victims in negative terms, in addition to or even instead of perpetrators.56 More efforts need to be made toward education that has a specific emphasis on taking responsibility for one’s actions as well as eliminating blame for the victims. Further research into factors related to taking responsibility, such as motivations behind the sexual violence and feelings of remorse or regret, could be crucial to effective primary prevention for youths at risk for perpetration as well as secondary and tertiary prevention for those who have already committed such acts. Adolescent health care professionals, including those providing counseling for birth control and sexually transmitted diseases, are ideally situated to educate youths about sexual violence. Efforts to equip professionals with referral information and talking points would likely increase the uptake of this recommendation in the clinical setting.
Encouraging progress has been made in the field of bystander intervention of sexual assault, which emphasizes the range of opportunities where friends can intervene.58 Such efforts shift the focus of prevention to peers and the community from efforts reducing risk among potential perpetrators and victims. Although most of the research has been focused on college campuses, similar efforts at the high school level are emerging.59,60 Given that most adolescent perpetrators meet their victim at school, this may be an especially important venue through which to provide prevention programming.
As one of the first reports of national rates of sexual violence in adolescence, findings should be interpreted cautiously. Results need to be replicated. Certainly, however, links between perpetration and violent sexual media are apparent, suggesting a need to monitor adolescents’ consumption of this material, particularly given today’s media saturation among the adolescent population. Because victim blaming appears to be common while perpetrators experiencing consequences is not, there is urgent need for high school (and middle school) programs aimed at supporting bystander intervention. Policies that enhance the likelihood that perpetrators are identified also need to be considered. Findings suggest important differences between males and females in the etiology of adolescent sexual violence; it appears to emerge earlier for males than females, perhaps suggesting different developmental trajectories. More work is needed in this area. Based on the individual2-4 and social1 costs associated with sexual violence, our findings highlight the importance of identifying perpetrators early in their adolescence before they become repeat offenders.
Corresponding Author: Michele L. Ybarra, MPH, PhD, Center for Innovative Public Health Research, 555 El Camino Real, Ste A347, San Clemente, CA 92672 (firstname.lastname@example.org).
Accepted for Publication: April 4, 2013.
Published Online: October 7, 2013. doi:10.1001/jamapediatrics.2013.2629.
Author Contributions: Dr Ybarra had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Ybarra.
Acquisition of data: Ybarra.
Analysis and interpretation of data: Ybarra, Mitchell.
Drafting of the manuscript: Ybarra.
Critical revision of the manuscript for important intellectual content: Ybarra, Mitchell.
Statistical analysis: Ybarra, Mitchell.
Obtained funding: Ybarra.
Conflict of Interest Disclosures: None reported.
Funding/Support: The survey was supported by grant R01 CE001543 from the Centers for Disease Control and Prevention.
Role of the Sponsor: The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Disclaimer: The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Additional Contributions: We thank the entire Growing Up With Media Study team from Internet Solutions for Kids, Harris Interactive, Johns Hopkins Bloomberg School of Public Health, and the Centers for Disease Control and Prevention, who contributed to different parts of the planning and implementation of the study. We also thank the families for their time and willingness to participate in this study.
Correction: This article was corrected online October 16, 2013, for an error in Table 2 and on November 15, 2013, for an error in Funding/Support.
Create a personal account or sign in to: