Abstract
The present study sought to examine features of sexual abuse cases among a U.S. nationally representative sample of 13,052 children and adolescents, ages 0–17 years. The National Survey of Children’s Exposure to Violence was collected in three different years (2008, 2011, and 2014) via telephone interviews. Information about sexual abuse and assault was obtained from youth themselves (ages 10–17) or caregivers (for children ages 0–9) using the Juvenile Victimization Questionnaire. Results indicate most offenses are at the hands of other juveniles (76.7% for males and 70.1% for females), primarily acquaintances, and occurring more frequently for adolescents aged 14–17. Whereas girls are mostly abused by males (88.4%), boys are abused by both males (45.6%) and females (54.4%). In 15% of cases, penetration is part of the abuse. Victims report being very afraid in 37.5% of episodes but not at all afraid in 19.8%. Among 10- to 17-year-olds, 66.3% of episodes are not reported to parents or any adult. Police reports occur for 19.1% of all cases. The results in the present study indicate that children and youth are exposed to sexual abuse and assault in varied ways, which require moving beyond conventional stereotypes of the problem.
Child sexual abuse (CSA) is a worldwide health problem with long-term outcomes on survivors’ mental, psychological, physical, and sexual health. While researchers have focused extensively on estimating the prevalence of CSA (Barth, Bermetz, Heim, Trelle, & Tonia, 2013; Pereda, Guilera, Forns, & Gómez-Benito, 2009; Singh, Parsekar, & Nair, 2014; Stoltenborgh, Van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011; Vogeltanz et al., 1999) and assessing the long- and short-term consequences of the abuse on the survivor (Beitchman et al., 2002; Pérez-Fuentes et al., 2013), less attention has been given to characterizing the diversity of the problem. One element to the diversity is the mixture of adult-on-child episodes that comfortably fit into the term “sexual abuse” and child-on-child episodes that may not. To stress that CSA in the present study includes offenses conducted by juvenile/peer offenders, the word “assault” is added to the term CSA (CSAA).
The epidemiology of childhood sexual abuse and assault (CSAA) is based on agency and clinical samples (e.g., Agyapong et al., 2017), child protection services (e.g., Bailey, Powell, & Brubacher, 2017), the police/medical forensic examinations (e.g., Aydin et al., 2015), and schools (e.g., Barth et al., 2013). These cases show only features that come to professional attention. Information about cases unknown to authorities comes from community samples, which are comprised overwhelmingly of adults and older adolescents recalling the abuse retrospectively (Barth et al., 2013; Briere & Elliott, 2003; Pereda et al., 2009; Pérez-Fuentes et al., 2013). The information from these retrospective self-reports can be biased as it entails recalling the abuse over a long period of time (Langeland et al., 2015). An additional needed perspective is from representative community samples concerning the full spectrum of current children, asking about the recent sexual abuse that has not necessarily been revealed to authorities with a sample size large enough to accurately analyze episode characteristics.
The National Survey of Children’s Exposure to Violence (NatSCEV) offers such an opportunity with several unique features: It provides information over the full course of childhood (0–17). It is based on a representative sample of the general population using parent and youth self-reports and is quite different from the picture obtained from reports to the police or child protection services. It also has advantages over an adult retrospective study because it does not require a respondent to recall events from the distant past.
Existing studies with representative national samples (Pérez-Fuentes et al., 2013; Vogeltanz et al., 1999) and systematic reviews/meta-analyses (Barth et al., 2013; Putnam, 2003; Stoltenborgh et al., 2011) provide information about the characteristics of abused children (sex, age at onset, race, socioeconomic status, region, and education level) yet do not give a full and detailed picture about the characteristics of the abuse itself. National samples that do provide some information about the abuse can still involve some level of bias, as these studies are conducted among adult survivors, which are not always able to recall different aspects of the abuse (e.g., Pérez-Fuentes et al., 2013; Vogeltanz et al., 1999). The NatSCEV provides a unique opportunity to receive a full and detailed profile on the features of the abuse as it reported by children. Although the NatSCEV data have shown the majority of victimized children suffer from various types of abuses (e.g., Finkelhor, Ormrod, Turner, & Hamby, 2005), a detailed profile on the features of each type of victimization can help understand a specific crime.
