School Safety: Promising Initiatives for Addressing School Violence
(Letter Report, 04/25/95, GAO/HEHS-95-106).

Many schools throughout the United States are struggling with rising
levels of youth violence.  Schools have adopted a broad range of
solutions to curb violence.  The four programs GAO visited--in
California, Ohio, and New York--are examples of some of the promising
approaches schools have initiated to address violence.  Research
suggests that the most promising school-based violence-prevention
programs involve at least some of seven key characteristics, including a
comprehensive approach, starting early, and involving parents.  Although
few prevention programs have been evaluated, some federal agencies are
now funding evaluations to examine various violence-prevention program
approaches.  The results, which should be available in three to five
years, will help determine which programs work best at curbing violence.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-95-106
     TITLE:  School Safety: Promising Initiatives for Addressing School 
             Violence
      DATE:  04/25/95
   SUBJECT:  Juvenile delinquency
             Crime prevention
             Crimes or offenses
             Interagency relations
             Educational programs
             Education program evaluation
             Students
             Education or training
             Public schools
IDENTIFIER:  Anaheim (CA)
             Dayton (OH)
             New York (NY)
             Paramount (CA)
             Go to High School, Go to College Project (Atlanta, GA)
             Beacons Initiative (New York, NY)
             SMART Program (Anaheim, CA)
             PACT Program (Dayton, OH)
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to the Ranking Minority Member, Subcommittee on Children and
Families, U.S.  Senate

April 1995

SCHOOL SAFETY - PROMISING
INITIATIVES FOR ADDRESSING SCHOOL
VIOLENCE

GAO/HEHS-95-106

School Safety


Abbreviations
=============================================================== ABBREV

  ADHD - Attention Deficit Hyperactivity Disorder
  ATGM - Alternatives to Gang Membership
  AUHSD - Anaheim Unified High School District
  CDC - Centers for Disease Control and Prevention
  ESR - Educators for Social Responsibility
  IPS - Incident Profiling System
  JJDP - Juvenile Justice Deliquency Prevention
  NIJ - National Institute of Justice
  NIH - National Institutes of Health
  NIMH - National Institute for Mental Health
  NSBA - National School Boards Association
  NSSC - National School Safety Center
  OJJDP - Office of Juvenile Justice Deliquency Prevention
  PACT - Positive Adolescents Choices Training
  RCCP - Resolving Conflict Creatively Program
  SMART - School Management and Resources Training

Letter
=============================================================== LETTER


B-256397

Letter Date Goes Here

The Honorable Christopher J.  Dodd
Ranking Minority Member
Subcommittee on Children and Families
United States Senate

Dear Senator Dodd: 

The incidence of school violence--searches for weapons, shootings,
gang activity, fighting, and other instances of disruptive
behavior--has risen to unacceptable levels.  Nearly 3 million crimes
occur in or near the nation's 85,000 public schools every year\1
--about one every 6 seconds.  Further, the Centers for Disease
Control and Prevention (CDC) estimates that 105 violent
school-related deaths occurred during the past 2 years.  Such
conditions create environments that impede teaching and learning and
make parents reluctant to send their children to school.  In fact, a
recent national survey listed school order and safety as parents' top
priority--right along with teaching the basics (reading, writing, and
arithmetic).\2 Ending the epidemic of youth violence will not be
easy.  According to violence-prevention experts, no simple solution
exists.  However, individuals, groups, and community organizations
are beginning to work with schools to develop programs aimed at
stopping youth violence before it starts. 

This report responds to your request for information about some of
the programs used by schools to curb violence.  As agreed with your
staff, we (1) examined four promising prevention programs, obtaining
teacher and student views on these efforts; and reviewed evaluation
data on these programs; (2) identified key characteristics typically
associated with promising school-based violence-prevention programs;
and (3) identified federally sponsored evaluations of
violence-prevention programs operating in schools. 

To answer your questions, we reviewed the research literature on
youth violence-prevention, and interviewed violence-prevention
program directors, federal agency officials, and acknowledged
experts.  In addition, we visited four programs in four cities that
have shown signs of success.  During visits, we talked with students,
teachers, and school administrators.  In addition, we interviewed
federal agency officials concerning their efforts to evaluate
violence-prevention programs.  (See app.  VII for more details on our
scope and methodology). 


--------------------
\1 National Crime Victimization Survey, Bureau of Justice Statistics,
U.S.  Department of Justice (1991). 

\2 "First Things First:  What Americans Expect From the Public
Schools," Public Agenda, New York, New York (1994). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

The four school-based violence-prevention programs that we visited
all show initial signs of success.  Schools in the Anaheim,
California, program, stressing school management and order issues,
reported reductions in the incidents of student fighting, graffiti,
and defiance of authority.  Paramount, California, schools uses an
antigang curriculum to reduce gang membership among students who
participated in the program.  The Dayton, Ohio, program provides
students with social skills and anger-management training.  According
to program officials, student participants had fewer juvenile court
charges than a comparable group of students.  Similarly, a New York
City program has used conflict-resolution and peer-mediation training
to reduce student fighting.  For example, 71 percent of teachers
observed less physical violence among student participants. 

Violence-prevention literature and experts consistently associate at
least seven characteristics with promising school-based
violence-prevention programs.  These characteristics are (1) a
comprehensive approach, (2) an early start and long-term commitment,
(3) strong leadership and disciplinary policies, (4) staff
development, (5) parental involvement, (6) interagency partnerships
and community linkages, and (7) a culturally sensitive and
developmentally appropriate approach.  For example, teaching students
early about making positive choices and linking school-based programs
to community groups, such as law enforcement or service agencies, are
approaches associated with promising programs. 

Although few violence-prevention programs have been evaluated,
efforts are underway to identify successful approaches for curbing
school violence.  For example, during fiscal years 1992 and 1993, we
identified 26 federal grants (approximately $28 million) that help to
evaluate the effectiveness of various school-based
violence-prevention programs.  In addition, recent actions to
increase collaboration among federal agencies could also enhance
efforts to identify promising programs by providing opportunities for
sharing expertise and resources. 


   BACKGROUND
------------------------------------------------------------ Letter :2

During the past decade, the Congress has expressed its concern for
student safety and school security through numerous hearings and the
introduction of several bills aimed at reducing youth violence.  In
1994, a desire to help make schools safe prompted the Congress to
pass two bills explicitly targeting school violence: 

The Safe Schools Act of 1994 authorizes the Secretary of Education to
make grants to local school districts with high rates of youth
violence.  Schools may use these grants to support educational
activities to reduce violence and promote safety.  For fiscal year
1994, approximately $20 million was appropriated for this program. 

The Safe and Drug-Free Schools and Communities Act of 1994 authorizes
the Secretary of Education to make grants to states to prevent
violence in and around schools as well as to deter the use of illegal
drugs and alcohol.  Allowable activities include violence-prevention
and education programs for students, training and technical
assistance, and developing comprehensive violence and drug prevention
programs that involve parents and coordination with community groups
and agencies.\3 Fiscal year 1995 appropriations for this program were
approximately $482 million.\4

In addition, in 1994 the Congress passed the Family and Community
Endeavor Schools Act, which established the Schools and Community
Endeavor Schools Program and the Community Schools Youth Services and
Supervision Program.  This act authorizes the Departments of
Education and Health and Human Services to provide grants to improve
the overall development of at-risk children in communities with
significant poverty and violent crime.  Allowable activities include
developing after-school programs that provide homework assistance and
educational, social, and athletic activities.  The fiscal year 1995
appropriation for the Schools and Community Endeavor Schools Program
was $11 million, while The Community Schools Youth Services and
Supervision Program received $25.6 million. 


--------------------
\3 Formerly the Drug-Free Schools and Community Act, in 1994 its
authorization was expanded to include violence-prevention as a key
program element. 

\4 The Department of Education is scheduled for disbursement in June
1995.  However, in March 1995, all funds were proposed for rescission
by the House Committee on Appropriations. 


      SCHOOL-BASED
      VIOLENCE-PREVENTION PROGRAMS
---------------------------------------------------------- Letter :2.1

Schools use a wide variety of educational and noneducational
approaches and programs to address violence.  Many school-based
violence-prevention programs operate under the premise that violence
is a learned behavior.  In general, these programs focus on primary
prevention; that is, they seek to prevent violence before it occurs. 
Although school-based violence-prevention programs and strategies
vary, most fall within the three broad categories listed below. 

Educational and curricula-based:  These programs seek to teach
students the skills to manage their behavior and resolve conflict
nonviolently.  Programs in this category focus on conflict
resolution, gang aversion, social skills training, mentoring, and
law-related education. 

Environmental modification:  These programs focus on either the
social or physical environment.  Programs aimed at improving
students' social environment include home visitation and after-school
recreational and academic activities.  Programs that seek to modify
students' physical environment include installing metal detectors and
gates limiting access to building entrances and exits. 

School organization and management:  These programs focus on
establishing school discipline policies and procedures that pertain
to student behavior, creating alternative schools, and developing
cooperative relationships with police and other government agencies. 
Legal efforts to regulate the use of and access to weapons in schools
also fall into this category. 

A survey of its members by the National School Boards Association
(NSBA) showed the wide variety of violence-prevention programs
operating in schools.\5 According to the 1993 report, Violence in the
Schools:  How America's School Boards Are Safeguarding Our Children,
the responding school districts implemented more than 750 different
violence-prevention programs.  For reporting districts, overall
responses ranged from establishing alternative schools or programs
for disruptive students (66 percent) to implementing
conflict-resolution and peer-mediation training (61 percent) to
developing safe havens (alternative after-school programs) for
students (10 percent).  (See app.  I for a complete listing of school
district responses.)

In a 1993 report, The Prevention of Youth Violence:  A Framework for
Community Action, the CDC highlighted several community programs
designed to prevent youth violence, including those with activities
closely linked to schools.  For example, according to program
descriptions: 

In Atlanta, Georgia, the Go to High School, Go to College project has
paired 100 successful older men with adolescent African-American
males at four Atlanta high schools and one middle school.  Each
mentor meets weekly with a student who is struggling academically,
has discipline problems, or is at risk of dropping out of school. 
The mentors are provided with a 40-page curriculum of instructions
and ideas.  Mentors strive to increase the students' self-esteem and
improve their grades.  A local fraternity chapter provides
scholarships to students in the program who qualify and want to
attend college. 

In New Haven, Connecticut, the public school system in collaboration
with the Yale School of Psychology provides training in social skills
in all the district's middle schools.  The curriculum emphasizes
self-control, stress- management, problem-solving, decision-making,
and communication skills.  Once students have learned a general
problem-solving framework, they apply their critical-thinking skills
to specific issues, such as substance use. 

In Oakland, California, Teens on Target, a peer education and
mentoring group, was formed after two junior high students were shot
in school by other students.  Teens on Target grew out of a task
force, made up of a coalition of elected officials, parents, and
school and community agency representatives, who felt that students
would do a better job of dealing with the youth violence problem than
adults.  Selected high school students are trained in an intensive
summer program to be violence-prevention advocates, particularly in
the areas of guns, drugs, and family violence.  These students become
peer educators to other high school students and mentors to younger
students in the middle and elementary schools. 

Another innovative program--the Beacon Initiative--operates in New
York City, where about 36 schools now stay open 7 days a week from
early morning until late evening--providing "one stop shopping"
services such as counseling, tutoring, recreational activities,
vocational training, and a safe place for kids to "hang out."


--------------------
\5 The NSBA surveyed more than 2,000 national affiliate school
districts, of which 729 school districts responded.  The results are
not generalizable. 


   FOUR PROMISING PROGRAMS
------------------------------------------------------------ Letter :3


      PROGRAMS SEEK TO STOP
      VIOLENCE BEFORE IT STARTS
---------------------------------------------------------- Letter :3.1

Creating an orderly and disciplined school environment, free of
violence, is essential for learning to take place.  Consequently,
many schools and communities across the nation are working together
to develop solutions to prevent violence.  Although school violence
is a challenging issue, some school-based prevention programs appear
to be promising approaches for curbing violence.  Table 1 summaries
the key features of the four programs that we visited. 



                                     Table 1
                     
                        Overview of Four Programs Visited

               Star
                  t  Type of
Name           date  program        Program scope  Funding        Partnerships
-------------  ----  -------------  -------------  -------------  --------------
Anaheim,       1982  Incident       26,000         School         Local police,
California:          reporting      students       district and   parents, and
School               system         districtwide   National       business
Management                                         Institute of
and Resource                                       Justice (NIJ)
Teams

Dayton,        1989  Social skills  190 students   Federal,       Wright State
Ohio:                and anger      in one middle  state, and     University and
Positive             management     school per     private        parents
Adolescents                         year           sources
Choices
Training

New York, New  1985  Conflict       70,000         School         Parents and
York:                resolution     students       district,      community-
Resolving            and peer       districtwide   private        based
Conflict             mediation                     sources, and   organization
Creatively                                         Centers for    (Educators for
Program                                            Disease        Social
                                                   Control and    Responsibility
                                                   Prevention     )

Paramount,     1982  Antigang       Districtwide   City funds     Parents and
California:          curriculum     curriculum                    city
Alternatives                        for 2nd, 5th,
to Gang                             and 7th
Membership                          graders
--------------------------------------------------------------------------------
Each program has received national recognition for its innovative
approach in addressing school violence and illustrates the types of
partnerships schools and communities have formed to curb violence. 
Each is summarized below. 

Anaheim, California's SMART program has been applauded by federal
officials as a key effort in addressing school management and order
issues.  According to program officials, SMART has been disseminated
across the country as a model school management program that allows
administrators to determine at the push of a button the number and
location of policy violations, offenses, and crimes committed in each
school. 

The SMART program operates districtwide, using a computerized data
collection system to identify and address school and law violations. 
SMART teams analyze data from the system to identify and develop
solutions to discipline problems.  A key SMART program element is its
focus on school problems, not problem schools.  (See app.  II for
more details on this program.)

Dayton, Ohio's PACT program has received wide recognition as a model
program and valuable resource for addressing violence among
African-American youth by several national organizations and leaders
in the field of violence-prevention.  Program officials describe PACT
as a culturally sensitive training program developed specifically for
middle-school-age African-American youth to reduce their
disproportionate risk of becoming victims or perpetrators of
violence.  The program especially addresses the problem of violence
that involves loss of control between family, friends, or
acquaintances and represents the greatest threat to adolescents. 

PACT builds on research that suggests that prevention programs are
most successful for economically disadvantaged and minority youth
when developed with sensitivity to racial, ethnic, and cultural
issues.  The program's training materials are based on the rationale
that adolescents who lack skill in such areas as communication,
negotiation, and problem-solving have a limited range of alternatives
with which to solve interpersonal problems.  Consequently, PACT
provides structured training in specific behavioral aspects of social
skills that enhance the capacity of students to form and maintain
violence-free relationships.  (See app.  III for more details on this
program.)

New York, New York's RCCP program is widely regarded as one of the
most promising violence-prevention programs among public health
experts.  RCCP is a school-based program in conflict resolution and
intergroup relations that provides a model for preventing violence
and creating caring schools that are conducive for learning. 
According to the program officials, RCCP teaches students that they
have many choices besides passivity or aggression for dealing with
conflict, gives them the skills to make those choices real in their
own lives, increases their understanding and appreciation of their
own and others' cultures, and shows them that they play a powerful
role in creating a peaceful world. 

The program's primary strategy for reaching young people is
professional development of the adults in their lives--principals,
teachers, and parents.  RCCP works intensively with teachers,
introducing them to the concepts and skills of conflict resolution
and diversity.  Through ongoing staff development, teachers are
supported as they teach these concepts and skills in an ongoing way
to their students.  (See app.  IV for more details on this program.)

Paramount, California's ATGM program has been widely replicated and
officials have responded to almost 1,000 requests for information
about the program.  Cited as an exemplary gang membership prevention
program in several journals, the program has been visited by
individuals from Ireland, New Zealand, and South Africa, according to
program officials.  The ATGM program seeks to reduce the number of
gang members and their destructive actions within the community by
teaching students the harmful consequences of a gang lifestyle.  (See
Appendix V for more detailed information.)

ATGM's approach is based on the belief that interest in gangs begins
at a young age and that a successful antigang program must reach
students early.  Consequently, the program targets students (most of
whom are Latino) in the 2nd, 5th, and 7th grades.  The program uses
three approaches to achieve its objectives:  (1) an elementary school
antigang curriculum, (2) an intermediate school follow-up program,
and (3) neighborhood meetings with parents and residents.  (See app. 
V for more details on this program.)


      STUDENTS AND TEACHERS
      BELIEVE PROGRAMS MAKE A
      DIFFERENCE
---------------------------------------------------------- Letter :3.2

Nearly all discussion group participants at the four programs'
schools that we visited said their violence-prevention programs were
useful, helped de-escalate conflict, and taught important coping
skills that students need to make positive choices.  Most discussion
group participants agreed that students often lack the skills needed
to make productive and positive decisions.  For example, several
students who participated in conflict-resolution and anger-management
programs said that before their training, they had not considered
resolving disagreements and conflicts without fighting.  One student
told us: 

     "I didn't know any other way to resolve my problems besides
     fighting.  Now instead of fighting, I work it out and talk about
     it.  I've got some options."

In addition, discussion group participants generally agreed that
programs empowered students and helped them avoid violence by
teaching them to (1) identify and solve problems through
role-playing, group discussions, mediation, and other learning
strategies, such as videos; and (2) make positive decisions,
recognize that they have choices, and understand the consequences of
their actions (for example that fighting, joining gangs, or engaging
in other nonproductive behavior may result in suspension from school,
personal injury, or death). 

Students and teachers we talked with expressed concern that
disruptive behavior and violence prevent meaningful learning.  They
told us that disruptive students interrupt the learning process by
distracting other students and the teacher.  According to one
teacher: 

     "Sometimes, teachers can't teach because they've got to stop and
     deal with hostile situations.  .  .  we're teaching students
     skills they should learn at home."

In addition, participants expressed concern that moneys intended for
instructional materials, staff development, and other educational
needs were spent on security efforts. 

Although students and teachers perceived their programs as
successful, they unequivocally stated that no one program or approach
could curb school violence.  They generally agreed that curbing
school violence requires a combinations of approaches.\6 In addition
to the key characteristics discussed later, students and teachers
suggested strengthening efforts to curb violence by

increasing the availability of violence-prevention programs,

providing after-school activities (especially sports),

reducing class sizes,

addressing physical child abuse and neglect,

providing services for at-risk families,

establishing a uniform dress code,

penalizing students who watch fights,

reducing easy access to guns, and

improving economic conditions. 


--------------------
\6 These comments are consistent with violence and drug education
research literature that suggest that comprehensive approaches that
involve parents and the community as well as classroom instruction
and counseling programs are more likely to achieve desired changes. 


