Youth Victimization: Prevalence and Implications U.S

U.S. Department of Justice Office of Justice Programs APR. 03 National Institute of Justice

Research in Brief

Youth Victimization: Prevalence and Implications U.S. Department of Justice Office of Justice Programs 810 Seventh Street N.W. Washington, DC 20531

John Ashcroft Attorney General

Deborah J. Daniels Assistant Attorney General

Sarah V. Hart Director, National Institute of Justice

This and other publications and products of the U.S. Department of Justice, Office of Justice Programs, National Institute of Justice can be found on the World Wide Web at the following site:

Office of Justice Programs National Institute of Justice http://www.ojp.usdoj.gov/nij APR. 03 Youth Victimization: Prevalence and Implications

Findings and conclusions of the research reported here are those of the authors and do not reflect the official position or policies of the U.S. Department of Justice. This research was supported by the National Institute of Justice under grant number 93–IJ–CX–0023.

NCJ 194972 RESEARCH IN BRIEF / APR. 03

ABOUT THIS REPORT

Beyond cases reported to A clear relationship exists authorities, little knowledge between youth victimization exists on the types, amount, and mental health problems and effects of childhood vic- and delinquent behavior. For timization. Through a nation- example— al survey of adolescents, researchers examined the ■ Negative outcomes in vic- prevalence of sexual assault, tims of sexual assault were physical assault, physically three to five times the rates abusive punishment, and wit- observed in nonvictims. nessing an act of violence and ■ subsequent effects on mental Girls who witnessed vio- health, substance use, and lence were nearly twice delinquent behavior problems. as likely as boys to experi- Gender- and racial/ethnic- ence posttraumatic stress specific findings are translated disorder. into national estimates. What were the study’s What did the limitations? researchers find? The nationally representative Rates of interpersonal vio- sample did not include ado- lence and victimization of 12- lescents from homes without to 17-year-olds in the United telephones and certain high- States are extremely high, risk adolescents (i.e., those and witnessing violence is who were homeless or considerably more common. housed in jails, juvenile correctional facilities, or Black and Native American inpatient mental health treat- adolescents were victimized ment facilities). more than whites, Hispanics, and Asians in each type of victimization. Much of the Who should read violence experienced by this study? youths is perpetrated by Criminal justice practition- peers or someone the victim ers, policymakers, and knows well. Most sexual researchers. assaults (86 percent) and physical assaults (65 percent) went unreported. ii YOUTH VICTIMIZATION

Dean G. Kilpatrick, Benjamin E. Saunders, and Daniel W. Smith

Youth Victimization Prevalence and Implications

Rates of interpersonal vio- Findings from the National lence and victimization of Survey of Adolescents (NSA) youths are extremely high in validate these concerns.1 In the United States. As of 1995, the NSA interviewed 1995, approximately 1.8 mil- 4,023 adolescents ages 12 to lion adolescents ages 12 to 17 to obtain information on 17 had been sexually assault- substance use, abuse, and ed and 3.9 million had been dependence; delinquent severely physically assaulted. behaviors as dictated by the Another 2.1 million had been type and amount of Crime punished by physical abuse. Index offenses—as deter- Most pervasive is victimiza- mined by the FBI’s Uniform tion by witnessing violence, Crime Reports—committed with approximately 8.8 mil- during the previous year; lion youths indicating that demographic information; they had seen someone else family history of substance being shot, stabbed, sexually abuse; posttraumatic stress assaulted, physically assault- disorder (PTSD); and victim- About the Authors ed, or threatened with a ization history. This informa- Dean G. Kilpatrick, Ph.D., weapon. tion was then broken down is a professor at and by age, gender, and racial/ director of the National The emotional consequences ethnic group to allow com- Crime Victims Research that youths experience be- parisons across groups (see and Treatment Center cause of victimization, such “NSA Methodology” and (NCVC) and director of as psychological disorders, “Study Terminology”). the National Violence substance abuse and de- Clearly, victimization in early Against Women Preven- pendence, and delinquency childhood and adolescent tion Research Center. problems, are often over- years is the root of many Benjamin E. Saunders, looked in research. More problems later in life. Much Ph.D., is a professor at openly, however, media NCVC and director of of the knowledge gained accounts of children seriously NCVC’s Family and Child from the NSA raises crucial Program. Daniel W. harmed or even killed in vio- issues that cross the lines of Smith, Ph.D., is an associ- lent episodes have brought research, policy, and practice. ate professor and director such issues to the height of of training at NCVC. the Nation’s consciousness.

1 RESEARCH IN BRIEF / APR. 03

NSA METHODOLOGY

Many studies have examined the relationships between childhood victimization and various mental health problems. Others have evaluated relationships between substance use and delinquency, and still others have tested possible relationships between childhood victimization and delinquency. Nearly all of these studies, however, have been conducted with samples recruited from clinical settings, juvenile or psychiatric institutions, agency records (e.g., child protective service records), and/or geographically limited areas. These recruitment methods significantly restrict the ability to generalize results obtained from previous studies. The NSA was the first study to examine simultaneously all of the relationships between victimization experiences and various mental health problems, substance use and delinquency, and childhood victimization and delinquency. The NSA also proposed a theoretically and empirically derived explanatory model connecting these variables and tested the model with a nationally representative sample.a Between January and June 1995, NSA researchers used random-digit dialing and stratified sampling techniques to identify households that met participant eligibility requirements for the study: ■ The household had to be able to be reached by telephone.

■ An adolescent between the ages of 12 and 17 had to live in the household with a parent or legal guardian.

■ Both the parent or guardian and the adolescent had to speak English and/or Spanish. Interviews gathered several types of information about substance use, abuse, and dependence. Adolescents were asked a series of questions to determine whether they met diagnostic criteria for substance abuse or dependence.b Questions explored such topics as nonexperimental alcohol use, the use of prescription drugs in a way not prescribed by a physician, use of marijuana or hard drugs, and frequency of drug use during the year prior to the NSA interview. Delinquent behaviors were measured using a slightly modified version of a scale first used by Delbert Elliott and colleagues in the National Youth Survey.c Adolescents were classified as a delinquent if they had committed at least one Crime Index offense as defined by the FBI’s Uniform Crime Reports during the year prior to the NSA interview. Once identified, 4,023 adolescents were interviewed by trained employees of the survey research firm Schulman, Ronca, and Bucuvalas, Inc. Computer-assisted protocols ensured that every respondent was asked all pertinent questions. The data collected included demographic information, family history of substance abuse, delinquent behaviors, substance abuse and dependence, posttraumatic stress disorder, and victimization history.d

Notes a. For more detailed information regarding survey methodology, see Kilpatrick, D.G., and B.E. Saunders, Prevalence and Consequences of Child Victimization: Results From the National Survey of Adolescents, Final Report, Washington, DC: U.S. Department of Justice, National Institute of Justice, 1997, NCJ 181028. b. American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Washington, DC: American Psychiatric Press, 1994: 175–183. c. Elliott, D., D. Huizinga, and S.S. Ageton, Explaining Delinquency and Drug Abuse, Thousand Oaks, CA: Sage Publications, 1985. d. For more detailed demographic information, see Kilpatrick and Saunders, 1997 (note a).

2 YOUTH VICTIMIZATION

STUDY TERMINOLOGY

The following is a list of terms and definitions as they pertain to this study. Delinquent behavior: committing one or more of the following Crime Index offenses—aggravated assault; sexual assault; gang fights; motor vehicle theft; theft of something more than $50 in value; breaking into a building or vehicle; or strong-arming students, teachers, and others. Hard drug: a drug other than alcohol and marijuana, such as heroin, crack, cocaine, ecstasy, and so forth. Lifetime prevalence: the percentage of adolescents and youths who have been victimized at any time in their life. Nonexperimental alcohol use: the consumption of five or more drinks on a given day during the previous year. Nonexperimental drug use: the use of any illicit, hard drug four or more times during the previous year. Physical assault: being attacked with a weapon, being attacked without a weapon but with intent to kill or injure, being threatened with a gun or knife, or being beaten with an object or fists and seriously injured. Physically abusive punishment: a spanking that results in a youth’s need to see a doctor; a spanking that results in noticeable marks, bruises, cuts, or welts; or punishment that includes burning, cutting, or tying up a youth. Posttraumatic stress disorder (PTSD): a mental disorder diagnosed from a set of symptoms following exposure to an extreme life stressor and characterized by intense fear, helplessness, horror, or, among children, disorganized or agitated behavior. Sexual assault: a range of acts, including the sexual penetration of a youth’s vagina or anus by a penis, finger, or object; the placement of another person’s mouth on a youth’s sexual parts; the touching of a youth’s sexual parts by another person or the forcing of a youth to touch others’ sexual parts; or the unwanted penetration of others by a youth (asked only of male adolescents). Stranger: someone the victim had never seen before or had seen before but did not know well. Substance abuse: a maladaptive pattern of substance use that results in significant impairment or distress, as indicated by one or more of the following problems: recurrent substance-related legal problems, such as arrests; recurrent substance use in situations where it is physically dangerous, such as while driving a car or crossing the street in heavy traffic; substance use resulting in a failure to fulfill major role obligations at school, work, or home; and continued substance use despite having major social or relationship problems because of use. Substance dependence: a cluster of symptoms indicating a pattern of substance use that results in tolerance, withdrawal, and compulsive substance-taking behavior. Witnessed act of violence: any one of a variety of violent incidents, including seeing someone shot with a gun, stabbed or cut, sexually assaulted, mugged or robbed, or threatened with a weapon.

