 

man-A.» a. n-
~

«

3.».1.»

:w.)

a
x

3
'u‘

“25.51. 4;:
#5359543
.3 12 L1‘

513'

.o»....., . ”v.1 .
.4. u.-

m.

”Lid

.33... .

“izﬁi‘ﬂgl‘

, : 2'} "
,.

A
’v
at

1 3'

U ’ ‘ ’3‘ 7
2-3314:
3‘3f’.’s
9..

v
D
.

.o—.
w.“

I.

‘3.-

it
‘ I“; «Rip
«5563:; .i ’nﬁg.
» 1.); ' .132
top”; .
E "u!
l ‘

a- v“ ;

m.

...,

‘jfiz.
Yaw.

....‘ g":
.. ‘

mm.
u m

- 7 BJ‘ .
PL'.

~ g 2 v . ' J" '1"! is :
ﬂ: 9 3 e . :9: z 33 " (gifgx‘vd'
. ' I I ‘ . :— ‘ V N

"woo.-

“ 1

Annu- unor-

~
v-

\f

u. .vdv' wu- ~
\
,
pa
hurl-n

,» >4
A.
...N

. «‘35
-q p-
.135}?-
74‘3‘ "

..-2
iv?

 

79%.“.1'8

 

MICHIGAN 8 ATE U NERSWY UBRARIES

m ”mull 1 gr

 

 

 

 

 

 

Ill! Ill/UNI”!Hill/Ulllll/Wl ‘

3 1293 1389 0235

This is to certify that the

thesis entitled

Dispute Management Training and Adolescents'
Social Problem-Salving Competence, Social
Self-Efficacy and Outcome Expectations

presented by

Stacy Elizabeth Curtis

has been accepted towards fulﬁllment
of the requirements for

M-A. degree in P_svchologv

 

    

Major professor

Date April 10. 1996

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

“f“

 

 

 

 

 

 

. .0 .1 .

PLACE ll RETURN BOX to remove this duckout from your record.
To AVOID FINES mum on or baton date duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAn AfﬂmativoMlONE Oppommty- Institution
“ Wanna-o1

‘ ~— - l r .

DISPUTE MANAGEMENT TRAINING AND ADOLESCENTS'
SOCIAL PROBLEM-SOLVING COMPETENCE, SOCIAL SELF-EFFICACY
AND OUTCOME EXPECTATIONS
By

Stacy Elizabeth Curtis

A THESIS
Submitted to
Michigan State University

in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1996

ABSTRACT
DISPUTE MANAGEMENT TRAINING AND ADOLESCENTS
SOCIAL PROBLEM-SOLVING COMPETENCE, SOCIAL SELF-EFFICACY
AND OUTCOME EXPECTATIONS
By

Stacy Elizabeth Curtis

This study evaluated the effects of a violence prevention program on middle
school students in a Midwestern metropolitan area. It was proposed that conﬂict
resolution training would directly impact students’ perceived self-efﬁcacy and outcome
expectations of nonviolent conﬂict resolution, which would affect students’ social
problem-solving competence. Social problem-solving competence was proposed to
impact nonviolent conﬂict resolution. Interviews were conducted with 38 students who
received dispute management training, 65 students who attended an intervention school
but did not receive training, and 108 students who attended a control school. A model
was tested using hierarchical regression analyses. Results of this study do not appear to
support the efﬁcacy of this dispute management program. However, ﬁndings conﬁrm
previous research which suggests that both self-efﬁcacy and outcome expectations are
important indicators of pro-social behavior. These factors should be considered in

future program development and evalutation.

ACKNOWLEDGMENTS

I would like to thank my advisor Dr. William Davidson and my committee
members Dr. Deborah Salem and Dr. Timothy Bynum for their advice, time, and
support. I need to thank Charlene Baker and Martha Ehle for their time, willingness,
and intelligence in coding the many responses from which this thesis grew -- they were
truely reliable. I also need to thank Becki Campbell for her assistance and statistical
knowledge.

Next, I wish to thank my friends who lived the moments of this thesis (and
graduate school) with me. Alison, Cheryl, David, Susan and many other critical
supporters, without your understanding, empathy, help and Shared laughs this work
would never have been completed. I wish you all the best.

Thanks to my mother and father, Pam and Ross Curtis, for seeing me through.
I thank you for your unfailing love, support, and understanding. To my brother, Scott

-- I wish you the world.

iii

TABLE OF CONTENTS

LIST OF TABLES .......................................... vi
LIST OF FIGURES ........................................ viii
CHAPTER ONE
INTRODUCTION .......................................... 1
Violence Prevention and Conﬂict Resolution .......................... 8
Theoretical Background ................................... ll
Biogenetic ......................................... 12
Social Control ....................................... 13
Ecological ......................................... 15
Social Leaming Theory ................................. 16
Program Design - Conﬂict Resolution as Social Problem-Solving ...... 18
Social Problem-Solving Competence ...................... 23
Perceived Self-Efﬁcacy .............................. 26
Expectations of the Outcomes of Competent Behavior ........... 28
Summary ....................................... 30
We ................... 31
Goals and Hypotheses ....................................... 35
CHAPTER TWO
METHOD .............................................. 38
Setting ................................................. 38
Subjects ................................................ 40
Design ................................................. 43
Program Description ........................................ 44
Procedure ............................................... 47
Measures ............................................... 49
Development of Coding Manuals ............................. 51
Social Problem-Solving Competence ........................ 52
Outcome Expectations .................................. 56
Perceived Self-Efﬁeacy ................................... 58
Nonviolent Conﬂict Resolution .............................. 59

iv

CHAPTERTHREE

RESULTS .............................................. 62

Analysis of Variance ........................................ 62
Hypothesis 1 .......................................... 62
Hypothesis 2 .......................................... 64
Hypothesis 3 .......................................... 64

Hierarchical Regression Model Predicting Social

Problem-Solving Competence ................................. 65

Hypothesis 4 .......................................... 65
Summary of Hierarchical Regression Analysis ..................... 74

Summary of Results ........................................ 74

CHAPTER FOUR

DISCUSSION ............................................ 77
Summary ............................................ 82
Future Directions ....................................... 84

APPENDICES

Appendix A: Coding Manual for Scenarios -- Social
Problem-Solving Competence ............................... 86

Appendix B: Coding Manual for Scenarios -- Outcome Expectations ......... 95

REFERENCES .......................................... 101

LIST OF TABLES

Table l - Students Receiving Training in Dispute Management Skills -- Self-report

Data and Data of Homeroom Assignment ........................ 42
Table 2 - Content of Seven Interview Scenarios Presented to Students ......... 50
Table 3 - Reliability Analysis for Social Problem-Solving Competence ........ 55
Table 4 - Reliability Analysis for Expectations of the Outcome of

Competent Behavior ..................................... 55
Table 5 - Reliability Analysis for Perceived Self-Efﬁcacy ................ 55

Table 6 - Content of Questions Related to Conﬂict
and Antisocial Behaviors - Interview ........................... 61

Table 7 - Comparative Analysis of Social Problem-Solving Competence, Social Self-
Efﬁcacy, Outcome Expectations and Nonviolent Conﬂict Resolution by
Experimental Condition Using One-way Analysis of Variance ........... 63

Table 8 - Hierarchical Multiple Regression Anlysis and Intercorrelations Predicting
Perceptions of Self-Efﬁcacy - Equation 1 ........................ 68

Table 9 - Hierarchical Multiple Regression Anlysis and Intercorrelations Predicting
Outcome Expectations - Equation 2 ........................... 68

Table 10 - Hierarchical Multiple Regression Anlysis and Intercorrelations Predicting
Social Problem-Solving Competence - Equation 3 .................. 69

Table 11 - Hierarchical Multiple Regression Anlysis and Intercorrelations Predicting
Nonviolent Conﬂict Resolution - Equation 4 ...................... 71

Table 12 - Hierarchical Multiple Regression Anlysis and Intercorrelations Predicting
Nonviolent Conﬂict Resolution - Equation 4 Backwards .............. 73

vi

Table 13 - Content of Questions Included in Survey Used During Wave One of the
Overall Evaluation Project ................................. 79

vii

LIST OF FIGURES

Figure 1 - Proposed Model .................................... 37
Figure 2 - Revised Model ..................................... 76

viii

Chapter One

INTRODUCTION

Violence in schools has increased in recent years. We hear nightly on the news
about children carrying guns to schools, about drive-by shootings, and about critically
injured teenagers. During a six month period of the 1988-1989 school year, over
400,000 students nationwide were victims of violent crimes at school including assault,
rape, and robbery (National Center for School Safety, 1991). In a study conducted by
the Centers for Disease Control (1992), 50% of boys and 25 % of girls reported having
been physically attacked by someone at school. Furthermore, the National Crime
Victimization Survey of 21,554,092 students reported that during a Six month period
nearly one tenth of students, ages twelve to nineteen years old, were victims of crime
in or around school. Of these students, two percent (431,081) were victims of violent
crime (Bastian & Taylor, 1991).

The above statistics indicate that our schools have become violent, unsafe places
for children. Other research provides further evidence of such violence. Osofsky
(1995) conducted a survey with African-American parents and children living in an
inner-city environment with a high rate of police documented violence. Results
indicated that 35 % of parents felt that their children were not safe walking to school

and 54 % perceived that their children were not safe playing in their neighborhoods.

1

2
Only 17 96 of parents surveyed believed their children were very safe while

participating in these activities. The majority of parents (62 %) felt their children were
very safe at home, while 30% of parents felt that their children were very safe at
school. It is alarming to recognize that 70% of parents surveyed perceived that their
children were not very safe at school. Finally, 90% of parents in this inner-city sample
indicated that violence was a critical problem or crisis in their neighborhood.

According to ofﬁcial records of the Centers for Disease Control and Prevention
(CDC) (1990), homicide has become the second leading cause of death for 15 to 24
year-olds and the leading cause of death for black youth (Commission on Violence and
Youth, 1993; Hausman, Spivak, Prothrow-Stith, & Roeber, 1992). Homicide is
currently the leading cause of death for young African-American men ages 15 to 34,
with a rate ranging between ﬁve and ten-times higher than the rate for white men
(DeJong, 1994). For African-American men between the ages of 15 and 19, the rate of
homicides involving ﬁrearms more than doubled between 1984 and 1988 (DeJong,
1994). Rates of homicides in large cities are even more alarming. In Detroit, for
example, the homicide rate for African-American males ages 15 to 19 quadrupled
between 1987 and 1988 (DeJong, 1994). It is striking to note that while the FBI’S
Uniform Crime Report for 1994 indicates a national decrease in all types of crime for
the third year in a row, it reports an increase in crime among juveniles under age
eighteen (Smith, Beals, Brant, & Annin, 1995). A report of the Department of Justice
(as cited in Smith et a1. , 1995) showed that the number of adolescents arrested for

serious offenses rose from 83,400 in 1983, to 129,600 in 1992. It is beyond dispute

3
that violence presents a major crisis in our communities, among our youth, and in our
schools.

Aggressive and violent acts such as verbal threats, harassment, theft, and physical
assaults drastically disrupt the school and learning environment and can cause fear and
intimidation in students (Nuttal & Kalesnik, 1987). Anti-social and aggressive
behavior interferes not only with school learning but also inhibits the development of
positive peer relations in the school context (Commission on Violence and Youth,
1993). The National Crime Victimization Survey indicated that twenty percent of all
students surveyed feared being attacked at school (Bastian & Taylor, 1991). The CDC
(1992) also surveyed students and found that during the month prior to being surveyed
eight percent of students in the ninth through twelfth grades had been in at least one
physical ﬁght resulting in injuries that required medical attention. The CDC survey
found that an estimated eighteen ﬁghts per 100 students occurred per month and that
more male Students (28 %) than female students (7%) were involved in physical ﬁghts.
Furthermore, the highest rates of physical ﬁghts were reported among
African-American male students (47%) followed by Hispanic males (35 %) and
Anglo-American males (22%).

The Commission on Violence and Youth (1993) reported ﬁndings from the
National Research Council Showing that four characteristics of schools may contribute
to violence. These characteristics included: (1) relatively high numbers of students
occupying a limited amount of space, (2) reduced ability to avoid confrontations, (3)

imposed behavioral routines and conformity which may contribute to feelings of anger,

4
resentment, and rejection, and (4) poor building design features which may facilitate
the commission of violent acts. Although violence is not limited to urban
environments, and indeed in recent years has increased drastically in suburban schools,
trends Show that violence is most prevalent in large, urban schools in areas of lower
socioeconomic status that often experience overcrowding and a lack of resources
(Elliott, 1994; Nuttal & Kalesnik, 1987). Personal violence, furthermore, is most
prevalent in inner-city junior-high schools with large minority populations (I-Iausman et
al., 1992; Nuttal & Kalesnik, 1987). According to comparison studies of school
records, schools with higher incident rates of violence tend to have lower levels of
Student academic achievement, higher rates of absenteeism, and more school drop-outs
(Christie & Toomey, 1990; Commission on Violence and Youth, 1993; Hellman &
Beaten, 1986).

The comparatively high incidence of violence in inner-city schools seems to be
related to social factors such as poverty or low socioeconomic conditions,
discrimination, and oppression in the surrounding neighborhoods and does not seem to
be related to race or ethnicity (Commission on Violence and Youth, 1993; Elliott,
1994; Tolan & Guerra, 1994b). In addition, the high incidence of violence in inner-
city schools is also related to chronic exposure of school children to violence in the
surrounding communities (Osofsky, 1995). Furthermore, the availability of handguns
and the use of alcohol and other drugs are also considered to be Strong correlates of
inner-city adolescent violence (Commission on Violence and Youth, 1993; DeJong,

1994; Tolan & Guerra, 1994a).

5

Contrary to common perceptions and media portrayals that violence is random and
that victims of violence are victims of unknown assailants, most violent acts, including
homicides, occur between people who are acquainted or related to each other (DeJong,
1994; Tolan & Guerra, 1994b). In a school setting, therefore, most violent and
aggressive acts occur between students who know each other (Commission on Violence
and Youth, 1993; Hausman et al., 1992; Nuttal & Kalesnik, 1987). In a larger
context, the majority of interpersonal violence between adolescents involves same race
and age peers and most frequently involves people who are acquaintances or friends
(Hausman et al., 1992; Nuttal & Kalesnik, 1987).

As outlined above, the causes and characteristics of interpersonal violence in
schools are multiple, complex, and often hard to separate from the larger social and
cultural system (Elliott, 1994; Osofsky, 1995; Tolan & Guerra, 1994b). As a result,
high rates of violence in schools cannot be attributed to a single factor, but instead are
related to numerous environmental catalysts including poverty, access to guns, use of
alcohol and other drugs, and gang activity (Commission on Violence and Youth, 1993;
DeJong, 1994; Tolan & Guerra, 1994b). These factors have contributed to a situation
in which reducing school violence has become one of the most challenging tasks facing
educators, legislators, parents, psychologists, and students (Hranitz & Eddowes, 1990).
Studies have found that the responses and attitudes of a school' S administration can
promote and reinforce students' attitudes towards violence (N uttal & Kalesnik, 1987).
School administrators ean strive actively to remove violence from the school

environment by promoting and reinforcing students' nonacceptance of violence.

6

Similarly, student perceptions of an administration as consistent and ﬁrm in its rule
enforcement have been found to contribute to a reduction in rates of school violence
(Nuttal & Kalesnik, 1987).

In the past decade, attempts to reduce violence in schools have increased in
response to increasing violence. Schools have implemented a variety of programs (both
structural and educational) to enhance a sense of safety and to curtail violence on their
campuses (Tolan & Guerra, 1994b; Webster, 1993). Attempts to reduce violence in
school environments have included physical changes such as installing metal detectors
at entrances to schools, altering bathrooms and school hallways, introducing guards
into schools, and using intercom systems to speed communication between classrooms
and administration ofﬁces (DeJong, 1994; Nuttal & Kalesnik, 1987; Webster, 1993).

Programmatic attempts to reduce violence have included implementing conﬂict
resolution and violence prevention programs designed to teach and encourage students
to use nonviolent methods of resolving interpersonal disputes (N uttal & Kalesnik, 1987;
Prothrow-Stith, 1991). Such programs stress that conﬂict is a normal part of
interpersonal relationships and teach students Skills to manage conﬂict constructively.
Because Statistics indicate that violence occurs more often between adolescents who
know each other, programs aimed at reducing adolescent violence may be less effective
if they do not address the interpersonal relationship aspect of violence (Tolan &
Guerra, 1994b).

In response to the growing concern among educators, in recent years, there has

been a rapid rise in the implementation of violence prevention and conﬂict resolution

_"_'-__r-P

7

curricula and programs in schools (Johnson & Johnson, 1995; Tolan & Guerra, 1994a;
Webster, 1993). Currently, such programs exist in thousands of elementary, middle,
and high schools nationwide. Furthermore, some states are considering making
conﬂict resolution programs part of the mandatory curriculum in their public schools
(Webster, 1993). The widespread adoption of programs reﬂects belief in their apparent
successes (Johnson & Johnson, 1995 ; Webster, 1993). Furthermore, available ﬁndings
of research and program evaluations indieate promising directions for intervention and
prevention strategies (Roth, 1994). However, much of the support for intervention is
based on anecdotal accounts of programs and very little empirical evidence exists as to
whether conﬂict resolution or violence prevention programs are effective at reducing
violence (Johnson & Johnson, 1995 ; Tolan & Guerra, 1994a; Webster, 1993).
Therefore, further research is necessary to determine whether conﬂict resolution

programs are an effective means of reducing violent responses to interpersonal conﬂict.

'5.—

“0"“

This thesis describes an evaluation of a dispute management program. Before the
description of this study is presented it is important to review deﬁnitional issues related
to this ﬁeld of research. In the research literature, the terms conﬂict resolution,
violence prevention, conﬂict intervention, and dispute management are used
interchangeably to describe the same or similar intervention and prevention programs.
In addition, the terms violence and aggression are also used interchangeably (Tate,
Reppucci, & Mulvey, 1995). Finally, both aggression and violence can be found in the
literature on delinquency, which can be deﬁned as ”a heterogeneous concept including

behaviors as diverse as theft, burglary, robbery, vandalism, violence against persons

8
[and] drug use” (Farrington, 1987, p. 33). Research on all of these areas was included

in the literature review conducted for this investigation.