      PRELIMINARY EVALUATIONS SHOW
      SIGNS OF SUCCESS
---------------------------------------------------------- Letter :3.3

The authors of the preliminary evaluations that we reviewed concluded
that the four programs showed initial signs of success because
student participants' attitudes and behaviors had changed.  Reported
participant changes included:  (1) new attitudes toward violence and
gang membership; (2) less disruptive behavior, including fewer
fights; and (3) less contact with the criminal justice system.  Based
on pre- and post-tests of knowledge and attitudes, behavioral
observations, self-reporting, and in some cases tracking student
behavior, the evaluations reported the following signs of success: 

Anaheim, California's SMART--The number of police activity reports
filed dropped from 189 during the spring semester of 1993 to 93
during the fall semester of 1993, according to SMART's incident
profiling system.  In addition, student offenses dropped or stayed
the same in all but nine categories--despite increases in crime
within the community. 

Dayton, Ohio's PACT--For the 1992 school year, first semester PACT
participants had a 50-percent reduction in incidents of physical
aggression, while nonparticipants (the control group) had a
25-percent increase.  Similarly second semester PACT participants had
a 53-percent reduction, while nonparticipants had a 56-percent
increase. 

New York City's RCCP--A 1988-89 school year evaluation of RCCP
teachers revealed that (1) 66 percent observed less student
name-calling and fewer verbal put-downs, (2) 89 percent agreed that
the mediation program had helped students take more responsibility
for solving their own problems, and (3) 71 percent reported that
students demonstrated less physical violence. 

Paramount, California's ATGM--After participating in the program, 90
percent of the students responded negatively to the idea of joining a
gang.  Based on a follow-up survey, most students maintained that
response a year later.  Further, a 1993 cross-check of the names of
more than 3,500 past participants with the names of approximately
1,600 known gang members found only a 4-percent match. 

Collectively, the evaluations showed high levels of enthusiasm and
support for the programs.  Although these results are promising, the
limited nature of the evaluations prevent concluding that
improvements in student behavior resulted from program participation. 
Determining whether these programs reduce violence among students
over the long term requires more carefully designed evaluations that
focus on the programs' actual impact on behavior.  Such evaluation
design matters are discussed further below. 


      PRELIMINARY EVALUATIONS NEED
      STRONGER DESIGNS
---------------------------------------------------------- Letter :3.4

A general consensus exists that while preliminary evaluations of
violence-prevention programs are a useful starting point, most lack
the methodological rigor needed to determine their effectiveness.  To
improve the usefulness of future evaluations, greater emphasis should
be placed on designing stronger impact or effectiveness studies.\7
Design issues requiring particular attention include:  sampling
techniques, longitudinal assessment, random assignment, and
collection of impact (fewer fights) and outcome measures (reduced
injury). 

To determine whether violence-prevention programs really cause
observed behavioral change, evaluators should compare the outcomes
achieved by students in randomly assigned treatment and control
groups.\8 For example, concerning the Paramount, California, ATGM
program cross-checking mentioned earlier, how many (or how few)
students would have joined gangs in the absence of the program is not
clear.  Because the evaluation design did not use random assignment,
linking changes in student behavior to program participation is
difficult.  Finally, longitudinal follow-up on students for 4 to 5
years after participation (or until they leave school) would show to
what extent initially observed effects persist over time. 

The program officials that we talked with acknowledged the need for
stronger evaluation designs that focus on program effectiveness. 
However, they said that the high cost of program evaluation, coupled
with a lack of skilled research and evaluation staff, generally
precluded developing stronger designs.  They said that implementing
programs (which many perceive as doing something about the problem of
violence) has more prestige than evaluating programs.  Consequently,
although funders want data on participant outcomes, they focus on
program implementation rather than evaluation. 

According to program officials, doing impact evaluations with
stronger designs depends on obtaining grants or private funds
specifically for that purpose.  For example, PACT has been able to
develop comparison groups and has longitudinal outcomes based on
funding received from private sources for its evaluation efforts. 
However, at the time that we completed our work, only one of the
other four programs we visited had received such funding.  In
September 1993, the National Center for Injury Prevention and Control
at CDC awarded the New York RCCP program a 3-year, $729,000 grant to
conduct a detailed impact evaluation.  The three-pronged evaluation
will include (1) a longitudinal process and impact study, (2) an
in-depth observational and qualitative study, and (3) development of
a management information system. 


--------------------
\7 Impact evaluations focus on program effectiveness.  Impact
evaluations are methodologically rigorous studies that use scientific
research methods to estimate to what extent participant outcomes (for
example, reductions in the prevalence of violence) occur because of
program participation. 

\8 True experimental design requires random assignment to treatment
and comparison group.  Although this is an analytically strong
methodology, it can be costly and require administrative control over
the program.  In some circumstances, quasi-experimental designs may
be adequate when a true experimental design is not feasible. 


   PROMISING PROGRAMS INCORPORATE
   SEVEN KEY CHARACTERISTICS
------------------------------------------------------------ Letter :4

Conclusive evidence concerning the effectiveness of specific programs
is not yet available.  Nevertheless, some indications of promising
program characteristics exist, especially in school settings.  Based
on our literature review, discussions with school safety experts, and
visits to violence-prevention programs, we identified seven
characteristics associated with promising school-based
violence-prevention programs (see fig.  1). 

   Figure 1:  Promising Programs
   Incorporate Seven Key
   Characteristics

   (See figure in printed
   edition.)

Comprehensive approach:  Recognizes violence as a complex problem
that requires a multifaceted response.  Consequently, these programs
address more than one problem area and involve a variety of services
that link schools to the community. 

Early start and long-term commitment:  Focuses on (1) reaching young
children to shape attitudes, knowledge, and behavior while they are
still open to positive influences, and (2) sustaining the
intervention over multiple years (for example, from kindergarten
through 12th grade). 

Strong leadership and disciplinary policies:  Show strong leadership
at the school level.  Principals and school administrators need to
sustain stable funding, staff, and program components and most
important, they must collaborate with others to reach program goals. 
In addition, student disciplinary policies and procedures are clear
and consistently applied. 

Staff development:  Provides training for key school administrators,
teachers, and staff that equips them to handle disruptive students
and mediate conflict as well as understand and incorporate prevention
strategies into their school activities. 

Parental involvement:  Seeks to increase parental involvement in
school efforts to reduce violence by providing training on
violence-prevention skills, making home visits, and using parents as
volunteers. 

Interagency partnerships and community linkages:  Seeks community
support in making school antiviolence policies and programs work by
developing collaborative agreements in which school personnel, local
businesses, law enforcement officers, social service agencies, and
private groups work together to address the multiple causes of
violence. 

Culturally sensitive and developmentally appropriate:  Considers (1)
racial ethnic students' cultural values and norms by using bilingual
materials and culturally appropriate program activities, role models,
and leaders; and (2) participants' age and level of development in
designing program materials and activities. 

Although researchers advocate the use of these key characteristics,
few violence-prevention programs incorporate all seven
characteristics. 


   FEDERAL EVALUATION EFFORTS
------------------------------------------------------------ Letter :5

Our analysis of program descriptions contained in the Federal Agency
Juvenile Delinquency Development Statements for the 3-year period
1990 to 1992, showed that 11 agencies funded 115 research programs or
evaluations (ranging from $108 to $133 million annually) that
addressed youth violence.  However, only one grant provided funding
to evaluate a school-based violence-prevention program. 

More recently, some federal officials have acknowledged the need to
focus available resources on identifying violence-prevention programs
and strategies that work.  For example, three research-based
agencies--CDC, the National Institute of Justice (NIJ), and the
National Institute for Mental Health (NIMH)--have taken the lead in
funding impact evaluations to study the effectiveness of specific
school-based interventions.  These three agencies awarded about 26
grants totaling approximately $27 million for this purpose during
fiscal years 1993 and 1994 (see app.  VI for a list of these
studies). 

The Safe and Drug-Free Schools and Communities Act of 1994 presents
an opportunity for the federal government to strengthen efforts to
determine the effectiveness of various violence-prevention programs
and approaches.  The act authorizes up to $1 million for a national
impact evaluation.  The act also authorizes up to $25 million in
discretionary funding for national programs--evaluation is one of
many allowable activities. 


      GREATER COLLABORATION NEEDED
      AMONG FEDERAL AGENCIES
---------------------------------------------------------- Letter :5.1

Increased collaboration among federal agencies could significantly
leverage federal funds and enhance efforts to identify and
disseminate information on promising violence-prevention programs. 
At least three federal departments--Education, Health and Human
Services, and Justice--support school-based violence-prevention
research and programming.  However, these individual departments have
not mounted a comprehensive strategy for addressing school violence. 
Further, although these departments occasionally work together, a
formal mechanism to facilitate collaboration and coordination on
school violence issues has not yet been developed.  Agency officials
cite budget constraints as a major factor necessitating greater
collaboration. 

According to federal officials, effective collaboration requires (1)
sharing resources (both staff and funding) and information (such as
research and programmatic plans and priorities) and (2) developing
clear roles and responsibilities for each federal department to
ensure a comprehensive approach that avoids overlapping and
duplicative efforts.  Consequently, the mechanism for coordination
and collaboration should facilitate planning and program integration,
not merely information sharing.  Although the Federal Coordinating
Council on Juvenile Justice and Delinquency Prevention is responsible
for coordinating the efforts of federal agencies concerning juvenile
delinquency and delinquency prevention; the group only recently
convened--after a 2-year hiatus.\9

Further, federal officials told us that in the past the actual level
of participation on the Coordinating Council was insufficient to
effectively facilitate the type of coordination and collaboration
needed.  However, efforts are now underway to revitalize the
Coordinating Council since a new Administrator for the Office of
Juvenile Justice Delinquency Prevention (OJJDP) has been selected. 

Some federal collaborative efforts concerning violence-prevention
programming have started.  For instance, the National School Safety
Center (NSSC) and the SMART program represent successful interagency
collaborations.\10 In both instances, the Departments of Justice and
Education shared funding and staff to implement these programs.  In
addition, for the past 2 years, these agencies and several others
have collaborated in sponsoring a national violence-prevention
conference.  Expected benefits from increased collaboration among
federal agencies include development of

a comprehensive long-term federal violence-prevention research
agenda,

stronger evaluation designs that result in firmer conclusions about
program effectiveness,

a database with information summarizing the cumulative results of
program evaluations across agencies,

a process for identifying successful programs, and

a central on-line database for disseminating information on
successful programs. 