3 RESEARCH IN BRIEF / APR. 03

Types and prevalence victimization among black of violence and Native American adolescents. Notably, more than Lifetime prevalence of four half of black, Hispanic, and types of violence was exam- Native American adolescents ined: sexual assault, physical had witnessed violence in assault, physically abusive their lifetimes. Also, Native punishment, and witnessing American adolescents had an act of violence. Exhibit 1 the largest prevalence rate presents the number, per- for sexual assault victimiza- centage, and population esti- tions; whites and Asians mates of adolescents who reported the lowest. With experienced each type of vio- respect to physical assault, lence and the racial/ethnic Native Americans, blacks, characteristics of the victims and Hispanics reported the of each form of violence. highest victimization preva- Particularly salient among lences—20 to 25 percent of these findings is the higher each group reported experi- prevalence of all types of encing at least one physical assault.

Exhibit 1. Prevalence of violence types in the NSA sample and estimates for U.S. adolescents ages 12 to 17 and across racial/ethnic groups

Physically Sexual Physical abusive Witnessed assault assault punishment violence

Number of adolescents in sample (N = 4,023) 326 701 376 1,586 Percent of total sample 8.1 17.4 9.4 39.4 Estimated number of victims in U.S. adolescent population (in millions)* 1.8 3.9 2.1 8.8 White (%) 6.7 15.6 7.9 34.3 Black (%) 13.1 24.2 15.4 57.2 Hispanic (%) 10.0 20.9 8.4 50.5 Native American (%) 15.7 27.3 15.1 55.7 Asian (%) 6.5 6.5 6.5 26.1

Note: Percentages do not equal 100 because not all respondents experienced or witnessed a type of violence. *Population estimates are based on 1995 U.S. Census data indicating there are 22.3 million adolescents in the U.S. population. Source: National Survey of Adolescents, 1995

4 YOUTH VICTIMIZATION

Exposure to violence differed Sexual assault. The 326 according to gender. Girls adolescent sexual assault were at greater risk of sex- victims experienced 462 sep- ual assault than boys (13.0 arate cases of sexual assault. percent versus 3.4 percent).2 In these cases, nearly three Boys, however, were at sig- in four (74 percent) reported nificantly greater risk of phys- that the assault was commit- ical assault than girls (21.3 ted by someone they knew percent versus 13.4 percent). well. Almost one-third of sex- A substantial number of all ual assault cases (32.5 per- adolescents (43.6 percent of cent) involved perpetrators boys and 35 percent of girls) who were friends, and more reported having witnessed than one-fifth (23.2 percent) violence. Physically abusive were strangers. Another one- punishment was similar for fifth (21.1 percent) were boys (8.5 percent) and girls committed by a member of (10.2 percent). the youth’s family, including fathers or stepfathers, broth- ers or stepbrothers, sisters or Characteristics of stepsisters, grandparents, sexual and physical other adult relatives, and assaults other child relatives. In terms of location, more than half of For adolescents who report- all sexual assaults occurred ed experiencing sexual or either within the victim’s physical assaults, additional home (30.5 percent) or in the information about the charac- victim’s neighborhood (23.8 teristics of the assault were percent). Another 15.4 per- assessed, including the cent of sexual assaults oc- identity of the perpetrator, curred at the victim’s school. the location of the assault, whether the youth experi- Another important character- enced a physical injury during istic is the extent of actual the assault, whether the and potential violence in- youth perceived that his or volved. Slightly more than her life was in danger, and one in four sexual assault vic- whether the assault was ever tims (28.1 percent) said they reported to authorities. feared death or serious injury (Similar characteristics about during their sexual assault. physically abusive punish- However, the majority (69.5 ment and witnessed violence percent) said they had no events were not gathered.) such fears. With respect to

