This is an archive page. The links are no longer being updated.

Statement on School Violence by Mark L. Rosenberg, M.D., M.P.P.
National Center for Injury Prevention and Control
Centers for Disease Control and Prevention
U.S. Department of Health and Human Services

Good morning. I am Dr. Mark Rosenberg of the Centers for Disease Control and Prevention (CDC), where I direct the National Center for Injury Prevention and Control. The CDC's mission is to promote health and quality of life by preventing and controlling disease, injury, and disability. I am honored to be here today to explore what schools can do to help prevent violence among our youth. I would like to talk with you about (1) the framework we use in public health to address the problem of violence; (2) what we know about violence among our youth and in schools; (3) the characteristics of successful strategies and programs that work in preventing violence; and (4) future directions for moving forward together to address the problem of school violence.

The Public Health Approach to Violence Prevention

In public health, we approach the problem of youth violence by asking four questions:

1. What is the problem? (Surveillance). We collect data on the problem that we can use to better understand the problem and to do something about it. We ask "who, what, where, when, and how did it happen?"
   
   
2. What are the causes? (Risk Factor Research). We seek to discover what puts people at risk or protects them.
   
   
3. What works to help prevent the problem? (Intervention Evaluation). We use the knowledge we have of the pattern of the problem to develop and test interventions that might work to prevent it.
   
   
4. How do you do it? (Program Implementation). We look at how we can accelerate the dissemination of research findings more quickly and effectively. We also explore how we apply the proven effective interventions broadly in the community.

Violence among our youth and in our schools is an urgent problem. Sometimes, the urgency of a problem forces us to take action before we have time to complete all the steps, and we must learn as we go how to refine our efforts and to provide the science base for our programs and programs that others are implementing. Therefore, our progress along this model is not always linear, i.e., following in an orderly fashion from numbers one through four. Congress; schools throughout the Nation; those of us in State, Federal, and local agencies who deal with violence; our partners in the private sector; parents; and others have had to respond at the same time that we are learning what works best. CDC's public health efforts focus on prevention, are science based, and integrate the approaches and findings of many different fields and disciplines. We have learned a number of things in public health that can help.

What Do We Know About the Problem of Youth Violence and Violence in our Schools

Surveillance

It is important for us to begin by taking a closer look at this problem based on what we have learned from surveillance and research. Prior to the tragedies at the schools in Jonesboro, Paducah, Pearl and Springfield, CDC collaborated with the Departments of Education and Justice and the National School Safety Center in a study of violence-related deaths in schools. This is important because it gives us a baseline for comparing and analyzing what has gone on more recently. We looked at any homicide, suicide, or firearm-related death in the United States , from January 1992 through June 1994 that occurred: 1) on the property of a functioning public or private elementary or secondary school, 2) on the way to or from regular sessions at such a school, or 3) while attending or on the way to or from an official school-sponsored event. We found that 105 deaths occurred during the two-year period. More than half of the victims

• (65 percent) were students, 11 percent were teachers or other staff, and the remaining victims
  
  
• (23 percent) were community members killed on school property. The majority of victims
  
  
• (83 percent) were males. The deaths included in this study occurred in 25 different States across the country. They affected primary and secondary schools and communities of all sizes. Fewer than one third (28 percent) of the fatal injuries happened inside the school building, approximately one third (36 percent) occurred on school property but out-of-doors, and about one third (35 percent) occurred off campus.

We are extending the study on violent deaths in schools to examine the time period from July 1994 through June 1998 to determine what differences there may be in violence in schools since the beginning of the study. We have analyzed the data relating to the events in Jonesboro, Paducah, Pearl and Springfield, and are examining how this compares to statistics from all other incidents of lethal violence in schools where multiple deaths occurred since 1992. Our analysis shows that whereas there were only two recorded multiple-death shooting events in schools between 1992 and 1994 (each involving two deaths), preliminary data show that between July 1994 and June 1998, there have been 18 multiple-victim events in schools. In addition to the cases of Jonesboro, Paducah, Pearl and Springfield, there were more than six other schools that experienced violence involving two or three deaths. Many also have been shocked at the profile of those involved: in the Jonesboro, Paducah, Pearl and Springfield shootings, all of the perpetrators were white males with an average age of 13.5 and all of the fatalities were white females with an average age of 14.5. Moreover, all of the perpetrators and victims knew each other. We are extending our analysis in order to determine whether these multiple-death incidents in schools represent an increasing trend, and whether there is an increasing trend in overall school deaths or other aspects of school violence.

During 1992-1994, when the 105 school-related deaths were reported from the initial phase of the CDC study, more than 8,000 children and adolescents were victims of homicide in this country. The unconscionable violence in school settings is an extension of the violence that occurs among children and adolescents in communities throughout our Nation. No school is an island -- violence prevention programs in the schools must include and encompass violence prevention in the community. The effects of school and community violence fall disproportionately among the poor and minorities and take a great toll on their education. Homicide remains the second leading cause of death for young Americans between the ages of 15 and 24 and the leading cause of death for African Americans in this age group. For children ages 10-14 years, homicide is the third leading cause of death. What has changed most dramatically over the past 20 years has been the emergence of fatal youth violence -- children killing children. Although there has been a slight decrease in national youth homicide rates since 1993, the number of young people who die violently remains unacceptably high, and the recent multiple shootings in schools underscores the importance of addressing youth violence.

In addition to the school studies, CDC continually monitors behaviors among youth that have an impact on their health through a surveillance system, the Youth Risk Behavior Survey (YRBS), conducted in collaboration with the States. This survey is administered to high school students in grades 9-12. Part of this survey covers risk behaviors for violence-related injuries and deaths, including weapon-carrying and physical fighting. In 1997 (the latest year for which these data are available), the YRBS showed that during the month preceding the survey: 8.5 percent of students in grades 9-12 carried weapons on school property; 14.8 percent of students had been in a physical fight on school property; more than 36 percent had been in a fight in the community, with 3.5 percent sustaining injuries that required medical treatment; and 4.0 percent of students had missed a day of school because they felt unsafe at school or traveling to and from school.

Risk Factor Research Findings

From analysis of data on Youth Violence, we know that:

• Violence involving youth is more lethal, i.e., a greater proportion of violent interactions among youth result in death. The use of firearms in violent interactions was associated with a dramatic increase in youth homicide between 1985 and 1993. More than 90 percent of the increase in youth homicide rates between 1985 and 1993 was due to homicides committed with firearms. Reacting violently to circumstances is a learned behavior that can be unlearned.
  
  
• Children who witness or experience violence in the home are more likely to become perpetrators or victims of violence later on. This means that children who live in homes where their mother's partner beats or sexually assaults her are at higher risk of being victims of the same violence themselves or of committing violence later on. It is important that our violence prevention efforts at school and in the community recognize and intervene in this cycle.
  
  
• 25 percent of 8^th and 9^th grade male and female students in a recent study, conducted by the University of North Carolina as part of a CDC-sponsored Safe Dates project, indicated they had been victims of physical and psychological dating violence and 8 percent had been victims of rape or other sexual dating violence.

In reviewing research on Youth Violence, we have identified the following risk factors:

Individual factors

• History of early aggression
• Beliefs supportive of violence
• Attributing hostility to others
• Social cognitive deficits

Family factors

• Problem parental behavior
• Low emotional attachments to parent/care givers
• Poor monitoring and supervision of children
• Exposure to violence
• Poor family functioning

Peer/school factors

• Negative peer influences
• Low commitment to school
• Academic failure
• Certain school environments/practices, such as undisciplined classrooms, lax enforcement of school rules and policies, and crowded physical space

Environment/neighborhood factors

• High concentrations of poor residents
• High levels of transience
• High levels of family disruption
• Low community involvement/participation
• Diminished economic opportunity
• Access to firearms

Intervention Evaluation

In their book, Violence in American Schools, Dr. Del Elliott (a prominent youth violence researcher who leads the Center for the Study and Prevention of Violence (CSPV) at the University of Colorado at Boulder), and his colleagues say, "Most of the violence prevention program currently being employed in the schools, e.g. conflict resolution curricula, peer mediation, individual counseling, metal detectors, and locker searches and sweeps, have either not been thoroughly evaluated or have been evaluated and found to be ineffective."At CDC, our primary efforts have been directed toward evaluating these types of programs to build a body of knowledge about programs that are effective.