--------------------
\9 The Coordinating Council on Juvenile Justice and Delinquency
Prevention was established through Section 206 of the Juvenile
Justice and Delinquency Prevention (JJDP) Act.  The Chairman of the
Coordinating Council is the Attorney General of the United States;
the Vice Chairman is the Administrator for the Office of Juvenile
Justice Delinquency Prevention (OJJDP). 

\10 NSSC represents a partnership between the Departments of Justice
and Education and Pepperdine University.  The NSSC, which serves as a
clearinghouse, focuses national attention on school safety issues and
solutions. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

Many schools throughout our nation are struggling with rising levels
of youth violence in and around schools.  Schools have adopted a
broad range of solutions to curb violence.  The four programs we
visited--in California, Ohio, and New York--represent examples of
some of the promising approaches schools have implemented to address
violence. 

Research suggests that the most promising school-based
violence-prevention programs will involve at least seven key
characteristics.  Among these characteristics are a comprehensive
approach, starting early, and involving parents.  Although few
prevention programs have been evaluated, some federal agencies are
funding evaluations to examine several violence-prevention program
approaches.  The results, which should be available in about 3 to 5
years, will help determine which programs work best at curbing
violence. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

We discussed a draft of this report with responsible agency officials
at the Departments of Education, Health and Human Services, and
Justice, and we have included their comments where appropriate.  In
general, these officials agreed with the report. 


---------------------------------------------------------- Letter :7.1

As arranged with your office, unless you publicly announce its
contents beforehand, we plan no further distribution of this report
until 10 days after the date of this letter.  At that time, we will
send copies to the appropriate congressional committees and the
Departments of Education, Health and Human Services, and Justice.  In
addition, we will make copies available to others upon request. 
Please contact me or Cornelia Blanchette, Associate Director, on
(202) 512-7014 if you have any questions.  Major contributors to this
report are listed in appendix VIII. 

Sincerely yours,

Linda G.  Morra
Director, Education and
 Employment Issues


SUMMARY OF SCHOOL DISTRICTS'
RESPONSES TO VIOLENCE
=========================================================== Appendix I

The following table summarizes the responses received by the National
School Boards Association from a survey of its members.  NSBA
surveyed more than 2,000 school districts, of which 729 school
districts responded.  The results are not generalizable.  The
responses are contained in NSBA's 1993 report, Violence in the
Schools:  How America's School Boards Are Safeguarding Our Children. 



                          Table I.1
           
           School Districts' Responses to Violence


Response
------------------------------------  ----  ====  ====  ----
Suspension                              78    85    78    75
Student conduct/discipline code         76    87    79    70
Collaboration with other agencies       73    93    73    62
Expulsion                               72    85    68    70
School board policy                     71    76    69    71
Alternative programs or schools         66    85    66    57
Staff development                       62    74    66    52
Conflict resolution/mediation           61    82    63    49
 training/peer mediation
Locker searches                         50    64    43    49
Closed campus for lunch                 44    46    48    37
Mentoring programs                      43    65    44    31
Home-school linkages                    42    55    45    32
Dress code                              41    52    42    33
Law-related education programs          39    57    36    33
Multicultural sensitivity training      39    62    49    18
Parent skill training                   38    51    39    28
Search and seizure                      36    51    35    28
Security personnel in schools           36    65    40    18
Support groups                          36    47    37    28
Student photo identification system     32    41    39    20
Gun-free school zones                   31    46    26    26
Specialized Curriculum                  27    48    25    18
Drug-detecting dogs                     24    27    18    27
Work opportunities                      23    34    21    19
Telephones in classrooms                22    31    21    16
Metal detectors                         15    39    10     6
Volunteer parent patrols                13    17    14     8
Closed-circuit television               11    19     8     8
Establishing safe havens for            10    16     9     6
 students
------------------------------------------------------------
Source:  NSBA. 


ANAHEIM, CALIFORNIA:  SCHOOL
MANAGEMENT AND RESOURCE TEAMS
PROGRAM, ANAHEIM UNION HIGH SCHOOL
DISTRICT
========================================================== Appendix II


   DISTRICT PROFILE
-------------------------------------------------------- Appendix II:1

The Anaheim Union High School District's (AUHSD) enrollment in grades
7 through 12 increased about 5 percent over the past 3 years to the
1993-94 level of about 26,000.  AUHSD students come from seven
different cities--La Palma, Cypress, Stanton, Buena Park, Anaheim,
Garden Grove, and Fullerton.  AUHSD has eight high schools, eight
junior high schools, and a variety of alternative education sites. 
Students with limited or no English skills account for 35 percent of
the student population.  In the 1993-94 school year, over 34 percent
of the students were eligible for free or reduced price lunches. 

Like many school districts nationwide, AUHSD began experiencing
problems with drug abuse, crime, and gangs in the late 1970s.  The
number of identified gangs and gang members increased significantly
from 1985 to 1994.  In 1985, AUHSD communities had eight gangs with
an estimated 179 members.  Today, local police departments have
identified over 50 gangs with about 2,100 members.  AUHSD has
initiated several strategies to combat crime in and around their
schools.  AUHSD has a zero tolerance policy for gangs, weapons, and
drugs on campuses.  Also, AUHSD has an antigang dress code and
closed-campus policy.  AUHSD schools have nonuniformed community
volunteers to help with security, but do not use metal detectors. 
The Anaheim Police Department has assigned two officers to work
full-time on gang prevention activities in the district.  AUHSD
officials are also experimenting with other methods of school
security.  For example, they have placed mobile homes on several
school campuses.  Retired persons live in the homes rent-free in
exchange for helping to deter after-hours school vandalism. 


   PROGRAM OVERVIEW
-------------------------------------------------------- Appendix II:2

In the late 1970s, AUHSD officials sought ways to combat the rising
levels of crime in their communities.  They participated in training
to identify, categorize, and log incidents on campus and design
appropriate interventions.  These training efforts turned into the
SMART program.  In 1983, the U.S.  Department of Justice, National
Institute of Justice (NIJ), and the U.S.  Department of Education
jointly funded SMART as a pilot in AUHSD and two other sites. 
SMART's goal is to improve school safety and discipline by combining
information technology with concepts in school team-building and
interagency coordination. 

Since 1983, AUHSD administrators and staff have developed and
integrated the SMART program districtwide.  SMART consists of five
program components that operate together and provide a unified
approach to school safety and discipline: 

Commitment.  The primary SMART requirement is the commitment of the
superintendent and principals to improve school safety and
discipline. 

Safety and security audit.  Program officials conduct an audit of
district policies and practices affecting drugs, crime, discipline,
and student/faculty safety to help clarify responsibilities for
school officials in dealing with different types of crime and
discipline problems. 

Incident Profiling System.  The Incident Profiling System (IPS) is a
computerized procedure by which each school records disciplinary
infractions and criminal acts.  IPS generates reports describing
patterns of disruptions and crimes.  It tracks the types of
incidents, locations, times, and persons involved. 

SMART teams.  SMART teams include students, parents, teachers,
support staff, law enforcement/security, and administrators.  SMART
teams meet monthly to analyze current IPS data, set priorities,
devise actions to reduce problems, and monitor results.  Each team
prepares a SMART plan to reduce priority problems. 

Interagency coordination.  AUHSD officials meet with representatives
from juvenile justice, social service, and law enforcement agencies
to coordinate their responses to youth who commit crimes or have
behavior problems. 

The core of the SMART program is the statistical information gathered
through IPS.  This computerized system enables AUHSD officials to
collect and analyze a wide range of information about student
discipline.  Over the years, AUHSD officials have developed their own
computer program to record and track disruptive events and
individuals.  School officials are responsible for completing a
machine-scannable form on any disruptive event or individual they
encounter.  IPS data include rule violations such as a failure to
serve detention and law violations such as robbery, sex offenses,
drug or weapons possession, assaults, and property crimes. 

AUHSD principals and SMART teams compile and analyze IPS data at the
school level to dispel rumor, to identify and characterize discipline
problems, and to assess the consequence of actions taken.  Principals
can identify the areas of the school and the period of the day when
the most violence or disruption occurs and the individual students
who are causing a disproportionate amount of disruption.  Once the
principal has this information, he or she convenes a SMART team. 
This team produces and monitors a school SMART plan concentrating on
one topic at a time.  For example, a team may concentrate on locker
thefts.  A district SMART teams follows a similar process when
analyzing districtwide information. 

AUHSD principals, counselors, and staff have developed an integrated
service model to handle students or systemic problems identified by
SMART data.  This model provides a mechanism for schools to work with
the community, governmental agencies, and local businesses to meet
the needs of these at-risk students.  Services provided through this
integrated approach include peer tutoring, alcohol- and
drug-prevention programs, crisis intervention, and
conflict-resolution training. 


   PROGRAM ACCOMPLISHMENTS AND
   EVALUATION EFFORTS
-------------------------------------------------------- Appendix II:3

AUHSD officials have conducted extensive internal evaluations of the
SMART program that compare trends in IPS data.  In general, district
statistics show that incidents on campuses have increased gradually
while levels of crime in the surrounding communities have increased
at a much faster rate.  AUHSD officials cited recent outcome
information generated from IPS:  (1) 55 percent of the 37 main
categories of incidents have declined since 1993,\11 (2) police
activity on campus is down 51 percent from the spring semester 1993
to 1994, and (3) the reported total costs of incidents have dropped
66 percent from the fall semester 1991 to 1993. 