The focus of this study is to understand how dispute management training can
reduce violence in schools. The goals of this Study are (l) to evaluate the effects of a
dispute management training program, and (2) to explore the relationships between
dispute management training and the theoretical processes and factors thought to be
salient in reducing violent conﬂict. These factors include students' prosocial
competence in social conﬂict Situations, perceptions of self-efﬁcacy for competent
behavior, and expectations of the outcomes of competent behavior in interpersonal
conﬂict Situations. In the following sections, an overview of violence prevention and
conﬂict resolution programs is presented. This discussion is followed by a review of
important theories related to violence and conﬂict, with a strong focus on Social
Learning Theory (Bandura, 1977). The constructs of prosocial problem-solving
competence, perceived self-efﬁcacy, and the expectations of the outcomes of competent
behavior are discussed, and research related to these constructs is presented. Finally,
the hypotheses of the current study are presented.

Violence Prevention and Conﬂict Resolution

Multiple school studies have determined that a majority of interpersonal conﬂicts
in schools results from gossip, disputes over property, invasion of privacy, verbal
arguments, and dirty looks (Araki, 1990; Cameron & Dupuis, 1991; Prothrow-Stith,
1987). It is often these types of altercations which escalate into violent confrontations

between students and may cause severe injuries or even death. Arguments are the

9

leading precipitant of homicide -- approximately 50% of all homicides follow
arguments, compared to only two percent that involve gang activity or sexual assault
(Prothrow-Stith & Spivak, 1992).

The Violence Prevention Project (VPP), a community-based outreach and
educational project in Boston, conducted a survey in 1987 with 815 students (57.6%
African—American, 25.3 % Latino) from nine high schools across the United States. It
found that 9.3 % of the students had been physically attacked and hurt at school during
the previous month (DeJong, 1994). More than half of the students reported being in a
physical ﬁght with someone their own age during the past six months; approximately
one in ten Students reported being in a physical ﬁght during the previous week,
although 38.7% said that they had been in a situation during the past week where they
might have gotten into a ﬁght. The interpersonal conﬂict referred to by students most
often involved a friend (23.9%), a family member (14.1%), or someone else they knew
(30.3%) (DeJong, 1994).

Although nationally the actual percentages of students experiencing these types of
violence at school may be relatively low, a few violent acts can have deleterious effects
for a large number of students (Short, 1990; Wolfgang, Figlio, & Sellin, 1972). For
example, one study found that students who bring weapons to school report doing so in
order to protect themselves from being victimized (National Center for School Safety,
1991).

The perceived need for protection and the availability of guns has led to an

increase in the number of ﬁrearm related youth homicides from 1984 to 1989

10
(Fingerhut, Ingram, & Feldman, 1992). A national study conducted in 1990 by the

CDC revealed that nearly 20% of all students in grades 9 through 12 reported carrying
a weapon at least once during the past month. AS a result of the prevalence of guns
and interpersonal violence, many schools struggle to provide a safe learning
environment. Conﬂict resolution and violence prevention programs are a step toward
increasing Student safety by trying to reduce school violence.

Conﬂict resolution programs are based upon the tenets that (l) violence is a
product of interpersonal conﬂict and (2) that violence is a learned response to conﬂict.
A major goal of violence prevention and conﬂict resolution programs is to teach and
foster students' use of nonviolent and constructive methods of resolving interpersonal
disputes. In order to do so, programs present students with the skills required to
manage conﬂicts constructively along with the opportunities to practice and master
these skills. Programs differ in their speciﬁc approaches to teaching this information.
Many programs present students with information on the risks of victimization while
challenging students' attitudes that support the use of violence by teaching means-end
relationships, consequences of social actions, and alternative strategies for resolving
interpersonal problems (Crary, 1992; Prothrow-Stith, 1987; Webster, 1993).

Although empirical evaluations of intervention efforts are lacking, schools that
have implemented conﬂict resolution or dispute management programs report fewer
incidences of escalation of interpersonal conﬂicts between students (Hranitz &
Eddowes, 1990; Prothrow—Stith, Spivak, & Hausman, 1987; Webster, 1993). Schools

with these programs also report positive changes in the overall school environment.

11

Among these changes are increased satisfaction with the school climate reported by
students and teachers, increased feelings of safety on school property, increased levels
of self-esteem among students, and student reports of feeling more able to resolve
conﬂicts constructively and nonviolently (Hranitz & Eddowes, 1990; Prothrow-Stith et
a1. 1987; Webster, 1993). Research has Shown that the instructional use of conﬂict can
lead to increases in achievement, higher level reasoning, and creative problem solving
(Webster, 1993). Despite these ﬁndings, Johnson, Johnson, Dudley, and Acikgoz
(1994) noted that few evaluation studies of conﬂict management programs have been
conducted without methodological weaknesses, and therefore results are tenuous.

Given that the evaluation of conﬂict management programs is still developing it is
relevant to consider multiple approaches towards examining the effects of such
programs. Furthermore, when trying to understand and reduce violence on school
campuses it is relevant to explore theories on the underlying causes of violent behavior.

WW

There are multiple theoretical approaches which attempt to explain the risk factors
and causes of violence. The terms violence and aggression are used interchangeably in
the literature (Tate et al. , 1995) and, therefore, theories of aggression are relevant to a
discussion of violence. Furthermore, both aggression and violence are considered to be
behaviors characterized as delinquent (Farrington, 1987) and, therefore, theories on the
development of delinquency are also relevant.

Many theories explain violence in terms of a multitude of individual and social

factors operating in the development of violence, delinquency, and other related

12

problems (Tolan & Guerra, 1994b). The numerous suggested psychosocial risk factors
of violence include: poverty, exposure to violent or criminal activity, lack of positive
role models, lack of commitment to social norms, pregnancy or birth complications,
harsh or erratic early family experiences, hyperactivity or impulse control problems,
inability to defer gratiﬁcation, impaired cognitive functioning and low academic
achievement, poor peer relations skills, and poor coping skills (Clark, 1994; Elliott,
1994; Tolan & Guerra, 1994b). Theorists incorporate different psychosocial risk
factors into their explanations of violence, aggression, and delinquency. The theories
included here review the biogenetic, social control, ecological, and social learning
views of violence.
Biogenetic

Several different biological and neurological processes are hypothesized to be
linked to violent behavior (Elliott, 1994; Tate et al., 1995). Among these biological
factors are genetic inﬂuences, neurological impairments, and the effects of neurotoxins
(Elliott, 1994; Tate et al., 1995). It is difﬁcult to separate the respective roles of
inherited factors, acquired biological factors (e.g. , birth trauma, head injuries), and
learned psychosocial factors in the development of violence (Commission on Violence
and Youth, 1993; Zigler, Taussig, & Black, 1992). Acquired biological deﬁcits, such
as the effects of low birth weight and other prenatal and perinatal complications,
including exposure to neurotoxins such as lead, have all been linked to or associated
with lower intelligence, decreased attention levels, increased physical activity, and the

prevalence of violent behavior (Commission on Violence and Youth, 1993; Elliott,

13

1994; Zigler et al. , 1992). Furthermore, antisocial personality traits, attention deﬁcit or
hyperactivity disorders, a fearless and impulsive temperament, and serious head injuries
have also been Shown to be related to violent behaviors (Commission on Violence and
Youth, 1993; Elliott, 1994; Tate et al., 1995).

Researchers of biological and pharmacological interventions argue that such
interventions may improve the effectiveness of psychological approaches to treatment
and, therefore, may be effective in helping to reduce violence (Tate et al. , 1995).
However, current understandings of the biological relationships to violence are limited
(Tate, et al. , 1995), and many questions remain as to the interaction of nature and
nurture in the development of violence. Therefore, biological considerations may be
most useful when considered as part of a multidisciplinary approach to violence
prevention (Zigler et al., 1992).

mm

According to control theory, low social control leads to delinquency because it
provides the adolescent with the freedom to deviate from socially desirable behaviors
(Agnew, 1993). Social control theories assume that delinquency results when an
individual's bonds to society are weak or broken (Hirschi, 1971). According to Hirschi
(1971), there are four forms of social control or social bonding, including: ( 1)
attachment, deﬁned as the affection and respect an adolescent holds toward signiﬁcant
others, such as parents and teachers; (2) commitment, which refers to an adolescent's
actual or anticipated investment in conventional activities; (3) involvement, the amount

of time an adolescent spends engaged in conventional activities; and (4) beliefs, deﬁned

14

as an adolescent' s commitment to the central value system of the society. According to
social control theory there is a negative correlation between these four forms of social
bonding and delinquency (Agnew, 1993).

Research on control theory indicates that individuals low in attachment,
commitment, involvement, and belief are more likely than other individuals to engage
in delinquent behavior (Agnew, 1993). Most control theorists assume that all
individuals possess unfulﬁlled needs (Agnew, 1993) and, therefore, have the potential
to act in a delinquent manner. Most individuals, however, do not engage in delinquent
behaviors. It is when control is low (i.e. , when an individual is low in attachment,
commitment, involvement, and belief) that he or She is free to satisfy unfulﬁlled needs
in the most readily available way, which may include engaging in delinquent activity
(Agnew, 1993; Hirschi, 1971).

Most adolescents have nondelinquent, or nonviolent, ways of achieving their goals
and are effective at handling most social situations (Elliott, 1994). Furthermore, most
adolescents are part of social networks of family members and peers and are committed
to conventional norms and values which lower the potential for them to engage in
violent or delinquent behavior. Under such circumstances, violent behavior would be
irrational as it could cause negative ramiﬁcations (Elliott, 1994). However, when an
adolescent lacks such bonds violent behavior may appear to be more rational. Many
modern control theorists have reconoeptualized control theory as a partial explanation

for delinquency or violence and have combined control theory with other theoretical

15
explanations, including social learning theory (Agnew, 1993; O'Donnell, Hawkins, &
Abbott, 1995).
W131

Bronfenbrenner' S (1979) theory of the ecology of human development states that in
order to understand human development it is necessary to go beyond the immediate
situation of the individual to consider the ecological environment as an expanding series
of contexts that surround the individual's immediate setting (i.e. , microsystem,
mesosystem, exosystem, and macrosystem). It is the interrelationships between the
environment, the individual’s behavior, and the social processes that are important to
development.

Ecological models of violence characterize violent actions as a result of individual,
interpersonal, and social risk factors (Tolan & Guerra, 1994b). Bronfenbrenner
(1979), and others (e.g. , Tolan & Guerra, 1994b), described delinquency and violence
as dependent on multiple inﬂuences across the biopsychosocial systems including the
individual, family and peers (interpersonal relations), proximal social settings (e.g. ,
school, community), and broader social macrosystem inﬂuences (such as gun control).
Ecological models assume that the multiple levels inﬂuencing an individual have direct
effects on an individual's risk for delinquent or violent activity (Tolan & Guerra,
1994a). The ecologieal model has important implications for conﬂict resolution and
violence prevention programs. According to this model, interventions must be
designed to correspond to the biopsychosocial system which the program is intended to

inﬂuence (Tolan & Guerra, 1994a; Zigler, et a1. , 1992). For example, school conﬂict

l6

resolution programs often target the individual as the level of intervention (Tolan &
Guerra, 1994b) while community-based intervention programs may target multiple
systems for intervention (Commission on Violence and Youth, 1993; Tolan & Guerra,
1994a).

S . l I . I]

Social learning theory describes the acquisition of behavior as a result of an
interplay between an individual and his or her social experiences (environment).
Bandura's (1977) Social Learning Theory describes how people use cognitive
operations to interpret their social experiences and how these interpretations then come
to inﬂuence their development and future behaviors. In social learning theory, learning
occurs through instrumental conditioning (Winfree, Backstrém, & Mays, 1994).
Accordingly, an individual is conditioned by positive and negative social experiences
and mechanisms. Behavior, therefore, is acquired through the effects, outcomes, or
consequences it has on a person's environment (Akers, 1985), and the resulting
cognitive appraisals a person associates with the behavior.

The primary processes by which conditioning is achieved are reinforcement and
punishment. Behavior is said to be reinforced when repeated episodes are met with a
response that inﬂuences the. actor to engage in the behavior again under similar
circumstances. As a result, behavior increases. Behavior is punished when the
response is such that the actor is discouraged from reengaging in the behavior under
similar circumstances. As a result, the particular behavior decreases (Winfree et al. ,

1994). The principles of reinforcement and punishment have been researched in

l7

qualitative studies of gangs and gang members, and have been found to explain the
process by which gang members regulate the behaviors of their peers (as described in
Winfree et al., 1994).

Individuals learn to evaluate behaviors through contact with others. Violent
behavior is more likely to occur when adolescents develop, through reinforcement and
punishment, orientations which are favorable to violence (Winfree et al. , 1994). The
social interactions and networks of an individual provide the setting in which favorable
(or unfavorable) behavior is deﬁned. Speciﬁcally, social interactions and networks,
through reinforcement of favorable behavior, and punishment of unfavorable behavior ,
promote a deﬁned orientation towards behavior. In a school setting, an individual' 3
social interactions and networks usually consist of peers who hold similar values and
ideas about behavior (Elliott, 1994; Luthar, 1995; Winfree et al., 1994). As a result,
adolescents are likely to engage in behaviors similar to those of their friends (Luthar,
1995; Waegal, 1989). Associations with delinquent or violent peers, therefore, may
increase the likelihood that an individual would engage in violent or delinquent
behavior (Waegal, 1989). If violence is considered by a peer group to be a favorable
response to conﬂict, individuals who are part of this group are more apt to resolve
conﬂicts violently.

The process of learning from a peer group (or other social networks) can occur
vicariously, i.e. , by seeing similar others perform a behavior. Vicarious learning can
result in behavior if an individual observes that the behavior is reinforced and

cognitively appraises the behavior as positive (Bandura, 1982). Therefore, aggression

18

or violence is one of many responses available to an individual and can be learned
through observation and imitation of aggressive or violent models. Subsequently,
violence can be retained and maintained through social reinforcement.

Along with social reinforcement or punishment, two additional cognitive
processes, perceived self-efﬁcacy and the expectations of the outcome of the behavior,
are important determinants of an individual’s decision to act. These appraisals are the
result of direct past experiences, vicarious or modeled experiences, verbal persuasions,
and psychological states (Bandura, 1982). Aggressive and violent responses to
interpersonal conﬂict can be reinforced when the actor: (1) has previous experiences
(direct or vicarious) in which aggressive behavior led to positive results; (2) has
positive perceptions of his or her capability to accomplish the perceived course of
action needed to meet situational demands (self-efﬁcacy) through aggressive behavior;
(3) has lower perceived capabilities to accomplish the desired ends through alternative,
nonviolent actions (social problem-solving competence); and (4) is in a social setting
which does not reinforce the use of nonaggressive responses (outcome expectations).
In sum, violence may occur in a Situation in which an actor lacks experience and/or
perceived self-efﬁcacy to generate multiple problem-solving approaches to resolve an
interpersonal conﬂict and to evaluate their consequences.

The components of social learning theory described above appear to underlie
prominent models and training curricula that aim to foster the use of nonviolent conﬂict

resolution techniques. Because social learning theory states that learning occurs

19

through the cognitive processes of observation, modeling, and response information,
including the evaluation of outcomes and efﬁcacy expectations (Bandura, 1977) ,
altering those components that previously reinforced violent conﬂict resolution
techniques may reduce violence in interpersonal conﬂict situations. Therefore, a
successful conﬂict resolution program should (1) create an environment that reinforces
(through modeling and direct experience) nonviolent means of responding to
interpersonal conﬂict, (2) foster the cognitive development of alternative social
problem-solving and evaluation of consequences of action, (3) boost perceptions of
self-efﬁcacy for socially competent behavior, (4) improve the expectations of the
outcomes of competent behavior, and (5) increase the actual social competence for
acting in speciﬁc interpersonal conﬂict Situations.

Applying the concepts of social learning theory to conﬂict resolution programs, as
prosocial conﬂict resolution techniques are shown to produce positive outcomes and to
avoid punishing ones, and as individuals’ perceptions of personal efﬁcacy are
enhanced, prosocial conﬂict resolution techniques are likely to be retained and
reinforced in social interactions. As a result, prosocial problem-solving competence in
interpersonal conﬂict Situations can be increased.

Programs based upon the principles of social learning theory have successfully
promoted social problem-solving competence, enhanced perceptions of self-efﬁcacy and
outcome expectations, and reduced rates of high-risk behaviors (Allen, Leadbeater, &
Aber, 1990; Caplan, Weissberg, Grober, Sivo, Grady, & Jacoby, 1992). Studies on

the effects of these programs (Allen et al., 1990; Caplan et al. , 1992) have

20

demonstrated that social skills training programs can lead to the development of social
competence, self—efﬁcacy, and outcome expectations. They have shown, for example,
the relationships between these constructs and positive growth in coping Skills and
reductions in self-reported substance use. Because social skills training has successfully
reduced other high-risk behaviors, it is relevant to consider and discuss these same
constructs, i.e. , social problem-solving competence, perceptions of self-efﬁeacy, and
outcome expectations, in relation to reducing violent conﬂict resolution. The following
sections include discussions of prosocial problem—solving competence, perceptions of
self-efﬁcacy in social conﬂict situations, and the expectations of the outcomes of
socially competent behavior.

Yeates and Selman (1989) described social problem-solving as a process that
involves teaching students how to evaluate a situation, develop and implement a
speciﬁc action, and evaluate the effects of their action. In a social problem-solving
intervention, participants are trained to follow a sequence of steps when solving social
problems (Tolan & Guerra, 1994b). Social problem-solving interventions have been
used effectively to alter the speciﬁc behaviors of targeted groups in both primary
prevention programs and intervention efforts (Tolan & Guerra, 1994b). A study
conducted with 112 youth referred to a diagnostic center for treatment of antisocial
behavior compared the effectiveness of three different intervention conditions.
Participants were randomly assigned to one of three conditions, social problem-solving
skills training, social problem-solving skills training with in-vivo practice, or client-

centered relationship therapy (Kazdin, Bass, Siege], & Thomas, 1989). Participants in

21
both social problem-solving conditions participated in 25 individual sessions designed
to teach problem—solving Skills in generating alternative solutions, means-ends and
consequential thinking skills, and perspective taking. Participants in the in-vivo
condition also took part in related activities outside of the 25 sessions. Results
indicated that participants in both of the problem-solving conditions showed higher
prosocial and lower aggression scores (as evaluated by participant, teacher, and parent
reports of behavior) one year after the intervention than did participants in the
relationship therapy condition (Kazdin et al. , 1989). Scores for participants in the
social problem-solving plus in-vivo condition were higher than scores for participants
in the social problem-solving condition (Kazdin et al. , 1989). These results indicate
that not only is social problem-solving a useful approach to enhancing prosocial
behavior, but opportunities to practice new skills further strengthen program
effectiveness.