In fiscal year 1993, Congress provided funds to CDC for research to determine what interventions work to prevent violence among youth, both in schools and in communities, and how best to implement those interventions. The initial thirteen projects, funded for three years with an investment of over $6.9 million, were located in 11 cities and one county and targeted elementary schools, middle schools, and older youth. These projects have targeted predominately urban, high-risk youth and may be applicable to other areas of the country; however, these projects have not been evaluated in rural settings. Results indicate:

• Effective strategies include school-based curricula that emphasize the development of problem-solving, social and communication skills, and anger management. In addition, parenting programs that promote strong bonding between parents and children and that teach parents skills in managing conflict in the family are also very promising.
  
  
• Even theoretically sound strategies do not succeed in schools with chaotic environments or where the administration does not actively support the project. Examples are schools where administrative policies are unclear, communication among school staff or between the school and parents is confused or where teachers are unable to manage their classrooms. Established partnerships between intervention agents (schools, community-based organizations and health departments) work better than newly established efforts.
  
  
• Outcomes varied across age groups whether examining elementary vs. middle vs. high school or even between grades in the same elementary school. It is important to design interventions that are appropriate for the developmental level of the students who will receive them. In addition, outcomes varied by gender for students in the same intervention showing the importance of tailoring the interventions as specifically as possible for girls and for boys.
  
  
• Timing and duration of interventions make a significant difference in the effect of a program. There is wide disparity in the amount of time some children are exposed to interventions. In some cases there is only one class period in the whole year compared to those which occur several hours a week for more than a semester. The emphasis should be on more interventions that have intensive and longer exposure rather than on interventions of shorter, less intense duration.

CDC recently funded a three-year follow-up effort, which will conclude in the Fall of 1999, with some of the most promising projects to see if these results are appropriate for replication and long term application. Below are brief summaries of some of our current projects (including the follow-up projects):

• Peace Builders in Tucson, Arizona, work in elementary schools to reduce physical and verbal aggression by creating a "culture of peace" within the school. Throughout the school year, counselors or other specially trained instructors, using various methods such as modeling, role play, and self-monitoring, teach students to interact socially in a positive way. The main messages are to praise others, avoid insults, seek the advice of wise people and speak up about hurt feelings. The intervention has achieved significant behavioral improvement. Teachers reported an overall decline in individual problem behaviors such as fighting and destruction of property belonging to others. The follow-up study will also assess whether these results will continue to hold up over time as the students progress through middle school.
  
  
• The University of Michigan is evaluating a three-level intervention to prevent violence. Participants are high-risk, urban youth, ages 7-13, residing in high-violence areas of Chicago and Aurora, Illinois. The first level consists of classroom-based training to increase awareness and knowledge about factors that influence peer and other social relationships. The next level includes activities in the first level plus training conducted through small groups and peer relationships for high­risk children. The third level adds a family intervention for the high-risk children and their families. The results show that a combination of classroom, small group, and family interventions significantly reduced aggressive behavior among study participants who had demonstrated higher levels of aggression.
  
  
• In Richmond, Virginia, the Youth Violence Prevention Program is a school-based project to reduce aggressive behaviors among 6^th graders. The 16-session curriculum teaches students how to deal with violence and anger. For example, the students participate in role-playing and practice the conflict resolution skills such as using humor or leaving the scene of a dispute. Some other programs simply discuss, but don't rehearse, these skills. The program also has a peer mediation program that uses a problem-solving approach to reinforce the skills students learn in the curriculum. The results show a significant, 50 percent reduction in fight-related injuries requiring medical attention, 56 percent fewer fights at school, 74 percent fewer incidents of carrying weapons to school, 59 percent fewer suspensions, a lower frequency of threats to hurt a teacher, and significant improvements in self-esteem. Students in the program had fewer incidents of disruptive behavior and defiance of school authority.
  
  
• In Portland, Oregon, Self Enhancement, Inc., provided students from low-income, high-crime neighborhoods in grades 7-9 in the Portland Public Schools with adult mentors and programs that included training in conflict resolution and social skills, peer education in violence prevention, recreational opportunities, and academic tutoring. Students spent at least 1 hour per month interActing with their mentor. Services were provided about 13 hours per week during the school year and 25 hours per week during the summer months. The training drew extensively upon unique cultural foundations and experiences that were relevant to the target population of students. Approximately 120 students from four schools (three middle and one high school) were enrolled in the program. Approximately 200 students from the same school and with similar school performance, behavioral problems, and peer relationships served as the comparison group. After two years, rates of weapon carrying decreased by 65 percent and fights at school decreased by 66 percent; no significant decreases in these behaviors were noted in the control group.
  
