TITLE: National Evaluation of G.R.E.A.T.
AUTHOR: Finn-Aage Esbensen and D. Wayne Osgood
SUBJECT: gangs, juvenile delinquency prevention, conflict resolution
SERIES: Research in Brief
DATE: November 1997
PAGES: 15
BYTES: 31,000

Figures, charts, forms, and tables are not included in this
ASCII plain-text file. To view this document in its entirety,
download the Adobe Acrobat graphic file available from this
Web site or order a print copy from NCJRS at 800-851-3420.

-------------------------------------------------------

National Institute of Justice
Research in Brief
Jeremy Travis, Director

National Evaluation of G.R.E.A.T.

 by Finn-Aage Esbensen and D. Wayne Osgood

-------------------------------------------------------

ISSUES AND FINDINGS

Discussed in this Brief: Early findings and research methods
pertaining to a national evaluation of the effectiveness of the
Gang Resistance Education and Training (G.R.E.A.T.) program,
a school-based gang prevention strategy taught to middle school
students.

Key issues: During the 1980s and 1990s, gang affiliation by
youths and their involvement in criminal activity became a major
concern for law enforcement and the public. The G.R.E.A.T.
program was developed to reduce adolescent involvement in
criminal behavior and gangs. The national evaluation of the
program consists of a two-pronged research approach: (1) a
preliminary study comparing students who completed G.R.E.A.T.
with others who either had not participated or had enrolled but
failed to finish, and (2) a longitudinal quasi-experimental design
assessing both the short- and long-term effectiveness of
G.R.E.A.T. 

Key findings: Early findings from the national evaluation are
based on a cross-sectional survey of 5,935 eighth graders from
42 schools in 11 locales where G.R.E.A.T. is taught. Researchers
are also assessing the training of police officers who teach the
program.

Preliminary results indicate that students who completed the
G.R.E.A.T. lessons reported more prosocial behaviors and
attitudes than their peers who did not finish the program or
failed to participate in the first place. Among other findings:

o Lower rates of self-reported delinquency and gang
membership.

o More communication and attachment with parents.

o Greater commitment to school and lower levels of perceived
obstacles to academic achievement.

The questionnaire administered to the eighth graders used five
background characteristics--sex, race, age, family status, and
parental education--to determine whether significant differences
existed between students who completed the G.R.E.A.T. program
and students who comprised the comparison group. Differences
between the groups were small, and initial findings of the
program's positive impact are not a product of preexisting
differences between the G.R.E.A.T. participants and comparison
students.

Target audience: Gang, delinquency prevention, and juvenile
justice specialists and researchers; law enforcement agencies;
school administrators; and State and local policymakers.

-------------------------------------------------------

Youth delinquent gangs continue to generate concern among
criminal justice professionals and the general public. Gang
membership and related criminal activity increased in the late
1980s and early 1990s, and the availability of firearms has led to
more gang-related homicides. One way to address these
problems is to find ways to prevent youths from joining gangs.

In 1991 police officers from the Phoenix Police Department and
from Mesa, Glendale, and Tempe, Arizona, and special agents of
the Bureau of Alcohol, Tobacco and Firearms developed Gang
Resistance Education and Training (G.R.E.A.T.) to reduce
adolescent involvement in criminal behavior and gangs.
G.R.E.A.T. is a national, school-based gang prevention program
in which uniformed law enforcement officers teach a 9-week
curriculum to middle school students. As of June 1997, more
than 2,400 officers from 47 States and the District of Columbia
had completed G.R.E.A.T. training.

Given this rapid program expansion, the National Institute of
Justice, in cooperation with the Treasury Department's Bureau of
Alcohol, Tobacco and Firearms, sponsored a comprehensive,
multisite evaluation to assess G.R.E.A.T.'s effectiveness. Initial 
findings indicate the program is having a positive effect on
student attitudes and behaviors and is deterring them from
involvement in gangs.

This Research in Brief discusses the evaluation's design and
methodology, G.R.E.A.T.'s program and officer training, and
preliminary findings of a cross-sectional study.

EVALUATION DESIGN

Context. The research design for the national evaluation
considered previous research and public policy on gangs.
Consensus is lacking about the magnitude of the gang problem,
the extent and level of gang organization, and the action needed
to address the issue. Some of the epidemiological and etiological
issues can be traced to different methodologies and theoretical
perspectives. Policy differences can be attributed to competing
government priorities and to the limited number of evaluations
of programs undertaken to address the gang phenomenon.
However, a number of suppression, intervention, and prevention
programs with evaluative components have been implemented in
the past few years at local and national levels.[1]

Knowledge about gangs traditionally has come from one of three
sources: observational or case studies,[2] law enforcement
records,[3] and surveys.[4] On one point there is considerable
consensus among researchers: the high rate of criminal offending
among gang members.

Two objectives and two strategies. The national evaluation of
G.R.E.A.T. has two primary objectives: (1) to perform an
outcome analysis examining G.R.E.A.T.'s short- and long-term
effects on students and (2) to conduct a process evaluation
assessing the quality and effectiveness of officer training (see
"G.R.E.A.T. Officer Training").

Two strategies were developed to determine program
effectiveness. The first is a cross-sectional study of students in
11 locales where G.R.E.A.T. is taught; group questionnaires
were administered to a sample of eighth-grade students. The
second strategy, which recognizes the limitations of
retrospective, cross-sectional designs, is a prospective
longitudinal study initiated at six sites.[5] A quasi-experimental
research design guided the assignment of classrooms to
experimental and comparison groups. Students in both groups
completed pretests and posttests during the first half of the
1995-96 school year and will be administered questionnaires
annually through fall 1999.

CROSS-SECTIONAL SURVEY

For the first study, a cross-sectional survey of 5,935 eighth-grade
students was completed in spring 1995. Survey results were used
to create a treatment group and a comparison group to assess
G.R.E.A.T.'s effectiveness in the 11 cities where the survey was
administered. These cities had delivered the G.R.E.A.T. program
during the 1993-94 school year, when the targeted students were
seventh graders. Surveying these students as eighth graders
permitted a 1-year followup to their program 
participation and also guaranteed that none of the survey sample
were currently enrolled in the program.

Site selection. In selecting the 11 sites, consideration was given
to geographic location, population characteristics, and population
size. The cities selected were Phoenix, Arizona; Torrance,
California; Orlando, Florida; Pocatello, Idaho; Will County,
Illinois; Kansas City, Missouri; Omaha, Nebraska; Las Cruces,
New Mexico; Philadelphia, Pennsylvania; Providence, Rhode
Island; and Milwaukee, Wisconsin.

In those cities, questionnaires were administered to all eighth
graders in attendance on the specified day at schools that had
offered G.R.E.A.T. during the previous 2 years. This resulted in
a final sample of 5,935 eighth-grade students from 315
classrooms in 42 schools.

Measures. The questionnaire was designed to assess the
G.R.E.A.T. curriculum. The goal was to include questions that
would assess specific aspects of the G.R.E.A.T. program while
also measuring dominant criminological theories.[6] Several
measures also were developed to reflect the curriculum's
cognitive aspects. For example, lesson 3 of the program
introduces students to six steps and five personal prerequisites
for conflict resolution. A sample measure for this lesson was to
ask students to respond to the following statement: "Violence
interferes with a person's basic right to feel safe and secure."
(See "The G.R.E.A.T. Curriculum.")

Another key measure concerns gang membership and
involvement in gang activity. Questions were designed to elicit
self-reports of illegal activity. This technique has been used
widely during the past 30 years and has provided a good
measure of actual behavior rather than a measure of police
response to behavior.[7] (See "Measuring Gang Affiliation.")

Comparison group. A primary question was whether students
who completed the G.R.E.A.T. program were comparable to
those who did not complete it--either because they never
participated or dropped out of the program. The treatment group
and comparison group were defined through answers to the
question, "Did you complete the G.R.E.A.T. program?" Of the
5,836 respondents who answered the question (99 students did
not respond), 2,629 (45 percent) reported they had completed the
program and thus were the treatment group. The 3,207 who had
not became the comparison group.

The schools varied substantially, however, in the number of
students who had completed and who had not completed the
G.R.E.A.T. program. Since the precision with which program
impact can be established at each school depends on the number
of students in both treatment and comparison groups, schools
with few students in one of the groups could contribute
relatively little to the evaluation. Therefore, analysis of the
treatment and comparison groups was replicated in a restricted
sample of 28 schools where at least 15 students comprised each
group.