The effectiveness of social problem-solving programs may be due to their greater
scope than other cognitive interventions and because they often include training in self-
control, anger management, perspective taking, and attitude change (Tolan & Guerra,
1994b). Multiple social problem-solving programs are being implemented in schools to
reduce the prevalence of violent conﬂict resolution (Tolan & Guerra, 1994b). One
such program is the WWW (Prothrow-Stith,
1987).

The Commission on Violence and Youth (1993) included in its summary report

that evaluations of school social problem-solving programs indicate that they can

22

"improve prosocial competence and reduce at-risk behavior among youth who are not
seriously violence-prone by promoting nonviolent norms, lessening the opportunity for
and elicitation of violent acts, and preventing the sporadic violence that emerges
temporarily during adolescence" (p. 55). Reports of the impact of the We:
WWW (Prothrow-Stith, 1987) and other conﬂict
resolution programs indicate that participants show gains in targeted cognitive skills and
beliefs, but most program evaluations have not included appropriate measures of
behavior outcomes, and therefore, the effectiveness of such programs is uncertain
(Tolan & Guerra, 1994b; Webster, 1993).

Although empirically substantiated outcomes of violence prevention programs are
few, research on the use of social problem-solving programs to alter speciﬁc behaviors
lends support for the use of such programs to reduce violence. Systematic approaches
to social problem-solving have been shown to foster social competence,
problem-solving efﬁcacy, and pro-social behavior (Caplan et al. , 1992; Yeates &
Selman, 1989) in other targeted behavioral areas. Studies demonstrate that when
students are taught effective, relevant solutions to interpersonal conﬂicts, they Show
enhanced social competence and pro-social behaviors including the ability to use
constructive techniques to de-escalate interpersonal conﬂicts (Johnson, Johnson, &
Dudley, 1992; Shure & Spivak 1988; Yeates & Selman, 1989). Violence prevention
and conﬂict resolution programs which incorporate these aspects may Show positive
gains in the reduction of violence in interpersonal conﬂict situations. The following

sections will discuss three of the primary elements of social leaming theory that have

23
been promoted through social problem-solving training programs, social problem-
solving competence, perceived self-efﬁcacy, and the expectations of the outcomes of
competent behavior.

WW Social competence, success at meeting
particular age-appropriate societal expectations (Luthar, 1995), is a complex construct.
Social competence is often deﬁned in relation to success in solving interpersonal
conﬂicts (Eisenberg & Harris, 1984) and in research has been operationalized as
positive peer reputations, favorable ratings by teachers, and/or high grades in school
(Luthar, 1995). Ford (1982) deﬁned social competence as the ”attainment of relevant
social goals in speciﬁed social contexts, using appropriate means, and resulting in
positive developmental outcomes" (p. 323). As related to social learning theory, social
competence includes an evaluation of consequences and action in a manner that
increases the possibility of beneﬁcial outcomes and limits the possibility of punishing
outcomes. In terms of interpersonal conﬂict situations, socially competent behavior is
behavior that allows for an effective resolution of an immediate conﬂict reducing the
chances that an individual will face similar problems in the future. AS stated by the
Commission on Violence and Youth (1993), an increase in prosocial competence is
correlated with a reduction in at-risk violent behavior. In this current study, social
competence refers to behaviors that embody both social skills and a prosocial
orientation (Ford, 1982).

Deﬁcits in social competence have been linked to high-risk adolescent behaviors

including substance abuse, school failure, delinquency (Caplan et a1. , 1992), aggressive

24

behavior (Pepler, King, & Byrd, 1991), and unprotected sexual intercourse (Allen et
al. , 1990). Social problem-solving programs, aimed at promoting social competence in
areas including impulsitivity, drug use, and aggression, have been shown to be
successful at reducing the frequency of high-risk behaviors (Caplan et al. , 1992).

These studies seem to indicate that social problem-solving competence may be a critical
mediator in the relationship between problem—solving interventions and the reduction of
high-risk behaviors.

Social problem-solving programs, also referred to in the literature as social
competence promotion programs (e.g. , Caplan et al. ,1992), are designed to enhance
personal and interpersonal effectiveness and to prevent the development of maladaptive
behavior, such as violence, drug use, and school failure. Social problem-solving
programs impact social competence by teaching students developmentally appropriate
skills and information. Such programs foster prosocial and health-enhancing values
and beliefs, and create environmental supports to reinforce the real-life application of
Skills (Weissberg, Caplan, & Sivo, 1989, as cited in Caplan et al., 1992). Such
programs also act to reinforce or punish behavior through social mechanisms.

Behavior is reinforced when other people respond in a way that causes the actor to
engage in the behavior again under similar circumstances. Behavior is punished when
the response from others is such that the actor is discouraged from engaging in the
behavior again under similar circumstances (Winfree et al., 1994). This process is one

way in which social problem-solving programs alter targeted behaviors.

25

In a study conducted by Caplan et al. (1992) relating social competence promotion
to alcohol use by inner-city Sixth and seventh graders (72 program and 134 control
students, 90% African-American), researchers found that students trained in social
competence skills indicated gains in self-reported problem-solving efﬁcacy, improved
constructive conﬂict resolution skills with peers, and increased impulse control and
popularity with peers. Furthermore, students trained in social competence did not
show an increase in excessive alcohol use. However, self-report data from control I
Students indicated that the number of times they reported drinking too much, the
frequency of having three or more drinks, and the usual amount of alcohol they
consumed on a single occasion increased (Caplan et al. , 1992).

In the study by Caplan et al. (1992), social competence training consisted of a
20-session curriculum composed of lessons on stress management, self-esteem,
problem-solving (including problem identiﬁcation, alternative solution generation,
evaluation of consequences, and plan implementation), information on substances and
health, assertiveness, and social networks. Lessons were presented through didactic
instruction, class discussion, videotapes, diaries, small-group role-plays, and work
Sheets. Results indicated that training improved problem-solving efﬁcacy for inner-city
program students including enhanced social adjustment and interpersonal effectiveness.
No gains were reported for the control students. The researchers concluded that the
information and competence training may have equipped students with the critical
thinking skills necessary to recognize the outcomes of their behaviors. Caplan et al. ,

also indieated that competence promotion, when applied to a Speciﬁc domain in a

26

developmentally appropriate content area, may be particularly instrumental in
preventing psychosocial problems. Because behavior and perceptions are difﬁcult to
alter in a short period of time, researchers report being encouraged by the ﬁndings of
this study.

Empirical evidence from social problem-solving programs related to violence
reduction is scarce (Tolan & Guerra, 1994b; Webster, 1993). Furthermore, social
competence has not been directly investigated with respect to school violence
prevention programs. However, given the ﬁndings that social competence training has
been shown to be effective in improving the problem-solving skills and social behavior
of trained adolescents in other domains, it is likely that social competence may act as a
mediator between conﬂict resolution training and nonviolent conﬂict resolution.
Furthermore, the results of research on social problem-solving programs demonstrate
that training can foster the positive development of self-efﬁcacy and outcome
expectations of competent behavior. These ﬁndings suggest that social problem-solving
approaches, such as conﬂict resolution training, may be effective in fostering the
development of self-efﬁcacy and positive outcome expectations of competent behavior.

W As deﬁned by Bandura (1977), an individual's
self—efﬁcacy expectation is one' s conviction in his or her ability to execute a behavior
successfully to achieve a desired outcome. Self-efﬁcacy is a contextual phenomenon
that varies from situation to situation depending upon an individual's experience with
similar situational demands and the individual's perceptions of his or her chances of

success or failure (Bandura, 1977; Wheeler & Ladd, 1982). Beliefs about self-efﬁcacy

27

are dependent upon an individual' S past experience of achievement in a domain,
observation of what others are able to accomplish, attempts of others to mold feelings
of self-efﬁcacy through persuasion, and an individual' 8 own physiological State during
a task (Bandura, 1982; Grusec, 1992). Self-efﬁcacy theory been applied to research in
a variety of domains, including academic achievement, health-related behavior,
parenting styles, self-concept, and social behavior (Grusec, 1992).

Self-efﬁcacy has been shown to be an important mediator of behavior (Bandura,
1982). As described by Bandura (1982), skills and self-beliefs about efﬁcacy are
required for successful functioning in any activity. If perceptions of self-efﬁcacy are
low, individuals tend to behave ineffectively, even when they know what to do in the
given situation (Bandura, 1982). Furthermore, individuals who perceive themselves as
efﬁcacious are more likely to persist in their efforts until they succeed (Bandura, 1982).
As such, an individual who does not perceive him or herself as able to act socially
competent in a given situation may decide to discontinue efforts in that Situation
(Connolly, 1989; Wheeler & Ladd, 1982).

Perceptions of self-efﬁcacy inﬂuence an individual's choice of which behavior to
initiate in a Situation. Beliefs about self-efﬁcacy also inﬂuence the persistence and
degree of effort exerted by an individual in trying to perform a behavior (Bandura,
1977; Lyman, Prentice-Dunn, Wilson, & Bonﬁlio, 1984). Perceived self—efﬁcacy is
highly related to a variety of social abilities including making friends, developing
intimate bonds, being socially assertive, and participating in social groups (Connolly,

1989; Ford, 1982).

28
Caplan et al. (1992) found that along with fostering social competence, social

problem-solving techniques inﬂuenced the development of adolescents' perceptions of
self-efﬁcacy. In other studies, the enhancement of perceived self-efﬁcacy has led to
the improvement of Skills or competencies (Grusec, 1992). Grusec stated that
competencies and self-efﬁcacy are acquired through experience, but in turn they also
regulate the individual's experience in such a way that they are maintained. Research,
such as that by Caplan et al. (1992) and Allen et al. (1990)., and described by Grusec,
has suggested that there are strong relationships among self-efﬁcacy, social
competence, and Speciﬁc behaviors.

WWW Along with one's
perceptions of self-efﬁcacy, one’s evaluations of the outcomes of behaviors may play a
mediating role in how he or she chooses to respond to a speciﬁc Situation. Both
self-efﬁcacy and outcome expectations are potential key elements in understanding how
adolescents act and perceive themselves in interpersonal interactions.

Outcome expectations as deﬁned by Bandura (1977) refer to an individual’ s
estimate that a given performed behavior will lead to a desired outcome. While both
perceived self-efﬁcacy and outcome expectations of speciﬁc behaviors have been
related to social competence (Allen, Weissberg, & Hawkins, 1989), it is important to
differentiate between the two (Bandura, 1977). An individual's belief in his or her
ability to perform a given behavior (perceived self-efﬁcacy) lacks meaning unless the
individual also believes that performing the behavior will lead to desired outcome

(outcome expectation). Furthermore, an individual may believe that a given outcome

29

could improve a situation but may feel unable to execute the behavior appropriately and
may choose an alternative response. It may be that in order to enhance an adolescents'
social competence, one must also foster a sense of self-efﬁcacy and positive
expectations of the outcomes of competent behavior.

Allen et al. (1990) stated that perceptions of self—efﬁcacy, when considered in
reference to Speciﬁc situations and behaviors, have been related to a range of adaptive
behaviors (Bandura, Adams, Hardy, & Howells, 1980; Allen et al., 1990). Existing
research shows that past experiences inﬂuence an individual's perceptions of
self-efﬁcacy and outcome expectations. These ﬁndings provide evidence that prior
positive experiences can improve perceptions of self-efﬁcacy and outcome expectations
and thereby also increase the degree of effort and persistence placed on performing
competent behaviors (Allen et al. , 1990; Bandura, 1977). Therefore, low efﬁcacy
expectations and negative appraisals of behavioral outcomes can be extinguished by
providing experiences that boost perceptions of self-efﬁcacy and outcome expectations
and disconﬁrm previous negative experiences (Bandura, 1977).

Allen et al. (1990) found that adolescents' expectations inﬂuenced their self images
in social Situations. Further, results indicated that adolescents' perceptions of
self-efﬁcacy are highly correlated to their social problem-solving competence. These
researchers found that social problem-solving competence was negatively correlated
with delinquency and drug use for both male and female research participants.

Allen et al. (1990) suggested that past failures with competent behaviors (i.e. ,

prosocial behaviors) may lead to low perceptions of self-efﬁcacy and to reduced

30

persistence in carrying out competent behaviors. They suggested that reduced
persistence in carrying out competent behavior ean lead, over time, to further
reductions in perceptions of self-efﬁcacy until only less competent behaviors are
perceived to be available to the adolescent (Allen et al. , 1990).

Results of Allen et al. (1990) indicated the importance of considering adolescents'
perceptions of self-efﬁcacy and outcome expectations for competent behavior in
designing and implementing a program aimed at promoting competent behavior in
adolescents. Their study also suggested the need to allow for successful experiences in
social conﬂict situations if adolescents are to gain positive perceptions of their
self-efﬁcacy and positive expectations of the outcomes of competent behavior.
Furthermore, fostering self-efﬁcacy and promoting positive expectations of the
outcomes of competent behavior may be essential to the development of social
competence and nonviolent conﬂict resolution strategies.

Summant. Perceptions of self-efﬁcacy and expectations of the outcomes of
competent behavior are linked to adolescent social competence and may mediate the
relationship between social problem-solving training, such as conﬂict resolution
training, and social competence. Furthermore, social competence may mediate the
relationship between adolescents' expectations and nonviolent interpersonal conﬂict
resolution.

If, through violence prevention training, adolescents are provided with the
knowledge and opportunities for positive experiences and reinforcement for resolving

interpersonal problem situations nonviolently, their perceptions of self-efﬁcacy and

31

expectations of the outcomes of competent behavior may be enhanced. This
enhancement, furthermore, may foster adolescents' social problem-solving competence,
which in turn will promote nonviolent interpersonal conﬂict resolution strategies. A
violence prevention training curriculum which may provide these opportunities for
adolescents is the W (Prothrow-Stith,
1987).

ll' 1 E . C . l E E l l

The W15, developed by
Prothrow-Stith (1987), is a program consisting of ten sessions designed to provide
students with information on the risks of violence and homicide, to teach various
alternatives to violence such as conﬂict resolution techniques, and to create a classroom
and school environment that is nonviolent and that values violence prevention behavior
(Prothrow—Stith et al., 1987).

The Wm is didactic and cognitive and combines
information delivery and situation role-plays. The curriculum stresses that while anger
is normal, adolescents can be taught to change their responses to anger from violent or
destructive to nonviolent or constructive. Through social problem-solving techniques,
the curriculum attempts to teach students to be creative in their responses to anger and
to determine for themselves the risks and beneﬁts of ﬁghting. Fighting is presented as
one choice among many to resolve conﬂict. The content of the ten-session Violence
WW (Prothrow-Stith, 1987) is described in more

detail in the following chapter (Methods).

32

The W is based upon the principles
of social learning theory. As described earlier, in social learning theory, behavior
(direct or modeled) is reinforced when it is met with a response that inﬂuences the
actor to engage in the behavior again under similar circumstances. AS a result,
perceptions of self—efﬁeacy in social situations increase, expectations of the outcomes of
competent (prosocial) behavior increase, and behavior increases. Behavior is punished
when the actor is discouraged from reengaging in the behavior under similar
circumstances. AS a result of punishment, the particular behavior decreases (Winfree
et al. , 1994).

It is through this process of reinforcement and modeling of behavior that conﬂict
resolution programs, such as the WW
attempt to increase students’ use of nonviolent conﬂict resolution techniques. The
program attempts to provide adolescents with effective prosocial means of resolving
conﬂict though information provision and modeling of speciﬁc behaviors. Role play
situations are used frequently during training sessions to allow for the acting out and
modeling of appropriate ways to resolve conﬂicts nonviolently. Students can learn
vicariously of the reinforcement and reward of one of their peers in such a situation.
These behaviors are then reinforced through discussions in the classroom that provide
social contexts and interactions in which students can evaluate the behavior.
Furthermore, students are encouraged to talk about conﬂicts they have seen or
experienced during the previous week. This interaction allows for direct feedback

(reinforcement or punishment) from other students and from the program facilitator to

33

the person who was involved in the conﬂict and also provides vicarious learning
experiences to other students in the class.

The W was ﬁrst implemented in
several Boston high schools in 1984. It has since gained acceptance throughout the
nation and has been implemented in various other cities. The curriculum has also been
used as part of a larger community-wide intervention in Boston that included a mass
media campaign and the involvement of community agencies (Hausman, Spivack,
Prothrow-Stith, Roeber, 1992).

Results of school evaluations of the WW
Adolescents have been mixed. A study conducted in 1987 showed signiﬁcant gains in
students' knowledge about violence. School suspension data suggested that suspensions
due to violence were reduced (Prothrow-Stith, et al. , 1987). However, a study
conducted by Spiro, et al., (1989) and discussed in Webster (1993) of tenth-graders at
six inner-city schools around the U.S. reported that no signiﬁcant changes were found
in students' total post-test knowledge about violence; their attitudes about ways to
handle conﬂicts; acceptance of violence; violence locus of control; self-esteem; or
self-reported ﬁghting, drug use, or weapon carrying. However, when stratiﬁed by
school and compared to the control group, the treatment group in one school had
greater gains in knowledge about violence, and the treatment group in another school
demonstrated greater gains in self-esteem. In twoschools, results indicated that students
were less likely to believe that people other than themselves were responsible for

preventing ﬁghts (Webster, 1993). Researchers attributed the apparently contradictory

34

results to the stylistic differences of the individual program implementers.
Furthermore, researchers attributed the overall lack of program effectiveness to
targeting the program at high school students and not at the preferred middle school
population.

Despite inconclusive results, the WWW
(Prothrow-Stith, 1987) has received highly positive anecdotal results and continues to
be one of the most widely implemented and promising conﬂict resolution programs in
the country (DeJong, 1994; Webster, 1993). The rate at which schools have adopted
programs and the rate at which programs are being funded are both increasing rapidly.
While programs may be inexpensive to implement at an individual school or school
system, when considered in aggregate, the program costs add up to considerable
personnel and monetary resources (Webster, 1993). In consideration of these efforts
and costs, programs need to be closely evaluated to determine the effects they have on
our schools and students. To date, there have been few empirical evaluations
conducted on programs developed to reduce violence in schools (Commission on
Violence and Youth, 1993; Elliott, 1994; Johnson et al., 1994; Johnson, Johnson, &
Stevahn, 1995).

During the 1993-1994 school year a county ofﬁce on violence reduction in
conjunction with a large, metropolitan public school district in the Midwest
implemented a dispute management training program based upon the Meme
WWW (Prothrow-Stith, 1987). This current study is

an investigation of the effects of this program.

35
Goals and Hypotheses

The current investigation was exploratory research on the effects of a dispute
management program. The goals were to determine the effects of the intervention on
adolescents' social problem-solving competence, perceptions of self-efﬁcacy for
performing competent behaviors, and expectations about the outcomes of competent
behavior in interpersonal conﬂict situations. This investigation also attempted to
determine the relationship between social problem-solving competence, perceptions of
self-efﬁcacy, outcome expectations for competent behavior, and nonviolent conﬂict
resolution.

It was eXpected that because the WWW
provided students with knowledge and opportunities for positive experiences in
resolving interpersonal problem situations nonviolently, it would foster the positive
development of adolescents' self-efﬁcacy and outcome expectations for competent
behavior in interpersonal conﬂict situations. As a result, adolescents’ prosocial
problem-solving competence would also be enhanced and would mediate the increased
use of nonviolent conﬂict resolution strategies. It was hypothesized that students who
participated in the dispute management training would: (1) perceive themselves as
more efﬁcacious, (2) have more positive expectations for the outcomes of competent
behavior, (3) report more socially competent responses to interpersonal conﬂict
Situations, and (4) resolve interpersonal conﬂict situations nonviolently more frequently

than students who did not participate in the program.

36

A model was developed to test the effects of the dispute management program (see
Figure 1). In the model, perceived self-efﬁcacy and the expectations of the outcomes
of competent behavior were hypothesized to mediate the relationship between dispute
management training and social problem-solving competence. In addition, social
problem-solving competence was hypothesized to mediate the relationship between

perceived self-efﬁcacy and the expectations of the outcomes of competent behavior.

37

 

 

ooze—08M

8580
32228 Z

No.

 

 

 

 

0833800
328-335

38.0.
k

 

ac.

an.

 

80388me
080350

 

 

seamsum
328qu

 

EB: Emacs s ”Ema

co.

 

 

means
Deoaowmomz
83me

 

 

Chapter Two
METHOD
Setting

A comprehensive violence prevention program was initiated by a Midwestem
metropolitan County Ofﬁce on Violence Reduction in conjunction with its public school
system. As part of this program, seventh grade homerooms at two middle schools were
chosen to participate in dispute management training for a ten-week period starting the
end of March, 1994. Homeroom classes were chosen to participate by each school's
administration in conjunction with the homeroom teacher. With the exception of a
homeroom of honor students at one school, and a classroom of students with learning
disabilities in another school, students were assigned to homeroom classrooms to reﬂect
a cross-section of academic ability. Students at all levels of academic ability
participated in the dispute management training. Approximately 100 students per
school participated in the training.

The County Ofﬁce on Violence Reduction and the public school system chose the
schools receiving training from approximately 33 schools with sixth, seventh, and
eighth grades. They also selected two middle schools with similar demographics,
student population, and rates and patterns of violent or illegal acts to be comparison
schools. Information on the incidence and type of violence was obtained from a central
computerized data system maintained by the public school system for reporting

violations of the school offense code.

38

39

In the school system’s records, offenses are reported individually and are included
in one of the following three categories: type "A" offenses include those that would be
considered criminal or delinquent acts such as ﬁghting, possession of a weapon, and
breaking and entering; type "B" offenses include non—criminal type activities such as
truancy; and type ”V” offenses are violent offenses such as use of a ﬁrearm or hitting a
teacher. A review of the Student Code Violations report for the 1992 - 1993 school
year indicated that 1610 students attended the two treatment schools combined. At
these two schools, seven violent (type V) acts and 197 illegal acts (type A) were
reported in the Student Code Violations report. The report indicated that 1197 students
attended the two control schools combined. Reported violations in the control schools
indicated that four violent acts and 122 illegal acts were committed. A chi-square test
for independence indicated that there were no signiﬁcant differences between the
number of students who were violent or who committed illegal acts in the treatment and
control schools, X2 (1, N = 2,807) = 3.03, NS. The results of the chi-squared
analysis indicated that the the treatment and control schools were statistically similar in
terms of the number and type of violent acts.

The violence reduction project was designed to take place in the following three
stages: (1) dispute management training for students, (2) peer mediation, and (3)
teacher dispute mediation training. The long term goal of the County Ofﬁce on
Volence Reduction is to establish ongoing dispute management programs in all of the
middle schools in the metropolitan public school system and throughout the rest of

county. These plans would include providing training opportunities in the two

40
comparison middle schools that participated in this study. An evaluation of the entire

program was conducted by investigators associated with the Institute for Public Policy
and Social Research at Michigan State University. The overall evaluation addressed the
following four questions: (1) Have students learned non-violent methods of conﬂict
resolution? How are these techniques utilized? (2) What is the change in the students'
feeling of safety at school? (3) Has the climate of the school been altered to support
the intervention? (4) What is the impact of the intervention on the numbers and types
of school behavioral violations?

The student population in the public school system used in this study is
approximately 95 % African-American. The system-wide teacher and staff population
is approximately 60 to 70% African-American. The student population in the schools
participating in this study was close to 100% African-American. The student to teacher
ratio at each of the schools was approximately 25 to 30 students to one teacher.

Subjects

At the conclusion of dispute management training at the participating schools,
seventh graders at the two treatment and two comparison schools were recruited to be
interviewed. Researchers had a list of students from each of the seventh grade
homeroom classrooms at each school. In each of the treatment schools, twenty-ﬁve of
the approximately 100 seventh grade students who participated in the dispute
management training were randomly selected to be interviewed. Twenty-ﬁve students
in homeroom classrooms that did not receive training were also selected at random to

be interviewed.

41

In each treatment and control school a member of the school' 8 staff was given a
list of the students who were selected at random. This staff member managed the
process of recruiting students and delivering them to the interview site. At the
beginning of each interview students were informed that their participation was
voluntary. In addition, students were told that they could stop the interview at any
time. While some students may have chosen not to answer speciﬁc questions, no
student refused to participate. Two-hundred and eleven interviews were completed at
the four schools.

At the completion of the interview at the treatment schools, students were asked to
identify whether they had participated in the conﬂict resolution training. In one
treatment school twenty-four students interviewed identiﬁed themselves as having
participated in training; in the other treatment school fourteen students responded that
they had been trained. A total of 38 Students interviewed indicated that they had been
trained in dispute management skills. Sixty-ﬁve students interviewed at the treatment
schools identiﬁed themselves as not participating in the conﬂict resolution training
program. Table 1 indicates that the numbers of students who identiﬁed themselves as
being trained in dispute management skills is far less than the number of students who
gave a homeroom number that corresponded with homerooms receiving training.
According to homeroom numbers reported by students during interviews, 67 students in
the treatment schools reported that they were assigned to homerooms that received
training. The discrepancy between self-report data and data of students in Speciﬁc

homerooms may indicate that students were not aware of the kind of program in which

42

 

 

Table 1
I"! ‘ :-.,.-.;£.- n”..- '. e'a. _- U .. ... -- --l-.,.,.. a. . ... L.
W
_Sel£-rencn_ W
School Condition
Treatment Trained 38 67
Not trained 65 36
Control Not applicable 108

 

43

they were involved. Because of the discrepancy, it is difﬁcult to know the true number
of interviewed students who participated in training. It was decided to use the count
obtained through self-report in all analyses, that is, 38 students.

At each of the control schools a staff member was given a list of 50 seventh grade
students selected at random. Fifty-two students at one control school were interviewed,
and 56 at the other control school were interviewed. A total of 108 Students at the
control schools participated in the interview process.

Design

Student interviews were used to collect information about the relationships among
dispute management training, social problem-solving competence, perceived social
self-efﬁcacy in performing competent behavior, and the expectations of the outcome of
competent behavior. In addition, interviews included questions on self-reported pro-
and anti-social behaviors, including the use of nonviolent conﬂict resolution strategies.

The current study was an independent groups, posttest-only investigation of the
effects of dispute management training on middle school students. Students trained in
dispute management were compared to students who attended the same school but were
not trained, and to students who attended a control school who also did not receive
training in conﬂict resolution skills.

Schools were selected due to their similar Size, student population, demographics,
and number and type of violent incidents reported in school code violations (as
described previously). Matching schools for these characteristics provided similar

student populations that could be considered equivalent in nature at the onset of the

44

dispute management intervention. Furthermore, the results of chi-square analysis of
data included in the School Violations Report indicated that there were no Signiﬁcant
differences in the frequencies of violent and illegal acts between the treatment and
control schools. This investigation, therefore, was well-suited for a post-only
measurement design. The dispute management training program was the independent
variable in this study. Social problem-solving competence, perceived self-efﬁcacy for
performing competent behaviors, outcome expectations of competent behavior, and
self-reported pro- and anti-social conﬂict resolution behaviors were the dependent
variables.
Program Description

The dispute management training program was delivered to students in the form of
weekly one-hour sessions for a period of ten weeks. Two independent consultants were
hired by the County Ofﬁce on Violence Reduction to teach the dispute management
training in the two treatment schools. The curriculum used in the training, the
W8, (Prothrow-Stith, 1987), was designed
to stress that anger is a normal emotion experienced by all people. Furthermore, the
curriculum recognized that aggression and violence are reactions to anger and conﬂict
that are learned through interaction with other people. The curriculum aimed to teach
students how to alter their reactions to interpersonal conﬂict through lessons on self-
control, anger management, perspective taking, and attitude change.

Most of the sessions included role play situations and frank conversations about

interpersonal conﬂicts students experienced from one session to the next. The role

45

plays and discussions allowed students the opportunities to learn new responses to
interpersonal conﬂict from their peers and from the facilitator. More speciﬁcally, the
use of role play situations provided students the opportunity to learn how interpersonal
conﬂict can escalate into violence and the possible avenues for diverting the conﬂict
away from a violent conclusion. Students who acted in the role plays had the
opportunity to learn through direct action, and students who watched the role plays had
the opportunity to learn vicariously through the modeled behavior. Furthermore, all
students were witnesses to reinforcement (or punishment) from student and facilitator
feedback on the resolution of conﬂict portrayed in the role plays. Discussions of
conﬂicts students experienced from one week to the next allowed for direct and
vicarious learning and feedback from students and the facilitator. The following
paragraphs provide a brief overview of the sessions included in the W
W (Prothrow-Stith, 1987).

Session One of the Wis, “There is a
Lot of Violence in Society," aims to determine what information and misinformation
students have about violence. Topics discussed during Session One include the causes
and effects of violence and the extent and types of violence in society. The overall
focus of the Wm (Prothrow-Stith, 1987) is interpersonal
violence. The curriculum deﬁnes interpersonal violence as accidental or intentional
violence that occurs between two or more persons. Intentional violence, as deﬁned in
Session One, includes stranger violence, that is, violence between persons who do not

know each other, as might occur incidentally during the commission of a crime; sexual

46

violence, such as rape; and acquaintance violence, that is, violence among people who
know each other, such as friends, peers, or family members (Prothrow-Stith, 1987).

Session Two, “Homicide: Statistics and Characteristics," presents statistical data
on homicide and the characteristics of homicide. Students are challenged to think about
the statistical information and the relationships between homicide and weapons,
alcohol, and arguments. Homicide is presented as the second leading cause of death
for young people ages ﬁfteen to twenty-four (Prothrow-Stith, 1987).

Session Three, “Exploring Risk Factors,” presents violence-related risk factors
introduced during Session Two, and describes the physiological effects of alcohol on
the brain and alcohol’s relevance to interpersonal violence. Information on weapons
and homicide is presented in detail. In addition, the curriculum provides facilitators
with information suitable for a mini-lecture on poverty, race, and homicide. In sum,
the information presents poverty as another factor that contributes to honricide and as a
more salient factor than race.

Session Four, "Anger is Normal," describes anger as a normal and natural part of
life. Information is provided on the physiological changes that occur when someone is
angry, including the concept of ”ﬁght or ﬂight” during an interpersonal confrontation.
Students are encouraged to talk about what makes them angry.

Session Five, ”There are Healthy and Unhealthy Ways to Express Anger, " builds
on the idea presented in Session Four that anger is normal and illustrates the healthy
and unhealthy ways to express anger. The session aims to help students understand that

there are constructive and destructive ways to deal with anger.

47

Session Six, "There's More to Lose than to Gain from Fighting," challenges
Students to compare the positive and negative consequences of ﬁghting. The session
attempts to demonstrate that the negative consequences of ﬁghting outweigh the
positive. To do so, students are asked to make a list of the positive consequences of
ﬁghting (e.g. , winning, proving a point) and a list of the negative consequences (e. g. ,
being hurt, being embarrassed.)

Session Seven, ”What Happens Before, During, and After a Fight,” illustrates the
steps that precede many ﬁghts situations. Included in this section is a discussion of the
roles of peer pressure, increased emotions, and nonverbal indicators in a ﬁght situation.
Students create role plays of ﬁght situations and are asked to dissect the scenes for cues
to how the ﬁghts escalated, how emotions were expressed verbally and nonverbally,
and the role of peer pressure in the ﬁghts (Prothrow-Stith, 1987).

Session Eight, “Preventing Violence, " helps students determine ways that violence
may be prevented through analyzing a ﬁght. In addition, students discuss the
differences between prevention and intervention in a ﬁght situation, and outline
methods to prevent violence in a school environment.

Session Nine, "Fighting - What Else is There, " emphasized that there are many
choices available to students other than ﬁght or ﬂight when confronted with a conﬂict.
Students are asked to identify obstacles to nonviolent resolutions of conﬂict, and to
describe how violence is glamorized in our society by television, etc. The curriculum

includes multiple situations that could lead to conﬂict and provides students the

48

opportunity to identify the choices available to try to resolve the situations
nonviolently.

Session Ten, "Practice Throwing a Curve, " provides students with an opportunity
to practice the Skills of nonviolent conﬂict resolution. This ﬁnal session encourages
students to be empathetic with their opponents and to identify alternatives to the ﬁght
or ﬂight concept discussed in previous sessions. Role play situations are again
encouraged. Students are challenged to summarize the perspectives of both parties in
the conﬂict situations, to use the Skills from the program in a role-play situation, and to
recognize that ﬁghting is only one of the several choices in a conﬂict situation.

Procedure

Student interviews were conducted in the auditoriums, libraries, or cafeterias of
each school during times when the rooms were not occupied by other students.
Interviews were conducted in these areas to allow students and interviewers adequate
privacy and quiet space.

Student interviews were conducted by graduate students from the Criminal Justice
and Psychology departments at Michigan State University and by local residents who
were familiar with the public school system. Interviewers were trained by the
investigators in interviewing techniques and were familiar with the instruments prior to
conducting their ﬁrst interview. AS much as possible, interviewers reﬂected the ethnic
background of the student population in the schools.

Each interview participant was interviewed for approximately thirty nrinutes during

school hours. Interviewers were asked to write verbatim each response given by the

49

students for each scenario and to circle the response students indicated for the number
of times they had engaged in the activities included in the different items on self-report
delinquency, victimization, and positive conﬂict resolution techniques. In exchange for
the cooperation of the school adrrrinistration and the students' participation, each school
received ﬁve dollars per student interviewed.

As part of an agreement between the county ofﬁce of violence reduction and the
public school system, individual guardian permission for students to be interviewed was
not required. Student participants were made aware that their responses were
conﬁdential and were reminded that they should decline to answer any questions that
they felt uncomfortable answering honestly.

Measures

Revised scenarios from the Adolescent Problem Inventory (API) for boys
(Freedman, Rosenthal, Donohoe, Schlundt, & McFall, 1978) and the Problem
Inventory for Adolescent Girls (PIAG) (Gaffney & McFall, 1981) and scenarios
developed for this study were used to measure adolescents’ social problem-solving
competence, expectations of self-efﬁcacy in performing competent behaviors, and
outcome expectations for competent behavior in interpersonal conﬂict Situations
(procedure is similar to that used by Allen et al., 1990) (see Appendix A).

A total of seven scenarios were used, three adapted from the API or PIAG and
four created for this investigation (see Table 2). The interpersonal conﬂict scenarios
included social conﬂicts or potential conﬂicts with parent, teachers, and peers in

Situations that an adolescent may experience directly or may be able to imagine

Table 2

C [51.5.2151

Scenario
Scenario one

Scenario two

Scenario three

Scenario four

Scenario ﬁve

Scenario Six

Scenario seven

50

Scenario Content
You’ve been disrupting a substitute teacher's class all week, and
She sends you up to the principal' s ofﬁce again. The principal
meets you at the door and says, "You have been sent here three
times this week and I'm excluding you this time!"

Your mother has been nagging you about getting home
immediately after school. One day you get home an hour late
and your mother yells at you, ”Do you know what time it is?
Where have you been?”

A kid you know is a drug dealer. He found out that school
security is going to check his locker after the next class. He asks
you to keep his stuff in your locker.

You think that one of your classmates has recently been Spreading
nasty rumors about you. You don't know this student well, so
you don't know why she (he) is saying these things. You're
upset about these rumors, and you want the rumors to stop.

You are walking home alone down a dark street. Two boys
(girls) are following you. You don't know them, but they are
about your age. They catch up with you and say, "We want your
jacket. "

You are Sitting with your friends in the lunch room. A girl (boy)
you don't know spills her (his) drink on you. One of your
friends tells her (him) to apologize, but she (he) refuses.

You are at school and you see another student talking to your
girlfriend (boyfriend). They are laughing and smiling and seem
to be having fun. Suddenly, your girlfriend (boyfriend) looks at
you and then turns quickly away.