  
• In Seattle, Washington, Second Step - A Violence Prevention Curriculum has been implemented for elementary school students. Evaluation of the Second Step program showed a positive impact in reducing aggressive physical acts and increasing pro-social behavior. The effect was modest, but combined with a growing body of evidence from other evaluated interventions, indicates that, just as violence is a learned behavior, it can be unlearned.

Program Implementation

We have put in place some programs that work, and we know some strategies for reducing youth violence. However, we need a more complete response to this problem. Parents, teachers, school Administrators and others need to have ready access to the best, most current, science-based information. The demand for solutions in youth violence prevention is a direct consequence of the increasing realization that youth violence can be prevented. More and more communities and schools are designing and implementing violence prevention programs. To benefit from new information, they will need to know: (1) how to access understandable and practical information on developing violence prevention programs that have demonstrated success elsewhere; (2) how to localize intervention strategies for specific community needs; and (3) how to determine if the violence prevention program is working. Based on our experience in implementing the first 13 projects, CDC published, "Youth Violence Prevention - - Descriptions and Baseline Data from 13 Evaluation Projects," a supplement to the American Journal of Preventive Medicine, and, "Measuring Violence-Related Attitudes, Beliefs, and Behaviors Among Youths: A Compendium of Assessment Tools."We also have produced a manual to help communities develop prevention activities, "Prevention of Youth Violence: A Framework for Community Action," and are in the process of developing recommendations for "best practices" that characterize promising violence prevention strategies. We are providing a copy of these documents to the Subcommittee.

CDC is conducting the Best Practices Project to capture and disseminate new knowledge gained from the CDC/outside experts and other evaluation projects, to the specific community, practitioners and lay people interested in violence prevention. Over the past two years we have consulted with over 70 experts in the field of youth violence prevention and reviewed the results of research to make recommendations on state-of-the-art prevention strategies. Four strategies have been identified that form an effective, manageable approach for communities to implement in their efforts to prevent youth violence: (1) social-cognitive skills training; (2) parent/family interventions; (3) nurse-home visitation; and (4) mentoring. Technical assistance materials on preventing youth violence will be developed for different audiences, including schools, and made available based on this compilation of current knowledge of the state of the art. We anticipate that the first products from this effort will be ready by the end of 1999.

In another effort to disseminate information about what works based on rigorous scientific standards and to provide technical assistance to schools and communities, CDC is supporting the work of the Center for the Study and Prevention of Violence (CSPV), headed by Dr. Del Elliot, through an interagency agreement with the Office of Juvenile Justice and Delinquency Prevention in the Department of Justice. The CSPV operates the Blueprints program to determine effective youth violence prevention activities. The CSPV has identified four scientific criteria which they use to determine effectiveness. The intervention must (1) be theoretically sound; (2) have demonstrated a reduction in violent behavior; (3) have demonstrated a long-term effect; and (4) have been replicated in at least one other site. Blueprints programs use strategies such as mentoring, intensive counseling or therapy for troubled youth, and bullying prevention. The CSPV works with communities and schools to assist them in choosing the most appropriate intervention and provides technical assistance in implementation.

The criteria mentioned above are also consistent with the principles of effectiveness outlined by the Safe and Drug Free Schools Program. To strengthen collaboration between CDC and the Department of Education, CDC has placed a staff person within the Department of Education to work with the Safe and Drug Free Schools Program.

