Welcome to the California Department of Social Services

Right Column

Youth Violence Prevention Article

by, Melanie Barker, MPH/MSW, Education Coordinator, Center for Child Protection

This is the first in a three-part series including defining the youth violence problem; aggressive behavior in adolescents, including prevention programs; and politics and policy for youths in the juvenile justice system.


Less than two years have passed since a school shooting at Columbine High in Littleton, Colorado left 15 dead and many more injured. Recently, a similar shooting at Santana High in Santee, California left 2 dead and 13 wounded. Within three weeks of Santana’s school shooting, another shooting occurred within the same school district at Granite Hills High. While school shootings are rare, such high-profile tragedies spark public fear and often fuel inappropriate social policies. With limited or no empirical evidence, policymakers have focused preventive resources on increasing physical security, hiring school security, and developing tactical response plans once a shooting has occurred (Reddy, et al 2001). Others have focused efforts on campaigning to have violent youth offenders transferred from the juvenile system into adult criminal courts or prisons. These policies have not been empirically tested, and deflect attention from prevention activities grounded in an understanding of the causes of this violence.

Defining the Problem

Despite these horrific school shootings, crimes in schools are on the decline. In 1993, there were 164 crimes for every 1,000 students ages 12-18. In 1996, the number decreased to 128 per 1000 (Hanson-Harding, 1999). According to the CDC, as cited by Hanson-Harding (1999), less than 1% of all homicides among children ages 5-19 happen in or around school grounds. According to Vossekuil, et al (2000), "school is one of the safest places for our nation’s children." The Secret Service Safe School Initiative (Vossekuil, et al, 2000) distinguishes "targeted violence" in school, referring to any violent incident where a known perpetrator selects a specific target prior to the attack, such as incidents having occurred in Colorado and California, from other types of youth violence. This distinction is an important one, as different types of violence have different antecedents and require different assessment and intervention methods (Reddy, et al, 2001).

According to Reddy, et al (2001), those youth at risk for targeted violence may not have many of the risk factors generally associated with juvenile delinquency. The etiology and intervention may differ significantly. Cornell, et al (as cited in Reddy, 2001) found that juveniles referred for evaluation after having committed homicide were less likely to have prior mental histories, arrests, poor school adjustment, or placements in juvenile facilities, vis-à-vis those juveniles referred for evaluation after having committed larceny. Findings revealed a substantial heterogeneity among juvenile homicide offenders. Reddy, et al (2001) posit that juvenile offenders of targeted school violence may differ considerably not only from non-violent crime youth offenders, but also from those who commit other acts of homicide. Hence, they argue that for purposes of preventing school violence, assessment requires identifying whether a particular student poses a threat to another identified individual(s) at school, as opposed to assessing whether an individual poses an increased risk for committing some act of aggression. Further, any inquiry should also include an investigation into a student’s grievances about school or potential targets.

Key Findings from the Safe School Initiative

In collaboration with the U.S. Department of Education, the U.S. Secret Service developed and implemented the Safe School Initiative project in an effort to bring their expertise in research and prevention of targeted violence to address the problem of school violence, providing accurate information to those multidisciplinary stakeholders challenged with school safety. To that end, The Secret Service National Threat Assessment Center studied 37 school shootings, involving 41 assailants who were current or recent students at the school, and where the assailant(s) chose the school for a particular purpose, rather than as an opportunistic site. Exclusion criteria included gang- or drug-related shootings, as well as those related to an interpersonal or relationship argument. For each occurrence, researchers reviewed primary source materials, including investigative, school, court, and mental health records. Information gathered included facts related to the assailant’s motives and development of an idea and plan to attack a particular target.


  • Targeted school violence incidents are rarely impulsive. There is a visible process of thinking and behavior that results in an attack, typically comprised of an idea to harm, a plan of attack, and a motive for revenge.
  • In most incidents, the attacker told someone about his/her idea and/or plan prior to the event. In almost all cases, the person told was a peer–a friend, schoolmate, or sibling. In only two cases did such a peer notify an adult of the idea or plan.
  • There is no accurate or useful profile of "the school shooter." The attackers varied with respect to age, ethnicity, family constellation, academic performance, quality of peer relationships, and behavioral histories.
  • Most attackers had previously used guns and had access to them–in most cases, the gun(s) came from the attacker’s home or that of a relative, and in some cases the weapons had been gifts to the attackers from their parents.
  • Most shootings were not resolved by law enforcement intervention. In over 50% of the cases, faculty, other students, or the attacker stopped the assault. In only three cases did law enforcement discharge any weapons during the assault.
  • In almost 50% of the cases, other students were involved in some capacity–influencing or encouraging the assailant in some manner.
  • In several cases, bullying played a pivotal role in the assault. In over 2/3 of the cases, the attackers felt bullied, persecuted, threatened, attacked, or injured by others preceding the attack. Several had experienced severe and longstanding bullying and harassment, and in these cases, the bullying experience was a motivating factor for the attack.
  • Most attackers had demonstrated some behavior that warranted alarm or a need for intervention. Almost 50% of the attackers had histories of depression, and nearly 75% had demonstrated suicidal ideation, gestures, or attempts.