51

experiencing. The content of the scenarios was reviewed by school ofﬁcials including
teachers and principals, students, and psychologists who agreed that the Situations were
similar to those a student may experience.

Adolescents' self-reported responses to the conﬂict Situations presented during the
student interview were rated to measure social problem-solving competence. Students
were asked to consider what they would have done in the conﬂict situation in
comparison to a sample response of how another adolescent responded in the same
situation. Perceived self-efﬁcacy and outcome expectations of competent behavior
were obtained by rating these responses. Social problem-solving competence and the
expectations of the outcomes of competent behavior were rated using coding manuals
developed for this study. Perceived self-efﬁcacy was rated on a Likert—type scale.

In the next section, background information related to the development of the
coding manuals for social problem-solving competence and the expectations of the
outcome of competent behavior is discussed. The sections following present
information on the measurement of the constructs social problem-solving competence,
perceived self-efﬁcacy, and the expectations of the outcomes of competent behavior.
Finally, nonviolent conﬂict resolution behavior is discussed.

Dcxclapmmmtmmmnuals

Two undergraduate psychology majors seeking independent research credits
worked with the principal researcher to develop a coding manual for social problem-
solving competence for the six interpersonal conﬂict scenarios and to code the

interview information according to the developed manuals (see Appendix A). The

52
three coders met on a weekly basis from May to August, 1994. One of the

undergraduate students continued with the project to assist in developing a coding
manual for outcome expectations for the six scenarios (see Appendix B). The
development of the coding manual for outcome expectations and the coding process
was accomplished from September to December, 1994. Each undergraduate student
received two independent research credits for each semester of her participation.

Responses collected from the ﬁrst wave of data collection (used in the overall
evaluation project) were used as the basis for the development of the coding manuals.
The process used to develop the coding manuals ensured that all responses included in
the coding manuals are ones that were actually given by students asked to respond to
the scenarios. Of the original seven scenarios presented to students, six scenarios were
coded using the developed manuals. The remaining scenario (Scenario One as
indicated in Table 2) was discarded because initial review of students' responses
indicated that there was no distribution of answers. Instead, the majority of students
gave the same or a very Similar response. Student interviews were organized by a four
digit student number which indicated the school the student attended (one through
four), the wave of data collection (wave two of the overall evaluation), and their
interview number.
5 . l E l l -S l . [2

Social problem-solving competence is deﬁned as a response that effectively
resolves the interpersonal conﬂict situation at hand and makes it less likely that the

subject will experience more problems of this type in the future. As deﬁned by Ford

53
(1982), and applied to this investigation, social competence embodies both social skills
and a pro-social orientation. To measure social problem-solving competence, each
student was presented with the interpersonal conﬂict scenarios one at a time and was
asked to consider his or her response to the Situation. Students were asked to answer
the question, ”What would you say or do now?” Student responses were written in the
interview booklet and were used to code social problem-solving competence.

The manual developed to code social problem-solving competence is similar to a
manual used by Allen et al. (1990). In the manual, social competence is rated on a
zero to eight scale with a score of zero corresponding to a very incompetent or very
ineffective response, and a score of eight corresponding to a very competent or very
effective response. Maximally competent responses are those which were identiﬁed by
school administrators and teachers as most likely to resolve the conﬂict and reduce the
chance that the adolescent would experience a Similar situation in the future. The
midpoint on this scale is a four which corresponds to a response that is neither
competent nor incompetent, i.e. , that neither directly helps nor hurts the Situation or
individual involved in the interpersonal conﬂict.

Responses to the scenarios gained from the ﬁrst wave of data collection (used in
the overall project evaluation) were reviewed by each coder who used them to create a
preliminary ranking of responses using the zero to eight scale. Coders worked on one
scenario at a time and reviewed and discussed preliminary coding schemes for each
scenario at weekly meetings. Meetings provided the opportunity for the coders to

discuss openly their reasons for placing each response at each level on the scale. This

54

process initially led to the development of a category for each scenario at all nine
levels. As more responses were reviewed, the coding manual was revised to reﬂect
more accurately the responses given and the level of social problem-solving competence
represented by the response. levels within scenarios were combined and collapsed if
the coders evaluated that two categories represented the same level of social
competence. Collapsing levels within scenarios in this manner led to the development
of scales with different coding levels per scenario. For example, if a scenario had a
level six that was determined to represent the same level of social competence as
responses originally coded as level ﬁve, the two category levels were combined.
Because this on-going process led to each scenario having different possible coding
levels, the ﬁnal scores for social problem-solving competence were standardized using
the z-statistic procedure.

In the beginning of the coding process, all coders rated the same interviews
starting with groups of ten interviews. Any discrepancies on scenario coding levels
were discussed at the weekly meetings. Each coder presented her argument for why
she coded the response at a given level. Once interrater reliability was established at
over 90% across scenarios for the ten selected interviews, the three coders rated all six
scenarios for all student interviews. Percent agreement interrater reliabilities for
coding all of the interviews were calculated for each scenario and ranged from 89.6 to
97.2 percent (see Table 3). The primary researcher assigned the ﬁnal coding score to
the scenarios in which the three coders were discrepant. In the end, each student

interview had codes for each scenario response given. Responses such as ”no answer”

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3
EH]. 5 !'§ § .12“ 25.1. E I

ResrrltL
Scenario Mean Standard Corrected Item

Deviation Total Correlation

Scenario 2 -3.17 1.49 .18
Scenario 3 -.19 .69 .23
Scenario 4 .34 .88 .23
Scenario 5 .46 .91 .33
Scenario 6 -.29 1.07 .14
a = .43
Percent agreement interrater reliability range 89.6 to 97 .2
Table 4
{‘W-Is at. ° 0 .I.‘-. .IH o 1")..011'0 0||'.'.‘1:.‘l-0

Results
Scenario Mean Standard Corrected Item

Deviation Total Correlation

Scenario 3 3.04 1.62 .26
Scenario 4 3.05 1.49 .45
Scenario 5 3.03 1.53 .36
Scenario 6 3.09 1.25 .27
Scenario 7 3.64 1.59 .35
a = .58
Percent agreement interrater reliability range 88.6 to 94.8
Table 5
E l. H. 3 l . E E . l§ lE‘EEE

Results
Scenario Mean Standard Corrected Item

Deviation Total Correlation

Scenario 2 4.56 .88 .19
Scenario 4 4.45 .83 .27
Scenario 5 3.97 1.49 .29
Scenario 6 3.35 1.47 .25
Scenario 7 3.86 1.47 .25

 

56

were excluded as missing data when analyses were preformed. The response, "I don't
know, " corresponded to a zero or a very ineffective response. A response of "nothing"
was coded at different level for each scenario because in some scenarios doing nothing
is an effective response to the situation while in others it is not.

Scale reliability analysis was calculated for social problem-solving Competence
scores on the six scenarios (a = .41). A review of the corrected item-total correlations
indicated that Scenario Seven (as indicated on Table 3) was not highly correlated with
the other ﬁve scenarios. This scenario was discarded for social problem-solving
competence. Reliability analysis was run on the remaining ﬁve scenarios ((1 = .42).
An average social problem-solving scale score was created to account for any missing
data in the remaining ﬁve scenarios. A total of 204 student interviews were included in
the coding process and the ﬁnal social competence scale; the remaining seven student
interviews represent missing data. Missing data may be the result of students declining
to answer Speciﬁc questions. The average scale score was used for all subsequent
analyses.

QumeExpectations

As deﬁned by Bandura (1977), outcome expectations refer to an individual's
perception of the liklihood that a given performed behavior will lead to a desired
outcome. Adolescents' expectations about the outcomes of competent behavior were
obtained by presenting the students with the seven interpersonal conﬂict situations a
second time, followed by a competent hypothetical response of another adolescent to

the conﬂict situation (similar to the process used by Allen et al. , 1990). Students were

57

asked to answer the open-ended question, "If you tried this, what do you think would
happen?” The response of the hypothetical adolescent was derived from competent
responses to the scenarios as evaluated by school ofﬁcials and psychologists as most
likely to resolve the conﬂict at hand and to reduce the chance that the adolescent would
experience a similar situation in the future.

The process for developing the coding manual for outcome expectations was
Similar to the process described above for social problem-solving competence.
Responses designating different levels were again derived from the responses collected
during wave one interviews. Outcome expectations were rated on a one to ﬁve scale,
with one corresponding to an expectation of a very negative outcome, and ﬁve
corresponding to an expectation of a very positive outcome. Each scenario had a
response category at each of the ﬁve possible levels. The two coders working on the
manual initially worked independently to determine the placement of possible responses
on the coding scale and met weekly to discuss response levels.

All responses were coded by each coder to assure interrater reliability. Percent
agreement interrater reliabilities were calculated and ranged from 88.6 to 94.8% (see
Table 4). Scale reliability analysis was conducted and the initial scale alpha for the Six
scenarios was .53. The corrected item-total correlations indicated that Scenario Two
(as indicated in Table 4) was not highly related to the other scenarios. Scenario Two
was removed from the scale producing a more reliable scale of ﬁve scenarios (a =
.5 8). To account for missing data, an average scale score was computed using the

remaining ﬁve scenarios. A total of 138 student interviews were included in the

58

coding process; the remaining 73 student interviews represent missing data. Missing
data may be the result of students declining to answer Speciﬁc questions. The average
scale score was used in the subsequent analyses.

The validity of the two coding manuals was tested using a group of graduate
students in psychology and psychology professors as independent judges. Each judge
was presented with a scenario and with the possible responses to that scenario, i.e. the
responses collected during wave one of interviews. Judges were asked to rank order
the responses for social problem-solving competence and outcome expectations
according to the deﬁnitions used in the coding manuals. As such, judges were asked to
order the social problem-solving competence responses from zero to eight, with a zero
corresponding to a very incompetent or very ineffective response, and a score of eight
corresponding to a very competent or very effective response. Responses related to the
expectations of the outcomes of competent behavior were rank ordered from one to ﬁve
with a one corresponding to an expectation of a very negative outcome and a ﬁve
corresponding to an expectation of a very positive outcome. Judges ordered the
responses in the same order as was decided by the original coders. This exercise
provided a check on the ordering system used by the original coders.

Welt-Efﬁcacy

As deﬁned by Bandura (1977), an individual's self-efﬁcacy expectation is one's
conviction in his or her ability to execute a behavior successfully to achieve a desired
outcome. To evaluate social self-efﬁcacy in performing competent behaviors,

adolescents were asked to consider the hypothetical response of the other teenager and

59

to respond to the question, "Do you think you could do what he (she) did if you tried?”
Adolescents were asked to rate themselves on a Likert-type scale ranging from zero to
ﬁve where a score of zero corresponded to ”deﬁnitely could not do it, " and ﬁve
corresponded to ”deﬁnitely could do it. "

Scale reliability analysis was run on self-efﬁcacy for the six scenarios (a = .44).
A review of the corrected item-total correlations indieated that Scenario Three (see
Table 5) was not highly related to the other ﬁve scenarios and was therefore removed
from the scale. Seale reliability analysis was conducted on the remaining ﬁve scenarios
((1 = 46). An average perceived self—efﬁcacy scale score was computed to account for
any missing data in the ﬁve remaining scenarios. A total of 209 student interviews
were included in the coding process; the remaining two student interviews represent
missing data. Missing data may have resulted from students declining to answer
Speciﬁc questions. The average scale score was used for all subsequent analyses.

11 . l [2 H' E l .

Students were asked to respond to a series of ten questions aimed at deterrrrining
the number and kinds of interpersonal conﬂicts and antisocial behavior they have been
involved in during the ten week period prior to being interviewed. The ten week time
period corresponds to the ten weeks of dispute management training at the two
participation schools. Students were asked to indicate how many times Since winter
break (ten weeks prior to being interviewed) they had: (1) been in a ﬁst ﬁght, (2)
talked their way out of a ﬁght, (3) messed up school property, (4) been sent to the

principal's ofﬁce for bad behavior, (5) carried a gun/knife to school, (6) stopped

60

people from ﬁghting each other, (7) threatened to hurt someone, (8) been suspended or
excluded from school, (9) been physically hurt by someone at school, and (10) had
someone take something from them using physical force. These items included
questions on self-report delinquency, victimization, and positive conﬂict resolution
techniques. Reliability analysis run on the ten items indicated that ﬁve of the items (1 ,
3, 4, 7, 8) were highly related to each other (a = .74) (see Table 6). The direction of
items was reversed as necessary. An average score for these ﬁve items was computed
to create a nonviolent scale score for each interviewee. This scale was used in the

hierarchical regression analyses discussed in the following chapter.

61

O . I O O . O C
out: I 0 '.HI \.-..'.. .0 Hi -Ha1l.0.- -.‘I.Ao - tier.

Question
Since winter break (beginning of March), how many times have you:

1. Been in a ﬁst ﬁght

3. Messed up school property

4. Been sent to the principal' s ofﬁce for bad behavior
7. Threatened to hurt someone

8. Been suspended or excluded from school
a = .74

Chapter Three
RESULTS

A series of analyses of variance was conducted to investigate the effect of dispute
management training on adolescents' social problem-solving competence, perceived
self-efﬁcacy in social conﬂict situations, expectations of the outcome of competent
behavior, and nonviolent conﬂict resolution. Scheffé post hoc multiple comparison
tests were used to examine group differences further. A series of hierarchical multiple
regressions was used to analyze the proposed model predicting nonviolent conﬂict
resolution from dispute management training, perceived self-efﬁcacy, expectations of
the outcomes of competent behavior, and social problem-solving competence. The
following results are presented in order of hypothesis.

Analysis of Variance
Hymthesisﬂne

Dispute management training will foster the positive development of adolescents'
problem-solving competence in interpersonal conﬂict Situations.

A one-way analysis of variance (ANOVA) was conducted using experimental
condition (i.e. , students trained in dispute management skills, students who attended an
experimental school who were not trained in dispute management skills, and students
who attended a control school) as the independent variable and average social problem-
solving competence as the dependent variable. This AN OVA was Signiﬁcant (E (2,
201) = 4.23, p < .05; n2 = .04) (see Table 7). Scheffé post hoc multiple comparison

procedure indicated that the mean social problem-solving competence scale score for

62

 

63

82 H ZV Ge" 5 8m" 7Q cove—80M

 

 

 

8°. 3v 8a was 8. 8d 8. a; S. ”an 688022382
Size 8qu £qu 2858?.—
So. m2 3 a. 3. ea 8; can R. NS“ 283.5
serve @qu amaze
8o. mz 8n 6. P. as 8. 8.... as. o3. seem—sum 38m
@2qu gaze 5qu
3. 3v 8m m3. 3. 8.- mm. 8.- me. an. 88358 38m
ad a \e ouch em 2 am 2 on 2
Season 8:58:
628 8.280 32.8 geostaxm 358,—. Dogmeou

 

 

 

....o. a ..4 .. ..«... .o-..o x..:.T.... ...- 9.... “.0

....‘..1... ...x....... -:... a. 3 I -. .r... ...-........ ..‘ . I... 13‘. .r... . ..m. 7&5...

A. 035.

64
students trained in dispute management skills (M = .22, SD = .48) was signiﬁcantly
higher than either group of students not trained in dispute management skills. There
were no signiﬁcant differences between the mean social problem-solving competence
sale scores for students attending an experimental school who were not trained in
dispute management skills (M = -.08, SD = .55) and students attending a control
school (M = -.03, SD = .51) (see Table 7).
W

Dispute management training will foster the positive development of adolescents'
self-efﬁcacy for performing competent behaviors in interpersonal conﬂict situations.

A one-way analysis of variance was conducted using experimental condition as the
independent variable and perceived self-efﬁcacy as the dependent variable. Results,
reported in Table 7, indicated that there were no signiﬁcant differences between
experimental groups with respect to perceived self-efﬁcacy scale scores (E (2, 206) =
.61, NS; 112 = .006).

Hmhesiﬂhmc

Dispute management training will foster the development of adolescents'
expectations about the positive outcomes of performing competent behavior in
interpersonal conﬂict situations.

A one-way AN OVA was utilized to examine the relationship between experimental
condition as the independent variable and outcome expectations as i the dependent

variable. The analysis, reported in Table 7, indicated that dispute management training

65

was not signiﬁcantly related to adolescents' expectations of the outcomes of competent
behavior (E (2, 135) = .12, NS; 112 = .002). *
Hierarchical Regression Model Predicting Social Problem-Solving Competence
W

There is a relationship between dispute management training, social problem-
solving competence, perceptions of self-efﬁcacy, outcome expectations for competent
behavior, and nonviolent conﬂict resolution.

A model, depicted in Figure l, to predict nonviolent conﬂict resolution was
evaluated. Previous literature has suggested that dispute management skills training is
positively related to perceived self-efﬁcacy and to higher expectations of the outcomes
of competent behavior. Furthermore, perceptions of self-efﬁcacy and expectations of
the outcomes of competent behavior are positively related to social problem-solving
competence. Finally, social problem-solving competence is positively related to
nonviolent conﬂict resolution. The proposed model suggests only mediated
relationships between dispute management training and nonviolent conﬂict resolution.

Four hierarchical multiple regression equations were used to test the proposed
model. These equations tested the following components:

Equation 1: Dispute management training was evaluated for its predictive effect

on perceived self-efﬁcacy in interpersonal conﬂict situations.

Equation 2: Dispute management training was evaluated for its predictive effect

on the expectations of the outcomes of competent behavior in interpersonal conﬂict

situations.

66

Equation 3: Perceived self-efﬁcacy and the expectations of the outcomes of
competent behavior were evaluated for their predictive effects on social problem-
solving competence (controlling for dispute management training).

Equation 4: Social problem-solving competence was evaluated for its predictive

effect on nonviolent conﬂict resolution in interpersonal conﬂict situations

(controlling for dispute management training, perceived self-efﬁcacy, and outcome

expectations).

In addition, when using hierarchical regression to test mediated effects, equations
must be inverted and run backwards to double check the effects of variables or blocks
of variables entered into the equation. To do this, Equation 4 was reversed: social
problem-solving competence was entered ﬁrst into the equation with nonviolent conﬂict
resolution as the criteria; perceived self-efﬁcacy and outcome expectations were entered
in the second block; and experimental condition was entered into the equation in block
three.