5 RESEARCH IN BRIEF / APR. 03

physical injuries, only 1.3 per- members, adult or child rela- cent of sexual assault cases tives, neighbors, or cowork- resulted in serious injuries ers. In terms of location, outside the sexual assault physical assaults were most itself, and 11 percent result- likely to occur somewhere ed in minor injuries. within the victim’s neighborhood (34.2 percent) or home Thirteen percent of sexual (27.9 percent). Another 20.2 assault cases were reported to percent occurred at the vic- police, 5.8 percent to child pro- tim’s school. tective services, 5 percent to school authorities, and 1.3 per- With respect to the extent of cent to other authorities. The actual and potential violence majority of sexual assaults (86 involved, physical assaults percent), however, went unre- were reported to involve ported. (These percentages more perceived life threats total more than 100 percent and injuries than were sexual because some cases were assaults. In more than half reported to more than one (52.4 percent) of physical type of authority.) In 4 percent assaults, victims said they of the cases, victims either feared being seriously injured were not sure whether cases or killed. Although the largest had been reported or refused group of physical assault vic- to answer the question. tims reported that they had not sustained any physical Physical assault. The 701 injuries (47.5 percent), close adolescent physical assault to half (45 percent) reported victims experienced 1,054 minor injuries. Only a small separate cases of physical percentage (4.5 percent) assault. In these cases, 6 in reported serious injuries. 10 (62 percent) reported that the assault was committed Similar to sexual assaults, by someone they knew well. physical assaults were gener- Slightly more than one-third ally unlikely to be reported to (36.4 percent) reported authorities. Of all physical that the perpetrator was a assaults, 65 percent were stranger. Of perpetrators never reported. Of the cases known by the victim, friends that were reported, most (20.5 percent) were the most were reported to police (16.9 commonly reported perpetra- percent) or school authorities tor. Other less commonly (16.3 percent). The remaining reported perpetrators in- cases included reports to cluded immediate family other authorities (3.8 percent)

6 YOUTH VICTIMIZATION

or child protection agencies substance abuse or depend- (2.8 percent). Adolescents in ence, and delinquent behav- 2.8 percent of cases either ior. Exhibit 2 presents the were not sure if reports had percentage of all respon- been made or refused to dents who reported each of answer the question. these types of problems, as well as the percentages across age, gender, and Mental health racial/ethnic group. problems and PTSD. In the overall sample, delinquent behavior the lifetime prevalence of The NSA examined the rates PTSD was 8.1 percent, indi- with which adolescents cating that, as of 1995, reported mental health approximately 1.8 million problems, such as PTSD, adolescents had met the

Exhibit 2. Percentage of self-reported adolescent lifetime mental health and delinquent behavior problems in total sample and by age, gender, and racial/ethnic group

Posttraumatic Alcohol Marijuana Hard drug stress abuse/ abuse/ abuse/ Delinquent Group disorder dependence dependence dependence behavior

Gender Male 6.2 6.3 4.7 1.2 17.7 Female 10.1 4.9 4.2 1.1 6.7 Age 12 3.4 0.3 0.1 0.0 6.4 13 6.9 1.5 1.0 0.4 8.9 14 6.8 2.7 2.4 0.1 9.5 15 8.2 4.5 6.6 1.3 14.8 16 10.1 11.3 8.7 2.8 17.1 17 13.1 14.2 8.3 2.2 17.8 Race/ethnicity White 7.3 6.0 4.7 1.1 9.9 Black 11.0 4.4 2.4 0.2 18.8 Hispanic 11.6 6.4 6.1 1.6 16.8 Native American 7.1 5.7 7.9 2.9 25.9 Asian 6.5 2.2 2.1 0.0 8.5 Total sample 8.1 5.6 4.5 1.2 12.3