The present study sought to address the features of the abuse from a gender perspective, including the relationship and the identity of the perpetrator, age at onset, the location in which the abuse took place, did the abuse include penetration, the level of fear during the offense, and whether it was followed by missing school, injuries, and/or medical treatment.
Sex of the Survivor
Although public awareness regarding CSA is increasing, research on the sexual abuse of boys is still scarce. Previous surveys tend to focus on women survivors (e.g., Campbell, Dworkin, & Cabral, 2009; Coles, Lee, Taft, Mazza, & Loxton, 2015; Fanslow, Robinson, Crengle, & Perese, 2007; Vogeltanz et al., 1999) or not to include enough male survivors to richly characterize the diversity and differences between male and female survivors. Male survivors of CSA differ in many ways from female survivors. For example, male survivors often delay disclosure of the abuse for years or even decades (Easton, Saltzman, & Willis, 2014; O’Leary & Barber, 2008). The impact of the abuse can also differ between men and women survivors. For example, the abuse has less of an effect on males’ sexual function than it has for female survivors (Dunlop et al., 2015; Najman, Dunne, Purdie, Boyle, & Coxeter, 2005). However, while the gender differences in long-term consequences of the abuse are investigated in many studies (e.g., Abajobir, Kisely, Maravilla, Williams, & Najman, 2017; Daigneault, Vézina-Gagnon, Bourgeois, Esposito, & Hébert, 2017; Krahé & Berger, 2017), little is known about the features of the abuse itself and whether boys have a different experience of abuse than girls. For example, are there any differences in same-sex perpetration, penetration, and level of fear. In the present study, all variables and characteristics of the abuse will be examined by sex, in an attempt to fill the gap on abused males.
The Relationship to the Perpetrator
The relationship to the perpetrator is an important factor as it has a many impact on the feelings of the child toward the abuse (e.g., fear and guilt), the disclosure process, the intervention needed, and even the prognosis given by professionals (Murray, Nguyen, & Cohen, 2014). Studies from the United States indicate that the majority (41–68%) of children are sexually abused by an immediate or extended family member (e.g., Briere & Elliott, 2003; Robin, Chester, Rasmussen, Jaranson, & Goldman, 1997; Vogeltanz et al., 1999) or by someone known to the child (Finkelhor, Hotaling, & Lewis, 1990; Vogeltanz et al., 1999; Young, Grey, & Boyd, 2009). No gender differences were found regarding incest cases (abuse within the immediate or extended family; Briere & Elliott, 2003). Similarly, studies conducted outside of the United States report that in most cases of CSA, the perpetrator is an acquaintance (Aydin et al., 2015; Edgardh, 2000; Mohler-Kuo et al., 2014) or a family member (Fanslow et al., 2007) rather than a total stranger. However, studies are inconsistent in the categorization of perpetrators, which can, later on, lead to inconsistent findings. Finally, growing understanding of contemporary CSA acknowledges peer assaults. Findings from a survey conducted in the UK found that sexual victimization by peers is the most prevalent from other types of offenders (Radford, Corral, Bradley, & Fisher, 2013), with males experiencing more victimization by peers than females. High prevalence of peer offenders is also reported later by Finkelhor, Shattuck, Turner, and Hamby (2014) and Kloppen, Haugland, Svedin, Mæhle, and Breivik (2016).
The relationship to the perpetrator can predict survivors’ reaction after the abuse and their mental health. Ullman (2007) found that more negative reactions such as disbelief were observed for those victimized by relatives compared with acquaintance and stranger survivors, especially for those disclosing in childhood. In addition, survivors of relatives had more post-traumatic stress disorder (PTSD) symptoms if they delayed disclosure, received more negative reactions in childhood, and engaged in self-blame at the time of the abuse. Survivors reporting CSA by a family member were at 2.6 odds of a current alcohol use disorder, 2 times higher odds of a substance use disorder, and 2.7 times higher odds of reporting an STI[Please replace “STI” with its expansion, if appropriate.] in the past year (Boroughs et al., 2015). Finally, information about the identity of the perpetrator can help direct prevention measures and education programs more effectively.
The Sex of the Perpetrator
Studies in the United States found that males are more likely than females to report at least one sexual abuse by a female perpetrator (Briere & Elliott, 2003; Finkelhor, Hotaling et al., 1990). Also, males are more likely to be abused by a perpetrator from the same sex (Briere & Elliott, 2003). In a U.S. national study of adult men and women, only 1% of women reported being abused by another female compared to 83% of men who reported being abused by another male (Finkelhor et al., 1990). The sex of the perpetrator is important as it may affect the well-being of male survivors and the disclosure process of boys, as they report fearing being labeled as gay or a threat to their masculinity (Easton et al., 2014; Kia-Keating, Grossman, Sorsoli, & Epstein, 2005). The NatSCEV provides information on the relationship and sex of the perpetrator, enabling the examination of female perpetration and same-sex perpetration, both of which are less known and studied.
Age at Onset
Studies from the United States vary widely in estimating the age of which a child first experienced CSA. Different distributions of age groups, various definitions of CSA, and nonrepresented or small samples, all make it difficult to draw a solid conclusion about the age at first onset. CSA at the period of infancy and toddler periods (under the age of 5 or 6) is estimated in the United States between 10.7% and 14% (Finkelhor, Hotaling, Lewis, & Smith, 1989; Vogeltanz et al., 1999). Robin, Chester, Rasmussen, Jaranson, and Goldman (1997) indicate almost one fifth (17%) of children are first abused when toddlers (ages 3–5), 47% between the ages of 6 and 9, 28% between the ages of 10 and 12, and 8% between the ages of 13 and 15. In a 10-year research review, Putnam (2003) provided a distribution of the age at first onset, indicating that at the time of the abuse 10% of victims are between ages 0 and 3 years, 28.4% are between ages 4 and 7 years, 25.5% are between ages 8 and 11 years, and 36.9% are 12 years and older. More recent studies point out that the risk of experiencing CSA increases with age, with adolescence being a period of increased risk of CSA for both genders (Finkelhor, Shattuck, Turner, & Hamby, 2014; Kloppen, Haugland, Svedin, Mæhle, & Breivik, 2016; Radford et al., 2013) probably due to a significant proportion of assaults that occur within the peer context. The NatSCEV is limited in a way in discussing ages and lifetime rates, as most of the children in this sample have not completed childhood yet, and particularly the high-risk stage of later adolescence. Therefore, the present study focuses on the age at earliest abuse.
Location
Little is known about the location where sexual abuse occurs. In a study conducted among American adolescents on peer sexual victimization, the majority reported the abuse occurred at school (54.3%) or somewhere else (38.3%), while only a small number of assaults were at the victim’s house (7.4%; Young et al., 2009). The location of where the assault takes place can vary by the age and identity of the perpetrator but also by the type of assault. For example, while CSA accompanied with physical contact but not penetration is likely to occur in public places, CSA including penetration is more likely to occur in a private domain (e.g., perpetrator’s or victim’s house; Mohler-Kuo et al., 2014). The location of where the abuse takes place is an important feature to examine, as it can assist in the decision of where prevention efforts should be directed.
Penetration
Penetrative sexual abuse usually refers to vaginal intercourse, oral sex (mouth/genital), and anal intercourse. Studies conducted in the United States vary widely on the prevalence of penetration, estimating penetration occurs between 9.5% and 80% of abused males and 14.6% and 68.5% of abused females (Finkelhor et al., 1990; Robin et al., 1997; Roesler & McKenzie, 1994). In the United States, penetrative sexual abuse is more extensive for female youth, with approximately 10 million U.S. females (8.4%) experiencing rape before the age of 18 years and 1.9 million U.S. males (1.6%) that are made to penetrate someone during youth (Merrick, Basile, Zhang, Smith, & Kresnow, 2018). However, according to a worldwide systematic review (Barth et al., 2013), and various studies from outside the United States (Aydin et al., 2015; Kloppen et al., 2016; Najman et al., 2005), the majority of CSA cases do not include penetration. Priebe and Svedin (2009) found that the prevalence of penetrative CSA declined for both genders between 2004 and 2009 in Sweden, yet Kloppen et al. (2016) show this is not the case in Norway, where penetrative CSA increased between 2004 and 2007 (Mossige & Abrahamsen, 2007).