Controlling for other differences. Because data were gathered on
one occasion only, a year after students had completed the
program, the researchers had to compare the treatment and
comparison groups using statistical controls to rule out the
possibility that differences between them were attributable to
various background characteristics.
BACKGROUND CHARACTERISTICS

Questions were asked in the survey to determine five background
characteristics that could be associated with the outcome
measures.[8] The analysis controlled for the following:

o Sex.

o Race/ethnicity (white, African-American, Hispanic, Asian-
American, and other).

o Age (there was little variation in age, because only eighth-
grade students participated in the evaluation).

o Family status (as reflected in the adults with whom the youths
resided).

o Parental education (defined as the highest level attained by
either parent).

Not surprisingly, differences surfaced among the 42 schools in
racial composition and socioeconomic status (as reflected by
family status and parental education).[9] The analysis, which
controlled for differences between schools, found a few small
but statistically significant differences in background
characteristics between treatment and comparison groups.

Ideally, the treatment and comparison groups would have been
matched, but this could not be expected in a post hoc evaluation
such as this study. The pattern of group differences in
background characteristics is ambiguous but does not appear
especially problematic to determining the impact of the
G.R.E.A.T. program.

Comparisons of treatment and nontreatment groups revealed no
systematic bias. Both groups had demographic characteristics
indicating high or low risk for delinquency, gang membership, or
both. In the comparison group, 15-year-old students were
overrepresented; in the treatment group African-American youths
were overrepresented. The comparison group had fewer females
but more youths from single-parent homes. Given this
inconsistent pattern and the small size of group differences, it
was concluded that the outcome measures were not a product of
preexisting differences between the G.R.E.A.T. and comparison 
students.[10]

Program impact was thus determined through a model that
controlled for school and the five background characteristics.
Although the results are consistent, restricting the analysis to the
28 schools tends to strengthen the magnitude of the program's
effect.[11]

INITIAL RESULTS

Early findings indicate that G.R.E.A.T. appears to be meeting its
objective--to reduce gang affiliation and delinquent activity. The
students completing G.R.E.A.T. reported lower levels of gang
affiliation and delinquency than did comparison students. These
differences are small but statistically significant. (See "Statistical
and Substantive Differences.") Not only is the aggregate measure
of delinquency lower for the G.R.E.A.T. group but so are most
of the subscales, i.e., drug use, minor offending, property crimes,
and crimes against persons. No differences between the groups
were found for rates of victimization or selling drugs.

A number of differences also were found for attitudinal
measures. G.R.E.A.T. lessons are aimed at reducing impulsive
behavior, improving communication with parents and other
adults, enhancing self-esteem, and encouraging students to make
better choices. The cross-sectional survey results (see exhibit 1)
reveal that 1 year after completing the program, G.R.E.A.T.
students (in contrast to the comparison group) reported the
following:

o Lower rates of delinquency.

o Lower rates of gang affiliation.

o More positive attitudes toward the police.

o More negative attitudes about gangs.

o More friends involved in prosocial activities.

o Greater commitment to peers promoting prosocial behavior.

o Higher levels of perceived guilt at committing deviant acts.

o More commitment to school.

o Higher levels of attachment to both mothers and fathers.

o More communication with parents about their activities.

o Fewer friends involved in delinquent activity.

o Less likelihood of acting impulsively.

o Lower likelihood of engaging in risky behavior.

o Lower levels of perceived blocks to academic success.

The cross-sectional survey also yielded findings about gang
membership that are contrary to popular perceptions and other
research results. For example, white youths comprised a larger
share of the gang population (25 percent), in contrast to previous
studies that found that gangs were predominantly composed of
minorities. (See "Differences Between Gang Members and
Nonmembers" and exhibit 2.)

CONCLUSIONS AND POLICY IMPLICATIONS

G.R.E.A.T. is one of myriad gang prevention efforts employed to
reduce adolescent involvement in crime and gangs. The
preliminary findings of this study support continuation of
G.R.E.A.T.; other prevention programs await evaluation results.

Results from the 1995 cross-sectional survey suggest that
students who participated in G.R.E.A.T. reported significantly
more prosocial behaviors and attitudes than students who did not
take part in the program. This 1-year followup survey supports
the idea that trained law enforcement personnel can serve as
prevention agents as well as enforcers of the law.

These cross-sectional results need to be viewed with caution,
however. Some differences existed between treatment and
comparison groups prior to the introduction of the program.
Although most of these differences were controlled through
available statistical techniques, a quasi-experimental design such
as that being implemented in the longitudinal phase of this
evaluation will provide a better assessment of program
effectiveness. This longitudinal design also will allow for
examination of long-term effects.

-------------------------------------------------------

 G.R.E.A.T. OFFICER TRAINING

Currently, the Bureau of Alcohol, Tobacco and Firearms, the
Federal Law Enforcement Training Center, and the Phoenix
Police Department coordinate officer training and, with the
Orange County, Florida, Sheriff's Office, share management
responsibility for the G.R.E.A.T. program.

Evaluators attended five officer training sessions as well as one
G.R.E.A.T. management training session during fiscal year 1995.
Despite some shortcomings, G.R.E.A.T. officer training has
many strengths that prepare officers to become successful
classroom instructors. Primary among them is the supportive
learning environment the training staff creates for the officers.
Instructors deal with officer students in an enthusiastic,
engaging, and encouraging manner. The instructional format
provides a spirit of camaraderie and cooperation, and a repeated
emphasis on professionalism creates a context of mutual respect.
Further, the strategy of modeling each lesson of the curriculum
and requiring officers to present a lesson is the keystone to the
training process, which repeatedly exposes officer students to
material they themselves will soon be teaching in their own
classrooms.

Other strengths of the training program include its use of
occasional role-play techniques and group exercises. In addition,
the curriculum focuses on several important skills--including
meeting basic needs, resolving conflict, taking responsibility, and
setting goals--that can be taught to middle school students and
may be instrumental in achieving the goal of crime-free
adolescents. Its graduation event serves as a motivator and
culminating activity.

-------------------------------------------------------

 THE G.R.E.A.T. CURRICULUM

To meet its objectives of reducing gang activity and teaching the
consequences of gang involvement, the curriculum consists of
nine lesson plans to be offered once a week to middle school
students, primarily seventh graders. Each detailed lesson plan
contains clearly stated purposes and objectives. In addition to the
nine lesson plans, the curriculum calls for the teaching officers
to discuss gangs and how they affect the quality of people's
lives. The nine G.R.E.A.T. lessons are:

1. Introduction. Students get acquainted with the G.R.E.A.T.
program and the presenting officer.

2. Crime, Victims, and Your Rights. Students learn about crimes,
their victims, and their impact on school and neighborhood.

3. Cultural Sensitivity and Prejudice. Students explore how
cultural differences affect their school and neighborhood.

4.,5. Conflict Resolution (two lessons). Students are taught how
to create an atmosphere of understanding that enables all parties
to better address problems and work on solutions together.

6. Meeting Basic Needs. Students learn how to meet their basic
needs without joining a gang.

7. Drugs and Neighborhoods. Students are educated about how
drugs affect their school and neighborhood.

8. Responsibility. Students examine the diverse responsibilities
of people in their school and neighborhood.

9. Goal Setting. Students learn the need for goal setting and how
to establish short- and long-term goals.

For information about G.R.E.A.T., contact the Bureau of
Alcohol, Tobacco and Firearms at 800-726-7070.

-------------------------------------------------------

 MEASURING GANG AFFILIATION
 
What constitutes a gang member? In the current research, two
filter questions introduce the gang-specific section of the
questionnaire: "Have you ever been a gang member?" and "Are
you now in a gang?" Of the total sample, 994 youths, or 17
percent, indicated they had belonged to a gang at some point in
their lives. In contrast, slightly more than half of these
youths--522, or 9 percent of the sample--indicated they were
currently gang members.

In an attempt to limit the sample of gang members to
"delinquent gangs," two different measures were employed. First,
a restrictive or conservative definition limited gang status to
those respondents who stated they were current gang members
and that their gangs engaged in at least one type of delinquent
behavior (fighting other gangs, stealing cars, stealing in general,
or robbing people). This resulted in identification of 451 gang
members, or 8 percent of the sample. Second, a more liberal, yet
still somewhat restrictive, definition included youths who
indicated they "had ever been a gang member" and whose gang
had been involved in at least one of the four illegal activities.
This more liberal definition produced 623 gang members,
representing 10.6 percent of the sample. The latter, more liberal,
definition was used for this research.