In hierarchical regression the relevant testis the signiﬁcance of the change in the
squared multiple R (R2 change), which indicates that the new variable (or block of
variables) entered into the analysis adds signiﬁcant predictive power, or variance, to
the model above and beyond what other variables entered into the analysis already
explain. For these analyses experimental condition was dummy coded along two
dimensions: (1) whether students received training in dispute management skills, and
(2) school setting. Therefore, the group that received dispute management training in

an experimental school was coded (1,1). The group that received no training but was

67

in an experimental school was coded (0,1). Finally, the group that received no training
and was in a control school was coded (0,0).

As listed previously, Equation 1 tested the direct effect of dispute management
training on perceptions of self-efﬁcacy. Results indicated that dispute management
training did not have a signiﬁcant effect on perceived self-efﬁcacy (E = .605, n = .55;
R2 = .006, p = .55). Equation 2 evaluated the direct impact of dispute management
training on the expectations of the outcome of competent behavior. Again, no
signiﬁcant effects were found (E = .116, p = .89; R2 = .002, p = .89). Tables 8 and
9, respectively, summarize these equations for the model and present the beta weights
and their signiﬁcance for variables in these equations.

Equation 3 tested the direct effects of perceived self-efﬁcacy and the expectations
of the outcomes of competent behavior (the entrants) on social problem-solving
competence (the criteria) with dispute management training as the control variable.
This block of entrants had a signiﬁcant direct effect on social problem-solving
competence as indicated by a signiﬁcant change in squared multiple R (E = 18.65, p
< .0001; R2 = .365; R2 change = .305, p < .0001). Simultaneous testing of the two
variables as indicated by the beta weights (Table 10) showed that perceived self-
efﬁcacy was a signiﬁcant predictor of social problem-solving competence whereas
expectations of the outcomes of competent behavior was not. These results indicated
that perceived self-efﬁcacy accounted for a signiﬁcant portion of the variance
associated with this block of entrants. In addition, there is a direct effect between

dispute management training and social problem-solving competence for students who

68

 

 

 

Predictor Blocks 1’ B" t of t
Block 1: Experimental Intervention
F = .605 , p = .547

R’ = .006

R2 change = .006, p = .547

Trained condition .06 .063 .858 .392
Untrained condition -.07 .067 .921 .358

 

' Correlations are corrected for attenuation due to unreliability in the measure.
" Standardized regression coefﬁcients are reported for the step in which they were
entered into the model.

 

 

 

Table 9
o‘r. ' .. u H‘ ("rs H a u. ' .H at .ur‘ «III int-.0 Ir-
Q E . _ E . Z
Olltmmncctations—
Signiﬁcance

Predictor Blocks 1‘ [3" t of t
Block 1: Experimental Intervention

F = .ll6,p = .890

R’ = .002

R’change = .002, p = .890

Trained condition .04 .038 .420 .675

Untrained condition .03 .033 .365 .716

 

‘ Correlations are corrected for attenuation due to unreliability in the measure.
" Standardized regression coefﬁcients are reported for the step in which they were
entered into the model.

69

Table 10

 

Signiﬁcance

Predictor Blocks 1‘ B" t of t
Block 1: Experimental Intervention

F = 4.17, p < .05

R2 = .059

R2 change = .059, p < .05

Trained condition .30 .260 2. 89 .005

Untrained condition -. 15 .087 .97 .333
Block 2: Predictive Variables

F = 18.65,p < .0001

R2 = .365

R2 change = .305,p < .0001

Outcome expectations .50 .093 1.27 .207

Perceptions of self-efﬁcacy 1.09 .519 7.09 .000

 

‘ Correlations are corrected for attenuation due to unreliability in the measure.
" Standardized regression coefﬁcients are reported for the step in which they were
entered into the model.

70

receive training in nonviolent conﬂict resolution, but not for either control condition.
Table 10 summarizes this equation for the proposed model and presents the beta
weights and their signiﬁcance for variables in the equation.

Finally, Equation 4 evaluated the direct effects of social problem-solving
competence on nonviolent conﬂict resolution. In Equation 4, nonviolent conﬂict
resolution was tested as the criteria with social problem-solving competence as the
entrant, and the other variables, perceived self-efﬁcacy, outcome expectations, and
dispute management training were control variables. If it is true that social problem-
solving competence is the most proximal variable to nonviolent conﬂict resolution as
depicted in the proposed model, when social problem-solving competence is entered
into a hierarchical regression equation it should have signiﬁcant predictive power on
nonviolent conﬂict resolution as demonstrated by a signiﬁcant change in the squared
multiple R and a signiﬁcant beta weight for the relationship between social problem-
solving competence and nonviolent conﬂict resolution.

Results indicated that social problem-solving competence had no impact on
nonviolent conﬂict resolution above and beyond the variance accounted for by the
control variables entered into the equation in the two previous blocks (E = 6.89, p <
.0001; R2 = .212; R2 change = .0003, p = .81). However, results indicate that both
perceived self-efﬁcacy and outcome expectations were signiﬁcant predictors of
nonviolent conﬂict resolution. Furthermore, dispute management training has a direct

effect on nonviolent conﬂict resolution. Table 11 summarizes this equation of the

71

 

 

 

Table 11
u‘vlt’al I»: V II ‘ ("In ‘II : I. ° .II l.‘ .II‘ AM he; '1' \01 II M
. . _ .
W4 11 . 1 ﬂ' 1 .
Signiﬁcance

Predictor Blocks 7‘ B" t of t
Block 1: Experimental Intervention

F = 5.89,p < .05

R2 = .083

R2 change = .083, p < .05

Trained condition .14 .242 2.71 .007

Untrained condition .21 .260 2.92 .004
Block 2: Predictive Variables

F = 8.67,p < .0001

R2 = .212

R’change = .129,p < .0001

Outcome expectations .44 .201 2.45 .016

Perceptions of self-efﬁcacy .45 .247 3.01 .003
Block 3: Predictive Variables

F = 6.89,p < .0001

R2 = .212

R’change = .0003,p = .814

Social competence .43 .023 .236 .814

 

' Correlations are corrected for attenuation due to unreliability in the measure.
" Standardized regression coefﬁcients are reported for the step in which they were
entered into the model.

72

proposed model represented in Figure 1 and presents the beta weights and their
signiﬁcance for variables in the equation.

As a double check to the four equations described above, when using hierarchical
regression to test a model the equations must be run backwards. To do so, Equation 4
was reversed, i.e. , social problem-solving competence was entered ﬁrst into the
equation with nonviolent conﬂict resolution as the criteria. If social problem-solving
competence is the most proximal variable to nonviolent conﬂict resolution, and if it
mediates the relationship between the other variables included in the model and
nonviolent conﬂict resolution, results would indicate a signiﬁcant direct effect between
social problem-solving competence and nonviolent conﬂict resolution. Also, if social
problem-solving competence is the most proximal variable, when other variables are
entered into the equation after social problem-solving competence, i.e. , block two and
three variables, there should be no signiﬁcant changes in squared multiple R. In other
words, if social problem-solving competence is the most proximal variable, it should
explain enough variance that there is no variance leﬁ to be explained by variables
entered in subsequent blocks.

This analysis indicated that when entered ﬁrst into the that the equation, social
problem-solving competence had a signiﬁcant effect on nonviolent conﬂict resolution
(E = 8.97, p < .05; R2 = .064; p < .05) (see Table 12). However, when perceived
self-efﬁcacy and outcome expectations were entered into the equation in block two,
there was a signiﬁcant change in squared multiple R, (E = 7.24, p < .05; R2 = .143;

R2 change = .079, p < .05) and the relationship between social problem-solving

73

Table 12

Hierarchical Multiple Regression Analysis and Intercorrelations Predicting Nonviolent
Conﬂict Resolution - Equation 4 Backwards
ll . l C [1' B l .
Signiﬁcance

Predictor Blocks 1‘ B” t of t
Block 1: Predictive Variables

F = 8.97 , p < .05

R2 = .063

R2 change = .063,p < .05

Social competence .43 .252 2.99 .003
Block 2: Predictive Variables

F = 7.24, p < .05

R2 = .143

R2 change = .079, p < .05

Outcome expectations .44 .201 2.35 .020

Perceptions of self-efﬁcacy .45 .206 2.07 .040
Block 3: Experiemental Intervention

F = 6.89,p < .0001

R2 = .212

R2 change = .069, p = <.05

Trained condition . 14 .214 2.47 .015

Untrained condition .21 .252 3.00 .003
‘ Correlations are corrected for attenuation due to unreliability in the measure.
" Standardized regression coefﬁcients are reported for the step in which they were
entered into the model.

74

competence and nonviolent conﬂict resolution became nonsigniﬁcant. Perceived self-
efﬁcacy and outcome expectations accounted for variance above and beyond what
social problem-solving competence and each demonstrated a signiﬁcant direct effect on
nonviolent conﬂict resolution. Finally, when experimental condition was entered into
the equation there was a signiﬁcant change in squared multiple R (E = 6.89, p <
.0001; 32 = .212; 32 change = .069, p <.05).

S EH' 1 . l B . ! l .

The equations described above provide evidence that the mediated relationships in
the hypothesized model are not supported. Instead, there are direct effects between
perceived self-efﬁcacy and nonviolent conﬂict resolution; between outcome
expectations and nonviolent conﬂict resolution; and between experimental condition
and nonviolent conﬂict resolution for students who are trained in dispute management
skills and for students who attend an experimental school who are not trained in
conﬂict resolution. Furthermore, there is a direct relationship between dispute
management training and social problem-solving competence. Figure 2 shows a
revised model of the direct effects between experimental school, perceived self-
efﬁcacy, outcome expectations, and nonviolent conﬂict resolution.

Summary of Results

An additional AN OVA conducted to investigate the effects of dispute management
training on nonviolent conﬂict resolution was signiﬁcant (E (2, 206) = 7.24, p < .05;
n2 = .066). Scheffé post hoc multiple comparison procedure indicated that students

trained in dispute management skills (M = 3.48, SD = .67) and students who attended

75

a treatment school who did not receive training (M = 3.49, SD = .65) were
signiﬁcantly more nonviolent than students who attended a control school (M = 3.07 ,
SD = .93). There were no signiﬁcant differences between groups of students who
attended the treatment schools with respect to nonviolent conﬂict resolution.

The ﬁndings from analyses of variance and the results of hierarchical regression
are consistent. Instead of the mediated relationships as hypothesized in the proposed
model, results indicate that there are direct effects between variables as shown in

Figure 2.

76

 

 

neuaemom

85.50
332282

J

K

382 3:39: ”N 2:5

 

83:38me
08038

 

 

% 58533

 

 

f 838qu

 

 

Begum
535235

 

Chapter Four
DISCUSSION

Results of this study do not appear to support the efﬁcacy of the dispute
management training program which was evaluated. Although receiving training in
conﬂict resolution skills was signiﬁcantly related to higher social problem-solving
competence, this relationship was not related to nonviolent conﬂict resolution.
Furthermore, attending a school with a dispute management training program was
signiﬁcantly related to nonviolent conﬂict resolution. However, it is precarious to
attribute these ﬁndings solely to the presence of the training program in the school. It
is possible that both trained and untrained students in the experimental schools
displayed less violent conﬂict resolution techniques because students who were not
trained learned nonviolent conﬂict resolution techniques from their peers who received
training. In other words, the dispute management program could have had radiating
effects whereby students who received training positively inﬂuenced the school
environment and their friends who did not directly participate in training. It appears
more likely, however, that this ﬁnding can be explained by the fact that there were
differences between the experimental and control schools prior to the implementation of
the dispute management program.

The chi-squared test for independence presented earlier indicated that there were
no signiﬁcant differences between the number of students who were violent or who
committed illegal acts in the treatment and control schools. However, in contrast to

this analysis, an analysis of variance run on survey data collected during wave one of

77

 

78

the overall evaluation indicated that there were differences between the schools prior to
implementation of the dispute management intervention (for content of questions see
Table 13). This AN OVA revealed that students at the control schools reported
participating in more delinquent or antisocial behavior in their schools than did students
at the treatment schools (M = 1.2, SD = .79; M = 1.1, SD = .75 respectively); (E
(1,1457) = 8.9, p< .05; I12 = .006). Based upon these ﬁndings, it is difﬁcult to
attribute the signiﬁcant difference between treatment and control schools as effects of
the dispute management training program.

Although the ﬁndings of this study do not support the utility of the intervention
evaluated in this investigation they do support previous research on the importance of
efﬁcacy and outcome expectations and shed light on the processes involved in
promoting nonviolent conﬂict resolution. Consistent with previous research on other
high-risk behaviors, this study conﬁrms the theoretical importance of perceived self-
efﬁcacy and outcome expectations when trying to understand how adolescents act and
perceive themselves in interpersonal interactions. Results indicated that both perceived
self-efﬁcacy and outcome expectations were signiﬁcantly related to nonviolent conﬂict
resolution. These ﬁndings conﬁrm previous research which suggests that both self-
efﬁcacy and outcome expectations are important indicators of behavior (Bandura, 1982;
Connolly, 1989; Wheeler & Ladd, 1982).

Self—efﬁcacy has been related to a variety of social abilities including making
friends, developing intimate bonds, being socially assertive, participating in social

groups (Connolly, 1989; Ford, 1982) and now to nonviolent conﬂict resolution. Self-

 

79

Table 13

IIII I .1 '.‘III I "le a‘ .‘v.I I I" A- ‘CJI' I I‘ hm:

Question

Since the beginning of this January, how many times have you:

Got into a ﬁst ﬁght

Talked your way out of a ﬁght

Hit someone who made fun of you

Messed up school property

Been sent to the principal's ofﬁce for bad behavior
Carried a gun or knife to school

Skipped a class

Skipped school for a whole day

Stopped people from ﬁghting each other

10. Went out of your way to help someone

11. Threatened to hurt someone

12. Been suspended or excluded from school

13. Helped a teacher during class

14. Had someone physically assault and hurt you at school
15. Had someone take something from you using physical force
16. Had something stolen from your locker

17. Had someone threaten to hurt you at school

a = . 8406

PPHP‘E’W‘P’PI‘

80
efﬁcacy has also been linked to social problem-solving competence (Allen, et al. 1990;

Caplan et al. 1992), a relationship which is supported by these current ﬁndings. This
study indicates that an individual’s evaluations of his or her capability to act
nonviolently are important to his or her ability to effectively resolve an immediate
conﬂict in a manner that also reduces the chance that the individual will be faced with a
similar situation in the future (social competence). Individuals who perceive
themselves as efﬁcacious are more likely to resolve conﬂicts nonviolently. Previous
research (e.g. , Allen et al. 1990), furthermore, indicates that an individual’s belief in
his or her ability to perform a given behavior (self-efﬁcacy) lacks meaning unless the
individual also believes that performing the behavior will lead to a desired outcome
(outcome expectations). Consistent with this research, outcome expectations in this
study had a signiﬁcant direct effect on nonviolent conﬂict resolution. These ﬁndings
indicate that adolescents who positively evaluated the outcome of resolving
interpersonal conﬂicts nonviolently were more apt to behave nonviolently. Given that
perceived self-efﬁcacy and outcome expectations demonstrated signiﬁcant direct effects
on nonviolent conﬂict resolution it is important to consider these constructs when
planning programs to reduce violent behavior.

When seeking to understand the apparent lack of efﬁcacy of this intervention it is
important to consider factors related to program conceptualization or implementation
failure. Program failure can occur when the theory behind a program does not ﬁt with
the processes underlying targeted behavior change. Although the current study did not

provide support for the hypothesized model there is evidence which indicates that the

81

underlying relationships described in the model may still explain the processes whereby
dispute management training can inﬂuence nonviolent behavior. It is likely, therefore,
that faulty theoretical explanations are not a problem for this current study or for the
conﬂict resolution training program. However, multiple issues regarding the
implementation of the training program should be considered.

Implementation failure occurs when a program is not implemented effectively.
Such failure can occur when a program modeled after a pilot program is not replicated
the way the original program intended. Implementation failure and threats to program
integrity are many and include: the relevance of information being presented to the
target population; attributes of the target population including demographic
characteristics such as age, sex, and grade in school, and environmental aspects such as
the socio-economic conditions of the area surrounding the school; level of support from
program sponsors; capabilities of the program facilitators; and consistency with which
the curriculum is delivered to students. One of the major concerns regarding the
implementation of this program centers around the lack of consistency of program
delivery. On multiple occasions throughout the ten-week training session the
facilitators were absent on training days. Interruptions in training can lead to decreased
effectiveness as students forget information from one training session to the next. In
addition, lack of consistency and reliability on the part of the facilitator may give
students the impression that the training is not important. One indication of
implementation failure in this study is the difference between the number of students

who indicated by self-report that they had received training in conﬂict resolution skills

 

82

(38 students) and the number of students who provided a homeroom number which
corresponded to a homeroom receiving training (67 students), see Table 1. This
discrepancy suggests that students who received training were unaware of the program,
an indication that the program may not have made a signiﬁcant impression upon the
students. Because of such implementation issues it is difﬁcult to draw conclusions about
the potential effectiveness of the dispute management training.

Current ﬁndings demonstrate that this dispute management training program did
not enhance adolescents’ self-efﬁcacy and outcome expectations. It may be, however,
that more effectively implemented conﬂict resolution programs are able to positively
affect adolescents’ perceived self-efﬁcacy, outcome expectations, and social problem-
solving competence, and by doing so reduce their violent conﬂict.

Summary

This study provides information on variables that are important to understanding
nonviolent conﬂict resolution. The results of this study indieate the underlying
theoretical importance of self-efﬁcacy and outcome expectations in nonviolent conﬂict
resolution. These ﬁndings are consistent with the literature on social learning theory
and research on reducing high-risk behaviors and provide an important contribution by
bridging previous knowledge to the understanding of nonviolent conﬂict resolution.

A number of caveats should be considered in relation to this investigation. Issues
related to measurement may have been problematic. It is important to note that the
signiﬁcant relationship between perceived self-efﬁcacy and social problem-solving

competence may be difﬁcult to interpret because of the high intercorrelation and

83

multicolinearity of these two variables (when corrected for attenuation r = 1.09).
Although the interview protocol and scenarios utilized in this study are based upon
previous instruments used by Allen et al. (1990) future research could beneﬁt by
further developing these constructs to limit high intercorrelations and by increasing
scale reliabilities of constructs used in the model. Also, because the data used in this
evaluation are all based upon self-report measures they are subject to possible
participant biases. The validity of self-report data can be threatened by a respondent’s
desire to report socially acceptable, rather than actual, information (Crano & Brewer,
1986). Future evaluation efforts may avoid such biases by including other sources of
data, for example, the number of conﬂicts reported to teachers and administrators as
well as the number of conﬂicts reported by students on self-report measures.