In a U.S. sample, no gender differences were found regarding oral, anal, or vaginal penetration by a penis or other objects (Briere & Elliott, 2003; Roesler & McKenzie, 1994). However, in studies conducted in Sweden and New Zealand, abuse that includes vaginal intercourse was found more likely among girls, and anal intercourse was found significantly more common for boys (Edgardh, 2000; Najman et al., 2005). Nonetheless, studies show an abuse that involves penetration is associated with more significant long-term harm (Beitchman et al., 2002), psychopathology (Aydin et al., 2015), sexual dysfunctions (Najman et al., 2005), PTSD, sexual risk behaviors (Boroughs et al., 2015), and revictimization in adulthood (Merrick et al., 2018). As penetration may have a significant effect on the victim’s experience and well-being, further research is required to determine the prevalence and gender differences in penetrative CSA.
Level of Fear
Survivors with intense fear have increased odds for PTSD during adulthood (Boroughs et al., 2015). However, little is known of the level of fear during the abuse, possibly due to the challenge of recalling the level of fear. Fear is often discussed in studies as an outcome of the abuse (e.g., fears and phobias; Pérez-Fuentes et al., 2013) or as a barrier for disclosure (e.g., fear of not being believed, fear of being judged or blamed; Alaggia, Collin-Vézina, & Lateef, 2017; Easton, Coohey, O’leary, Zhang, & Hua, 2011).
Injuries, Harm, and Medical Treatment
Injuries, harm (e.g., missing school), and medical treatment have an important role in understanding the abuse and its complexity. Previous studies report survivors who suffered a physical injury during the abuse have increased odds for PTSD during adulthood (Boroughs et al., 2015; Campbell et al., 2009). It is possible an injury increases the perceived life threat, which is predictive of PTSD (Ozer, Best, Lipsey, & Weiss, 2003).
The Present Study
Various studies show individual-level factors (of the survivors, the perpetrator, and the abuse) are associated with the long-term effects of CSA. Yet many of the national representative studies on CSA do not include all factors. In addition, information about the individual-level factors can be seriously biased due to the need to recall the abuse among adult survivors. The present study offers a unique perspective from a national sample of children aged 0–18, which provides a detailed picture of their abuse. Children are asked about the sexual abuse they experienced and about the relationship to the perpetrator, the sex of the perpetrator (examining same-sex abuse), the age at onset, the location in which the abuse took place, disclosure (and to whom), the level of fear during the abuse, and whether the abuse included penetration, physical injury, and harm, and followed by receiving medical treatment. The information provided in this study can be extremely useful for preparing policy makers, physicians, and health-care and child-serving professionals to help children, patients, and families concerning possible needs for protection and counseling regarding CSAA.
Method
Participants and Procedure
The data for this analysis come from the aggregation of three separate cross-sectional representative samples of U.S. children: NatSCEV, carried out in 2008, 2011, and 2014. At 2008, 4,549 children participated in the survey. Another 4,503 children participated in the survey conducted in 2011, and 4,000 children participated in the survey of 2014. Response rates were relatively high in all three waves yet varied by year. In 2008, the cooperation rate for the random digit dialing (RDD) was 71%, and the response rate (in households in which no one could be contacted) was 54%; the rates for the oversample were 63% and 43%, respectively (Finkelhor, Turner, Ormrod, & Hamby, 2009). In 2011, the cooperation rate was 60%, and the response rate was 40.4% (Finkelhor, Turner, Shattuck, & Hamby, 2013). In 2014, response rates differed across the four sampling frames: (1) an address-based sample of households from which cell phone and residential numbers could be dialed, (2) a prescreened sample of households with children from recent national RDD surveys, (3) a listed landline sample, and (4) cell phone numbers drawn from a targeted RDD sample frame. Response rates in the third wave in 2014 were relatively low compared to the early two samples, ranging from 9.7 in the phone RDD to 67% who replied to the study mailing (Finkelhor, Turner, Shattuck, & Hamby, 2015).
All three were telephone surveys conducted about the abuse, crime, and victimization experiences of children and youth aged 1 month to 17 years. Youth aged 10–17, deemed capable of providing information in a phone survey, were interviewed directly about their experiences, while information about the experiences of children aged 0–9 was obtained through interviews with a caregiver. Respondents were promised complete confidentiality and were paid US$20 for their participation. Respondents who disclosed a situation of serious threat or ongoing victimization were recontacted by a clinical member of the research team, trained in telephone crisis counseling, whose responsibility was to stay in contact with the respondent until the situation was appropriately addressed locally. Human subject participation was reviewed and approved by the University of New Hampshire’s Institutional Review Board.