-------------------------------------------------------

 STATISTICAL AND SUBSTANTIVE DIFFERENCES

 An important distinction exists between statistically significant
differences and substantively important differences. Sample sizes
and statistical approaches can affect the level of statistical
significance, sometimes exaggerating an effect and other times
underestimating an effect.

One alternative to relying solely on statistical significance is to
examine relative effect sizes. Effect size (ES) can be defined as
"a measure of change due to the treatment as a proportion of the
standard deviation for each measure employed."* Thus, an ES of
-1 indicates that the treatment group performed one standard
deviation lower than the comparison group; an ES of +1
indicates that the treatment group performed one standard
deviation unit higher than the comparison group. The larger the
ES, the greater the measurable impact of the program. In one
review of delinquency treatment and prevention programs, the
author found average effect sizes of .17 and argued that even a
small ES of .10 may have practical value when dealing with
criminal activity.**

One way of interpreting an effect size is to convert it to a
percentage. This can be done by dividing the effect size by two.
For example, an effect size of .10 represents a 5 percent
difference. In the current research, effect sizes were in the .10
range, indicating modest program effects.

*Gottfredson, Denise C., "School-Based Crime Prevention," in
Larry W. Sherman, et al., Preventing Crime: What Works, What
Doesn't, What's Promising: A Report to the United States
Congress, Washington, D.C.: U.S. Department of Justice, Office
of Justice Programs, 1997.

**Lipsey, Mark W., "Juvenile Delinquency Treatment: A Meta-
Analytic Inquiry into the Variability of Effects," in Meta-
Analysis for Explanation, ed. T.D. Cook, et al., Beverly Hills,
California: Sage, 1992.

-------------------------------------------------------

DIFFERENCES BETWEEN GANG MEMBERS AND 
NONMEMBERS

Contrary to much of the prevailing literature about the male-
dominated nature of gangs, 38 percent of gang members in the
sample were females. Although this figure still indicates that
females are underrepresented among gang members, it is to a far
lesser extent than is commonly assumed.*

The racial composition of gang members in this sample reveals
that white youths were proportionately less involved in gangs
than African-American and Hispanic youths, but not to the
extent that prior research (often based on case studies of
minority populations) has suggested: 25 percent of the gang
members in this study are white. In fact, if some of the "other"
category--which comprises white youths who identified
themselves as American, Italian, German, Portuguese, and the
like--is included, the proportionate difference is reduced even
further.

Consistent with earlier assessments of the demographic
characteristics of gangs, this sample reveals that younger youths
are underrepresented in gangs, and gang members are more
likely to live with a single parent and have parents with lower
levels of educational attainment. Even within this limited age
sample, the youths who were 13 and younger accounted for only
17 percent of gang members, although they represented 31
percent of the nongang sample. At the other extreme, 23 percent
of gang members were 15 years old or older, although only 9
percent of nongang members were in this age bracket. A
minority of youths lived in single parent homes, but gang
members reported living in single parent homes more frequently
(40 percent) than nongang youths (30 percent). Gang members'
mothers, fathers, or both were more likely not to have finished
high school (20 percent for gang members, 11 percent for
nongang youths). These demographic characteristics suggest
there may be qualitative differences in the living situations
between gang and nongang youths.

*This discrepancy in rates of female participation in gangs may
be due to a combination of methodological issues. First,
relatively few studies have sampled youths as young as 12 and
13. Second, few studies have used general surveys of adolescent
populations.

-------------------------------------------------------

Finn-Aage Esbensen is a professor in the Department of
Criminal Justice, University of Nebraska at Omaha; D. Wayne
Osgood is a professor in the Department of Sociology,
Pennsylvania State University.

-------------------------------------------------------

NOTES

1. For a review see Howell, James C., Youth Gang Violence
Prevention and Intervention: What Works, Washington, D.C.:
U.S. Department of Justice, 1996.

2. See, for example, the works of Campbell, Anne, The Girls in
the Gang, 2d ed., Cambridge, Massachusetts: Basil Blackwell,
1991; Hagedorn, John M., People and Folks: Gangs, Crime and
the Underclass in a Rustbelt City, Chicago, Illinois: Lakeview
Press, 1988; Miller, Walter B., "Lower Class Culture as a
Generating Milieu for Gang Delinquency," Journal of Social
Issues, 14(1958):5-19; Thrasher, Frederick M., The Gang: A Study 
of One Thousand Three Hundred Thirteen Gangs in Chicago, 
Chicago: University of Chicago Press, 1927; Vigil, James D., 
Barrio Gangs: Street Life and Identity in Southern California, 
Austin, Texas: University of Texas Press, 1988.