Source: National Survey of Adolescents, 1995

7 RESEARCH IN BRIEF / APR. 03

criteria for PTSD at some Hispanic adolescents, a find- point during their lifetime. ing somewhat surprising Girls were significantly more among Native Americans likely than boys to have life- because of their high level of time PTSD (10.1 percent exposure to violence. This versus 6.2 percent). The finding suggests that re- prevalence of PTSD rises sig- sponses to victimization may nificantly with increasing age, be different among racial/ as older youths have more ethnic groups. time for exposure to poten- tially traumatic events. It is Substance abuse or noteworthy that the rate of dependence. Overall, 9.1 lifetime PTSD among 17- percent of American adoles- year-olds was 13.1 percent. cents reported meeting diag- This means that more than nostic criteria for lifetime one in eight 17-year-olds has substance abuse or depend- had PTSD at some point in ence for alcohol or any drug.3 his or her lifetime. This means that as of 1995, approximately 2 million With respect to race and adolescents had met diag- ethnicity, whites, Native nostic criteria for substance Americans, and Asians had abuse or dependence at significantly lower rates of some point in their lives. PTSD than did black and More specifically, the NSA found that the lifetime prevalence of alcohol abuse or dependence among adoles- THE NSA IN SCHOLARLY JOURNALS cents was 5.6 percent, the lifetime prevalence of mari- Acierno, R., D.G. Kilpatrick, H.S. Resnick, B.E. Saunders, M. de Arellano, and C.L. Best. “Assault, PTSD, Family Substance juana abuse or dependence Use, and Depression as Risk Factors for Cigarette Use in was 4.5 percent, and the life- Youth: Findings From the National Survey of Adolescents.” time prevalence of hard drug Journal of Traumatic Stress 13 (2000): 381–396. abuse or dependence was Crouch, J.L., R.F. Hanson, B.E. Saunders, D.G. Kilpatrick, and 1.2 percent. H.S. Resnick. “Income, Race/Ethnicity, and Exposure to Violence in Youth: Results From the National Survey of Boys were significantly more Adolescents.” Journal of Community Psychology 28 (2000): likely than girls to have met 625–641. diagnostic criteria for lifetime Kilpatrick, D.G., R.E. Acierno, H.S. Resnick, B.E. Saunders, abuse of or dependence on C.L. Best, and P.P. Schnurr. “Risk Factors for Adolescent alcohol, but not marijuana or Substance Abuse and Dependence: Data From a National hard drugs. Each type of Sample.” Journal of Consulting and Clinical Psychology 68 (2000): 19–30. abuse or dependence was significantly related to age:

8 YOUTH VICTIMIZATION

■ Rates of alcohol abuse or approximately three times dependence essentially more likely to have ever com- doubled each year be- mitted an Index offense than tween ages 12 and 16. girls (17.7 percent and 6.7 percent, respectively). In ■ Rates of marijuana abuse regards to race and ethnicity, or dependence more than white and Asian adolescents doubled each year be- reported significantly lower tween ages 12 and 15. rates of Index offenses than did black, Native American, ■ Rates of hard drug abuse or Hispanic adolescents. or dependence increased dramatically from age 14 to 16. Victimization and These findings reflect the mental health extent to which the risk problems/delinquency of alcohol and drug prob- NSA results indicate clear lems increases dramatically relationships between the throughout adolescence. experience of youth victim- Rates of problem substance ization and mental health use were very similar across problems (i.e., PTSD and sub- racial and ethnic groups. The stance abuse or dependence) only significant difference and delinquent behavior. was that black and Asian adolescents were significant- Relationships with sexual ly less likely to report lifetime assault. Among boys who marijuana abuse or depend- had experienced sexual ence than were white, His- assault, 28.2 percent had panic, and Native American PTSD at some point in their youths. lives. The rate of lifetime PTSD among boys who had Delinquent behavior. Slightly not been sexually assaulted more than 12 percent of ado- was 5.4 percent. Similarly, lescents acknowledged com- sexually assaulted girls had a mitting at least one Index lifetime PTSD rate of 29.8 offense at some point in their percent, compared with only lifetime. This percentage 7.1 percent of girls with no translates to an estimated sexual assault history. There- 2.7 million adolescents in the fore, a history of sexual United States who have assault was associated with engaged in serious delin- a four- to fivefold increase in quent behavior. Boys were the prevalence rate of PTSD.