Measurement
All three cross-sectional samples were combined into a single sample totaling 13,052 children and youth. In each survey, information on children’s exposure to violence was collected using the Juvenile Victimization Questionnaire (JVQ; Finkelhor, Hamby, Ormrod, & Turner, 2005; Finkelhor, Ormrod, et al., 2005; Hamby, Finkelhor, Ormrod, & Turner, 2004). Additional information on sampling methods and procedures are available elsewhere (Finkelhor, Turner, et al., 2009, Finkelhor et al., 2013, 2015).
For this analysis, we used four sexual abuse and assault questions common to all three surveys. If a child experienced any of the types of victimizations (as reported by the child directly or via proxy by the caregiver), follow-up questions were asked to gather information about their relationship with the perpetrator, injury, and penetration (respondents were asked the follow-up questions for each JVQ item endorsed). The 4 items were as follows: (S1) At any time in your (your child’s) life, did a grown-up you know touch your private parts when they should not have or make you touch their private parts? Or did a grown-up you know force you to have sex? (S2) At any time in your life, did a grown-up you did not know touch your private parts when they should not have, make you touch their private parts, or force you to have sex? (S3) Now think about other kids, like from school, a boyfriend or girlfriend, or even a brother or sister. At any time in your life, did another child or teen make you do sexual things? (S4) At any time in your life, did anyone try to force you to have sex that is sexual intercourse of any kind, even if it did not happen?
The follow-up questions gathered information on the features of the abuse (respondents were requested to refer to the most recent if there were multiple episodes) including (1) the age of the child at the first incident; (2) the location where the abuse occurred (inside/near house/school, day care or after-school programs/somewhere else/not sure); (3) the relationship with the perpetrator (open-ended question, which was then coded into one of the following: family/acquaintance/stranger and adult or peer); (4) whether penetration was part of the victimization (yes/no); (5) whether the incident was reported to the police (yes/no); (6) reported to the teacher (yes/no); (7) reported to a parent (yes/no); (8) the level of fear (not afraid/little afraid/very afraid); and (9) whether the child missed school as a result of the incident (yes/no). For these questions, CSAA was defined as encompassing forced and unwanted contact sexual acts with anyone, including inappropriate sexual acts with adults. The study report lifetime experiences contact experiences and perpetrators of any age. A “not sure” option was available to all these questions.
Data Analysis
The analysis and results for the current study are reported for the 506 exposed to CSAA from the entire sample of 13,052. To ensure that the estimates derived from the combined frames were representative of the target population of children in the United States aged 17 and under, a four-step process, detailed elsewhere (Finkelhor et al., 2013), was used to construct the analysis weights. These weights were combined into a single variable in the pooled data set, and this variable was used to conduct the pooled analyses. Rates were reported with 95% confidence bounds. To examine differences in the rates by various categorical measures (e.g., the occurrence of sexual abuse, sex of victim, sex of abuser), we conducted a series of weighted χ2 tests for independence of measures, with Monte-Carlo significance tests, followed by relative risk scores. To examine differences in age of the earliest sexual abuse by victims’ sex and/or age of abuser, we conducted a series of weighted independent samples t tests.
Results
Rates of Sexual Abuse and Assault by Sociodemographic Characteristics
Overall rates of lifetime sexual abuse and assault in the pooled sample of 0- to 17-year-olds were 5.6% for girls and 1.9% for boys (Table 1). However, it should be noted that most of the children in this sample have not yet completed childhood. In fact, the mean age of children exposed to CSAA at the time of the survey (N = 506) was 8.63 (SD = 5.24). The analyses indicated that sexually abused children were more likely female, Black (non-Hispanic), of low socioeconomic status (SES), and residing in a large city. They more likely have either a single parent, single parent and a stepparent, or other adults as their legal guardians (rather than living with both biological parents). Table 2 indicates the age of the earliest victimization is highest for 14- to 17-year-olds. The mean age is lower for male victims and adult-perpetrated abuse.
Rates of Lifetime Sexual Abuse and Assault Sociodemographic Characteristics.