3. Maxson, Cheryl L., and Malcolm Klein, "Street Gang
Violence: Twice as Great or Half as Great?" in Gangs in
America, ed. C. Ronald Huff, Newbury Park, California: Sage
Publications, Inc., 1990; Spergel, Irving A., "Youth Gangs:
Continuity and Change," in Crime and Justice: An Annual
Review of Research, ed. Norval Morris and Michael Tonry,
Chicago: University of Chicago Press, 1990.

4. Esbensen, Finn-Aage, and David Huizinga, "Gangs, Drugs,
and Delinquency in a Survey of Urban Youth," Criminology,
31(1993):565-589; Esbensen, Finn-Aage, David Huizinga, and
Anne W. Weiher, "Gang and Non-gang Youth: 
Differences in Explanatory Variables," Journal of Contemporary
Criminal Justice, 9(1993):94-116; Fagan, Jeffrey, "The Social
Organization of Drug Use and Drug Dealing Among Urban
Gangs," Criminology, 27(1989):633-669; Thornberry, Terence
P., Marvin D. Krohn, Alan J. Lizotte, and Deborah Chard-
Wierschem, "The Role of Juvenile Gangs in Facilitating
Delinquent Behavior," Journal of Research in Crime and
Delinquency, 30(1993):55-87; Winfree, L. Thomas, Teresa V.
Backstrom, and G. Larry Mays, "Social Learning Theory, Self-
Reported Delinquency, and Youth Gangs: A New Twist on a
General Theory of Crime and Delinquency," Youth and Society,
26(1994):147-177.

5. The six sites are Las Cruces, New Mexico; Lincoln, Nebraska;
Omaha, Nebraska; Philadelphia, Pennsylvania; Phoenix, Arizona;
and Portland, Oregon.

6. Winfree, L. Thomas, Finn-Aage Esbensen, and D. Wayne
Osgood, "Evaluating a School-Based Gang Prevention Program:
A Theoretical Perspective," Evaluation Review,
20(1996):181-203.

7. Hindelang, Michael J., Travis Hirschi, and Joseph G. Weis, 
Measuring Delinquency, Beverly Hills, California: Sage, 1981;
Huizinga, David, and Delbert S. Elliott, "Reassessing the
Reliability and Validity of Self-Report Delinquency Measures,"
Journal of Quantitative Criminology, 2(1986):293-327; Huizinga,
David, "Assessing Violent Behavior with Self-Reports," in
Neuropsychology of Aggression, ed. Joel Milner, 
New York: Kluwer Academic Publishers, 1991.

8. For a discussion of the correlates of delinquency and gang
membership, consult Covey, Herbert C., Scott Menard, and
Robert J. Franzese, Juvenile Gangs, Springfield, Illinois: Charles
Thomas Publisher, 1992; and Esbensen, Finn-Aage, David
Huizinga, and Anne W. Weiher, "Gang and Nongang Youth:
Differences in Explanatory Variables," Journal of Contemporary
Criminal Justice, 1993.

9. "Dummy" variables were created to control for the mean
differences between schools so that the assessment of program
impact is an average of the within-school differences between
G.R.E.A.T. students and comparison group (weighted according
to sample size), adjusted for any differences in the five
background characteristics.

10. An additional analysis examined only students from schools
at which no preexisting differences were found between the
treatment and comparison groups. While some of the findings
differ from those reported here, the same pattern of program
effect was found. These findings are reported in Esbensen, Finn-
Aage, and D. Wayne Osgood, "Promising Results from a Gang
Prevention Program," Annual Meeting of the Academy of
Criminal Justice Sciences, Louisville, Kentucky, 1997.

11. A copy of the technical report of this study is available from
Finn Esbensen, Department of Criminal Justice, University of
Nebraska at Omaha, Lincoln, Nebraska 68588.

-------------------------------------------------------

This research is supported under grant number 94-IJ-CX-0058
from the National Institute of Justice, Office of Justice Programs,
U.S. Department of Justice. Points of view in this document are
those of the authors and do not necessarily represent the official
position of the U.S. Department of Justice.

-------------------------------------------------------

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

-------------------------------------------------------

NCJ 167264

-------------------------------------------------------