  • Any inquiry should also include an investigation into grievances about school or potential targets that a student may have.
  • It is imperative to differentiate between making and posing a threat. The former refers to expressing intent to harm, and the latter refers to engaging in behaviors that indicate planning and preparation for an attack. Threat assessment and intervention also requires eliminating barriers that prevent individuals who have information of concern from disclosing it.
  • Profiling is not an effective means for identifying high risk students nor for assessing the risk that a particular individual may pose for school-based targeted violence–the heterogeneity of personality and social characteristics of shooters clearly speaks to the possibility of over-identification of students who have characteristics in common with previous assailants but clearly pose no threat of violence. In contrast, an effective method for identification is to focus on a student’s behaviors and communications to determine if a student is actually planning an attack.
  • Efforts must focus on determining a student’s access to weapons. Inquiries should include investigating a student’s history of weapon use, as well as access to and communications about weapons.
  • Investigations should include information from a student’s friends and classmates, and information on peer influences that may be operating in the attack planning/preparation.
  • Strong efforts need to be made to eliminate bullying in our schools. See Bully Online, a UK National Workplace Bullying Advice Line website, available at http://www.successunlimited.co.uk/serial.htm
  • Prevention requires early intervention–it is important not only to identify students who are planning an attack, but also to identify appropriate interventions for those children who are already in trouble. Inquiries should direct attention to a student’s difficulty coping with losses or perceived failures.

Some Indicators of Troubled Youth

  • Lack of school interest
  • Lack of age-appropriate anger management skills
  • Disregard for or refusal to follow rules
  • Cruelty to animals
  • Bleak or violent artwork/writings depicting isolation or anger
  • Obsessions with violent themes or weapons
  • Depressed mood or mood swings
  • History of bullying or being bullied
  • Isolation from family or friends
  • Involvement with or interest in gangs

(US Dept. of Education and US Dept. of Justice, 1996).

Troubled Youth–More Facts

Other key indicators, such as arrest and victimization data, also depict a decline in youth violence since the 1993 epidemic peak. Conversely, gleaning information from youth self-reports of violent behaviors indicate a prevalence approximating 1993 levels, which is unacceptably high (Youth Violence: A Report of the Surgeon General, available at www.surgeongeneral.gov/library/youthviolence/chapter1/sec2.html).

Findings of an Office of Juvenile Justice and Delinquency Prevention (OJJDP)- funded study of children in Rochester, New York revealed that there was a 24% increased probability that children who had been victims of violence within their families were likely to report violent behavior as adolescents compared to those who had not been maltreated in childhood (Marans and Berkman, 1997).

Far too often children experience victimization, as illustrated in the following statistics:

  • In 1994, approximately 2.6 million youth ages 12-17 were victims of crime–simple and aggravated assaults, rape, and robbery (Bureau of Justice Statistics, National Crime Victimization Survey, as cited in Marans and Berkman, 1997).
  • In a study conducted at Boston City Hospital, 1 out of 10 children seen by their primary care clinic had witnessed a shooting or stabbing prior to age 6–50% of these incidents occurred in home and 50% in the street. These children averaged 2.7 years in age (Taylor, et al as cited in Marans and Berkman, 1997).
  • In a survey of 5th and 6th graders in Washington, D.C., 31% reported having witnessed a shooting; 17% reported witnessing a homicide; and 23% reported having seen a dead body (Richters and Martinez, as cited in Marans and Berkman, 1997).

Children’s exposure to violence and child maltreatment is highly correlated with depression, anxiety, posttraumatic stress, anger, substance abuse, and poor school performance (Garbarino et al; Martinez and Richters; Singer et al; and Ciccetti and Carlson, as cited in Marans and Berkman, 1997). These symptoms are clear indicators, however, the Surgeon Generals’ report states that abuse and neglect are "relatively weak predictors of violence," and "most children who are abused and neglected will not become violent offenders during adolescence." Conversly, the role of family and parental factors, influenced by social learning, modeling, and identification with the aggressor, with respect to aggressive, physically abusing, and criminal parents is well cited in the literature as predictive of aggressive/antisocial behavior (Fry 1988; Hall and Cairns 1984; West and Farrington 1977; as cited in Shaw & Campo-Bowen, 1995).

Next quarter, look for part two in the series on aggressive behavior in adolescents .


Bureau of Justice Statistics (2001). Homicide trends in the U.S. U. S. Department of Justice Bureau of Justice Statistics [ On-Line  ].

Cooper, W.O., Lutenbacher, M., Faccia, K. (2000). Components of effective youth violence prevention programs for 7- to 14-year-olds. Archives Pediatric Adolescent Medicine, 154 , 1134-1139.

Fanning, K. (2001) Kids in the legal system. School violence: Staying safe.

Hanson-Harding, A. (1999). Ending School Violence. Junior Scholastic [ On-Line  ].

Marans, S. & Berkman, M. (1997). Child development–community policing: partnership in a climate of violence. Office of Juvenile Justice and Delinquency Prevention Bulletin, March, 1-8.

Reddy, M., Borum, R., Vossekuil, B., Fein, R., Berglund, J. & Modzeleski, W. (2001) Evaluating risk for targeted violence in schools: Comparing risk assessment, threat assessment, and other approaches. Psychology in the Schools, 38 (2), 157-172.

Shaw, J.A. & Campo-Bowen, A (1995). Aggression. In Sholevar, G. Pirooz (Ed.), Conduct Disorders in Children and Adolescents . Washington, D.C., American Psychiatric Press.

Spoth, R.L., Redmond, C., & Shin, C. (2000). Reducing adolescents’ aggressive and hostile behaviors. Archives Pediatric Adolescent Medicine, 154 , 1248-1257.

U.S. Depts. of Education and Justice (1996). Creating Safe and Drug-Free Schools: An Action Guide. Washington, D.C.
On-line at www.ed.gov/about/offices/list/osdfs/index.html  or www.ncjrs.gov/spotlight/school_safety/Summary.html

Vossekuil, B., Reddy, M., Fein, R., Borum, R., & Modzeleski, W. (2000) U.S.S.S. Safe School Initiative: An Interim Report on the Prevention of Targeted Violence in Schools . Washington, DC: U.S. Secret Service, National Threat Assessment Center