Copyright 2011 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2011
To obtain national estimates of the degree to which school, police, and medical authorities are involved after children experience violence, abuse, and crime victimizations.
A cross-sectional, national telephone survey involving a target sample of 4549 children and youth conducted from January 1, 2008, through May 31, 2008.
Contiguous United States.
Children and adolescents aged 10 to 17 years and the parents of children aged 0 to 9 years.
Conventional crime, maltreatment, abuse by peer and siblings, sexual abuse, and witnessing and indirect exposure to violence.
A total of 45.7% of children and adolescents who had experienced violence in the past year had at least 1 of their victimization incidents known to school, police, or medical authorities. For serious victimizations, such as sexual abuse by an adult, kidnapping, and gang assaults, authorities knew about 70.1% or more of the incidents. Awareness, however, was low for peer and sibling victimizations, dating violence, and completed and attempted rape. In general, school authorities knew about victimizations more often (42.3%) than police (12.7%) or medical authorities (1.8%). However, police were the most likely to know about kidnapping, neglect, and sexual abuse by any adult. Medical authorities were most likely to know about sexual abuse by any adult, gang assault, physical abuse by a caretaker, and assault with a weapon.
More incidents of victimization and abuse appear to be known to authorities currently than was the case in a comparable 1992 survey, but officials should improve at identifying a large quantity of victimizations of children and adolescents that appear to go undetected.
Childhood/adolescent abuse is frequently described as a hidden problem, and victimization studies regularly have shown that much abuse goes undisclosed. The National Crime Victimization Survey, one of the largest population surveys, finds that violent crimes against children are less likely to be disclosed to authorities than are adult crimes and are particularly unlikely to be known to the police.1 A study2 of childhood victimization incidents in 1992 found that only 25% of all incidents and only 6% of the sexual assault and abuse episodes had been disclosed to schools or police.
The hidden nature of childhood victimization has multiple sources.3 Clearly, children and adolescents are easily intimidated by offenders and fear retaliation. However, families, children, and adolescents often wish to deal with crime and victimization informally. They sometimes fear the consequences of disclosure to authorities in the form of interviews and police and court involvement. In other cases, they do not perceive the victimizations as something that would be of interest to authorities.
One of the major public policy efforts of the last generation has been to increase the proportion of abuse and victimization cases known to authorities. The mandatory reporting statutes in the wake of the Child Abuse Prevention and Treatment Act of 1974 had this as a goal.4 The message of virtually all education programs dealing with child maltreatment, bullying, dating violence, and a host of other problems has been to “tell someone and get help.”5 In addition, criminal justice and child protection agencies have instituted various reforms to attempt to increase confidence in those agencies on the part of individuals who have experienced victimization and their families as a way to bolster disclosure.
Despite these policy initiatives, efforts to promote disclosure and track its patterns are unfortunately hampered by limited research on this topic. Cited research frequently refers to studies completed decades ago or based on adult retrospective recollection. It is not clear that such data reflect current experience after a generation of mobilization and increased awareness about childhood victimization in its many forms. The present study attempts to remedy this problem by looking at the degree to which authorities know about victimization in a contemporary cohort of children and adolescents.
The National Survey of Children's Exposure to Violence, designed to obtain 1-year and lifetime prevalence estimates of a wide range of childhood/adolescent victimizations, was conducted from January 1 through May 31, 2008, concerning the experiences of a nationally representative sample of 4549 children and adolescents aged 0 to 17 years living in the contiguous United States. The interviews with parents, children, and adolescents were conducted over the telephone by the employees of an experienced survey research firm.
Sample households were drawn from a nationwide sampling frame of residential telephone numbers through random-digit dialing. This nationally representative cross-section yielded 3053 of the 4549 completed interviews. To ensure that the study included an adequate number of racial/ethnic minority and low-income respondents for more accurate subgroup analyses, there was also an oversampling of telephone exchanges that had high concentrations of African American or Hispanic households or those with low socioeconomic status. This second oversample yielded 1496 of the completed interviews. Sample weights were generated to correct for disproportionate sampling procedures when combining the 2 samples. Additional information on sampling methods and procedures is available elsewhere.6
A short interview was conducted with an adult caregiver in each household to obtain family demographic information. One child or adolescent was randomly selected from all eligible living in a household by selecting the boy or girl with the most recent birthday. If the selected individual was aged 10 to 17 years, the main telephone interview was conducted with him or her. If the selected individual was younger than 10 years, the interview was conducted with the caregiver who “is most familiar with the child’s/adolescent's daily routine and experiences.”
Respondents were promised complete confidentiality and were paid $20 each for their participation. The interviews, each averaging 45 minutes in length, were conducted in English and Spanish. Respondents who disclosed a situation involving 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 resolved or brought to the attention of appropriate authorities. All procedures were authorized by the institutional review board of the University of New Hampshire.
This survey used an enhanced version of the Juvenile Victimization Questionnaire (JVQ), an inventory of childhood victimization.7- 9 The JVQ obtains reports on 48 forms of youth victimization covering 5 general areas of interest: conventional crime, maltreatment, victimization by peers and siblings, sexual victimization, and witnessing and indirect exposure to violence.10 Examples of individual items include robbery, assault with a weapon, peer or sibling assault, neglect by a caregiver, bullying, sexual assault (subdivided by perpetrator identity), sexual harassment, living in an area with shooting or other violence, and witnessing family violence, among others. Specific wording for all the JVQ victimization screeners is available elsewhere.6
Follow-up questions for each victimization item gathered additional information about each event, including whether it occurred in the past year, perpetrator characteristics, weapon use, injury, and whether the event was known to school or police authorities, among other characteristics. If injury resulted, respondents were asked whether medical help was sought (“ . . . go to the hospital, a doctor's office, or some kind of health clinic to get treated for this injury?”). Not all screeners included the same follow-up questions.
This analysis examined victimizations that had occurred in the past year. Specific types were identified and counted in 2 ways. First, each screener was treated as an individual type of victimization based on its described characteristics. Victimizations were flagged as known or not known to school, police, and/or medical authorities based on relevant follow-up questions.
Second, 2 aggregate types of victimization were identified. In particular, assault with a weapon, assault with no weapon, attempted assault, threatened assault, kidnapping, bias-motivated attack, physical abuse, gang or group assault, peer or sibling assault, nonsexual genital assault, and dating violence were each identified as a version of physical assault. Similarly, sexual assault by a known adult, sexual assault by a nonspecified adult, sexual assault by a peer, completed or attempted rape, sexual exposure/being flashed, and sexual harassment are examples of sexual victimizations.
Specific instances of physical assault and sexual victimization were further distinguished from one another using follow-up questions and screener descriptions. Thus, physical assaults and sexual victimizations were characterized as (1) causing no injury, minor injury, or major injury; (2) having an adult or a juvenile perpetrator; (3) having a male or female perpetrator; (4) having a family, known or acquaintance, stranger, or unidentified perpetrator; or (5) occurring at home, in a school or day care setting, or elsewhere. They were also characterized by how afraid and how badly the victim felt as a result of the event (not at all, a little, or very).
Furthermore, physical assault incidents were differentiated by whether the event involved a weapon, bias motivation, attempted assault only, or a threat only. Sexual victimizations were distinguished by whether no sexual penetration, only attempted penetration, or completed penetration occurred.
Child and household information was obtained in the initial parent interview. Measurements included in the analysis were the child's sex, age (5 groups), race/ethnicity (4 groups [white non-Hispanic, black non-Hispanic, other non-Hispanic, and any Hispanic]), and socioeconomic status based on income and parental educational level. Family structure was categorized into 4 groups: children living with (1) a pair of biological or adoptive parents, (2) a single biological parent plus partner (spouse or nonspouse), (3) a single biological parent, and (4) other caregiver. Place type was distinguished as (1) a city of at least 100 000 population, (2) a suburb of a city, (3) a town of less than 100 000 population, or (4) rural.