Other potential limitations of this study include low power (i.e. the probability of
correctly rejecting the null hypothesis) which may have compromised the ability to
recognize effects that would have been signiﬁcant if power were higher. Small sample
size relative to the effect sizes of the independent variables on the dependent variables
contributed to decreased power. Future studies should include a larger sample. In
addition, the use of a post-only design warrants discussion. The posttest—only design
was chosen because of the expense involved in tracking students in a pretest-posttest
design. Unfortunately, data collected from surveys administered during the pre-test
phase of the overall evaluation and analyses of variance indicate that the schools may
have been different prior to the intervention. These differences confound conclusions

drawn from this study.

34

Future Directions

The prevalence of violence in our society and its impact on schools and students is
undeniable. To date there is a lack of empirical evidence that school conﬂict resolution
programs effectively reduce violent resolutions to interpersonal conﬂict (Tolan &
Guerra, 1994a; Webster, 1993). However, there is anecdotal evidence that social
problem-solving programs can promote gains in targeted cognitive skills and beliefs
about nonviolent conﬂict resolution (Tolan & Guerra, 1994a). Because personnel and
monetary resources continue to be devoted to conﬂict resolution programs in schools
(Webster, 1993), and because of the far-reaching deleterious effects of violence in
schools, further research which includes strong outcome measures and which
documents the process and ongoing implementation of training programs is vital.

This exploratory investigation attempted to explain the process of reducing
conﬂict by applying new constructs to the evaluation of a dispute management
program. These constructs, social problem-solving competence, perceived self-
efﬁcacy, and outcome expectations, have been found to be important in understanding
change with respect to other antisocial behavior (e. g. substance abuse, unprotected
sexual intercourse). This current investigation provides support for the theoretical
importance of perceived self-efﬁcacy and outcome expectations in understanding
nonviolent conﬂict resolution. Further research and evaluation on nonviolent conﬂict
resolution should consider these variables when attempting to reduce violent conﬂict.
Although the dispute management training program did not impact either of these self-

expectations, the lack of inﬂuence may be due to failures in program implementation.

85

Other programs may have more impact on fostering perceptions of self-efﬁcacy and the
outcome of competent behavior and by doing so may be able to directly increase the
use of nonviolent techniques to resolve interpersonal conﬂict.

One of the more promising developments in the ﬁeld of conﬂict resolution is the
implementation of community-based outreach and edueation programs which aim to
combat violence in all areas of an adolescent's life (Commission on Violence & Youth,
1993; DeJong, 1994; Elliott, 1994; Tolan & Guerra, 1994a). These programs are
comprehensive in nature and attempt to bring together all primary institutions that serve
youth, including: families, health agencies, schools, employers, and juvenile justice
systems (Elliott, 1994; Tolan & Guerra, 1994a). The inclusion of all relevant
institutions is a common-sense approach to reducing violence in our communities and
may be more effective in reducing violence programs which target only one institution,
such as school-based programs. As with school conﬂict resolution programs, however,
there is no empirical evidence to support the effectiveness of these programs (Elliott,
1994; Tolan & Guerra, 1994a). It is clear that more research and evaluation is needed
to effectively reduce violence in our communities. This current study may provide
insight to variables critical to understanding nonviolent conﬂict resolution. However,
only through stringent evaluations that include clear behavioral outcome variables and
thorough documentation of program implementation and delivery will the true

effectiveness of programs be realized.

APPENDICES

APPENDIX A

Appendix A

Coding Manual for Scenarios -- Social Problem-Solving Competence

115" [5.12“ -51' D
Social problem-solving competence is often deﬁned in relation to success in resolving
interpersonal conﬂicts. In these scenarios, a maximally competent response is one that
effectively resolves the immediate social conﬂict and reduces the chances that the
subject will have to face similar problems of this type in the future. Conversely, an
incompetent response is one that does not effectively resolve the immediate social
conﬂict and may be likely to cause more problems for the subject in the future.

mm

8 -- a very competent or very effective response

6 -- a competent or effective response

4 -- a response that is neither competent nor incompetent, a response that neither
directly helps nor hurts the situation or individual

2 -- an incompetent or ineffective response

0 -- a very incompetent or ineffective response

NOTE: Various coding levels require that one or more conditions be met in a subject's
response in order to be coded at that speciﬁc level. These areas are indicated with
AND between the different qualiﬁcations. Others will indicate OR. In these instances
different responses qualify for the coding level.

W: Unless otherwise directed a response of ”Don't Know” should be coded
as very incompetent or very ineffective and is therefore coded as a zero. "Don't
Know" is different from a response in which a subject says he or she would do nothing.
In many instances doing nothing is an understandable approach.

Mmethanmmsponse; If the subject says that he/ she would do tWo or more
responses which would normally receive different scores and does not choose one over
the other as a ﬁnal answer, the subject should receive the score of the lowest response.
If the subject gives one or more responses in a series or sequence of responses (e. g. ”I
would ﬁrst try to talk it out and if that did not work I would try something else”) the
response should be coded as a totality of its parts according to the elements given.

PhysicaLharm; Any response in which the subject says he/ she would use physical
force or ﬁght should receive a score of zero.

86

87

Scandal Your mother has been nagging you about getting home immediately after
school. One day you get home an hour late and your mother yells at you, "Do you
know what time it is? Where have you been?”

NOTE: For 4 - 8, student may state a speciﬁc place or activity he! she was doing and
therefore was late. If the response seems that this is a normal and acceptable activity
then it should not be taken as an excuse or a lie. For example, "I stayed at school to
help a teacher or I was at a friends house. So she wouldn't worry.” Instead of saying
" I would tell her where I was, " this ﬁrst student gave examples of why he/ she could
be late. In this case these reasons are not excuses or lies but examples.

8 - Student explains why he or she is late. AND
Student is empathetic towards the mother and shows concern for her wishes or feelings.
Student may apologize.

Students may state his/ her concern with telling the truth as the reason for explaining.
The main or only reason the student is explaining should not be solely to avoid getting
into further trouble.

Examples: I'd tell my mother where I was and what I was doing because I know she
was worried. I‘d say I got caught in the wrong place and was not able to call home
right away. She won't think I was getting hurt or didn't care about coming home, I
didn't want her to worry.

I'd tell her where I was because it is the truth.

4 - Student gives an explanation but does not show concern for mother's feelings or
wishes.

Student simply answers the mother' s question with no mention of his/ her tardiness
being a problem. OR Student states desire to not get into further trouble. OR
Student states that he! she would answer the mother just to answer her question. OR
Student apologizes but offers no explanation of where he/ she has been.

Examples: Tell her where I've been because she asked.
Tell her where I was so I won't get into more trouble.
Tell the truth so I won't get into trouble.

3 - Student gives an explanation but shows no concern for mother' s feelings and
wishes.

Student may sound rude and state the reason for explaining is to have the mother stop
asking.

Myles: Tell her where I was so she would stop hollering.
Tell her I was with my friends so she would quit yelling at me.
Explain the situation to get her off my back.

V

88

0 - Student does not give an explanation for tardiness.
Student ignores mother.

Student says he/ she would not do anything.

Student states that it “would depend. "

Student lies or makes up excuses to avoid getting in trouble.
Student says "Don't Know”

Examples: I don't say anything. I just go to my room.

"I've been out. "

Nothing. I'd take my punishment.

Make up a story because I was probably somewhere my mother wouldn't want me to
be.

I lost track of time because I don't have a watch.

Say that the bus was late so I wouldn't get in bigger trouble.

Seem A kid you know is a drug dealer. He found out that school security is
going to check his locker after the next class. He asks you to keep his stuff in your
locker.

8 - Student assertively refuses to keep the drugs in his or her locker.
Student refuses and explains why not as being a refusal to be involved with drugs or
people who are involved with drugs.
Student may say that he or she would report it to an authority (teacher, principal,
security ofﬁcer)
NOTE: If a student says that he/ she would tell an authority and wishes to avoid
trouble score as a 8.
Student says N O and that he or she would change friends.

Examples: 1' d say noandwalkawaybecauseldon' tlikedrugs.

I' (1 go tell the principal because it is bad to have it. My friends wouldn' t have it or we
would quit being friends.

I' (1 tell him no and tell him to turn the drugs 111. I wouldn t want something to happen
to him.

I'd tell him no and tell the security ofﬁcer what he asked me to do. I wouldn't want to
get in trouble or him to get in trouble.

4 - Student refuses to keep the drugs in his or her locker and expresses fear of being
eaught by an authority ﬁgure as the reason for not keeping the drugs. Student tells the
dealer to ﬁnd someone else or to ﬁnd another place to store the drugs.

89

Examples: I'd say no because I don't want to get caught with it.

I'd say no because it is stupid, I'd be getting in trouble for someone else.

No, he has to keep his own stuff, they would probably check my locker and I don't
want to get in trouble.

I'd probably say to give it to someone else. I wouldn't want to get caught.

I'd say no because he might give someone else my locker combination.

3 - Student says yes but expresses intentions of turning the drugs in, not of really
storing them.

Examples: I'd let him keep it in there and then tell the counselor. So I wouldn't get in
trouble and he wouldn't bring it anymore.

0 - Student says that he or she would keep the drugs in his/her locker.

Student says that he or she will hide the drugs elsewhere for the dealer.

Student agrees to keep the drugs but only if he or she can have some of it or some of
the proﬁts of sale for his or her help.

Student says ”Don't Know”

Examples: I'd say yes, I wouldn't want him getting busted.

If it's a cousin I'd say yes. If it's just a friend I'd say no. If it's a cousin I wouldn't
want him to get into trouble. If it's just a friend, it's his responsibility.

I'd say yes but only if he would give me some.

Samaria}: You think that one of your classmates has recently been telling rumors
about you. You don't know this student well, so you don't know why he (she) is

telling these rumors. You're upset about the rumors and you want the student to stop
telling them.

8 - Student says that he/ she would like to talk over the situation with the other student.
Student says that he! she would ask the other student why he/ she was telling the rumors.
Student says that he! she would not pay attention to the rumors because rumors are not
worth it. Ignores the situation.

Student states that he! she would tell an authority only after one of the above means of
resolving the situation is attempted and fails, e. g. only if necessary.

Examples: I'd ask him why he keeps telling these rumors on me and if he didn't stop
I'd tell the teacher. I don't see why he's spreading rumors about me because most of
them aren't true.

I’d go to the person and ask them about it and talk it over because if she's spreading
rumors I'd want them to stop.

I really don't care about rumors because they are not true. What's the use. If they're
not true why worry about them.

90

6 - Student would ASK the other student to stop but does not say he/ she would talk
about the situation, only that he/ she would ask the other student to cease telling rumors.
Student states that he/ she would tell an authority only after one of the above means of
resolving the situation is attempted and fails, e. g. only if necessary.

Examples: I'd ask him to quit telling rumors, it's not polite.
I'd ask her politely to stop because I don't want to get in a ﬁght.
I'd ask him to stop and if not, ask a teacher to help.

4 - Student states that he/ she would TELL the other student to stop telling rumors.
Student states that he/ she would approach the other person in an aggressive manner as
if giving an order to stop telling rumors.

Student states that he! she would go to an authority or another person ﬁrst (without
trying to talk to the other student).

Student describes a roundabout way of dealing with the situation.

Examples: I'd confront her so she'd stop.

I'd say leave me alone and stop spreading rumors about me.

I'd tell the teacher or principal because it's wrong to spread rumors.

I'd go talk to someone and ask them to ﬁnd the student and see what they can do then I
would talk to the student and ask them why because I don't want nobody telling rumors
about me.

2 - Student says that he! she would threaten the other student.
Student states that he/ she would talk to the other student but then states that he! she
would hit the other student.

Examples: I'd ask him to stop to scare him.

I'd go up to him and let him know not to mess with me.

I'd ask him if he said it to make sure it's him and not someone else so I won't start a
ﬁght with the wrong person.

I'd confront him and ask him why and if he doesn't answer it's a ﬁst ﬁght.

0 - Student states that he/ she would ﬁght the other student.
Student responds with "Don't know.”

Examples: I'd hit him in the mouth because he talks too much.

91

W You are walking home alone down a dark street. Two boys (girls) are
following you. You don't know them but they are about your age. They catch up with
you and say, "Give us your jacket. "

8 - Student says that he/ she would give them the jacket.

Examples: I'd give them the jacket because I don't want to get hurt. A jacket is not
worth risking your life over.
I'd give them the jacket. Why should I ﬁght for it?

6 - Student would give them the jacket only after trying to determine if they had a
weapon or only after he! she sees a weapon.

Examples: If they had a weapon I'd give it to them.
I'd say something and if they pulled out a weapon I'd give them the jacket.

4 - Student says that he/she would walk away, ignore them, say no, run away, ask
them why (with no further action), etc..

Examples: I'd say no and then I'd keep on walking because I don't want to get hurt.
I'd ignore them and run away.

I won't give it, they won't kill me.

I'd say no, I need my jacket.

0 - Student states that he/ she would ﬁght the other teenagers.
Student responds with ”Don't know.”

Examples: I'd ﬁght and then run. To protect myself.
I'd hit them.

Scenario}: You are sitting with your friends in the lunch room. A boy (girl) you
don't know spills his (her) drink on you. One of your friends tells him (her) to
apologize and he (she) refuses.

8 - Student says that he/ she would walk away, let it go, ignore it, not worry about it.
Student says that he! she would get something to wipe it up.
Student tells his/her friend not to worry about it.

92

Examples: I'd tell him he doesn't have to apologize. I'll just clean it up because I
don't want to go through no trouble.

I'd forget about it because if it was a mistake I wouldn't care if it was on purpose I'd
be mad.

I wouldn't say anything but I'd be mad, it always happens in food ﬁghts.

6 - Student states that he! she would ask the other student to apologize or to help clean it
up.

Student would try to determine if it was an accident or on purpose and if on purpose
would tell an authority.

Student would ask why but does not state any further action.

Student says he! she would ask for an apology but if the other student refused it would
be okay.

Examples: Can you please apologize or get something to get this off of me. If a
person spills on you they're supposed to say sorry.

Ask her to get it off of me because she spilled it.

I'd ask her why she did it and go tell someone.

I'd ask her why.

I'd ask him to apologize but if he didn't I won't make a big deal about it because I
wouldn't want to get in a ﬁght.

4 - Student states that he/ she would TELL the other student to apologize.

Student states that he/ she would TELL the other student to get something to wipe it up.
Student states that he/ she would tell an authority ﬁrst before any other attempt at
resolution.

Examples: I'd tell him to get me something to wipe it up with.
I'd tell her to say sorry.
I'd tell on him because he spilled his juice on me.

3 - Student states that he/ she would threaten the other student.
Student states that he! she would yell at the other student.
Student states that he! she would TELL the other student to buy replacement clothing.

Examples: I'd get angry and yell.
I'd tell her that she better wash my shirt and apologize. I'd be mad at her.

2 - Student would determine it was on purpose or an accident and only if on purpose
would he/ she ﬁght.

Student states that he! she would ask for an apology or help cleaning and if none, only
then would ﬁght.

Examples: I'd tell him to get it off me and if he didn't I'd ﬁght.

93
If it was an accident, okay. If on purpose, I'd hit him.

0 - Student states that he/ she would ﬁght or throw a drink back at the other student.
Student responds with ”Don't know."

Examples: I'd ﬁght because he didn't apologize.
I'd throw it back at him.

Seemm You are at school and you see another student talking to your boyfriend
(girlfriend). They are laughing and smiling and seem to be having fun. Suddenly,
your boyfriend (girlfriend) looks at you and turns quickly away.

8 - Student states that he! she would wait until he/ she is in private with their
girlfriend/boyfriend to talk about the situation.

Student acknowledges that it is okay to be friends with other opposite sex people
without it being a problem.

Student states that he/ she would join the conversation BUT is not doing so because
he/she is mad, instead is just interested in joining the two.

Examples: I wouldn't do anything, it's okay to be friends.

I'd wait until he's alone and ask who she was.

Ask what they were talking about later in private to avoid a ﬁght.

I'd walk over there to see what they said.

I'd ask her what they are talking about.

I'd ask him what was that all about. It's better to ask, I might be thinking the wrong
thing.

4 - Student states that he/ she would ask the boyﬁiend/ girlfriend who the other person is
and what they are talking about AND expresses suspicion about the situation.

Student states that he/ she would be jealous or mad.

Student states that he/ she wouldn't do anything but is suspicious and assumes the worst
of the situation.

Student states that he/ she would not ﬁght over girls/boys.

Examples: I'd ask him what he was doing; if he was just talking I'd forget about it. If
more, I'd dump him.

After they get through I'd ask her why she turned away from me. The other boy might
be trying to turn her away from me.

I wouldn't do anything but I'd be mad.

I'd keep on walking. He's ignoring me and talking to her so I'll do the same.

I'd ask him why he's talking to her, he's not supposed to talk to other girls.

2 - Student states that he/ she would dump the boyfriend/ girlfriend.

94

Student does not consider the idea that they may just be friends and makes no mention
of gathering more information.

Student mentions being wronged by the boyfriend/ girlfriend.

Student states that he/ she would yell at the boyfriend/ girlfriend.

Student states that he/ she would try to make boyfriend/ girlfriend jealous.

W: I wouldn't do anything, I'm ﬁnished with him.
I'd tell her I don't like her anymore because she is talking to other boys.

0 - Student states that he/ she would ﬁght.
Student responds with "Don't know.”
Examples: I'd ﬁght who she is talking to.

APPENDIX B

Appendix B

Coding Manual for Scenarios - Outcome Expectations

Outcome expectations as deﬁned by Bandura (1977), refer to an individual's estimate
that a given performed behavior will lead to a desired outcome. Outcome expectations
are related to, but separate from, social competence and perceptions of self-efﬁcacy.

WSW:

5 -- expectation of a very positive outcome

4 -- expectation of a somewhat positive outcome
3 -- expectation of a neutral outcome

2 -- expectation of a somewhat negative outcome
1 -- expectation of a very negative outcome

Directions: For each scenario, use the following coding schemes to rate responses.
For a response to be coded at a level it must be similar to any of the responses
described by that level.