In 1992, NIJ and the Department of Education contracted with the
Center for Research and Evaluation in Social Policy\12

for an external evaluation of the SMART program.  The report focused
on SMART's usefulness--based on interviews, SMART data, and internal
documents.  The evaluators found strong support and encouragement for
the SMART program among school board members, the superintendent, and
district and school staff.  Further, they found that AUHSD officials
developed a depth of understanding and experience well beyond the
core elements of SMART.  They also found evidence that schools using
the SMART program had reduced problems with graffiti, fighting,
failure to attend detention, and defiance of authority.  The report
concluded that AUHSD officials have institutionalized SMART into its
daily operations and are taking positive actions as a result of the
SMART program.  The report also concluded that SMART and related
systems have increased the level of confidence of parents, students,
and teachers in school safety. 


--------------------
\11 Incident categories that have declined include assaults, battery,
robbery, possession of destructive devices, property crimes, forgery,
tardiness, weapons, failure to serve detention, throwing objects,
threats/intimidation, profanity, tobacco, and off-campus incidents. 

\12 Draft report No.  P-557, Graduate School of Education, University
of Pennsylvania, Philadelphia, Pennsylvania. 


   MAJOR SUCCESS FACTORS
-------------------------------------------------------- Appendix II:4

AUHSD officials believe that the following key factors contribute to
the SMART program's success: 

using a systematic problem-solving approach to resolve school crime
and discipline problems;

focusing on local control;

using existing resources with minimal additional funding;

developing positive working relationships among educators, parents,
students, local leaders, and community agencies; and

focusing on school problems, not problem schools. 


   IMPLEMENTATION BARRIERS
-------------------------------------------------------- Appendix II:5

The main barrier that AUHSD officials encountered was in designing
the computer data files, reports, and forms to support SMART.  AUHSD
officials have spent considerable time and effort in developing their
own computer system to fit their needs.  AUHSD officials stated that
they have experimented with three generations of computer programs
since SMART's inception.  School officials currently use a
machine-scannable form to record incidents.  They also developed a
SMART data dictionary that lists various identifying characteristics
of individuals and events and types of violations committed.  IPS
generates numerous computer reports to identify problems and make
comparisons.  This system has also been programmed to provide school
safety information required by California. 

AUHSD representatives have briefed other districts interested in
their SMART computer program.  They said that their system could be
easily replicated in other school districts.  However, dissemination
efforts are contingent on future funding. 


   FUNDING SOURCES
-------------------------------------------------------- Appendix II:6

Since 1983, AUHSD has had various levels of funding from NIJ and the
Department of Education.  Over the years, AUHSD used these federal
funds for a variety of activities, such as program dissemination and
site coordinator stipends, training, and conferences.  AUHSD has
funded most of SMART operating expenses out of its general funds
since SMART's inception.  Major ongoing expenses for 18 sites during
1993-94 totaled about $37,000.  This estimate includes site
coordinator stipends ($16,000), materials/supplies ($4,000), and a
part-time program specialist ($17,000).  In addition, NIJ and the
Department of Education awarded AUHSD a grant of $40,000 in 1993-94
to enhance program interventions with conflict-resolution training,
training packets, and training for new administrators. 


DAYTON, OHIO:  POSITIVE
ADOLESCENTS CHOICES TRAINING
PROGRAM ROTH MIDDLE SCHOOL
========================================================= Appendix III


   DISTRICT PROFILE
------------------------------------------------------- Appendix III:1

The Dayton public school system serves approximately 27,000 students. 
The district offers 5 high schools, 7 middle schools, 35 elementary
schools, and several centers with specialized programs of study. 
About 65 percent of the district's students are African-American. 
Most of the student population is educationally and economically
disadvantaged; all schools within the district qualify for Title I
funding.\13 In addition, about 70 percent of the students participate
in the free or reduced-price lunch program. 

Violence is a growing concern within the district.  During the 1990
school year, 152 students were referred to the principal for
disciplinary action because they carried weapons in school.  To help
maintain safe school environments, walk-thru metal detectors were
installed in all middle and high schools in November 1992.  During
the first 2 years of metal detector use, districtwide expulsions and
suspension rates for weapons-related incidents declined.  For
example, expulsions dropped from 200 in school year 1991-92 to 120 in
school year 1992-93.  Similarly, suspensions dropped from 3,483 in
school year 1991-92 to 3,311 in school year 1992-93. 


--------------------
\13 The Title I program (formerly Chapter I), authorized under the
Elementary and Secondary Education Act of 1965, as amended, serves
educationally deprived children--children whose educational
attainment is below the level that is appropriate for their age--in
relatively high-poverty areas. 


   PROGRAM OVERVIEW
------------------------------------------------------- Appendix III:2

PACT is a violence-prevention program directed specifically at middle
school African-American youth.  Staff from the School of Professional
Psychology at Wright State University implement the PACT program in
cooperation with Dayton Public Schools.  The program began operation
in 1989.  Since the beginning of the program, more than 130 youths
have been trained. 

PACT uses the rationale that adolescents who lack skills in such
areas as communication, negotiation, and problem-solving have a
limited range of alternatives with which to solve interpersonal
problems.  In addition, the program builds on research that suggests
that violence-prevention interventions with economically
disadvantaged and minority youth are most successful when developed
with sensitivity to racial, ethnic, and cultural issues.  Therefore,
the program uses African-American role models to create a learning
environment directly relevant to the lives of African-American
adolescents.  Teachers refer participants based on perceived skill
deficiencies in peer relations, behavior problems (particularly
aggression), or history of victimization by violence.  PACT training
uses small groups, composed of 10 to 12 students (12 to 15 years
old).  Training is provided twice a week at Roth Middle School during
regular school hours.  Groups generally receive about 38 training
sessions, each lasting about 50 minutes (the duration of a classroom
period).  Two doctoral-level clinical psychology students facilitate
each training session. 

The principal PACT program components include

training in three social skills (giving negative feedback, receiving
negative feedback, and negotiation),

training in anger management skills (techniques to control or express
anger constructively), and

education or information about violence (awareness-building lessons
on the nature and extent of violence). 

Three videotaped vignettes feature African-American role models
demonstrating how some of the target skills are used:  The "Givin'
It" video introduces skills on how to express angry feelings in a
calm, nonthreatening manner.  The "Takin' It" video introduces skills
on how to receive negative feedback without acting irrationally or
becoming overly upset.  The "Workin' It Out" video introduces skills
on how to solve problems or work out a compromise to a conflict
without resorting to aggressive or violent behavior. 

Principal techniques used during training sessions include modeling,
coaching, role-playing, feedback, and homework (practicing the skills
outside of class).  In addition, program facilitators use an
incentive system that rewards active participation and appropriate
behavior (such as being on time and following directions) during
training sessions.  For example, students receive success
dollars--paper money that can be exchanged for various gift items,
such as cassette tapes, candy or food, T-shirts, jewelry, or games. 


   PROGRAM ACCOMPLISHMENTS AND
   EVALUATION EFFORTS
------------------------------------------------------- Appendix III:3

PACT evaluations seek to measure both process and behavioral effects
of the program.  In addition, since 1989, staff conducted
longitudinal followup through juvenile court on all PACT-trained
students.  An evaluation of the 1989-90 project year showed that
youth who participated in the program demonstrated improvement in all
target skill areas (for example, giving and accepting negative
feedback, problem-solving, and resisting peer pressure).  A 1992-93
outcome study demonstrated that the PACT participants had less
involvement in fighting and fewer referrals to juvenile court in
comparison with a control group that did not receive training.  For
example, PACT-trained students

demonstrated a 50-percent reduction in physical aggression at school;

showed behavior improvement during the course of training, which was
maintained beyond participation in the program;

showed a greater reduction in levels of physical aggression than
similar nontrained students who did not receive the training; and

had more than 50 percent fewer overall and violence-related juvenile
court charges and a lower per-person rate of offending than students
in a nontrained control group. 

Other signs of success include (1) principal and teacher observations
concerning improvements in the behavior of individual students
attributed to PACT and (2) student testimonials that PACT training
made a difference in how they acted. 


   MAJOR SUCCESS FACTORS
------------------------------------------------------- Appendix III:4

PACT officials believe that the following key factors contribute to
the program's success: 

designing the program as a leadership club and providing a student
reward system,

using culturally sensitive videotapes that feature African-American
role models, and

using doctoral-level clinical psychology students as trainers. 


   IMPLEMENTATION BARRIERS
------------------------------------------------------- Appendix III:5

The PACT program overcame two major implementation barriers.  First,
PACT initially targeted a small group of chronically truant
adolescents at an alternative school-based dropout prevention
program.  Although these participants showed skill and behavior
improvements, the setting proved to be problematic.  For example, the
participants were often absent from school.  Consequently, absence
from PACT training was a major problem.  To address this issue, in
1990 the program relocated to the Roth Middle School where it
targeted a younger population.  Second, the program initially
operated as a pull-out model, with meeting times scheduled for the
same time as academic subjects.  Consequently, students missed some
regularly scheduled classes.  Since the program targeted high-risk
students who most likely were also experiencing academic problems,
this was of particular concern.  To overcome this barrier, PACT
violence-prevention training was made part of the regular curriculum
under the health education track. 


   FUNDING SOURCES
------------------------------------------------------- Appendix III:6

State, federal, and private sources provide funding for the PACT
program.  Funding to develop, test, evaluate, and disseminate program
information has come from a variety of sources, including the Ohio
Commission on Minority Health; the Ohio Governor's Office of Criminal
Justice Services; the Mathile Foundation; and the Bureau of Justice
Assistance, Department of Justice.  In addition, support for the
production of the PACT curriculum resource guide came from the
Department of Health and Human Services' Bureau of Maternal and Child
Health. 