Note. *p < .05. **p < .01. ***p < .001. Relative risk = effect size; SES = socioeconomic status. Values with different superscripts are significantly different at p < .05 (a is significantly different from b and c; b is significantly different from a and c; c is significantly different from a and b). Values in bold mark the reference group for the relative risk scores.
Age at Earliest Sexual Abuse and Assault.
Note. N = 506. *p < .05, **p < .01, ***p < .001. Values in bold are significantly different at p < .05.
Features of Sexual Abuse and Assault
Incident-level features of sexual abuse and assault are presented in Table 3 as percentages of the total sample of abuse cases (n = 506). The likelihood of sexual abuse increases by age (i.e., 0–1 < 2–9 < 10–13 < 14–17; see Figure 1 for differences between males and females). Sexual abuse and assault are the least likely to occur in day cares (or in after-school care) followed by schools and most likely to occur inside or near children’s houses and/or somewhere else. In 15% of cases, penetration was part of the abuse. In addition, although more than one third of victims felt very afraid during the episode, approximately one in five of the victims reported that they were not afraid at all. However, the level of fear was significantly higher when the abuse included a penetration attempt and involved an adult (rather than a peer) or a male (rather than a female) perpetrator (Table 4). Finally, 9.7% of victims missed school because of the abuse, 9.7% were injured during the abuse, and 11.5% received medical treatment.
Features of the Abuse.
Note. N = 506. *p < .05, **p < .01, ***p < .001. Due to an option to skip a question or report don’t know/not sure, answers that do not sum up to 100%.
* Reporting to parents was tested only among children aged 10–17 (N = 312).
Rates of sexual abuse by victims’ age and gender.
Level of Fear by Penetration, Adult Versus Peer and Female Versus Male Perpetrator.
Note. N = 506. *p < .05. **p < .01. ***p < .001. Values in bold are significantly different at p < .05. Due to an option to skip a question or report don’t know/not sure, answers that do not sum up to 100%.
Among the 10- to 17-year-olds, 66.3% did not disclose the abuse to a parent or any other adult. (The survey only knew about abuse for 0- to 9-year-olds if it was reported to the survey by a parent.) For the full sample, 19.1% was reported to the police and 21.8% to a teacher.
The relationship between perpetrators and victims is presented in Table 5. The majority of offenses were at the hands of other juveniles (76.7% for males and 70.1% for females). Females were more likely to be abused by boyfriends, male juvenile acquaintance, and male adult acquaintance. Males were more likely to be abused by their girlfriends or female juvenile acquaintances. The sex of the perpetrators and victims are presented in Table 6. Females were mostly abused by males (88.5%), males were equally abused by males (45.6%) and females (54.4%).
Identity of the Abuser by Relationship.
Note. BF = boyfriend; GF = girlfriend.
Same-Sex Perpetration.
Note. BF = boyfriend; GF = girlfriend.
Discussion
The NatSCEV approach to epidemiology offers some unique features: It provides information over the full course of childhood (0–17) and includes cases of sexual abuse and assault that were not reported to authorities or even, for the 10- to 17-year-olds, parents. Therefore, the picture of epidemiology presented in this study based on a representative sample of the general population is quite different from the picture obtained from reports to police or child protection services. It also has advantages over an adult retrospective study because it does not require a respondent to recall events from the distant past, and it includes episodes (reported in this case by parents of younger children) that might not even be available in an adult victim’s retrospective memory. The 506 cases available for analysis constitute one of the largest representative samples of cases in the literature.
The demographic analysis found sexually abused and assaulted children were more likely to be female, Black (non-Hispanic), of low SES, and residing in a large city. These features have been highlighted in other studies examining CSA and maltreatment (Mohler-Kuo et al., 2014; Pérez-Fuentes et al., 2013; Singh et al., 2014). Though the overall rates, 5.6% for girls and 1.9% for boys, 3.9% in total, may appear low in the context of other general population surveys, it needs to be remembered that most of the children in this sample have not completed childhood and particularly, the high-risk stage of later adolescence. A previous report based on just the 17-year-olds from these three samples combined (Finkelhor et al., 2014) showed lifetime rates of 26.6% for girls and 5.1% for boys, more typical of adult retrospective samples of those who had completed childhood.