Note that the difference between levels of responses often includes speciﬁc references
to how the student describes the outcome in relation to him/herself. For example, in
Scenario 2, level 3 includes, ”The dealer would be mad, " whereas level 4 includes,
”The dealer would be mad but I'd be okay. " The difference in level is reﬂective of the
student' s attention to him/herself. In the same vein, level 2 includes, ”Dealer would be
mad and not want to be friends.” Again, this response differs from level 3; in level 2
the student's speciﬁc reference to him/ herself is more negative than in level 3.

Morethanpneresppnse: If the subject gives two or more potential outcomes which
would normally receive different scores and does not choose one over the other as a
ﬁnal answer, the ﬁnal rating should be the score of the lowest response. If the subject
gives one or more outcomes in a series or sequence the response should be coded as a
totality of its parts according to the elements given.

Note: For the purposes of coding, responses containing words such as ”probably,”
”maybe,” "might," etc. will be coded the same as similar responses without such
words.

Ehxsicallrarm; Any response in which the subject states the outcome includes physical
harm to him/herself should be scored as a one.

95

 

96
Seenaﬁml Your mother has been nagging you about getting home immediately after
school. One day you get home an hour late and your mother yells at you, ”Do you
know what time it is? Where have you been?"
CODING CRITERIA

5 - Student expects mother to understand; to not punish; to forgive; to believe.
Student does not expect to get in any trouble. Student says nothing will happen.

Example; She'd say okay and I wouldn't get in any trouble.

4 - Student expects mother to warn him/her about future occurrences. Student does
not expect to get in any trouble.

Example; She'd tell me to get home earlier next time.

3 -- Student does not expect to get in trouble. Mother may/ may not be mad. Mother
will stop yelling.

Example; She may or may not be mad.

2 -- Mother may yell but student does not expect to get in too much trouble. Mother
may check out the truth of the story. Mother may not listen/believe student.

Example; She'd yell but I probably wouldn't get into too much trouble.
1 -- Student expects to get in trouble
Example; I'd be grounded.
Seem A kid you know is a drug dealer. He found out that school security is

going to check his locker after the next class. He asks you to keep his stuff in your
locker.

CODING CRITERIA

5 -- Student expects dealer to leave him/ her alone; to ﬁnd someone else. Student states
that the dealer may get in trouble but student is okay. Student will not get in trouble.
Nothing dealer could do. Student states that nothing would happen.

Example; He'd leave me alone.

97
4 - Student expects dealer would get mad but student would be okay. Student says

dealer may ”get an attitude" but leave student alone and ﬁnd someone else to keep the
drugs. Student states dealer would get in trouble.

Example; He'd get mad at me but I'd be okay.
3 -- Student says dealer would be mad.
Example; He'd be mad.

2 - Student says dealer would get mad and not want to be friends. Dealer would
continue to try to persuade student. Dealer would bully student.

Example; He'd keep pestering me to keep his stuff.

1 -- Student expects dealer would ﬁght or threaten to ﬁght. Dealer would tell an
authority that the drugs belonged to the student. Dealer would force student to keep
the drugs.

Example; He'd probably hit me and make me do it.

Samaria} You think that one of your classmates has recently been telling rumors
about you. You don't know this student well, so you don't know why he (she) is
telling these rumors. You're upset about the rumors and you want the student to stop
telling them.

CODING CRITERIA

5 -- Student expects the rumor-teller to apologize; to stop telling rumors. The student
states that the situation is rectiﬁed by the rumor-teller denouncing the truth of the
rumors.

Example; She'd apologize and stop telling the rumors.

4 -- Student expects to ﬁnd out why the rumor-teller was spreading rumors; expects
confession from the rumor-teller; gets an explanation.

Example; She'd tell me why.

98
3 -- Student does not expect situation to change; does not expect the rumor-teller to
listen. Student states the outcome would depend upon the person telling rumors. A

ﬁght would be avoided; the rumor-teller may or may not explain why he/ she is telling
rumors. Student states that nothing would happen.

Example; The rumors probably won't stop.
2 - Student expects that the rumor-teller would have a "bad attitude” and may start
cussing or arguing; the rumor-teller would lie or deny the rumor. Student states that
the situation might get worse. Student states that he! she would be humiliated or
embarrassed.

Example; She'd start yelling and cussing.
1 -- Student expects rumor-teller will try to ﬁght or will want to ﬁght.

Example; He'd probably hit me.
SeenapipA You are walking home alone down a dark street. Two boys (girls) are
following you. You don't know them but they are about your age. They catch up with
you and say, ”Give us your jacket. "

CODING CRITERIA

5 -- Student expects to keep the jacket; to be left alone by the two others; to be able to
leave the situation; to be safe; able to get home.

Examples; They'd leave me alone.
I'd be safe even if I don't have my jacket.

4 -- Student expects to be in a better position after giving them his/her jacket.
Example; I'd be better off.

3 -- Student says they would take the jacket and leave. Student states he/ she could get
another jacket. Student states that nothing would happen.

Example; They'd leave with the jacket.
2 -- Student is not physically hurt but the situation is not over. Student states that
he/ she is humiliated; that they laugh at him/her; that they bully him/her; that they
follow him/ her. Student states that he/ she would be cold.

Examples; They'd make fun of me.

99

I'd be cold.

1 - Student expects that they would try to ﬁght; beat him/her up; want something else;
push; knock student down. Student states that he/ she would get in trouble at home.

Examples; They would ask me for my shoes.
I'd get in trouble with my mom.

Seenarlpj You are sitting with your friends in the lunch room. A boy (girl) you
don't know spills his (her) drink on you. One of your friends tells him (her) to
apologize and he (she) refuses.

CODING CRITERIA

5 -- Student expects the drink-spiller to apologize; to offer to clean up the mess.
Student is satisﬁed with the outcome; expresses resolution of the situation.

Example; She'd apologize and help clean up.

4 -- Student does not expect an apology but states that it does not matter. Student states
that a ﬁght would be avoided. Student states that he/ she would be left alone.

Example; He'd go sit down and not apologize but I don't care.

3 - Student and spiller go separate ways; say nothing; student states that he/ she would
wipe up the mess him/herself. Student states that it depend upon the person who
spilled the drink; says he! she would forget it. Student states that nothing would

happen.
Example; He'd go sit down and I'd wipe up the mess.

2 -- Student states that he/ she would be humiliated; that the spiller or the student' 3
friends would make fun of him/her; that he/ she would be picked on by spiller or
friends. Student states that spiller does not apologize.

Example; My friends would call me weak.

1 -- Student expects a ﬁght; expects that spiller or his/her friends will spill on him/her
again.

Example; He might do it again.

100
W You are at school and you see another student talking to your boyfriend
(girlfriend). They are laughing and smiling and seem to be having fun. Suddenly,
your boyfriend (girlfriend) looks at you and turns quickly away.
CODING CRITERIA

5 -- Student expects to work out the situation with his/her girl/boyfriend. Talk things
out. Expects girl/boyfriend to stop talking to the other student. Student expects to
become friends with the third student.

Example; She'd wait after school so we could talk about everything.
4 -- Student expects to ﬁnd out who the other student is.

Example; My boyfriend would tell me who she was.

3 -- Student states nothing would happen. Says a ﬁght would be avoided. Situation is
not better and not worse.

Example; Nothing would get started in school.
2 -- Student says that girl/boyfriend would get angry; girl/boyfriend would say ”leave

me alone”; would walk away; not wait after school. Girl/boyfriend might " get an
attitude” or yell. They might or might not talk.

Example; She'd yell and be angry.
I -- Student expects to break up; ﬁght.

Example; She'd dump me.

REFERENCES

REFERENCES

Agnew, R. (1993). Why do they do it? An examination of the intervening
mechanisms between ”social control” variables and delinquency. leumalpLReseareh

erimeandJlelinquenstﬂO), 245-266

Akers,R- L. (1985). WWWGM ed.)-
Belmont, CA: Wadsworth.

Allen, J. P., Leadbeater, B. J ., & Aber, J. L. (1990). The relationship of
adolescents' expectations and values to delinquency, hard drug use, and unprotected

sexual intercourse. W 85-98.

Allen, J. P., Weissberg, R. P., & Hawkins, J. A. (1989). The relation between

values and social competence in early adolescence. We),
45 8-464.

Araki, C. T. (1990). Dispute management in the schools. Mediationﬂranmlx.
8(1), 51-62.

Bandura, A. (1977). Self-efﬁcacy: Toward a unifying theory of behavioral
change. W0), 191-215-

Bandura, A. (1982). Self—efﬁcacy mechanism in human agency. American
WC). 122-147.

Bandura, A. ,Adams, N. E. H,ardy, A. R. ,&Howells, G. N. (1980). Tests of
the generality of self-efﬁcacy theory. W 39—66.

Bastian, L. D., &Taylor, B. M. (1991). Schoolsrime:_A_national_caime

WW (Report No. NCJ 131645). Washington, DC: U. S.
Department of Justice.

Bronfenbrenner, U- (1979). W
hanamreandnesign. Cambridge, MA: Harvard University Press.

Cameron, J., & Dupuis, A. (1991). The introduction of school mediation to New
Zealand. W240). 1 13.

101

102

Caplan, M. ,Weissberg, R. P., Grober, J. 8., Sivo, P. J, Grady, K., & Jacoby,
C. (1992). Social competence with inner-city and suburban young adolescents:
Effects on social adjustment and alcohol use. .lnumamfﬁonsulnnLandEhnis‘al
Esxehologm 56-63.

Centers for Disease Control and Prevention. (1990). Homicide among young

black males- United States, 1978 1987. MorbidinLandMonalituMeeklrLRemn.
39(48), 869-873.

Centers for Disease Control and Prevention. (1992). Physical ﬁghting among
high school students- United States, 1990. IoumalpnheAmericanMcdit‘al
W02), 3009.

Christie, D. J ., & Toomey, B. G. (1990). The stress of violence: School,
community, and world. In L. E. Arnold (Ed.), Childhrxxljtress (pp. 297-323). New
York: John Wiley & Sons.

Clark, P. M. (1994). Preventing violence by building competent communities.
CommunithewsﬂienLlQ). 3-4

Commission on Violence and Youth. (1993). W
W Washington, DC: American Psychological

Association.

Connolly, J. (1989). Social self-efﬁcacy rn adolescence: Relations with self-
concept, social adjustment, and mental health. Canadianleumalpﬂﬁehayieuml
We), 258-269.

Crano, W. D.. & Brewer. M. B. (1986). ErinciplesandMethcdmLsttcial
Research (pp. 40-57). Newton, MA: Allyn and Bacon, Inc.

Crary, D. R. (1992). Community beneﬁts from mediation: A test of the ”Peace
Virus” hypothesis. W220), 24 -252.

DeJong,W. (1994). WWW
. Washington, DC: U.S.

 

Department of Justice.

Eisenberg, D. S., & Harris, J. D. (1984). Social competence: A developmental
PCISPOCtive. W30). 241-252-

Elliott, D. S. (1994). MW Boulder, CO: Center for
the Study and Prevention of Violence.

103

Farrington, D. P. (1991). Childhood aggression and adult violence: Early
precursors and later-life outcomes. In D. J. Pepler & K. H. Rubin (Eds.). The
development and treatment of childhool aggression (pp. 5-29). Hillsdale, NJ:
Erlbaum.

Fingerhut, L. A, Ingram, D. D., & Feldman, J. J. (1992). Firearm homicide

among black teenage males 1n metropolitan counties. IoumalpnheAmerieanMedical
W02), 3054-3058.

Ford, M. E. (1982). Social cognition and social competence in adolescence.
Dexelopmentalﬂrtehalogrtalﬁc), 323-340.

Freedman, B. J. Rosenthal, L, Donohoe, C. P, Schlundt, D. G., & McFall, R.
M. (197 8) A social-behavioral analysis of skill deﬁcits in delinquent and

nondelinquent boys. WWWAS. 1448-1462

Gaffney, L. R., & McFall, R. M. (1981). A comparison of social skills in
delinquency and nondelinquency adolescent girls using a behavioral role-playing

inventory. Wﬁ), 959-967.

Grusec, J. E. (1992). Social learning theory and developmental psychology: The

legacies of Robert Sears and Albert Bandura. WW (5).
776 - 786.

Hausman, A. J., Spivack, H., Prothrow-Stith, D., & Roeber, J. (1992). Patterns
of teen exposure to a community-based violence prevention program. Jpnmalpf
AdolescentﬂealthJZ. 668-675.

Hellman, D. A, & Beaton, S. (1986). The pattern of violence in urban public
schools: The inﬂuence of school and community. lpumalpresearehjnﬁrimeand
W0). 102- 127

Hirschi, T. (1971). W. Berkeley, CA: University of
California Press.

Hrarritz, J. R. ,& Eddowes, E. A. (1990). Violence: A crisis in homes and
schools. ChridhnmlJiducatronJl. 4-7.

Johnson, D. W, & Johnson, R. T. (1995). Why violence prevention programs
don’ t work -- and what does. we), 63-67.

Johnson, D. W. ,,Johnson R. T., & Dudley, B. (1992). Effects of peer
mediation training on elementary school students. Medratmnﬂnarterlrtamu), 89- 99.

104

Johnson, D. W., Johnson, R. T. Dudley, B. ,& Acikgoz, K. (1994). Effects of
conﬂict resolution training on elementary school students. IheleumalpLSpeial

Wm), 803- 817.

Johnson, D. W. ,Johnson, R. T., &Stevahn, L. (1995, July). W

WWW Paper presented at the annual meetings
of the National Association for Mediation 1n Education, Seattle, WA.

Kazdin, A. 13., Bass, D., Siegel, T, & Thomas, C. (1989). Cognitive-behavioral
therapy and relationship therapy 1n the treatment of children referred for antisocial

behavior. laumalnLConsulnnundﬂmrcaIPsxchologLﬂm). 522-535.

Luthar, S. S. (1995). Social competence in the school setting: Prospective cross-
domain associations among inner-city teens. W, 416-429.

Lyman, R. D., Prentice-Dunn, S, Wilson, D. R., & Bonﬁlio, S. A. (1984).
The effect of success or failure on self-efﬁcacy and task persistence of conduct-

disordered children. WW 516-519.

National Center for School Safety. (1991). W

W Washington, DC: U. S. Department of Justice. Ofﬁce of
Juvenile Justice and Delinquency Prevention.

Nuttal, E. V., & Kalesnik, J. (1987). Personal violence 1n the schools: The role
of the counselor. Ianmalntﬂaunsehnaandﬁexdapmentji 372- 375

Pepler, M. B. ,King, G. & Byrd, W. (1991). A social-cognitive based social
skills training program for aggressive children. In D. J. Pepler & K. H. Rubin (Eds. ).
. ... 'II.-.(pp361379) Hillsdale,NJ:

 

Erlbaum.

O' Donnell, J., Hawkins, J. D., & Abbott, R. D. (1995). Predicting serious

delinquency and substance use among aggressive boys. 1onmal_of_Consulting_and
WM), 529-537.

Osofsky, J. D. (1995). The effects of exposure to violence on young children.
Amensanlsxchalogrstamw), 782-788.

ProthrOW-Stith. D. (1987). YiolenmmenﬁonLCunieulumioLadolescems.
Newton, MA: Education Development Center.

Prothrow-Stith, D. (1991). Deadlmnsequenees. New York, NY:
HarperCollins.

105

Prothrow-Stith, D., & Spivack, H. (1992). Homicide and violence:
Contemporary health programs for America' s black community. In R. L. Braithwaite

&S. E. Taylor (Eds ),Hea1th.rssues_1n_the_blackcommumty(pp 132-143). San

Francisco: J ossey-Bass

Prothrow-Stith, D., & Spivack, H., & Hausman, A. J. (1987). The violence
prevention project: A public health approach. WW
We & 4), 67-69.

Ray, W. J. (1993). Methods(4th ed.). Belmont, CA: Brooks/Cole.

Roth, J. A. (1994). Understanding and preventing violence. Reseamhjnjﬁef.
(NCJ-145645). Washington, DC: National Institute of Justice.

Short, J. F. (1990). W Englewood Cliffs, NJ: Prentice-
'Hall.

Shure, M. B. & Spivack, G. (1988). Interpersonal cognitive problem solving.
InR. H. Price, E. L. Cowen, R. P. Lorion, &J. Ramos-McKay. (Eds..) Emmeen

ouncesmpmenﬁommmmpraetiﬁmm (pp. 69- 82). Washington, DC:

American Psychological Association.

Smith, V. E., Beals, G., Brant, M., & Annin, P. (1995, December 4). The lull
before the storm? Newsweek. 40—42.

Tate, D. C., Reppucci, N. D., & Mulvey, E. P. (1995). Violent juvenile
delinquents. WW9), 777-781.

Tolan, P., & Guerra, N. (1994a). Prevention of delinquency: Current status and
issues. AW 251-273.

Tolan, P. ,& Guerra, N. (1994b). WW2.
Anemmnmlmampimeﬁeld. Boulder, CO: Center for the Study and Prevention

of Violence.

Wacgel.W- 13- (1989). WWW
perspeet’me; Englewood Cliffs, NJ: Prentice-Hall.

Webster, D. W. (1993). The unconvincing case for school-based conﬂict

resolution. WW2“). 126-
141.

Wheeler, V. A., & Ladd, G. W. (1982). Assessment of children's self-efﬁcacy
for social interactions with peers. Wm), 795-805.

106

Winfree, L. T., Jr., Backstrom, T. V., & Mays, G. L. (1994). Social learning
theory, self-reported delinquency, and youth gangs: A new twist on a general theory
of crime and delinquency. WWI—25(2). 147-177.

Wolfgang, M. E., Figlio, R., & Sellin, T. (1972). W
Chicago: University of Chicago Press.

Yeates, K. O., & Selman, R. L. (1989). Social competence in the schools:
Toward an integrated developmental model for intervention. DexelopmemaLRerdew.
2. 64- 100.

Zigler, E., Taussig, C., & Black, K. (1992). Early childhood intervention: A

promising preventative for juvenile delinquency. AmedeanPsyehplogisLAlw),
997- 1006.

"I11111111111118?