At the school level, instituting the PACT program using doctoral-
level students as trainers costs approximately $55,200 per year or
about $287 per student.  If parent training is provided, the cost
increases by $23,400 to approximately $78,600 or about $409 per
student.  These estimates assume that the program serves 192
students.  Costs for program evaluation vary depending on the
complexity of the research design.  Current PACT evaluations cost
about $21,000 for a part-time evaluation consultant. 


NEW YORK CITY, NEW YORK: 
RESOLVING CONFLICT CREATIVELY
PROGRAM NEW YORK CITY PUBLIC
SCHOOLS
========================================================== Appendix IV


   DISTRICT PROFILE
-------------------------------------------------------- Appendix IV:1

New York City's public school system is the largest in the country,
with 1,052 schools and nearly 1 million students.\14

The ethnically diverse student population citywide consisted of 37
percent African American, 36 percent Hispanic, 18 percent white, and
9 percent Asian/Pacific Islander for the 1992-93 school year. 
Fourteen percent of the students were limited English proficient and
69 percent were eligible for free or reduced-price lunches. 

Officials are grappling with the increasing levels of urban crime and
violence that are invading their schools.  For the 1991-92 school
year, the New York Board of Education reported 4,955 serious
incidents in its schools, including homicide, robbery, sex offenses,
and controlled substance and weapons possession.  To increase the
level of school security, the board has responded by installing metal
detectors and X-ray machines and using student photo identification
cards in high schools.  The board has also funded conflict-resolution
and mediation programs and increased the number of school safety
officers in all schools.  The board employs about 3,000 security
officers.  If considered a police department, it would rank as the
ninth largest police department in the country, between the Baltimore
and Dade County Police Departments.  The budget for the Division of
School Safety, which includes security officers, 123 civilians and a
fleet of 90 vehicles, totaled nearly $73 million in 1993.  Also that
year, 95 percent of the schools had school safety officers and 41
high schools--or more than one-fourth of all high schools--had a
weapons-detection program. 


--------------------
\14 The Board of Education of the City of New York has oversight over
all the schools and direct responsibility for the high schools.  The
board selects superintendents to administer elementary and
intermediate/junior high schools in 32 separate community school
districts. 


   PROGRAM OVERVIEW
-------------------------------------------------------- Appendix IV:2

RCCP is a school-based program in conflict-resolution and
intercultural understanding, jointly sponsored by the New York City
Board of Education and Educators for Social Responsibility (ESR). 
ESR is a nonprofit organization dedicated to conflict- resolution and
multicultural education.  RCCP began in 1985 as a collaboration
between ESR and the board to provide a model for an instructional
program in school change and violence-prevention. 

RCCP's objectives include modeling nonviolent alternatives for
dealing with conflict and teaching negotiation, as well as other
conflict-resolution skills.  RCCP focuses on changing the school
climate and requires a strong commitment at the highest levels within
the school system.  School districts then approach individual
principals and teachers about joining the program.  Participation at
every level is voluntary; school districts, principals, individual
teachers, and parents take part because they choose to do so.  RCCP
is in place in 180 elementary, intermediate/junior high and high
schools in New York City, with 3,000 teachers and 70,000 students
participating.\15 According to RCCP officials, most of the programs
serve at-risk students located in lower-income neighborhoods in
Brooklyn, South Bronx, Manhattan, and Queens. 

The K-12 curriculum focuses on violence-prevention, conflict
resolution, and countering bias.  RCCP curricula include the
following:  active listening, assertiveness (as opposed to
aggressiveness or passivity), expressing feelings,
perspective-taking, cooperation, negotiation, and interrupting bias. 
Teaching strategies include role-playing, interviewing, group
discussion, and brainstorming.  These teaching strategies require
teachers to adopt a new style of classroom management.  This method
involves a sharing of power with students so that they can learn to
deal with their own disputes.  To create this change, RCCP has an
intensive staff development component.  RCCP staff visit classrooms
and conduct one-on-one consultations, demonstration lessons, and
after-school meetings with teachers and administrators.  In addition,
RCCP's parent component teaches parents how to lead workshops for
other parents on intergroup relations, family communication, and
conflict resolution.  RCCP has an administrator's component also. 


--------------------
\15 Currently, RCCP is being disseminated in over 300 schools
nationwide under the auspices of the RCCP National Center. 


   PROGRAM ACCOMPLISHMENTS AND
   EVALUATION EFFORTS
-------------------------------------------------------- Appendix IV:3

RCCP contracted with Metis Associates, Inc., a private consulting
group, to do an external evaluation for the 1988-1989 school year in
three community school districts.  The evaluation assessed the
various program components and measured the program's impact on
participating students, staff, and administrators.  To determine
RCCP's program impact, Metis Associates, Inc., primarily used
questionnaires to 200 participating teachers,\16 as well as various
school personnel and administrators; and administered achievement
tests to a sample of 176 participating students and a control group
of 219 nonparticipating students.  To determine the effectiveness of
the mediation program, Metis Associates, Inc., surveyed approximately
150 teachers, 11 school staff, and 143 student mediators in the five
schools using the program. 

Seventy-one percent of teachers responded that the program led to
less physical violence in the classroom and 67 percent observed less
name-calling and fewer verbal put-downs.  The test results of 4th,
5th, and 6th grade participants showed that they learned key concepts
of conflict resolution and could apply them in hypothetical
situations.  Over 98 percent of respondents in the five schools said
that the mediation component gave children an important tool for
dealing with conflicts. 

The Metis report concluded that RCCP was exemplary and that
participants' assessments were extremely positive.  The report cited
teacher's surveys that revealed a positive change in children's
attitudes and behaviors as a result of their participation in RCCP. 

In 1993, CDC awarded RCCP a 3-year grant to conduct an extensive
evaluation.  This evaluation will look at the impact of the program,
the readiness of teachers, and the importance of each program
component. 


--------------------
\16 Of the 200 teachers participating, about 130 or 65 percent
returned completed surveys.  For about 75 percent of the respondents,
1988-89 was their first year with RCCP. 


   MAJOR SUCCESS FACTORS
-------------------------------------------------------- Appendix IV:4

RCCP officials believe the following key factors contribute to the
program's success: 

an on-going long-term commitment with the school district,

strong support of the principal and school administrators, and

partnerships with parents and the community. 


   IMPLEMENTATION BARRIERS
-------------------------------------------------------- Appendix IV:5

RCCP officials stated that the major barrier they encountered was
struggling to work within a culture that glamorizes violent responses
to conflict.  RCCP bases its program on the premise that human
aggression is a learned behavior and that conflict itself is a normal
part of life.  According to RCCP officials, what must change,
therefore, is how students respond to conflict.  RCCP teaches
students that violence is not an acceptable means of resolving
conflict, that they can learn new nonviolent skills, and that they
have a choice to make when a conflict arises. 


   FUNDING SOURCES
-------------------------------------------------------- Appendix IV:6

Sponsoring school districts and foundation grants fund RCCP
operations.  For the 1992-93 school year, the budget for the New York
City RCCP program was about $2 million.  The board funded staff
salaries and teacher stipends in the amount of $700,000.  ESR was
responsible for $1.3 million, with $750,000 from contracts with the
City's participating school districts and $550,000 raised from
private sources.  As mentioned earlier, CDC has funded a 3-year
evaluation of the RCCP program, totaling approximately $729,000
($243,000 annually starting in 1993). 


PARAMOUNT, CALIFORNIA: 
ALTERNATIVES TO GANG MEMBERSHIP
PROGRAM, PARAMOUNT UNIFIED SCHOOL
DISTRICT
=========================================================== Appendix V


   DISTRICT PROFILE
--------------------------------------------------------- Appendix V:1

The Paramount Unified School District encompasses 13 schools that had
an enrollment of 13,879 in the 1993-94 school year.  About 73 percent
of the student population is Latino and 14 percent is
African-American.  In the 1993-94 school year, 46 percent of the
students had limited English proficiency and about 60 percent
received free or reduced-price lunches. 

Most of the 30 expulsions in school year 1993-94 were for weapons
possession or assault and battery.  Although the district does not
compile the reasons for suspension, school officials estimate that
most of the 4,254 suspended days that year involved possession of
drugs, fighting, or defiance of authority.  Currently, the district
contracts for one armed, uniformed sheriff's deputy at the single
high school.  As of March 1994, because of racial tensions that had
erupted on campus, the school board was considering the purchase of
hand-held metal detectors for the high school. 


   PROGRAM OVERVIEW
--------------------------------------------------------- Appendix V:2

ATGM seeks to reduce gang membership by teaching students the harmful
consequences of the gang lifestyle, how to not participate in it, and
how to choose positive alternatives.  The City Council, together with
the school district and the Los Angeles County Sheriff's Department,
recognized that recreation programs and law enforcement alone were
not enough to contain the growing gang problem in their community. 
The City of Paramount and the Paramount Unified School District
developed a partnership in 1982 and started ATGM.  The program bases
its approach on the belief that interest in gangs begins at a young
age and, therefore, the focus is on reaching children early. 

The ATGM program includes three parts:  an elementary school
curriculum, an intermediate school follow-up program, and
neighborhood meetings.  The elementary program includes a 2nd grade
curriculum, taught in 10 weekly 40-minute lessons and a 5th grade
curriculum, taught in 15 weekly 55-minute lessons.  The 7th grade
follow-up program consists of eight biweekly lessons.  These lessons
expand on the topics introduced in the elementary school curriculum,
such as peer pressure and drug abuse.  The program uses many guest
speakers and focuses on self-esteem, the consequences of a criminal
lifestyle, and higher education and career opportunities.  The ATGM
program includes all students in the 2nd, 5th, and 7th grades. 