One important conclusion from the analyses in this study is that the majority of offenses are at the hands of other juveniles (76.7% for males and 70.1% for females). This is far higher than the juvenile proportion that appears in most police or child protective system samples (Finkelhor, Ormrod, & Chaffin, 2009). Since many people assume that the term “child sexual abuse” implies an adult offender, it is essential to stress how much of the child sexual victimization problem stems from other juveniles, particularly acquaintances. This is the basis for our recommendation to refer to the problem as “child sexual abuse and assault” (CSAA), to refute the wrongful perception that children are only sexually assaulted by adults, and to include offenses conducted by juvenile offenders and particularly peer offenders. In the present study, we found peers represent a markedly larger proportion of the perpetrators. This finding corresponds with previous studies conducted in the United States (Young et al., 2009), the UK (Radford et al., 2013), and in Nordic countries (Kloppen et al., 2016) and has strong implications for prevention and treatment. The school context can be an ideal location to initiate prevention efforts of peer-on-peer sexual victimization. Schools should employ sexual education programs, which also address information about sexual offences between peers and stress the importance of disclose (Allnock & Atkinson, 2019). Prevention programs can also educate youth on the characteristics and warning signs of dating violence and describe positive and normal relationship behaviors. This is especially important as peer groups set powerful norms that influence the ability and willingness of children to report sexual offences (Allnock & Atkinson, 2019).
A second important finding is the differences in the victimization of boys and girls. Among boys, there was a mix of both male and female perpetrators, with female offenders comprising 54.4% of their perpetrators overall and 61.0% of the acquaintance perpetrators. Almost half of the boys reported being abused by males, and a little more than 10% of girls reported being abused by a female. The finding of female offending against boys is consistent with other surveys conducted in the United States (e.g., Dube et al., 2005; Ybarra & Mitchell, 2013) and U.S. Federal data (Stemple, Flores, & Meyer, 2017) indicating female perpetration is not rare. This finding underscores the complexity of boys’ experience facing both the stigma of homosexual behavior in episodes with male perpetrators and the lack of recognition that boys can be victimized at the hands of females. Dominant masculinity ideals of strength and control and the stigma of homosexuality can delay the disclosure of the abuse for years or even decades (Easton et al., 2014) and stress the need for a suited therapeutic approach and recovery journey (Forde & Duvvury, 2017). In the present study, we found boys are reportedly having a high rate of being sexually victimized by an acquaintance and especially another juvenile (rather than a boyfriend/girlfriend or an acquaintance adult) and experience higher rates of same-sex perpetration than girls. Same-sex peer sexual victimization among boys can possibly be explained by homophobic teasing among boys during early adolescence, which turns into sexual abuse or harassment in more advanced stages of adolescence (Espelage, Basile, De La Rue, & Hamburger, 2015). While boys’ victimization is highest in early childhood, girls’ victimization is at similarly high rates in early adolescence and then increases significantly during later adolescence. It is possible these patterns are explained mostly by sexual violence perpetrated in the teen dating context (in the present study, girls were 2 times more likely to report abuse by their romantic partner: boyfriend or girlfriend).
A third important finding from the study is that only a minority (37.5%) of CSAA induced high levels of fear. Low levels of fear are more prevalent in peer victimization, female perpetrators, and when penetration occurs. Low levels of fear can be reexplained by the low amount of CSA cases in the study that includes physical injury, harm and required medication treatment. This finding is important as studies show many of the cases of CSA are by the hand of peer perpetrators (Finkelhor et al., 2014; Kloppen et al., 2016; Radford et al., 2013). Adding to this, previous studies have found that the complexity of CSA and intense fear during the abuse are associated with increased odds for mental health problems (Boroughs et al., 2015). In addition, the present study found female survivors felt significantly more fear than male survivors during the abuse. Gender differences in the level of fear during the abuse can also reflect the wish of boys to try and live up to the normative gender roles and masculinity ideals and avoid stigmatization (Easton et al., 2014; O’Leary, Easton, & Gould, 2015).
The findings about the level of fear underline that sexual victimization is in many cases an offense that often does not occur through physical force but rather by the usage of pressure, surprise, and manipulation. This can sometimes contribute to guilt, self-blame, and failure to report when victims fail to see their encounters as meeting an imagined stereotype (Briere & Elliott, 1994). These findings also have implications for treatment, as sexually abused children are primarily conceptualized as suffering from a form of PTSD, whose etiology is thought to result from a sense of overwhelming fear and threat to bodily integrity (American Psychiatric Association, 2000). In fact, the literature suggests that a considerable portion of the harm of sexual abuse stems from shame, stigma, and inappropriate sexual socialization (Finkelhor & Browne, 1985; Kennedy & Prock, 2018; Mckenzie & Botts, 2018). This should discourage automatic inferences that episodes with less fear, such as peer perpetrator or nonpenetrative episodes, are less harmful.
Finally, the findings point to the continuing barriers for reporting episodes of victimization. Among the 10–17 years age-group, only 31.0% of the episodes were reported to parents, and only 33.7% were reported to other adults. Approximately one in five (19.1%) abuse cases were reported to the police. The literature is full of findings of the burdens children face in the wake of disclosures of sexual abuse and assault (Alaggia et al., 2017; Easton et al., 2014; Morrison, Bruce, & Wilson, 2018; Sivagurunathan, Orchard, MacDermid, & Evans, 2019). In a previous analysis, we found evidence that disclosure to authorities has increased over time (Finkelhor, Ormrod, Turner, & Hamby, 2011). However, there is much room to improve, suggesting the need for more making child protection and police involvement more child-friendly and providing education and anticipatory guidance from educators and health professionals, encouraging children to feel confident that they will receive a supportive response if they disclose.
Although this study has many strengths, including its nationally representative sample, it is important to be aware of its limitations. There were families in the sample frame who could not be reached at home or who refused cooperation for themselves or their children, and these may be families in which children have discrepant levels of exposure to violence or sexual abuse incidents compared with the cooperating families. Also, while it is highly unlikely that the same participant was recruited in more than one wave, given the number of children in the United States and the size of the waves, it is still a possibility. A few repeat participants cannot affect the results that are reported in this article but they can violate statistical assumptions of independence of observations. Approximately half of the interviews were with children themselves (10–17), while the other half was with a caregiver (usually a parent) on behalf of the child (0–9). Parents who reported on behalf of their children might not have disclosed all incidents for various reasons (e.g., they were not aware of an abuse that happened, they wanted to be portrayed as good and protective parents, or the abuse was by family members [or even themselves] whom they are trying to protect or feared). This could affect the relatively lower rates for abuse among younger children. Children may also fail to disclose all incidents for various reasons such as the fear of being blamed for the abuse, shame, guilt, or fear of the perpetrator. These challenges were addressed by telling the youth to make sure they are completing the interview in a private and safe place, that if they no longer feel safe they can stop the interview and resume later, and that most questions were yes/no which limits how much someone listening in can determine what’s being asked. Some incidents, especially over a longer time span, may have been forgotten or may have occurred before the memory capacity of victims was well formed, although given that the data are about childhood and from children this is less of a limitation than in adult retrospective samples. Also, while we have detailed information on the exact identity of the perpetrator (e.g., brother, girlfriend), these were collapsed into broader variables due to statistical reasons. Details are only provided on the most recent CSAA episode for each victim. Although cases of multiple sexual abuse are not common in the NatSCEV data set, it may have changed the results somewhat if all episodes were included. Finally, while the present study provides a detailed picture of children’s sexual abuse incident, further examination of this data is required. Further analysis from these data will examine how the specific characteristics of the abuse impact children’s mental health and self-concept. In addition, build upon these identified characteristics of the abuse, we sought to create clusters of children who experienced CSA that will assist in the examination and treatment of survivors.
Footnotes
Authors’ Note
What is known on this topic: Children and youth are exposed to sexual abuse and assault in varied and extensive ways
The study provides health-care professionals, policy makers, and parents with a detailed profile on child sexual abuse such as the identity and gender of the perpetrator, location where the offense occurred, and other features as child’s level of fear, penetration attempts, injury, medical treatment, and school absence.
The total amount of federal funding involved is US$2,848,809. For the purposes of compliance with §507 of Pub. L. No. 104-208 (the Stevens Amendment), readers are advised that 100% of the funds for this program are derived from federal sources.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by Grants 2006-JW-BX-0003 and 2009-JW-BX-0018 from the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, and U.S. Department of Justice.