Because of the nature of youth and school violence, much of our work in this area is collaborative. For example, within CDC, there are three Centers that work together on various parts of the public health approach to school violence. In addition to the work of the National Center for Injury Prevention and Control, which I have primarily addressed, CDC's National Center for Chronic Disease Prevention and Health Promotion supports the YRBS and a number of other school health-related activities, and the National Center for Health Statistics works with us on the surveillance of violence against women, including teenagers. Within the Department of Health and Human Services, we are jointly funding evaluation of an early childhood intervention to prevent youth violence with the National Institute of Mental Health at the NIH, working with the Health Resources and Services Administration on school safety guidelines, and will be working with the Substance Abuse and Mental Health Services Administration on the Safe Schools/Healthy Students initiative. In addition to our collaboration with the Department of Education, CDC collaborates with a number of offices in the Department of Justice. We also have initiated a Safe USA partnership that currently involves over 25 key organizations in the public and private sector who have joined together to address injury prevention and control, including violence.

Future Directions for Moving Forward Together

In, Violence in American Schools, Dr. Del Elliott, et al. make the point, "When their costs are high or there are alternatives that are known to be effective, the absence of any evidence of effectiveness is a sufficient basis for challenging the use of these programs or policies. If there are no alternatives known to be effective and the costs are modest, a case may be made for using untested prevention programs for a time, pending a careful evaluation."We believe that the public health approach can help focus investments in school-based violence prevention programs on programs that work -- and, in cases where proven programs are not adaptable or available, this approach can help determine if other programs are making a difference. Throughout my testimony, I have used the public health framework to address the problem of youth and school violence. I would like to return to this framework to set out some future directions that we can take together to address this problem.

Surveillance

• An ongoing system for describing school-associated violent deaths and injuries. We need to monitor school-associated nonfatal injuries as well as violent deaths on an ongoing basis to identify emerging trends. Previous studies have been periodic and require significant effort to update. One possible approach would be to make the school-associated violent death study an ongoing effort. In addition, we are exploring the use of sentinel schools to report non-fatal injuries from violence on a routine basis, similar to CDC's national notifiable disease reporting system for infectious diseases. We could provide information to communities and schools throughout the country to help them in identifying emerging problems and in planning and monitoring the success of their responses.

Risk Factor Research

• Further examination of school-associated violent deaths. We need to examine in further detail the circumstances and background of the school-related, violent deaths described in the school-violence study and compare the information with information from a control group which did not experience this level of violence. Based on this comparison, we can learn what puts kids at risk for violence. If we can gain a greater understanding of these circumstances, we will be able to help schools identify kids at risk and get them help before another tragedy occurs. For example, if we find that homicide offenders are from families in which abusive violence is present, family-based intervention strategies should be explored.

Intervention Evaluation and Implementation

• Evaluation of school-based violence prevention programs and assistance to schools in selecting and implementing successful program. We need to provide more support to schools in identifying and implementing violence prevention interventions of proven efficacy, and we need schools that can serve as demonstration sites for violence prevention programs. In order to do this we need to promote evaluation research to identify effective programs for widespread dissemination and to develop and test new interventions. There needs to be a way to communicate information about good programs systematically. The Best Practices manual that we are developing can help. Many schools have programs that are not based on the best scientific knowledge and are not being implemented as designed. In partnership with the Departments of Education and Justice, we are providing schools with guidance in selecting violence prevention programs. But we are just beginning to fill the needs of the schools.
  
  
• Quality School-based Health Education. Young people must acquire the skills needed to prevent future injuries and violence. To assist State and local educational agencies and schools in promoting safety and teaching students the skills needed to prevent future injuries and violence, CDC, in collaboration with other Federal and national non-governmental organizations, has recently begun developing evidence-based injury and violence prevention guidelines. As a model, the guidelines development process has been successfully employed for the topics of tobacco use prevention, HIV infection prevention, nutrition, and physical activity. It includes an extensive review and synthesis of the literature on effective program components and the creation of an expert panel to guide the process.
  
  
• Parenting Skills. Parenting programs are important because the behavior of adults in the home can have an enormous influence on children. Parenting skills training can make a difference in preventing violence and anti-social behavior in children. Many parents need assistance to develop structured environments, and they need to learn how to talk with their kids about the risks of weapons and fighting.
  
  
• Better Communication Between Parents and Schools. Schools can link parents to programs in the community to help children with behavior problems. We have seen that parent training programs can help, but there is a gap in getting knowledge to the parents. Schools can help make the link.
  
  
• Reach out to high-risk youth. Many youth are not accessible in traditional settings such as schools. We need to support the implementation of efforts to identify and recruit high-risk youth into programs intended to reduce the risk of violence.