Program officials consider the neighborhood meetings a major emphasis
of the ATGM program.  These efforts encourage parents to attend
meetings--held at schools, churches, parks, community centers and
private residences--to educate them about gangs.  These meetings,
which are bilingual (English/Spanish), also provide parents with
information, encouragement, and help in preventing their children
from joining gangs.  ATGM staff members hold an average of 50
neighborhood meetings each year.  An important part of these meetings
is the individual outreach and follow-up meetings with students and
their families.  ATGM staff members frequently meet one-on-one with
at-risk students who have been referred to them by teachers. 


   PROGRAM ACCOMPLISHMENTS AND
   EVALUATION EFFORTS
--------------------------------------------------------- Appendix V:3

City officials have conducted several evaluations of the ATGM program
over the years, which mainly used pre- and post-test questionnaires
of participants.  These studies showed that the ATGM program made a
difference in keeping children out of gangs.  For example, while some
5th grade students may have had neither positive nor negative
feelings toward gang membership before participating, after their
participation these same students tend to have a negative attitude
toward gangs.  According to ATGM staff members, ATGM made a
difference in keeping students out of gangs.  During a 4-year period,
officials gave several questionnaires to the original 1982-83 group
of 5th grade participants.  For example, in June 1984, officials
tested the original 5th grade group when they were in intermediate
school.  Of the 170 6th graders who previously participated in the
ATGM program, 90 percent responded positively when asked if the
program helped them to stay out of gangs. 

During the 1986-87 school year, ATGM officials tested a group of 9th
graders who were ATGM participants in the 5th grade.  Again, more
than 90 percent of these students said that they were staying out of
gangs.  In addition, in February 1993, program officials conducted a
different type of evaluation.  Working with the Los Angeles County
Sheriff's Department, city officials matched 3,612 names of ATGM
participants with a listing of identified gang members.  This match
identified 152 students, or 4 percent, as gang members, and 3,460
participants, or 96 percent, who were not. 


   MAJOR SUCCESS FACTORS
--------------------------------------------------------- Appendix V:4

According to program officials, the key factors contributing to the
program's success include

presenting information on the consequences of being in a gang in a
factual, no-nonsense manner that allows students to come to their own
conclusions about whether joining a gang is a good choice;

reaching out to children and their families through home visitation
and parent meetings in the community;

exposing children to the program early in their lives--in the 2nd,
5th, and 7th grades;

incorporating the program sessions as part of the regular school
curriculum and conducting lessons in both Spanish and English;

using bilingual staff who are familiar with the community and are
sensitive to Latino culture; and

exposing children to positive role models and alternatives to the
gang lifestyle. 


   IMPLEMENTATION BARRIERS
--------------------------------------------------------- Appendix V:5

When ATGM began about 13 years ago, the major barrier was the lack of
research, training, and materials about antigang programs, especially
for the Latino community.  Over the years, ATGM staff have developed
their own curriculum, through trial and error, and borrowed from
other prevention programs.  Program officials cite the lack of
bilingual audiovisual materials for both children and parents as a
continuing problem. 

Program and school officials cited the District's high transiency
rate of 33 percent\17 as a major problem.  Children who recently
moved to Paramount have not been exposed to the ATGM program.  A 1992
survey of Paramount high school students revealed that over one-half
(56 percent) did not attend elementary school in the district. 
Program officials speculated that many of transient students are
unable to develop strong ties to the school and are more likely to be
involved in gangs.  They believed that if area schools had similar
antigang programs, the problem would lessen. 


--------------------
\17 Average rate of children in the district who begin the school
year and leave before the school year ends. 


   FUNDING SOURCES
--------------------------------------------------------- Appendix V:6

Since its inception, Paramount has funded the ATGM program entirely
with its general fund.  For the 1992-93 school year, the program
budget was $150,000, which funded three neighborhood counselors and
supplies. 


FEDERALLY SPONSORED EVALUATIONS OF
SCHOOL-BASED VIOLENCE-PREVENTION
PROGRAMS
========================================================== Appendix VI

                                                              Fiscal    Fiscal
                                                              year      year
Grant title             Evaluation objective                  1993      1994
----------------------  ------------------------------------  --------  --------
National Institute of Justice (Department of Justice)
--------------------------------------------------------------------------------
Crime Prevention        To evaluate several programs that     Not       $150,000
Projects                provide a range of social services    applicab
                        to targeted youth 11-13 years old     le
                        and their families in four cities.
                        Services received by program
                        participants include educational
                        counseling, mentors, and summer
                        programs. Parents receive advice on
                        effective parenting, crisis
                        intervention, and drug counseling.
                        Target neighborhoods receive
                        additional public services in the
                        form of increased enforcement, safe
                        school routes, and community
                        policing teams.

Evaluation of           To evaluate two types of violence-    Not       Not
Violence-Prevention     prevention programs in New York       applicab  applicab
Programs in Middle      City. One program couples a           le\a      le
Schools (New York, New  traditional conflict-resolution
York)                   program with peer-mediation. The
                        other program combines traditional
                        methods with a victimization
                        curriculum, a schoolwide
                        antiviolence campaign, and a
                        counseling component.

National Evaluation of  To evaluate a gang-prevention         Not       183,318
the Gang Resistance     program in which uniformed officers   applicab
Education and Training  provide a nine week (one hour per     le
                        week) course about the negative
                        aspects of gangs.

Improving School        To evaluate a program where           120,841   Not
Safety by Empowering    students, teachers, and police work             applicab
Students in             together to identify and solve                  le
Educational Process     problems on a high school campus.
(Omaha, Nebraska)       The evaluation design calls for a
                        matched pair of high schools, one to
                        receive the program and the other to
                        serve as a control.

Reducing School         To evaluate a conflict-resolution     214,970   Not
Violence in Detroit     and violence-reduction intervention             applicab
(Detroit, Michigan)     program implemented in 10 middle                le
                        schools.

================================================================================
Subtotal                                                      $335,811  $333,318


National Institute of Mental Health (Department of Health and Human Services)
--------------------------------------------------------------------------------
Epidemiologic Center    To evaluate two interventions, one    $1,638,0  $1,292,0
for Early Risk          directed at shy and aggressive        00        00
Behaviors (Baltimore,   behaviors that predict later
Maryland)               antisocial behavior and heavy
                        substance use. The other is aimed at
                        learning problems, a predictor of
                        later psychiatric symptoms and,
                        possibly, disorders. Both
                        interventions target children in
                        grades 1-2.


Multisite Prevention of Conduct Disorder (MPCD): To evaluate a multisite
prevention program to implem intervention to prevent cond adulthood. It is
hypothesiz child behavior and family an lead to long-term prevention
--------------------------------------------------------------------------------
MPCD, Durham, North                                           1,594,00  1,754,00
Carolina                                                      0         0

MPCD, Nashville,                                              1,573,00  1,713,00
Tennessee                                                     0         0

MDCD, Seattle,                                                1,612,00  1,612,00
Washington                                                    0         0

MPCD, Rural                                                   1,358,00  1,454,00
Pennsylvania                                                  0         0

Prevention Research     To evaluate two intervention          100,000   100,000
with Aggressive,        programs. One is designed to have
Rejected Children       impact on parents and teachers as
(Durham, North          well as on aggressive, rejected
Carolina)               children. The second design involves
                        examining a combination of
                        intervention methods involving
                        students 7-17 years old.

Oregon Prevention       To evaluate several integrated        1,348,00  1,440,00
Research Center         components designed to prevent        0         0
(Eugene, Oregon)        conduct disorder. The interventions
                        focus on family behavior management,
                        parental monitoring of children's
                        activities, home-school liaison, and
                        an after-school program to improve
                        academic and social skills for
                        children in grades 1 and 5.

School-Based Secondary  To evaluate a prevention program for  378,000   327,000
Prevention for ADHD     young children with Attention
Children (Saint Paul,   Deficit Hyperactivity Disorder
Minnesota)              (ADHD) 8-14 years old at 18
                        elementary schools.

Preventing Antisocial   To evaluate the effectiveness of      864,000   898,000
Behavior in High-Risk   three complementary, multiyear
Children (Chicago and   preventive interventions for high-
Aurora, Illinois)       risk urban youth. Level A is a
                        classroom-based training program
                        combined with teacher training.
                        Level B uses the same curriculum as
                        in level A plus small group
                        training. Level C combines level B
                        interventions with a family
                        intervention.

================================================================================
Subtotal                                                      $10,465,  $10,590,
                                                              000       000


National Center for Injury Prevention and Control, Centers for Disease Control
and Prevention (Department of He
--------------------------------------------------------------------------------
Peace Builders          To evaluate the effectiveness of a    223,164   242,120
(Tucson, Arizona)       peace-builders curriculum at
                        reducing physical and verbal
                        aggression in elementary school
                        students.

System of Values        To evaluate the effectiveness of a    211,250   220,675
(Portland, Oregon)      program that provides students in
                        grades 7-9 with adult mentors and
                        training in conflict resolution and
                        social skills, peer education in
                        violence-prevention, recreational
                        opportunities, and academic
                        tutoring.

Reducing Dating         To evaluate the effectiveness of a    233,671   229,281
Violence (Johnston      school-based program designed to
County, North           reduce dating violence. Selected
Carolina)               students receive classroom
                        instruction about gender stereotypes
                        and social norms that contribute to
                        dating violence. Key individuals in
                        the community will be trained to be
                        resources for youth who seek
                        assistance about teenage dating
                        violence.

Students for Peace      To evaluate the effectiveness of a    261,523   226,941
(Houston, Texas)        project to reduce aggressive
                        behaviors among students in grades 6
                        to 8. The primary program components
                        are a school health promotion
                        council, a curriculum that provides
                        knowledge and skills, peer-
                        mediation training, and parent
                        involvement.

Conflict Resolution     To evaluate the effectiveness of a    118,697   208,141
Computer Module         computer-based instructional program
(Indianapolis,          designed to teach social skills and
Indianapolis)           conflict resolution. The computer
                        module will be used to teach 6th to
                        8th grade students nonviolent
                        interpersonal problem-solving
                        strategies.

Resolving Conflict      To evaluate a curriculum for          221,403   200,826
Creatively (New York,   students in grades K-12 covering
New York)               conflict-resolution and
                        intercultural understanding. The
                        curriculum includes teacher
                        training, classroom lessons on
                        conflict resolution, and training in
                        peer-mediation.

Preventing Antisocial   To evaluate three levels of           227,333   230,883
Behavior in High-Risk   preventive interventions for high-
Children (Chicago and   risk urban youth aged 7 to 13. Level
Aurora, Illinois)       A consists of classroom-based
                        training to increase awareness and
                        knowledge about the factors that
                        influence peer and other social
                        relationships. Level B consists of
                        the same treatments as in Level A
                        plus training conducted through
                        small groups and peer relationships
                        for high-risk children. Level C
                        consists of the same treatments as
                        in Level B plus a family
                        intervention for the high-risk
                        children and their families.

================================================================================
Subtotal                                                      $1,497,0  $1,558,8
                                                              41        67

Working Toward Peace    To evaluate the effectiveness of two  101,039   117,227
(Detroit, Michigan)     school-based primary prevention
                        programs. The first program, Skills
                        for Adolescence, is a comprehensive
                        skill-based curriculum that covers
                        self-discipline, responsibility,
                        problem-solving, setting goals,
                        critical thinking, service to
                        others, and prevention of drug
                        abuse. The second program, Working
                        Toward Peace, reinforces concepts
                        taught in the Skills for Adolescence
                        curriculum.

Attributional Bias and  To evaluate a program designed to     174,599   195,639
Reactive Aggression     decrease reactive physical and
(Los Angeles,           verbal aggression directed towards
California)             peers. This program uses role
                        playing, discussion of personal
                        experiences, and training to
                        interpret and properly categorize
                        the behavioral cues expressed by
                        others in social situations.

Richmond Youth          To evaluate a program designed to     210,711   224,200
Violence-prevention     reduce aggressive behaviors among
Program (Richmond,      6th grade students. The program
Virginia)               teaches conflict-resolution skills
                        and peer-mediation and modifies the
                        school environment by altering
                        school policies to support
                        nonviolence.

Peer Group Training     To evaluate a program to reduce       375,000   509,386
and Community           mortality and morbidity due to
Empowerment (Houston,   violence among African-American and
Texas)                  Hispanic youth in three middle
                        schools. The interventions include
                        training peer leaders in group
                        support, social skills, leadership
                        and violence-prevention; parenting
                        skills training for parents of the
                        youths in peer leader groups; and
                        training for 20 neighborhood
                        violence-prevention advocates.

Education, Counseling,  To evaluate a multifaceted            339,830   346,653
and Community           intervention approach that targets
Awareness (New York     students, families, and communities;
and Brooklyn, New       through a conflict-resolution
York)                   program, a counseling and education
                        program, a schoolwide antiviolence
                        campaign, and a big sibling program
                        to mentor younger children.

================================================================================
Subtotal                                                      $1,201,1  $1,393,1
                                                              79        05

================================================================================
Total                                                         $13,499,  $13,875,
                                                              031       290
--------------------------------------------------------------------------------
\a Funding for this project based on a fiscal year 1992 grant of
$215,378. 


SCOPE AND METHODOLOGY
========================================================= Appendix VII

To accomplish our objectives and further our understanding of key
issues, we interviewed acknowledged experts and federal agency
officials involved in violence-prevention programming and research. 
We also collaboratively sponsored a school safety symposium with
Harvard University's School of Public Health, which brought together
federal officials, program directors, and violence-prevention
researchers to discuss efforts to curb school violence. 

Based on this information, we compiled a listing of more than 250
violence-prevention programs.  From this listing, and after
considering the recommendations of violence-prevention experts, we
judgmentally selected four programs for site visits--two in
California and one each in Ohio and New York. 

Anaheim, California:  SMART,

Dayton, Ohio:  PACT,

New York, New York:  RCCP, and

Paramount, California:  ATGM. 

We selected these programs because they operated primarily in
schools, were located in different regions of the country, were
located in areas with nationally recognized violence problems, and
had completed evaluations showing initial signs of success in curbing
school violence. 

We then visited seven schools associated with these four programs. 
At each school, we gathered program information, including key
characteristics, success factors, implementation barriers, evaluation
outcomes, and funding sources.  In addition, we observed students
participating in the programs and interviewed program officials. 

To obtain student and teacher views on efforts to curb school
violence, we conducted 16 discussion groups--7 with teachers and 9
with students--at the seven schools.  The size of these groups
varied.  Generally, the teacher groups consisted of 4 to 9
participants, while the student groups ranged from 8 to 15
participants.  In total, 103 students and 35 teachers participated. 
Our discussion group results are descriptive, showing the range of
opinions expressed by participants. 

To identify the key characteristics associated with promising
violence-prevention programs we conducted a literature review.  In
addition, we held discussions with school safety and
violence-prevention experts and determined the key factors for
success at the programs we visited. 

Finally, to identify federally sponsored evaluations of
violence-prevention programs, we interviewed federal officials at the
Departments of Education, Health and Human Services, and Justice and
reviewed the 1994 Juvenile Delinquency Development Statements:  A
Report on Federal Programs.\18

We conducted our review between January 1994 and February 1995 in
accordance with generally accepted government auditing standards. 


--------------------
\18 This report lists federal programs that support the goals of the
Juvenile Justice and Delinquency Prevention Act.  These programs
generally concern juvenile delinquency, prevention, treatment,
diversion, education, training, and research, including alcohol and
drug programs and programs to improve the juvenile justice system. 


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
======================================================== Appendix VIII

GAO CONTACTS

Wayne Upshaw, Assistant Director, (202) 512-7006
Valerie Giles-Reynolds, Assistant Director, (202) 512-4552
Nancy Kawahara, Senior Evaluator, (213) 346-8082

ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  Mark Whittle, Evaluator;
Revae Steinman, Senior Evaluator; and Valerie Dumas, Evaluator. 


BIBLIOGRAPHY
============================================================ Chapter 0

Cohen, Stu, and Renee Wilson-Brewer, Violence-prevention for Young
Adolescents:  The State of the Art of Program Evaluation, New York: 
Carnegie Council on Adolescent Development, 1991. 

Dryfoos, Joy G.  Adolescents at Risk:  Prevalence and Prevention. 
New York:  Oxford University Press, 1990. 

"First Things First:  What Americans Expect From the Public Schools."
Public Agenda, New York, New York, 1994. 

Graduate School of Education, draft report No.  P-557.  Philadelphia: 
University of Pennsylvania, 1992. 

Juvenile Delinquency Development Statements:  A Report on Federal
Programs.  Department of Justice, 1994. 

Kids and Violence, National Governors' Association, Washington, D.C.,
1994. 

Losing Generations, Adolescents in High-Risk Settings.  National
Research Council Commission on Behavioral and Social Sciences and
Education, Panel on High-Risk Youth.  Washington, D.C.:  National
Academy Press, 1993. 

Lowry R., S.  Sleet, C.  Duncan, K.  Powell, and L.  Kolbe,
"Adolescents at Risk for Violence." Educational Psychology Review,
Vol.  7, No.  1 (1995), pp.  7-37. 

Mendel, R.A.  Prevention of Pork?  A Hard-Headed Look at
Youth-Oriented Anti-Crime Programs, Washington, D.C.:  American Youth
Policy Forum, 1995. 

Mercy, James A., Mark L.  Rosenberg, and others.  "Public Health
Policy for Preventing Violence," Health Affairs, Vol.  12, No.  4
(1993), pp.  7-29. 

Metropolitan Life Survey of the American Teacher 1993:  Violence in
America's Public Schools, New York:  Louis Harris and Associates,
Inc., 1993. 

National Crime Victimization Survey, Bureau of Justice Statistics,
Department of Justice, 1991. 

The Prevention of Youth Violence:  A Framework for Community Action,
Centers for Disease Control and Prevention.  Atlanta:  CDC, 1993. 

Prothrow-Stith, Deborah.  Deadly Consequences.  New York:  Harper
Collins Publishers, 1991. 

Reiss, Albert J., Jr., and Jeffrey A.  Roth, eds.  Understanding and
Preventing Violence.  Washington, D.C.:  National Academy Press,
1993. 

Safeguarding Our Youth:  Violence-Prevention for Our Nation's
Children, National School Safety Center, Forum Proceedings, held on
July 20-21, 1993, Westlake Village, California. 

Stephens, Ronald D., National School Safety Center, testimony before
the Senate Subcommittee on Education, Arts and Humanities, September
23, 1993. 

Tolan, Patrick, and Nancy Guerra.  "What Works in Reducing Adolescent
Violence:  An Empirical Review of the Field." Boulder, Colorado: 
Center for the Study and Prevention of Violence, 1994. 

"Violence and Youth:  Psychology's Response." American Psychological
Association Commission on Violence and Youth.  Washington, D.C.: 
American Psychological Association, 1993. 

Violence in the Schools, How America's School Boards Are Safeguarding
Our Children, National School Boards Association, Alexandria,
Virginia:  1993. 

Webster, Daniel W.  "The Unconvincing Case for School-Based Conflict
Resolution Programs for Adolescents," Health Affairs, Vol.  12, No. 
4 (1993), pp.  126-141. 