9 RESEARCH IN BRIEF / APR. 03

More than one-third (34.4 Relationships with physical percent) of boys who had assault or physically abu- been sexually assaulted sive punishment. Experi- demonstrated substance encing either a physical abuse or dependence at assault or physically abusive some point during their life- punishment was associated times. The rate of lifetime with a lifetime PTSD rate of substance abuse or depend- 15.2 percent for boys. The ence in nonsexually assault- rate of lifetime PTSD in boys ed boys was 9 percent. A who had not been physically similar pattern was noted for assaulted or abusively pun- girls: 27.5 percent of those ished was 3.1 percent. sexually assaulted reported Among girls who reported lifetime substance abuse or experiencing physical assault dependence, and only 5.4 or physically abusive punish- percent of those nonsexually ment, the rate of lifetime assaulted reported such PTSD was 27.4 percent, problems. compared with 6 percent among those with no history Almost half (47.2 percent) of of physical assault or physi- the sexually assaulted boys cally abusive punishment. reported engaging in delinquent acts, compared with Similar overall patterns were only 16.6 percent of those noted with respect to sub- not sexually assaulted. Sex- stance use. Approximately ually assaulted girls reported 25 percent of physically engaging in delinquent acts assaulted or abused adoles- less often (19.7 percent). cents reported lifetime sub- This rate, however, was stance abuse or dependence. approximately five times Rates of substance problems higher than the delinquency among nonphysically assault- rate of girls who had not ed or abused adolescents been sexually assaulted (4.8 were roughly one-fifth (approx- percent). Overall, victimiza- imately 6 percent) of those tion by sexual assault is clear- reported by assault or abuse ly associated with dramatic victims. Differences, how- increases in the rates of each ever, existed across genders. negative outcome among Substance abuse and depend- both boys and girls. The ence rates for each type of vic- increases were quite large, timization were similar among ranging from three to five boys (approximately 24 per- times the rates observed in cent in both physical abuse nonvictims. and physical assault groups)

10 YOUTH VICTIMIZATION

but not among girls. Fewer Relationships with wit- physically abused girls report- nessed violence. Although ed problems with substance witnessing violence was con- abuse or dependence than siderably more common than those physically assaulted personal victimizations in the (17.3 percent versus 26.4 per- NSA sample, rates of PTSD cent). Of note, the rate of sub- associated with this form of stance use problems among violence exposure were The rate of nonphysically abused girls notably lower. Among boys is substantially lower (7.2 who had witnessed violence, PTSD among percent). 11.2 percent had met diag- girls who had nostic criteria for PTSD at The relationship between a some time in their lives. Boys witnessed history of physical assault/ who reported never witness- violence was abusive punishment and ing violence had a lifetime nearly double delinquency was particularly PTSD rate of 2.3 percent. strong. The percentage of The rate of PTSD among girls that of boys. boys who were physically who had witnessed violence assaulted and had ever com- (20.2 percent) was nearly mitted an Index offense was double that of boys. Only 4.2 46.7 percent, compared with percent of girls who reported 9.8 percent of boys who no history of witnessed vio- were not assaulted. Similarly, lence had ever had PTSD. 29.4 percent of physically assaulted girls reported hav- Among boys who witnessed ing engaged in serious delin- violence, 17 percent reported quent acts at some point in lifetime substance abuse or their lives, compared with 3.2 dependence. Only 4.4 percent percent of nonassaulted girls. of boys who did not witness The results for physically abu- violence reported lifetime sive punishment resembled substance abuse or depend- those for physical assault, in ence. Among girls who did or that 44.6 percent of abused did not witness violence, the boys had engaged in delin- lifetime substance abuse or quent acts, compared with dependence rates were near- 15.1 percent of nonabused ly identical to those of boys boys. Of abused girls, 20 per- (17.8 percent and 3.1 percent, cent had ever participated in respectively). delinquent activities, compared with 5.2 percent of Approximately one-third (32 nonabused girls. percent) of boys who witnessed violence reported