Study goals were to estimate the relative number of juvenile victimizations that become known to authorities and to identify factors that appear to influence the likelihood of victimizations becoming known (with counts and statistical analyses based on weighted data). First, the number of children within the sample who had been victimized and whose individual victimizations were known to school, police, or medical authorities was counted. Follow-up questions that asked, “Do any of these people know about what happened? A teacher, counselor, or other adult at school or day care (Y/N); A police officer or some other law official (Y/N),” were used for this count. Having one's victimizations known or unknown to medical authorities was based on the follow-up question concerning seeking medical attention for an injury, described previously. Only screeners that involved physical violence included this follow-up item, somewhat limiting information about the awareness of medical authorities.
To look more closely at differences in victimizations and characteristics that might affect whether they are known or not known to authorities, we examined the physical assault and sexual aggregated victimizations. For this analysis, victimizations were treated and counted as incidents, that is, as a single event occurring at a single time and place regardless of the number of screeners that referred to the event. For example, a child might be robbed and at the same time be physically assaulted and/or threatened. This single event could be described by several screeners (eg, robbery, assault with a weapon or with no weapon, and having been threatened). Each screener in the JVQ includes follow-up questions that allow those events that are part of the same incident and those that are independent events to be identified.
These follow-ups were used to identify all incidents that included any physical assault or any sexual victimization, as previously identified. The separate physical assaults and sexual victimizations were then described in terms of the victimization characteristics (eg, injury, perpetrators, and place of occurrence) and linked to the demographic and household attributes of the child or adolescent who had experienced the incident(s).
Direct logistic regression was used to identify the victimization characteristics, child/adolescent characteristics, or household characteristics that were associated with incidents being known to authorities. Four models were constructed that predicted (1) physical assault incidents known to school authorities, (2) physical assault incidents known to police authorities, and (3 and 4) sexual victimization incidents known to school and police authorities, respectively. Incidents known to medical authorities were not analyzed because of the limited number of screeners reporting that information. Because a child or adolescent might have more than 1 victimization incident of each type, a dummy variable flagged those victimizations for which individuals had experienced multiple incidents. In addition, the models included a measure of each individual's total victimization experience—for physical assault models, the number of nonphysical assaults experienced in the past year, and for sexual victimization models, the number of nonsexual victimizations.
A considerable portion of children who had experienced victimization had those incidents known by some authority (school, police, or medical). A total of 58.3% of the sample reported at least 1 direct victimization in the past year (this included bullying but excluded events such as witnessing domestic assault). Of these victims, 45.7% had at least 1 victimization that was known to authorities.
The degree to which victimizations were known to authorities varied according to victimization screener (Table 1). The victimizations most likely to be known were typically of a more serious nature, such as sexual abuse by a known or nonspecified adult (69.0% and 76.1%), kidnapping (73.5%), and gang or group assault (70.1%). However, even emotional bullying (51.5%), neglect (47.8%), and theft (46.8%) were often known to authorities. The types of episodes least likely to be known were peer and sibling assault (16.9%), dating violence (15.2%), sexual exposure/being flashed (16.6%), completed and attempted rape (14.0%), and statutory rape (3.4%).
There was also considerable variation in the degree to which indirect victimization was known to authorities. Obviously, murders to which children were exposed and public offenses against schools were widely known. Also, witnessing of domestic violence by children was known in approximately half the episodes (48.9%).
School authorities tended to know about victimization events (42.3%) more than police (12.7%) or medical authorities (1.8%). However, police were more likely than school authorities to know about incidents for several types of victimization, including kidnapping (71.1% vs 46.0%), neglect (36.9% vs 29.2%), sexual abuse by a known or nonspecific adult (64.9% vs 30.2% and 76.1% vs 28.5%, respectively), and witnessing of domestic violence (42.3% vs 22.9%).
As noted, our information regarding events known to medical authorities was limited. Although few episodes involved medical authorities, the most common were sexual abuse by a known or nonspecified adult (7.4% and 19.1%, respectively), gang assault (8.7%), physical abuse (10.0%), and assault with a weapon (7.7%).
There was some variation by age in the degree to which authorities knew about victimizations. For direct victimizations, children and adolescents aged 6 years and older were more likely to have their incidents known by school authorities, but there was little difference between 6- to 9-year-olds, 10- to 13-year-olds, or 14- to 17-year-olds for this factor. For police and medical authorities, the victimizations of 14- to 17-year-olds were most likely to be known.
As noted, 4 logistic regressions were used to identify the features of children and adolescents and incidents that were associated with being known to authorities. We looked at physical assault and sexual victimization incidents separately, as well as incidents known to school or police, expecting the dynamics to differ for the kinds of victimizations and authorities (Table 2 and Table 3).
School authorities were more likely to know about a physical assault when it occurred in school, involved a serious injury, involved a stranger or an adult perpetrator, had a bias motivation, or made the individual who experienced it report feeling afraid or bad (Table 2). Attempts and threats were also more likely to be known than actual assaults. Several child/adolescent characteristics were also related to school authorities knowing about physical assault victimizations: female sex, younger than 13 years, having experienced other assaults in the past year, and being from a family with low socioeconomic status. By comparison, assaults experienced by Hispanic individuals were less likely to be known.
Sexual offenses were more likely to be known to school authorities when they occurred in school, involved a nonidentified perpetrator, occurred to a 2- to 5-year-old or a 6- to 9-year-old, and involved a child living with a stepparent or unmarried partner of a parent (Table 3).
Police were more likely to know about physical assaults when the offense occurred somewhere other than home or school or involved a serious injury, a nonfamily or adult perpetrator, or bias motivation or when the child reported being very afraid (Table 2). Also associated with greater police likelihood of knowing were when affected individuals were girls, from families with lower socioeconomic status, and living in a rural area. The only features associated with sexual incidents being known to police were the child feeling afraid and an adult perpetrator (marginal) (Table 3). Victims who were black, of mixed race/ethnicity, or other race/ethnicity were also more likely to have their sexual victimizations known to police.
According to children, adolescents, and caregivers interviewed in this study, the victimizations of many children and adolescents are known to authorities. Almost half (45.7%) of youth who had experienced at least 1 incident of victimization within the past year were known to school, police, or medical authorities. This overstates somewhat the level of knowledge in that some individuals experienced multiple victimizations and not all were disclosed. However, having 1 victimization known gives officials, with proper training, an opportunity to inquire about other victimizations. Moreover, the analysis also suggests that children and adolescents with multiple victimizations were among those most likely to have some events known by authorities.
Many of the most serious victimizations, such as sexual abuse by adults, kidnapping, and gang assault, were likely to be known to authorities, whereas others, such as attempted and completed rape and dating assault, were unlikely to be known. The contrasting pattern for different kinds of sexual crime may have to do with differences in perpetrator characteristics; for example, attempted and completed rape and dating assault often involve peer perpetrators. Incidents with adult perpetrators are more likely to be known, perhaps because the adult offenses are seen as more criminal or because peer allegiances may inhibit reporting of younger perpetrators.
The idea that disclosure to authorities has increased over time is suggested by comparison of the current study findings with another national survey2 of victimization published in 1994, using data from 1992. In the 1992 telephone survey with caretakers and children and adolescents aged 10 to 16 years, only 25% of all victimizations were known to police or school authorities (vs 50.6% in the present study for that age group), only 29% of the kidnapping incidents (vs 73.3% in present study), and only 6% of the sexual assault or sexual abuse incidents (vs 11.2% in the present study). This is good news and may reflect efforts by authorities and advocates to promote disclosure. Because early disclosure is believed to facilitate prevention, increased disclosure rates are consistent with the findings that childhood victimization rates have decreased considerably since the early 1990s.13,14
Consistent with 2 other studies,2,15 our findings indicate that school authorities are more likely to find out about victimizations experienced by children and adolescents than other authorities, which is understandable given how much time children and adolescents spend in school and interact with school professionals. Although police and medical authorities may conclude from these results that they are seriously underinformed about victimization, it is not clear how much victimization of children and adolescents merits the specialized involvement of those professionals.
The analysis also provides some clues about where to reinforce efforts to promote disclosure, one of the main avenues for authorities knowing about victimizations. Boys are less likely to have their victimizations known to authorities, probably reflecting the “boy code” of self-sufficiency that stigmatizes help seeking, a code that some educational programs are now trying to counteract.16 Hispanic children and adolescents are less likely to have their victimizations known, perhaps reflecting Hispanic concerns about how authorities will treat them. Children from families with higher socioeconomic status are also less likely to have their victimizations known, which perhaps reflects suspicion among those families about the negative effect of disclosure on children and adolescents, combined with having the resources and status to deflect the involvement of authorities. Efforts to emphasize the helpful rather than stigmatizing features of professional intervention might be useful to counteract some of the concerns in these groups.
The study also shows that authorities are less likely to know about victimizations involving family and peer perpetrators in contrast to nonfamily and adult perpetrators. Educators have long recognized the need to promote disclosures about such family and peer incidents. An important task is for authorities to persuade children and families that they have resources to help in these situations and that they can provide protection against retaliation for individuals who disclose their victimizations.
Although this study has many virtues, including its nationally representative sample and comprehensive spectrum of questions, it is important to be aware of some limitations. The kinds of victimizations not known to authorities might also be those that children, adolescents, and caregivers would not disclose to the survey. Thus, the survey may overstate the proportion of victimizations that become known to authorities. However, this limitation applies to all surveys and was certainly as true in earlier surveys as in this one. Therefore, the higher rates of disclosure found in this study relative to earlier surveys likely reflect real increases. Some might argue that when children and adolescents become adults, they are more willing to disclose, favoring adult retrospective interviews as the best method for identifying unreported abuse. However, adult retrospective interviews suffer from the fact that adults likely do not remember all the victimizations experienced during childhood and adolescence, particularly those that happened at a younger age and especially those that might have been less serious. In general, victimology studies17 suggest that contemporaneous research is more complete and accurate.
The current study suggests progress and challenge in the effort to identify children and adolescents who have experienced abuse and victimization. The higher rates of victimizations known to authorities found in the current study may mean that past efforts to promote disclosure have been working and should be sustained. However, the study also shows that a considerable portion of childhood/adolescent exposure to victimization is still unknown to authorities. The study suggests that outreach needs to be particularly enhanced toward boys, Hispanics, and higher-income groups. It also suggests that disclosure promotion should be directed toward episodes that involve family members and peer perpetrators.
The promotion of disclosure also gains justification as authorities develop interventions that are truly helpful in preventing future victimizations and treating the negative effects of victimization on development. Fortunately, more such interventions are being developed, tested, and deemed effective, including conflict resolution programs,18 parenting education programs,19 and cognitive behavioral treatments for victimization trauma.12,20- 22 So, in addition to more education and awareness to encourage disclosure, communities should also ensure that they have professionals trained in such evidence-based programs to provide care to children, adolescents, and families once victimization has been disclosed.
Correspondence: David Finkelhor, PhD, Crimes Against Children Research Center, University of New Hampshire, 126 Horton Social Science Center, 20 Academic Way, Durham, NH 03824 (email@example.com).
Accepted for Publication: May 21, 2010.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Finkelhor, Ormrod, Turner, and Hamby. Analysis and interpretation of data: Ormrod, Turner, and Hamby. Drafting of the manuscript: Finkelhor and Ormrod. Critical revision of the manuscript for important intellectual content: Ormrod, Turner, and Hamby. Statistical analysis: Ormrod and Turner. Obtained funding: Finkelhor and Turner. Administrative, technical, and material support: Finkelhor. Study supervision: Finkelhor.
Financial Disclosure: None reported.
Funding/Support: For the purposes of compliance with section 507 of Public Law 104-208 (Stevens Amendment), readers are advised that 100% of the funds for this program are derived from federal sources (this project was supported by grants 2006-JW-BX-003 and 2009-JW-BX-0018 awarded by the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, US Department of Justice). The total amount of federal funding involved is $2 848 809.
Disclaimer: Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the US Department of Justice.
Finkelhor D, Ormrod R, Turner H, Hamby S. School, Police, and Medical Authority Involvement With Children Who Have Experienced Victimization. Arch Pediatr Adolesc Med. 2011;165(1):9-15. doi:10.1001/archpediatrics.2010.240