11 RESEARCH IN BRIEF / APR. 03

ever engaging in delinquent behavioral problems. Given acts, compared with only 6.5 the strong associations percent of boys who did not between victimization and witness violence. Approxi- negative outcomes, research mately 17 percent of girls about adolescent alcohol and who witnessed violence drug use and delinquency reported lifetime delinquent should screen for history of behavior, compared with 1.4 violent assault and witness- percent of girls who did not ing violence. witness violence. The NSA findings indicate that the bulk of violent Implications assaults are perpetrated by someone the victim Future research. Longitudinal knows well rather than by research is needed to clarify a stranger. Further research the temporal sequence of vic- is needed about the circum- timization; PTSD; substance stances and behavioral use, abuse, or dependence; sequences that precede and and delinquent behavior follow such assaults. Such among adolescents. This is data might prove useful in particularly important given the design of violence pre- that the prevalence of violent vention programs. assault, witnessing violence, alcohol and drug use, and The roles of specific types of delinquent behavior increases victimization and particular dramatically between the characteristics of victimiza- ages of 12 and 17. Such tions should be evaluated in research should examine the the development of sub- temporal sequence of prob- stance use problems and lem development as well as delinquency, especially with risk and protective factors that gender and racial/ethnic sub- are related to victimization, groups. Results suggest that alcohol and drug use, PTSD, some types of victimization and delinquent behavior. are more important for pre- dicting problems in certain The NSA research demon- subgroups. Further refine- strates the feasibility of ob- ment and testing of these taining detailed information hypotheses are needed. from adolescents about their victimization experiences, Research also is needed to exposure to violence, and better understand the factors related mental health and

12 YOUTH VICTIMIZATION

that contribute to the dramat- substance use and delin- ic underreporting of crimes quency. Although all child against children and adoles- and adolescent victimizations cents. Although some re- cannot be prevented, at least search exists in this area, some of the long-term nega- most reasons offered for tive effects leading to sub- underreporting are simply stance use and delinquency conjecture. Intervention may be mitigated if more vic- cannot occur without case timizations are recognized identification. earlier and effective intervention is provided. Policy and practice. The NSA research demonstrates Mental health programs the feasibility of gathering designed to reduce common meaningful data directly from psychological problems as- adolescents and families sociated with child and ado- regarding these difficult top- lescent victimization are ics. Unfortunately, trends common, but few include cannot be detected without specific interventions delay- ongoing data collection. Fu- ing the onset of substance ture crime victimization sur- use and reducing substance veys might consider using abuse or delinquency. The methods such as telephone NSA findings may be benefi- surveys, interviews with indi- cial to mental health practi- vidual adolescents, and more tioners as they screen for sensitive screening ques- victimization and consider the tions. Such methodological types of prevention compo- changes may provide a more nents of their victimization detailed picture of adolescent treatment protocols. victimization trends over time. The findings also suggest that proactive, creative com- The NSA results strongly munity programs are needed suggest that victimization to encourage children and and its mental health corre- others to report crimes to lates play a role in the devel- law enforcement. The consis- opment of substance use tent underreporting of vio- and delinquency behavior lence against children and among adolescents. Policies adolescents points to a need that promote the prevention for collaborative efforts of child and adolescent vic- among law enforcement, timization also would pro- criminal and juvenile justice mote the prevention of youth professionals, victim service

13 RESEARCH IN BRIEF / APR. 03

providers, the news media, and Adolescent Victimization in educators, and allied pro- America: Prevalence and Implica- fessionals that focus on en- tions, NCJ 186167) can be obtained from NCJRS (800–851–3420 or couraging victimized youths 301–519–5500). Electronic versions to report crimes and provide of this Research in Brief can be them with appropriate and accessed online at http://www.ojp. safe venues in which to do so. usdoj.gov/nij or http://www.ncjrs.org.

2. In this report, female adolescents are referred to as “girls” and male Notes adolescents are referred to as “boys.” 1. This Research in Brief is a con- solidated version of the full report, 3. American Psychiatric Association, Prevalence and Consequences of Diagnostic and Statistical Manual of Child Victimization: Results From Mental Disorders, Fourth Edition, the National Survey of Adolescents, Washington, DC: American Psychia- Final Report (NCJ 181028). Copies tric Press, 1994. of the full report and an executive summary of the full report (Child

14 The National Institute of Justice is the research, development, and evaluation agency of the U.S. Department of Justice and is solely dedicated to researching crime control and justice issues. NIJ provides objective, independent, nonpartisan, evidence-based knowledge and tools to meet the challenges of crime and justice, particularly at the State and local levels.

NIJ is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime.