What are they thinking?

MISSING IMAGE

Material Information

Title:
What are they thinking? moral reasoning in elementary children with emotionalbehavioral disorders
Physical Description:
xi, 221 leaves : ill. ; 29 cm.
Language:
English
Creator:
Hardman, Elizabeth Long, 1949-
Publication Date:

Subjects

Subjects / Keywords:
Special Education thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Special Education -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 210-219).
Statement of Responsibility:
by Elizabeth Long Hardman.
General Note:
Printout.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 028812594
oclc - 50405691
System ID:
AA00013630:00001


This item is only available as the following downloads:


Full Text










WHAT ARE THEY THINKING?
MORAL REASONING IN ELEMENTARY CHILDREN WITH
EMOTIONAL/BEHAVIORAL DISORDERS











By

ELIZABETH LONG HARDMAN


















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

THE UNIVERSITY OF FLORIDA

2002

























Copyright 2002

by

Elizabeth Long Hardman























I dedicate this dissertation to a former student.

For William
A four-year-old who taught me about patience, persistence, and unconditional love













ACKNOWLEDGEMENTS

I would like to recognize the contributions of the members of my dissertation

committee: Dr. Stephen W. Smith, Chair; Dr. David Miller; Dr. Hazel Jones; Dr.

Maureen Conroy; and Dr. Cecil Mercer. Dr. Smith's feedback was a critical element in

the preparation and funding of two grant applications, which ultimately provided valuable

support in the completion of the pilot and dissertation study. Dr. Miller's careful attention

to my research and data analysis procedures directed me down the straight and narrow

path of a competent, ethical researcher while, at the same time, providing patient teaching

to a novice investigator. I would also like to thank Dr. Jones, Dr. Conroy, and Dr.

Mercer, for offering me guidance and support as my doctoral program unfolded and the

Graduate School of Education, which provided financial support and words of

encouragement throughout my program.

I also owe a great deal of gratitude to the Spencer Foundation faculty who

provided valuable support in the early stages of the development of my study. Dr.

Michael Olneck from the University of Wisconsin-Madison and Dr. Geneva Gay from

the University of Washington spent hours with me critiquing and challenging my

thinking as I defined the purposes of my study and developed my research hypothesis. In

addition, I give credit to Dr. Robert Enright, my national mentor at the University of

Wisconsin-Madison, for identifying the anger connection, which proved to be vital in

interpreting the relationship between children's experiences and the development of

moral orientation. Finally, I want to acknowledge the contribution of Michele Gregoire,








another Spencer Fellow representing the University of Florida. Michele patiently guided

me through the data analysis procedures and provided valuable feedback as I developed

my ideas.

Finally, I would like to acknowledge the sacrifices made by my family in support

of my studies. My husband, Chris, spent many nights at home alone and was forced to

learn how to cook, clean, and maintain a household in my absence. My children, Emily

and Jack, supported my studies by patiently waiting for spring breaks and Christmas

holidays to receive any attention from their spiritually and physically absentee mother. I

could not have made it through the doctoral process without love and support from my

family.














TABLE OF CONTENTS
page

ACKNOWLEDGEMENTS ................... ..... ....... iv

LIST OF TABLES ................. .......................... ix

ABSTRACT ......................... .............. x

CHAPTERS

I INTRODUCTION TO THE PROBLEM ........................ 1

Statement of the Problem ................................. 4
Current Approaches ................. .................. 4
The Student With EBD .................................. 7
The Cognitive-Developmental Approach ..................... 10
Purpose and Objectives of the Study ........................... 12
Rationale .............................................. 12
Cognition ........................................... 13
Behavior ...................... ................... 15
Emotion ........................................... 17
Definition of Terms .......................... ......... 20
Delimitations of the Study ................................... 24
Limitations of the Study .................................... 25
Summary ............................................... 25
Overview of the Remaining Chapters .......................... 27

II REVIEW OF THE LITERATURE .......................... 28

Theory Development ....................................... 28
Piaget's Theory ......................................... 29
Piaget's Study ............................. ........... 30
Kohlberg's Contribution .... ............................. 35
Relevant Research ....................................... 41
Evidence for Stages ..................................... 42
Evidence for Social Domains ............................. 46
Social Constructivists Challenges .......................... 49
A Cognitive-Developmental Approach to Moral Education ....... 50
Review of Seven Independent Variables ....................... 52
Socioeconomic Status ..................... ............. 53









G rade ....................... .............. 54
Gender ............ ... ........................... 55
A nger ............................. ................... 57
Emotional/Behavior Disorders (EBD) ....................... 59
Ethnicity .......................... .. .............. 61
Reading Comprehension ...................... .. ..... .... 63
Sum mary ........................ ..................... 64

III METHOD ................... ........................ .68

Introduction to the Pilot Study ........................ ....... 68
Statement of Research Hypothesis .......................... 69
Research Design ..................................... 69
Setting and Participants ........................ ........ 70
Pilot Instrumentation.................. ..... ............ 73
Moral Theme Inventory ......................... ....... 73
Moral Dilemma Interview ............................... 74
Data Collection .......................... ............. 75
Moral Theme Inventory ................................ 75
Moral Dilemma Interview .................. .......... .. 76
Pilot Study Results ....................................... 81
Introduction to the Present Study .......................... 82
Statement of Research Hypothesis .......................... 83
Design ...................... ...................... 84
Sampling ................ ...... ........... ........... 85
Setting and Participants .................................. 88
Setting .............................................. 88
Participants ................. ........................ 88
Instrumentation ............................ .............. 90
Moral Theme Inventory .................................. 90
Moral Dilemma Interview ................................ 92
Feelings Questionnaire ................................. 93
Research Procedures ................. ..................... 96
Data Collection and Analysis .................. ........... 96

VI RESULTS ............................................. 99

Causal Comparative Results .................... ... ... ... .. 99
Bivariate Correlations ................................. 100
Simultaneous Multiple Regression .......................... 102
Case Study Results ........................................ 103
Issue Choice ......................................... 105
Moral Judgement ......................... ............. 107
M oral Orientation ....................................... 114
Summary of the Results .................................... 120









V DISCUSSION ............................ ............. 123

Overview of the Study ... .............................. 124
Summary of the Findings ................ ........ .......... 125
Discussion ... .................................. 126
Causal Comparative Results .................................. 127
Case Study Results .................................... 130
Relationships .......... .............................. 132
Just and Unjust Relationships ......................... 134
Feelings .. ....................... .................. 139
Moral Orientation ........................ ............. 144
Experience and the Development of Moral Orientation ............. 150
Limitations ........................................... 150
Implications .. ......................................... 153
Professional Practice ................... ............... 154
Teacher Preparation ..................... .... .......... 155
Future Research ............................ ............. 156

APPENDICES

A UNIVERSITY OF FLORIDA INSTITUTIONAL REVIEW......... 157

B RESEARCH INSTRUMENTS ............................... 164

Moral Theme Inventory ..................................... 165
Feelings Questionnaire ................................... 190
Moral Dilemma Interview ................................... 197

REFERENCES ............................................ 210

BIOGRAPHICAL SKETCH .................................... 220








LIST OF TABLES


Table page

2.1 Piaget's Theory ...................................... 33

2.2 Kohlberg's Theory .................................... 41

3.1 Issues Presented in Eight Dilemmas ......................... 77

3.2 Norms Valued and Their Definitions ........................ 80

3.3 The Moral Elements .................................. 81

4.1 Means and Standard Deviations By Group and Grade .......... 100

4.2 Correlation Coefficients Among Seven Variables.............. 101

4.3 Simultaneous Multiple Regression Analysis .................. 103

4.4 Summary of Issue Choice By Group ........................ 106

4.5 Summary of Norms Valued By Interview for Typical Peers ...... 109

4.6 Summary of Norms Valued By Interview for Students with EBD 110

4.7 By Group Summary of Moral Elements and Value Elements ..... 114

4.8 Moral Orientation By Group .............................. 120












Abstract of Dissertation Presented to the Graduate School
of the University of Florida In Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

WHAT ARE THEY THINKING?
MORAL REASONING IN ELEMENTARY CHILDREN WITH
EMOTIONAL/BEHAVIORAL DISORDERS

By

Elizabeth Long Hardman

May, 2002

Chair: Stephen W. Smith
Major Department: Special Education

Moral education is the cornerstone of public education with the classroom

community providing the environment in which students learn to think and act as morally

responsible citizens. While most students seem to benefit from the moral lessons taught

in the classroom, students with emotional/behavioral disorders (EBD) are placed in

special education specifically because they do not. Therefore, the purpose of this study

was to employ a cognitive-developmental approach to examine moral reasoning in 3",

4th, and 5"' grade students with EBD and their typical peers. Specifically, I collected and

analyzed data to address the following hypothesis. When accounting for the variables

gender, ethnicity, trait anger, reading comprehension, socioeconomic status (SES),

atypical/typical group, and grade, there will be a significant positive relationship between

reading comprehension and moral theme comprehension and a significant negative

relationship between trait anger and moral theme comprehension.








Results of the correlation analyses showed that ethnicity, SES, reading

comprehension, and trait anger were significantly related to students' ability to

comprehend moral themes in stories. When accounting for all seven independent

variables, however, only trait anger was found to be a significant predictor of

respondents' scores. Case study results indicated that the reasoning of informants with

EBD and many of their typical peers with low SES appeared to be self-focused, while

only two middle-income typical peers voiced other-focused moral judgements. These

results seem to question the sufficiency of the developmental process to direct the

development of children's moral judgement and suggest that experience may play a more

influential role in its development.

Results indicate that the meaningful inclusion of all children in the learning

community may be vital to the formation of responsible citizens and may prevent the

development of EBD for some children. To understand the importance of including all

students in the learning community, preservice teacher preparation should include the

study ethics and preservice teachers should be given opportunities to participate in

democratic learning communities. Future research should include rigorous inquiries into

the causal relationships among behavior, emotion, and moral judgement, but such

inquiries will require the development of valid measures of emotion, behavior, and moral

judgement.














CHAPTER 1
INTRODUCTION TO THE PROBLEM

Moral education is the cornerstone of America's public education system.

Historian Edward McClellan (1999) described the moral education of America's children

as an "article of faith" central to the creation of the public education and, in 1916,

philosopher John Dewey proclaimed schools as environments framed with the expressed

purpose of shaping children's mental and moral dispositions. Although moral education

was originally conceived as an extension of the moral training children received at home

(McClellan), modem supporters of moral education charge that the school, not the

family, bears the responsibility for developing a moral citizenry. According to Emile

Durkheim (1961)

[T]he general principle that the domain of the genuinely moral life only
begins where the collective life begins--or, in other words, that we are
moral beings only to the extent that we are social beings. (p. 64) ... The
family, especially today, is a very small group of persons who know each
other intimately and who are constantly in contact with one another. As a
result, their relationships are not subject to any general, impersonal,
immutable regulation. (p. 147)

Hence, the classroom community becomes an appropriate, naturally occurring context in

which students can learn to think and act as morally responsible citizens.

While most students seem to benefit from the moral lessons experienced in the

classroom community, students with emotional/behavioral disorders (EBD) are placed in

special education specifically because they do not (Kauffman, 1995). Since 1987, the

number of students with EBD has increased by 20% and most of this increase occurred








during the elementary school years (Office of Special Education Programs [OSEP],

1998). Moreover, researchers warn that these figures may significantly underestimate the

actual prevalence of EBD among elementary school children (Kauffman). In a recent

report, OSEP addressed the import of this alarming trend by admonishing, "Failure to

address the needs of students with emotional disturbance is a portent for poor community

results as well as poor academic results" (OSEP, Section II, Module 5, p. 41). The

growing number of students who develop EBD during the elementary school years and

the ensuing adverse individual and community consequences enjoin continued focused

attention on the problem and its possible solutions.

Evidence suggests, however, that the school may be falling short of its

responsibility to develop moral citizens when it come to the education of students with

EBD. The warning signs of emerging antisocial behavior patterns are often present at the

point of school entry and appear to be elaborated, not remediated, during the elementary

school years (Golly, Sprague, Walker, & Groham, 2000). For example, young children

who present significant behavior problems at age 3 or 4 have a 50:50 probability of

continuing to show behavior problems as they mature (Webster-Stratton, 1997). In fact,

researchers have found that the stability of aggressive and antisocial behavior, a common

characteristic of students with EBD, is equal to that of IQ and provides the single best

predictor of delinquency in adolescence (Walker, Colvin, & Ramsey, 1995).

Researchers project that failure to meet the moral and mental development needs

of young children who exhibit antisocial and aggressive behavior may result in serious,

life-altering consequences (Walker et al., 1995). The first of these life-altering

consequences may occur when the child is designated eligible for a special education








program for students with EBD. Although the general education classroom has become

the primary educative context for the general population of students with disabilities,

students with EBD are more likely to be placed in the most restrictive educational

settings (OSEP, 1998). For example, recent OSEP figures indicate that 55% of all

students with disabilities are served in the general education classroom, but only 26% of

students with EBD are included in the classrooms with their typical peers. Although 18%

of all children with disabilities are served in special classrooms and only 2% are served in

special facilities, 40% of students with EBD are served in separate classes and 12% are

sequestered from the mainstream in special facilities. Moreover, as elementary students

with EBD continue through school and beyond, their chances of reentering the general

education classroom are slim (Kauffman, 1995) and they bear an increased risk for a

variety of long term negative consequences such as school dropout, low employment

rates, drug and alcohol abuse, relationship problems, higher hospitalization and mortality

rates (Walker et al., 1995), and incarceration (U.S. Department of Education, 1994).

Researchers advise early intervention as a critical element in the prevention and

effective treatment of EBD, but state that students with EBD are not benefiting from

early intervention and prevention because serious emotional/behavioral problems are not

identified early enough to prevent the chronic and likely life-long disability of EBD (e.g.,

Forness, Kavale, MacMillan, Asarnow, & Duncan, 1996; Kauffman, 1995) and general

and special classroom environments are not adequate to meet the needs of children who

present serious emotional/behavioral problems (e.g., Kauffman, Lloyd, Baker, & Riedel,

1995). Moreover, researchers warn that some children will develop EBD in spite of








educators' best efforts at early intervention and prevention (Landrum & Tankersley,

1999; Webster-Stratton, 1997).

Statement of the Problem

The classroom community may be falling short of its commission to positively

influence the moral education of elementary students with EBD. The task, however, is

complex and difficult to realize. An emotional/behavioral disorders is a multifaceted

disability whose dimensions and developmental course are not clearly understood

(Kauffman, 1995). Therefore, the sometimes elusive task of positively influencing their

social behavior underscores the need to reexamine current approaches to the moral

education of students with EBD and to explore the promise of employing a previously

unexplored paradigm, the cognitive-developmental approach, to advance current

knowledge and to improve practice.

Current Approaches

Special education programs for students with EBD began to develop rapidly in the

late 1950s and early 1960s. During this time, researchers focused their efforts on the

effectiveness of behavioral approaches that employed the principles of conditioning and

learning to develop prosocial behavior in students with EBD (Kauffman, 1995). In the

1980s, behavioral models evolved into more sophisticated integrative approaches that

address the realities of children's affective and cognitive experiences by exploring the

effects of consequences on shaping behavior, thoughts, and affect (Kauffman). As a

result, cognitive-behavioral approaches have recently gained prominence, especially in

the study of aggression and antisocial behavior (Furlong & Smith, 1994; Kassinove,

1995).








Cognitive-behavioral researchers look to distorted cognition (Beck, 1976; Dodge

& Coie, 1987; Ellis, 1995,) and/or social skill deficits (Davis & Boster, 1992;

Deffenbacher & Swaim, 1999; Lochman, 1992; Spivack, Platt, & Shure, 1976) as sources

of maladaptive behavior. The practical aim of this approach, therefore, is to replace

distorted/deficit cognitive schemas with more effective ones through modeling, verbal

rehearsal, practice, and feedback (e.g., Smith, Siegel, O'Connor, & Thomas, 1994). In

addition, practitioners are encouraged to employ techniques such as self-instructional

training, self-talk, and role play to facilitate the remediation and to enhance

generalization to other settings (e.g., Lockman, Lampron, Gemmer, & Harris, 1987).

Consistent with behavioral approaches, however, the learner is portrayed as a passive

participant and the environment is "a major source of influence on behavior" as a result

of biological "processes that are collectively called conditioning or learning" (Johnston &

Pennypacker, 1993, p. 4).

During this same time period, proponents of behavior analytic approaches also

began to shift their focus toward examining thoughts as private events shaped by the

principles of conditioning and learning. These researchers took aim at the communicative

intent of maladaptive behavior as well as its environmental triggers to explain the

functional relationship between behavior and its consequences. As a result, functional

assessment gained such prominence in the treatment of EBD that the 1997 amendments

to IDEA require functional behavioral assessments for any student with a disability who

exhibits maladaptive behavior (e.g., Armstrong & Kauffman, 1999). The outcome of the

functional assessment is a behavior intervention plan that represents the school's good







6

faith effort to maintain students with disabilities in the general education classroom (Yell,

1998).

Over the last fifty years, researchers have found both behavioral and cognitive

behavioral approaches to be effective in shaping prosocial behavior in students with

behavioral problems (cf. Cameron & Pierce, 1994; Robinson, Smith, Miller, & Brownell,

1999). Nevertheless, researchers have also defined several important limitations to these

approaches (Webster-Stratton, 1997). When practically applied, researchers have found

that (a) many teachers are opposed to some behavior modification procedures (e.g.,

Kazdin & Cole, 1981), (b) behavior shaped using the principles of behavior modification

may not generalize to natural settings (e.g., Cullinan, Epstein, & Lloyd, 1991; Kauffman,

1995), (c) functional assessment may be too difficult to implement in classrooms (e.g.,

Katsiyannis & Maag, 1998), (d) some cognitive-behavioral approaches may be too

sophisticated for young children (Webster-Stratton), and (e) behavioral and cognitive-

behavioral approaches may not alter the environment that is nurturing the development of

EBD (e.g., Webster-Stratton).

Additionally, researchers have found that classrooms are not characterized by the

kind of positive strategies known to be effective in shaping prosocial behavior; therefore,

they speculate that the classroom environment itself may contribute to the development

ofEBD (e.g., Kauffman et al., 1995). For example, teachers frequently become highly

critical of misbehaving students, thereby engaging in coercive power struggles. As a

result, teachers fall victim to negative reinforcement traps and may unintentionally

perpetuate undesirable behavior (e.g., Jack et al., 1996; Webster-Stratton, 1997). For this

reason, researchers suggest that some approaches based on behavioral principles may be






7

inappropriate for classroom settings (e.g., Jack et al.; Repucci & Saunders, 1974; Shores,

Gunter, Denny, & Jack, 1993; Shores, Gunter, & Jack, 1993; Winett & Winkler, 1972).

The Student With EBD

The emotional and behavioral concomitants of EBD are explicit in the

nomenclature, but, according to the federal definition, EBD is a multifaceted, complex

disorder that involves maladaptive behavior, disordered affect, or disorder cognition

(Kauffman, 1995). According to federal law, students with EBD are characterized by one

or more of the following: (a) an inability to learn which cannot be explained by

intellectual, sensory, or health factors, (b) an inability to maintain satisfactory

interpersonal relationships with peers and teachers, (c) inappropriate types of behavior or

feelings under normal circumstances, (d) a general, pervasive mood of unhappiness or

depression, and (e) a tendency to develop physical symptoms or fears associated with

person or school problems (Individuals with Disabilities Act [IDEA], 1997). Therefore,

to understand the student with EBD, one must examine how these behavioral, emotional,

and cognitive components are manifested in the disorder.

Behavior. According to Kauffman (1995), EBD is partially represented in

maladaptive behavior that is discordant with social-interpersonal environments and

precludes opportunities for gratifying social interactions and experiences of self-

fulfillment. Externalizing disorders such as attention and activity disorders and conduct

disorders make up the most prevalent types of EBD and characteristically include

maladaptive behavior in the form of aggression, disruption, and antisocial behavior

(Kauffman; Kauffman, Culllinan, & Epstein, 1987). Aggressive, disruptive, antisocial

students arouse negative feelings and induce negative behavior in others, resulting in








social rejection and alienation from adults and peers alike. In fact, the behavior of these

students can be so unreasonable that others sometimes perceive their motives as

purposefully seeking punishment or rebuke (Kauffman). As members of the classroom

community, teachers regard children with externalizing disorders to be the most difficult

to teach, the least likable (Kauffman & Wong, 1991), and a threat to the development of a

secure, safe classroom environment (Langdon, 1997).

Emotion. Federal law portrays the disordered affective component of EBD as

"physical symptoms or fears" or "a general, pervasive mood of unhappiness or

depression," suggesting that depression and/or anxiety are the primary affective

characteristics of EBD. Although depression and anxiety are frequent emotional

concomitants of internalizing disorders such as obsessive-compulsive disorders, anxiety

disorders, clinical depression, and schizophrenia (Kauffman, 1995), researchers

commonly recognize anger as the underlying affective component of externalizing

disorders (Walker et al., 1995). Anger in children, however, has received little attention

from the research community (e.g., Fabes & Eisenberg, 1992).

Anger and cognition are mysteriously woven together in the existing research

literature. Novaco (1975) conceived a reciprocal relationship between cognition and

anger. Current models, however, portray anger as a mediating influence between

distorted/deficit cognition and aggressive behavior. For example, cognitive-behavioral

researchers treat aggression by reducing attributional biases of hostile intent (e.g., Dodge

& Coie, 1987; Fortman & Feldman, 1994; Lochman et al., 1987) and cognitive

distortions (e.g., Ellis, 1995), while behavior analytic researchers seek to identify and

change the objective elicitor and consequences of anger (Kassinove & Sukhodolsky,








1995). If, however, the relationship between anger and cognition is reciprocal, then

current approaches appear to be limited because they do not allow for anger as a justified

response to an unjust, hurtful situation. From this perspective, anger might precede

disordered cognitive functioning and aggressive, disruptive, and/or antisocial behavior

(Enright & Fitzgibbons, 2000). In fact, Kassinove and Sukhodolsky state that to

understand anger, researchers "would be wise to learn how people form their

conceptualizations of how their friends, family members, colleagues, and others should

act" (p. 24).

Cognition. Although maladaptive behavior and disordered affect are described as

defining manifestations of EBD, they are not the only determining characteristic.

According to the federal definition disordered cognition is also an influential determinant

ofEBD (Kauffman, 1995). For example, the definition describes EBD "as an inability to

learn that cannot be explained by intellectual, sensory, or health factors," "the inability to

build or maintain satisfactory interpersonal relationships," "inappropriate behavior or

feelings under normal circumstances," "a general pervasive mood of unhappiness or

depression," and "a tendency to develop physical symptoms or fears." Thus, by

definition, EBD appears to be deeply rooted in the cognitive domain.

Present conceptions of best practice in the education of students with EBD

suggest that remediation of distorted/deficit thinking processes is necessary to mitigate

the cognitive dysfunction inherent in EBD (Kauffman, 1995). In other words, the child's

thinking must be remediated because the child's reality is distorted. In contrast, the

cognitive developmental approach targets the child's reasoning as a lived reality,

constructed as the child perfects tentative solutions for a problematic situations (Dewey,








1916/1944). Therefore, the child's reasoning reflects a complex interaction between

judgement and action (Colby & Kohlberg, 1987a; Piaget, 1932/1965) and provides

knowledge about the bearings, uses, and causes of the situation as the child knows it to be

(Dewey). Overt behavior influences judgement; judgement influences behavior (Colby &

Kohlberg). Both the child and the environment become actors in the meaning making

process.

The Cognitive-Developmental Approach

Current issues and trends in the education of students with EBD enjoin a

heightened level of attention to the empirical and conceptual foundations of special

education (Kauffman, 1995). By employing a cognitive-developmental approach,

researchers are provided an opportunity to conceptualize and examine EBD from a

previously unexplored perspective. From this perspective, researchers can study the

thinking of students with EBD as an expression of a subjective reality, couched in the

developmental process (Coby & Kohlberg, 1987a; Piaget, 1932/1965). As a result,

researchers might extend current knowledge about EBD and its treatment and, at the

same time, broaden the conceptual foundations of special education.

Piaget (1932/1965), the father of cognitive-developmental psychology (Siegler,

1998), advised that if we want to form moral men and women, we must study the laws

that govern their formation. With this advice, he enjoined researchers to begin the inquiry

by examining the development of morality in children. Piaget argued that a child's

actions do not reveal the child's motivations and urged researchers to move beyond

merely observing how precise the child is in respecting rules and examine "how [the

child] judges of [sic] good and evil in the performance of his own action" (p. 117). How








children think and feel about their behavior is as much a part of social behavior as is the

most overt cooperative or hostile act (Dewey, 1916/1944). The relationship between

reasoning and behavior, however, continues to be obscure (Colby & Kohlberg, 1987a).

Moreover, Piaget asserted that "Difficult children, whom parents and teachers send or

ought to send up for psycho-therapeutic treatment, supply the richest material for

analysis" (p. 112).Yet only two studies (Astor, 1994; Astor & Behre, 1997) have been

published that examined the reasoning processes of elementary children with EBD.

According to Higgins (1995), the guiding assumption of cognitive-developmental

theory is that children create knowledge as a result of a bi-directional relationship

between thinking and experience. From this perspective, the child is portrayed as an

active participant in the creation of a subjective reality whose dimensions are described in

the child's verbal reasoning. Therefore, moral development researchers study children's

verbal reasoning because it can provide insight into past experiences and thinking, as well

as present perceptions of how actions and consequences are connected.

The present study was designed to employ a cognitive developmental approach to

examine the relationships among cognition, emotion, and behavior. I included elementary

aged students because the number of students with EBD seems to increase rapidly during

the elementary years (OSEP, 1998), 3" 5t grade present the best opportunity for

intervention (e.g., Walker et al., 1995), and mid to late elementary years provide a

context for rapid growth in the development of moral judgement (Piaget, 1932/1965).

Because the population of students with EBD is characterized by a disproportionate

number of African American males and students with low SES (Kauffman, 1995), I

included gender, SES, and ethnicity in my analyses. I also included reading








comprehension as a cognitive variable, trait anger as an emotional variable, and the

student's grade in school to gauge developmental change.

Purpose and Objectives of the Study

The aim of the present study is examine moral reasoning in 3rd, 4h, and 5'h grade

students with EBD using a cognitive developmental approach. The results should provide

valuable information about the developmental course of moral reasoning in elementary

children with EBD and provide depth to current understanding about the relationships

among moral reasoning (cognition), trait anger (emotion), and disordered behavior.

Specifically, I collected and analyzed data to address the following research hypothesis:

When accounting for the variables gender, ethnicity, trait anger, reading comprehension,

SES, atypical/typical group, and grade, there will be a significant positive relationship

between reading comprehension and moral theme comprehension and a significant

negative relationship between trait anger and moral theme comprehension.

Rationale

The purpose of the present study is to provide depth to current understanding

about the relationships among moral reasoning, trait anger, and disordered behavior by

examining how moral reasoning in elementary students with EBD differs from that of

typical peers. Because EBD is a complex disability, it presents a unique opportunity to

increase knowledge about the relationships among behavior, emotion, and cognition.

Although only a few researchers have examined moral reasoning in elementary children

with EBD (Astor, 1994; Astor & Behre, 1997, Hardman & Smith, 2001), many have

examined moral reasoning in typical populations of elementary children charting its






13

developmental course (e.g., Colby & Kohlberg, 1987a; Piaget, 1932/1965) and evaluating

the influence of culture (e.g., Gilligan, 1982; Shweder, Nahapatra, & Miller, 1987).

Cognition

Piaget (1932/1965) described changes in children's moral reasoning as a

cognitively structured, maturational progression. Subsequent research, however, has

demonstrated that poverty (Colby & Kohlberg, 1987a), reading comprehension (Narvaez,

Gleason, Mitchell, & Bently, 1999), gender (Gilligan, 1982) and ethnicity (Edwards,

1987) may also influence the structural characteristics and maturational progression of

moral reasoning. Interestingly, students with EBD are overly represented by low

socioeconomic, African American males and as a population exhibit a slower rate of

achievement (Kauffman, 1995). Therefore, a review of researchers' findings regarding

the relationships among reading comprehension, poverty, SES, and ethnicity and moral

reasoning is warranted.

Reading comprehension. Studies of moral reasoning in the general population and

cross-culturally suggest that low achieving individuals will also exhibit depressed or

stagnant maturation in moral reasoning. For example, in a longitudinal study of moral

judgement in U.S. males, Colby, Kohlberg, Gibbs, and Lieberman (1987) found that

although age accounted for 60% of the variance in scores, there was also a moderate

relationship between the maturation of moral judgement and achievement. Further

examination of the results revealed, however, that differences in IQ and SES seemed to

control the strength of this relationship; therefore, the relationship between achievement

and moral reasoning appeared to be mediated through IQ and SES differences.








Recently, Narvaez and her coauthors (1999) studied moral reasoning in 3r, 5 ,

and university students by assessing their ability to understand themes of cooperation in

moral stories. Narvaez and her colleagues found that reading achievement and reading

comprehension were significantly related to participants' understanding of moral themes

of cooperation; however, when reading comprehension was controlled, the relationship

between moral theme comprehension and maturation remained significant. Therefore,

these authors concluded that while moral theme comprehension is related to reading

comprehension, participants' understanding of moral themes of cooperation seems to

require something beyond reading comprehension.

Socioeconomic status (SES). Colby, Kohlberg, Gibbs, and Lieberman (1987)

found that individuals who live in poverty reason at lower stages of maturation. These

authors interpreted the finding of social class differences in rate and terminus of

development in moral reasoning as a reflection of differential participation in and

identification with society and its institutions. Differential participation in social

institutions creates differential role-taking opportunities for middle-class and working

class children. Therefore, individuals from the middle class experience the role of

cooperative participants in society and, as a result, develop a social system perspective

that characterizes higher stages of reasoning (Colby & Kohlberg, 1987a). In a pilot study

of moral reasoning in elementary students with EBD, Hardman and Smith (2001) found a

significant difference in the ability of students with low SES to detect moral themes in

stories when compared with peers from middle and upper income families. Because SES

and EBD were confounded, these authors reached no conclusions regarding the influence

of EBD on moral reasoning. These findings seem to support the hypothesis that poverty







15

presents circumstances in which democratic education may be inconsistently applied and

opportunities for developing a moral disposition may be differentially perceived (Colby

& Kohlberg; Dewey, 1916/1944).

Gender/culture. Social constructivists propose the development of moral

reasoning as an entirely social phenomenon, arguing that the norms and values that guide

a child's moral reasoning are culturally dependent (Kurtines & Gewirtz, 1995).

Furthermore, social constructivists charge that stage theories preference the

individualistically focused values and norms of Western cultures and relegate reasoning

that preferences communitarian ideals to lower developmental stages (e.g., Brown,

Tappan, & Gilligan, 1995; Gilligan 1982; Shweder et al., 1987). For example, Gilligan, a

feminist critic of stage theories, found that moral reasoning in females is more likely to

be grounded in communitarian ideals that are nurtured by women's unique cultural role

as care givers. She argued that the Piagetian/Kohlbergian framework misinterprets the

stage development represented in a communitarian orientation as deficient when, in fact,

this orientation is only different.

Behavior

The nature of the relationship between reasoning and behavior in children under

the age of 12 is, for the most part, unexplored territory. Piaget (1932/1965) hypothesized

that children's moral reasoning represents a gradual coming into consciousness of past

actions; therefore, moral reasoning must lag behind moral action. Moreover, Colby and

Kohlberg (1987a) conjectured that as individuals attain more mature stages of reasoning,

they are more likely to demonstrate higher levels of moral behavior. These researchers

explained that the principles that guide higher stages of moral reasoning (e.g., it is wrong








to harm others) are unalterable and universal; therefore, moral relativism is not possible

and moral action ensues. Principled reasoning obligates moral behavior. On the other

hand, moral realism premised on conventional morals (i.e., cultural rules about dress,

sexual behavior, social behavior) encourages moral relativism because conventional

morals may be altered to suit the egoistic demands of a situation. Thus, conventional

morality is not always obligatory (Colby & Kohlberg).

According to Gibbs (1995), egocentric biases and moral realism are the primary

indicators of lower stages of moral reasoning and appear to be natural to early childhood.

The persistence of egocentrism into adolescence, however, may render the individual at

high-risk for serious antisocial behavior. Because egocentrism precludes meaningful

perspective taking, egocentric biases preempt any consideration for the expectations and

feelings of others. Therefore, egocentric, antisocial adolescents can develop moral

realism (rule oriented behavior) only with respect to their own needs, wants, and desires.

When compliance does not meet selfish needs, rule breaking is justified. Their thinking is

similar to that of a five-year-old, but because their size, needs, and level of freedom are

vastly different, egocentric biases in their moral reasoning present a clear danger.

Hardman and Smith (2001) examined moral reasoning in 21 students with EBD

and found that these students experienced greater difficulty comprehending moral themes

in stories than did their typical peers. This finding was supported and elaborated as a

result of an analysis of moral dilemma interviews with three students selected from the

participant sample of students with EBD. As a result, these authors found informants'

moral judgements to be replete with expressions of moral realism and egocentric biases.

The authors caution, however, that EBD and SES were confounded; therefore, the effects








for EBD could not be evaluated independent of the effects for SES. This finding bears

preliminary theoretical and empirical significance about the relationship between moral

reasoning and behavior and warrants further inquiry (Hardman & Smith).

Emotion

Anger. In most cases, anger is assumed to underlie the observable antisocial,

aggressive, and disruptive behaviors that limit social interactions of elementary students

with EBD (Kauffman, 1995; Walker et al., 1995). Aggressive, disruptive, antisocial

behavior arouses negative feelings and induces negative behavior in others, resulting in

social rejection and alienation from adults and peers (Kauffman). Anger, therefore,

appears to provide a motivational limit to the moral socialization of many students with

EBD (Gibbs, 1995).

Definition. Anger is a complex emotion (e.g., Deffenbacher & Swaim, 1999) that,

until recently, has received little attention from the research community (e.g., Boekaerts,

1993; Fabes & Eisenberg, 1992; Murphy & Eisenberg, 1996). Anger is defined as an

emotional response to perceived threats, insults, frustration, injustice (Lehnert,

Overholser, & Spirito, 1994), intentional harm (Levine, 1995) or conflicts (Murphy &

Eisenberg) and may represent a temporary emotional state or a personality trait (Jacobs,

Phelps, & Rohrs, 1989). Students with externalizing disorders may experience such

intense chronic anger that their angry feelings reach beyond state anger and may signify

an anger disorder (Kassinove, 1995) or personality trait (e.g., Deffenbacher, Lynch,

Getting, & Kemper, 1996; Walker et al. 1995), thereby adversely affecting the quality of

a child's social interactions.







18

Anger, however, is not always a sign of disordered affect. On the contrary, anger

may represent a normal emotional response in many instances. For example, anger is a

common denominator in situations in which children believe their personal values,

including those of fairness and justice, are challenged (Boekaerts, 1993) and represents a

culturally normative response to discrimination and injustice for minority youth

(Stevenson, Reed, Bodison, & Bishop, 1997). Conversely, years of discrimination and

injustice may also result in such serious assaults on personal values that perceptions of

injustice may become a source of intense, destructive trait anger (Chan, 1994).

Anger and other emotions. Recent research aimed at clarifying the relationship

between anger and other emotions has defined anger as a mysterious, complex emotion

that may play an instrumental role in the developmental course of other emotions, such as

depression (Berkowitz, 1990; Clay, Anderson, & Dixon, 1993; Enright & Fitzgibbons,

2000; Heavey, Adelman, Nelson, & Smith, 1989; Levine, 1995) and anxiety

(Deffenbacher et al., 1996; Hains & Szyjakowski, 1990). For example, Enright and

Fitzgibbons highlight cases in which anger emerges as an intense response to extreme

circumstances of injustice such as sexual, physical, or mental abuse and represents a

precursor to the onset of intense depression, poor health, and antisocial behavior.

Moreover, these authors argue that, in these cases, anger must be resolved before the

destructive emotional, physical, and behavioral expressions of anger can be effectively

addressed.

Anger and cognition. Cognitive models of emotion suggest that people's emotions

depend less upon actual events and more upon people's interpretations of those events

(e.g., Dodge & Coie, 1987; Levine, 1995); therefore, children's use of morally relevant








information may be an important moderating influence on children's tendencies to

become angry. Simply stated, the expression of anger signals moral judgement (Olthof,

Ferguson, & Luiten, 1989). For example, researchers have found that (a) violent children

show lower empathy and insight, leading to concerns over issues of fair play and evoking

aggressive responses (Davis & Boster, 1992), (b) ego development seems to be inversely

correlated with sadness and anger in emotionally disturbed adolescents (Hauser & Safyer,

1994), (c) students diagnosed as having significant learning disabilities exhibit higher

levels of anger and misbehavior when compared to those without learning problems

(Heavey et al., 1989), and (d) children who exhibit externalizing disorders become

aggressive because they misinterpret the intentions of others and are unable to manage or

cope with the situation in appropriate ways (e.g., Deffenbacher et al. 1996; Lockman,

1992).

Anger and socialization. Researchers have found evidence to indicate that there is

a significant relationship between anger and social competency. For example, Murphy

and Eisenberg (1996) examined the angry conflicts of 108 elementary aged children and

found that social competency predicts anger intensity even when controlling for sex, age,

and causes of angry conflicts. Downey, Lebolt, Rincon, and Freitas (1998) studied

rejection sensitivity and its relationship to anger and found that poor and minority

children who angrily expected social rejection behave more aggressively, experience

increased interpersonal difficulties, and decline in academic learning over time.

Moreover, Fabes and Eisenberg (1992) found that young children who were relatively

low in social status and competence seemed to "invite" aggressive rejecting conflict and








that popular, socially competent children were less likely to be involved in angry

conflicts.

Definition of Terms

Accommodation

Accommodation refers to the way in which individuals adapt their ways of

thinking when presented with new experiences. For example, an extreme case of

accommodation is imitation (Siegler, 1998).

Assimilation

Assimilation is the way individuals transform or interpret information to fit within

their existing way of thinking. If individuals are not able to understand new information

in concert with existing ways of thinking, then they cannot form a meaningful

representation of the new material. An extreme case of assimilation is fantasy play.

Accommodation cannot be present without assimilation and vice versa (Siegler, 1998).

Autonomous Morality

Autonomous morality is characterized by an independent, self-legislative stance

when making moral judgements (Piaget, 1932/1965).

Cognitive-Developmental Approach

The cognitive-developmental approach posits that an individual's thinking

undergoes a series of transformations as a result of an interaction between heredity and

experience (Kertines & Gerwitz, 1995). These transformations occur in universal,

invariant stages (Colby & Kohlberg, 1987a) and are facilitated by the basic

developmental processes of assimilation, accommodation, and equilibration (Siegler,

1998).








Constructivism

Constructivism refers to individuals' ability to construct meaning for themselves

by thinking about and acting on the world. When individuals invent or construct new

responses to each novel situation, the form of the meaning constructed is constrained by

the individual's current developmental level. Therefore, the present mode of construction

is an outgrowth of the prior mode. Individuals cannot simply internalize higher stage

reasoning, but instead are only able to move forward to the logical next step of cognitive

reorganization (Colby & Kohlberg, 1987a).

Distributive Justice

Distributive justice means to deal an equal, proper share to each group or

individual (Piaget, 1932/1965).

Deontic Reasoning

Deontic reasoning refers to what is right or duty bound and does not encompass

other types of ethical judgements such as judgements of moral worth or virtue of

particular persons or actions, judgements about the goodness of lifestyle, or judgements

about ideals of the good life (Colby & Kohlberg, 1987a).

Egocentrism

Egocentrism refers to a way of thinking about the external world in terms of one's

own perspective. This psychological limitation can be found in the thinking of many

children between the ages of 2 and 7 (Siegler, 1998).








Emotional/Behavioral Disorders (EBD)

Emotionally disturbed is defined as follows:

(i) The term means a condition exhibiting one or more of the following
characteristics over a long period of time and to a marked degree, which
adversely affects educational performance:
(A) An inability to learn which cannot be explained by intellectual,
sensory, or health factors; (B) An inability to build or maintain
satisfactory interpersonal relationships with peers and teachers; (C)
Inappropriate types of behavior or feelings under normal
circumstances; (D) A general, pervasive mood of unhappiness or
depression; or (E) A tendency to develop physical symptoms or
fears associated with personal or school problems.
(ii) The term includes children who are schizophrenic. The term does not
include children who are socially maladjusted, unless it is determined that
they are seriously emotionally disturbed. (34 C.F.R. 300.7 [b][9][1997])

Equilibration

Equilibration refers to the overall interaction between existing ways of thinking

and new experience. It is the essential element of developmental change. Equilibration

involves a three-phase process. First, equilibration is represented in a state of satisfaction

with present modes of thought. Then, the individual becomes aware of shortcomings in

their existing thinking and is dissatisfied. Finally, through assimilation and

accommodation, the individual adopts a more sophisticated mode of thought that

eliminates the shortcomings of the old one; that is, the individual reaches equilibrium

(Siegler, 1998).

Expiatory Punishment

Punishment employed to extract payment from wrong doers for their bad behavior

(Piaget, 1932/1965).

Heteronomous Morality

Moral behavior that is subject to external forces or controls (Piaget, 1932/1965).








Moral Dilemma

A moral dilemma is a story that poses a conflict in values. Perceptions and ideas

generated in response to moral dilemmas represent the individual's constitutive

knowledge about the bearings, uses, and causes of the situation (Dewey, 1916/1944).

For example, the Heinz dilemma is a story about a man who must steal drug to save his

wife's life. Hence, this story poses a conflict between the value of preserving life and the

value of upholding the law (Colby & Kohlberg, 1987a).

Moral Judgement

A moral judgement is an imperative derived from some rule or principle of action

that the speaker believes is binding on his own actions. Moral judgements include (a)

judgements of value, not fact; (b) social judgements involving people; (c) prescriptive or

normative judgements, and (d) value judgements of rights and responsibilities, rather than

value judgements of liking and preference (Colby & Kohlberg, 1987a).

Moral Reasoning

To perfect a tentative solution for a problematic situation, individuals will

carefully scrutinize existing conditions and the implications of various hypotheses. This

operation is called reasoning (Dewey, 1916/1944). Moral reasoning refers to the

hypothesis constructed when reasoning in the moral domain.

Phenomenalism

Phenomenalism refers to the development of moral judgement as an expression of

an individual's subjective reality that is developed through a complex interaction between

judgement and action. Causality is bidirectional. Overt behavior influences moral beliefs;

moral beliefs influence behavior (Colby & Kohlberg, 1987a).








Retributive Justice

Retributive justice requires "pay back" in equal measure. An eye for an eye, a

tooth for a tooth is an example of retributive justice (Piaget, 1932/1965).

Structuralism

Structuralism refers to the general organizing principles or patterns of thought

rather than specific moral beliefs or opinions. According to Colby and Kohlberg (1987a)

concepts are not learned or used independently of one another, but are bound together by

common structural features. A pattern of connections, a structure or set of relations and

transformations, develop within the subject's meaning. Therefore, structuralism is the

hermeneutic that guides the analysis of the organization of thought inherent in the

individual's responses to moral dilemmas.

Delimitations of the Study

The present study was designed to employ a cognitive developmental approach to

examine the relationships moral reasoning (cognition), trait anger (emotion), and

behavior. I included elementary aged students because the number of students with EBD

seems to increase rapidly during the elementary years (OSEP, 1998), 3rd 5* grade

present the best opportunity for intervention (e.g., Walker et al., 1995), and mid to late

elementary years provide a context for rapid growth in the development of moral

judgement (Piaget, 1932/1965). Sampling procedures took place in two moderately sized

school districts (approximately 27,000 30,000 students) in rural, Central Florida. I

purposively selected a sample of 12 students from the participant sample to participate in

moral dilemma interviews.








Limitations of the Study

The purpose of this study was to examine moral reasoning in 3rd, 4h, and 5th grade

students with EBD. I implemented a causal correlational research design, using sampling

procedures and statistical methods that yield generalizable results to the population of 3rd,

4', and 5th grade students with EBD and their typical peers in the two rural school

districts from which they were selected. The findings cannot be generalized to students at

other grade levels or to students with disabilities other than EBD. The results of case

study research are not generalizable to other students and informants' prescriptions about

what should be done may not predict the choices they make in natural settings.

Additional specific methodological constraints are described in Chapter III.

Summary

Efforts at prevention and intervention have failed to stem the increasing numbers

of students who develop EBD during the elementary school years. Since the inception of

special education classes for students with EBD in the 1950s, researchers and

practitioners have focused intervention and prevention efforts on the effectiveness of

behavioral and cognitive-behavioral approaches to meet the needs of students with EBD.

Proponents of these approaches assume that the principles of learning and conditioning

are the primary influences that shape behavior, cognition, and emotion and attribute

importance to environmental forces as the active influence on the learning process (e.g.,

Kauffman, 1995). In contrast, the cognitive developmental approach presents the child as

an active participant in creating a subjective, lived reality (e.g., Dewey, 1916/1944;

Piaget, 1932/1965) and allows for the study of EBD as a manifestation of a complex

interaction among cognition, emotion, and behavior.








The present study was designed to employ a cognitive developmental approach to

examine the relationships among moral reasoning (cognition), trait anger (emotion), and

behavior by targeting the development of moral reasoning in elementary students with

EBD. I included elementary aged students because the number of students with EBD

seems to increase rapidly during the elementary years (OSEP, 1998), 3" 5' grade

present the best opportunity for intervention (e.g., Walker et al., 1995), and mid to late

elementary years provide a context for rapid growth in the development of moral

judgement (Piaget, 1932/1965). Because the population of students with EBD is

characterized by a disproportionate number of African American males and students with

low SES (Kauffman, 1995), I included gender, SES, and ethnicity in my analyses. I also

included reading comprehension as a cognitive variable, trait anger as an emotional

variable, and the student's grade in school to gauge developmental change.

Because of the complexity of the disability, EBD present a unique opportunity to

increase current knowledge about the complex relationships among behavior, emotion,

and cognition and the influence of each of these elements on the development of

morality. Cognitive-developmental researchers have examined moral reasoning in typical

populations of elementary children charting its developmental course (e.g., Colby &

Kohlberg, 1987a; Piaget, 1932/1965) and evaluating the influence of culture (e.g.,

Gilligan, 1982; Shweder et al., 1987). On the other hand, the relationship between moral

reasoning and emotional/behavioral influences has received little attention. I found only

two studies (Astor, 1994; Astor & Behre, 1997) that examined moral reasoning in

students with EBD.







27

Overview of Remaining Chapters

In the remaining chapters, I further elaborate the theoretical framework, review

the relevant literature, describe the methods, report the results, and discuss the

implications of the findings. In Chapter II, I describe the theoretical framework of the

cognitive developmental approach and review the literature relevant to the development

of moral reasoning in typical and atypical populations. Chapter III provides a detailed

description of a pilot study and describes the research design, the setting and participants,

data gathering instruments, and the research procedures employed to implement the

study. In Chapter IV, I relate the results and in Chapter V, I synthesize the findings and

discuss the implications for researchers and practitioners.













CHAPTER II
REVIEW OF THE LITERATURE

Chapter II provides a synthesis of the literature on children's moral development

from a cognitive developmental perspective. I begin by relating the theoretical

assumptions that undergird the cognitive-developmental approach, followed by a review

of Piaget's seminal study (1932/1965). I then review Colby and Kohlberg's (1987)

reiteration of the theoretical assumptions and describe how these authors refined and

expanded Piaget's structural, or stage theory of moral development. Finally, I review

subsequent research, discussing both support for and challenges to the cognitive-

developmental conceptualization of the development of morality in elementary aged

children, giving detailed attention to the studies that investigated the seven independent

variables (i.e., gender, grade, ethnicity, atypical/typical behavior, trait anger, SES, and

reading comprehension) targeted in the present study.

Theory Development

The cognitive developmental perspective of children's moral development

assumes that children create knowledge as a result of a bi-directional relationship

between thinking and experience. When children see that a certain way of acting and its

consequence are connected, but they do not know how they are connected, reasoning

becomes an intentional endeavor to discover specific connections between the things

children do and the ensuing consequences (Dewey, 1916/1944). By examining children's

reasoning processes, the researcher seeks insight into past experiences and thinking, as








well as their present perceptions of how actions and consequences are connected

(Higgins, 1995; Siegler, 1998).

Piaget's Theory

Assuming a cognitive-developmental perspective about the creation of

knowledge, Piaget (1952) set forth three psychological processes that facilitate children's

progression toward higher levels of knowing: assimilation, accommodation, and

equilibration. According to Piaget, assimilation is the psychological process that defines

the way children transform incoming information so that it fits their existing way of

thinking. When new information is assimilated, children are able to form a meaningful

representation of the new material, but if new information can not be assimilated, the

meaning making process is thwarted. Siegler (1998) presents pretend play as an extreme

case of assimilation. When children pretend, they gloss over the physical characteristics

of objects and treat them as if they are what the children are momentarily interpreting

them to be. In contrast, Piaget presented accommodation as the psychological process

that defines the way children adapt their ways of thinking to new experiences. For

example, Siegler uses imitation as an extreme case of accommodation. When children

imitate, they minimize their interpretations and simply mimic what they see. According

to Piaget, assimilation and accommodation mutually influence each other, and

assimilation is never present without accommodation and vice versa.

Equilibration refers to the overall interaction between existing ways of thinking

and experience and is accomplished in three phases. First, children are satisfied with their

present mode of thought and are in a state of equilibrium. When they are confronted with

shortcomings in their present thinking, dissatisfaction or a state of disequilibrium,








develops. To restore a state equilibrium, children adopt a more sophisticated mode of

thought that eliminates the shortcomings of the old way of thinking (Piaget, 1952). In this

manner, equilibration provides the impetus for maturation.

Piaget's Study

In 1932, Piaget conducted a study of children's moral reasoning, employing a

methodology and establishing a theory firmly grounded in the epistemological

assumptions of cognitive-developmental psychology. Assuming that children construct

their moral realities as a result of their mental and physical actions, Piaget (1932/1965)

examined how nature, or the maturational process, and experience influence children's

understanding of morality.

"All morality consists in a system of rules, and the essence of all morality is to be

sought for in the respect which the individual acquires for these rules" (Piaget,

1932/1965, p. 13). Because adults teach children the rules for moral behavior, moral rules

are developed, not by children, but by a succession of generations (Dewey, 1916/1944)

and are transmitted to children before they are conscious of obligation (Piaget). To make

this developing consciousness visible, Piaget studied children's understanding of the

rules of the game of marbles because the rules of children's games are consciously

developed and elaborated by children. Specifically, Piaget targeted two phenomena to

observe: (a) practice; that is, the way in which children of different ages apply rules and

(b) consciousness; that is, whether they are obligatory and sacred or a matter of choice.

The rules of children's games. Piaget's (1932/1965) subjects were infants to 12-

year olds living in Geneva and Neuchatel, Switzerland. He began by observing and

noting children's variations in rule application and, then, he questioned them about








fairness of these variations. As a result, Piaget described four developmental stages that

define children's practice of rules: sensorimotor, egocentric, cooperation, and

codification of rules. Moreover, through the practice of rules, children develop a

consciousness of rules, or an understanding of the source of the obligatory, in three

stages: nonmoral; heteronomous morality; and autonomous morality.

The sensorimotor stage lasts from birth to approximately two years. During this

stage, children merely engage in sensorimotor explorations of a game's accoutrements

without regard for rules. Thus, stage 1 children are not conscience of moral obligations

and are described as nonmoral.

The sensorimotor stage is followed by the egocentric stage. At about age 2,

children enter a developmental period during which their thinking is limited by

egocentrism. Egocentric children think about the external world always in terms of their

own perspective and their own position within it (Siegler, 1998). Stage 2 children can

recite rules and state a respect for rules but, in practice, apply rules to suit their egocentric

wants and desires. Because of egocentrism, stage 2 children perceive the obligatory as

coerced by authority figures and entirely externally imposed. Moral realism develops,

leading to the practice of a heteronomous morality, a morality dependent on external

sources of obligation.

At about age 6, children develop a perspective of cooperation and enter the

cooperation stage. Egocentrism begins to lose its control on children's thinking allowing

insight into others' perspectives. This perspective taking ability predisposes them to an

attitude of cooperation even though moral realism persists. In fact, Piaget (1932/1965)

noted that stage 3 children accept rules with "mystic respect... [R]ules are eternal, due








to the authority of parents, of the Gentlemen of the Commune, and even of an almighty

God" (p. 61). Thus, stage 3 children continue to view obedience to authority as a primary

moral obligation; rules are sacred, untouchable and last forever.

At about age 10, experiences of cooperation with other children and authority

usher in an understanding of the rational rule, one that is self-imposed through mutual

consent. This move toward rationalism begins at about age 10 and signals the child's

transition into stage 4, the codification of rules stage. By age 12, the transition into stage

4 is complete and children understand that rules are self-imposed obligations that are

created through mutual agreement. Their moral judgements exemplify a perspective of

equal justice tempered by equity and the child becomes a moral being. Table 2.1

summarizes the salient points of Piaget's theory.

Moral dilemmas about clumsiness, lying, and stealing. Piaget's (1932/1965)

objective was to gain insight into morality as a developmental process and to determine

how this process relates to the practice and perception of the rules for moral behavior.

Identifying the stages that guide the application and consciousness of the rules of

children's games was only the first step toward achieving this objective. Piaget's second

step was to employ the technique of theoretical sampling to examine how these

developmental stages affected children's reasoning about moral problems.

To elicit children's moral reasoning, Piaget (1932/1965) presented 5 to 13-year

old children with moral dilemmas about clumsiness, lying, and stealing and discovered

that the same concepts that guide the practice and consciousness of the rules of children's

games also guide children's understanding of the rules for moral behavior. Piaget found

that themes of moral constraint coerced by an external authority distinguished young









Table 2.1

Piaget's Theory


Age Practice of Rules Consciousness of Rules


birth-2 years Stage 1

3-6 years Stae 2


7-9 years

10-12 years


Sensorimotor

Egocentric

Limited by egocentrism


Stage 1

Stage2


Stage3 Cooperation


12- adult Stage 4Codification of Rules Stage 3


Non-Moral

Heteronomous Morality

The moral is externally

coerced. Moral realism

prevails.

Develop perspective

taking

The moral is determined

by mutual consent.

Autonomous Morality

The moral is determined

from within


children's reasoning processes. As children matured, however, themes of cooperation and

social interaction began to emerge as the precursors of a developing autonomous morality

characterized by an independent, self-legislative stance.

Piaget (1932/1965) described children who practice heteronomous morality as

moral realists who perceive moral behavior as entirely externally imposed. From a moral

realist's perspective, rules do not require understanding, only obedience and conformity

regardless of the circumstances. Therefore, duty becomes self-subsistent and independent









of reason. Any act that shows obedience to a rule or to authority, regardless of the

command, is good. Any act that does not conform to the rules is bad. The good is

obedience and the letter rather than the spirit of the law represents an objective

conception of moral responsibility. As children mature, however, the development of

relationships among children and a perception of increasing equality with adults lead to

an attitude of mutual respect and children begin to understand that rules are imposed

through social contract.

Justice. To examine the development of justice reasoning, Piaget (1932/1965)

presented informants with dilemmas that stimulated their thinking about expiatory,

distributive, and retributive justice; collective and communicable responsibility;

immanent justice; equality and equity; and authority. He found that up until the age of 7

or 8, justice is subordinate to adult authority, but from ages 8 to 11, children

progressively develop the notion ofequlitarianism, or mutual respect. By the age 11 or

12, children adopt a perspective of purely equalitarian justice tempered by considerations

of equity. As a result, Piaget concluded that mutual respect is prerequisite to notions of

distributive justice and reciprocity and identified the ethics of mutual respect as the

moral.

The child sets forgiveness above revenge, not out of weakness, but
because 'there is no end' to revenge (a boy of 10)... [S]o in ethics,
reciprocity implies a purification of the deeper trend of conduct, guiding it
by gradual stages to universality itself. (p. 323-24)

Thus, morality develops out of reciprocity, when mutual respect is strong enough to make

the individual feel from within the desire to treat others as he wishes to be treated

(Piaget).








Kohlberg's Contribution

In 1987, Colby and Kohlberg replicated and refined Piaget's theory and authored

what might be one of the finest examples of a stage theory (Higgins, 1995). Their work

represented the end product of Kohlberg's lifelong search for the constituents of Piaget's

universal morality and the mechanisms that facilitate its development (Reed, 1997).

Moreover, Kohlberg attempted to transform theory into practice, by implementing a Just

Community approach to moral education in three high schools in Boston (Power,

Higgins, Kohlberg, 1989).

Colby and Kohlberg (1987a) reframed the epistemological assumptions of the

cognitive-developmental approach by defining the role of phenomenalism structuralism

and constructivism in the development of morality. The first, phenomenalism, refers to

moral judgement as a subjective reality, dependent on individual perceptions and

influenced by both experience and maturation. Thus, an individual's moral reality

develops as a result of a complex interaction between judgement and behavior as

mediated by maturation. Thus, the relationship between moral judgement and behavior is

bi-directional; overt behavior can influence moral beliefs just as moral beliefs can

influence behavior.

Structuralism sets forth the proposition that as children grow older their thinking

changes as a result of an interaction between experience and maturation (Higgins, 1995).

This change occurs in qualitatively distinct stages that begin and end with a brief

transitional period and may occur abruptly (Siegler, 1998). Thus, children's reasoning in

earlier stages differs qualitatively from their reasoning in later ones and at any given

point in development children reason similarly on moral problems. Therefore, concepts







36

such as moral obligation are not merely learned or used independently of other concepts,

but are bound together by common structural features (Colby & Kohlberg, 1987a). In

fact, the relations among ideas in the individual's thinking and the pattern of connections

within the individual's meaning provide the hermeneutic that guides the interpretive

process (Colby & Kohlberg).

Finally, constructivism defines individuals' active roles in constructing and

reconstructing reality. According to Colby and Kohlberg (1987a), by thinking about and

acting on the world individuals are creative because they are always inventing or

constructing new responses to new problems. Each response, however, is not simply a

creation of the moment, but is constrained and/or facilitated by the individual's current

developmental level and the individual's developmental history. In effect, each stage

provides a scaffold for the next. To reveal an individual's progressive coming into

consciousness as a result of action and reason, Colby and Kohlberg, like Piaget

(1932/1965), presented informants with moral dilemmas to stimulate their moral

reasoning.

Kohlberg's theory. Colby and Kohlberg (1987a) conceptualized a theoretical

framework based on deontic reasoning; that is, questions involving judgements of

rightness, duties, and rights. These researchers defined moral judgements as (a)

judgements of value, not of fact; (b) social judgements, involving people; and (c)

prescriptive or normative judgements, judgements of duty, or rights and responsibilities

rather than value judgements of liking or preference. Thus, moral judgements are

prescriptive because they command or oblige a prescribed course of action and are

derived from some rule or principle of action that the speaker takes as binding. Moral








judgements are different from social-conventional judgements (e.g., judgements about

dress, manners) because conventional judgements are relative to a particular situation, but

moral judgements are unalterable and universal.

To elicit the individual's constitutive knowledge about the bearings, uses, and

causes of moral problems, Colby and Kohlberg (1987a) presented informants with moral

dilemmas that represented conflicts in values. For example, in the "Heinz Dilemma"

informants' were presented with a conflict between upholding the law or stealing to save

someone's life. Informants' responses to these dilemmas were then evaluated by

identifying the relations among ideas in the individual's thinking and the pattern of

connections with the individual's meaning. Concepts such as the nature of duty, rules,

obligations, and the fairness of consequences were not perceived as concepts merely

learned or used independently of other concepts, but were envisioned as concepts bound

together by common structural features. Therefore, similar relations among ideas and a

generalized moral perspective define each level and stage of development. Moreover,

Colby and Kohlberg hypothesized that the developmental process they identified is

invariant and universal. As a result, Colby and Kohlberg identified six stages of moral

reasoning, grouped into three levels: (a) the preconventional level which includes stage 1:

punishment obedience and stage 2: personal reward moral orientations; (b) the

conventional level which includes stage 3: good boy/nice girl and stage 4: law and order

moral orientations; and (c) the principled or postconventional level which includes stage

5: social contract and stage 6: universal ethical principle moral orientations.

Preconventional level. The preconventional level describes the thinking of most

children under the age of 9, some adolescents, and many adolescent and adult criminal








offenders. Preconventional morality is characterized by reasoning that does not reliably

understand and uphold socially shared moral norms and expectations because the

reasoning process is constrained by na've moral realism and egocentrism. Stage 1

individuals define the right as following rules backed by punishment, obeying for its own

sake, and avoiding physical damage to persons and property. Thus, avoiding punishment

and obeying a superior authority supply the motivations for moral behavior. For example,

a stage 1 individual might reason that stealing is "wrong because you're not supposed to

break into stores. You'll get locked up" (Colby & Kohlberg, 1987b, p. 66).

In contrast, stage 2 is represented in a concrete, individualistic perspective. Stage

2 individuals follow rules only when it serves their immediate interests and they expect

others to do the same. For this reason, they have difficulty ordering or setting priorities

when presented with conflicting needs and interests, leading to relative morality and

indecisiveness in moral judgement. Colby and Kohlberg (1987b) present the following

moral judgement as an example of stage 2 reasoning.

Because if he steals the drug, he might get caught and then he'd have to
put it back, and he'd be in jail and then he wouldn't be able to raise the
money. (p. 70)

Conventional level. The conventional level includes most adolescents and adults

and refers to a morality guided by a socially shared systems of moral rules, roles, and

norms. During stage 3, the separate perspectives of individuals are coordinated into a

third person perspective, that of mutually trusting relationships among people embodied

in a set of shared moral norms. Stage 3 norms can be distinguished from Stage 2 rules

because they represent an integration of perspectives that is separate from strictly







39
individual interests. Colby and Kohlberg (1987b) present the following as an example of

a stage 3 moral judgement.

In reality he shouldn't steal because the man in the shop worked hard and
earned the money, so he doesn't have the right to steal from the shop. (p.
74)

At stage 4, the individual begins to take the perspective of a generalized member

of society and conceptualizes the social system as a consistent set of codes and

procedures that apply impartially to all members. The informally shared norms of stage 3

are systematized to maintain impartiality and consistency and an individual's interest is

considered legitimate only when it is consistent with the maintenance of the sociomoral

system as a whole. Moral judgement, therefore, is usually made in reference to

institutions or systems. The following example expresses a stage 4 moral judgement.

The law protects the druggist whether or not he is morally right, and in
order to maintain any kind of order in society the law should be followed
as much as possible. (Colby & Kohlberg, 1987b, p. 85)

Postconventional level. The postconventional level is reached after the age of20

and by only a small minority of individuals. The term postconventional morality refers to

a general acceptance of society's rules that is based on formulating and accepting the

general moral principles that underlie these rules. These general moral principles are

universal and unalterable and, at times, may come into conflict with society's rules. In

this case, the postconventional individual judges by principle rather than convention.

A stage 5 prior-to-society perspective is that of a rational moral individual who

reasons using the universal values and rights that anyone would choose to build into a

moral society. Laws and social systems are evaluated in terms of the degree to which they

preserve and protect fundamental human rights and values. Within this perspective, the








primary focus is either on rights or on social welfare. Rights cannot be abridged even

through freely chosen contracts and each person has an obligation to make moral choices

that uphold these rights, even in cases where they conflict with society's laws. Thus, the

perspective reflects a philosophy in which social institutions and laws are evaluated

according to their long-term consequences for the welfare of each person or group. For

example,

[A] society depends on a set of laws and regularities for its very existence
... [A] set of common agreements and understandings among people
which increases their ability to predict or control other people's behaviors.
(Colby & Kohlberg, 1987b, p. 100)

The sociomoral perspective of stage 6 is that "the moral" reflects a point of view

that all human beings should take toward one another as free and equal autonomous

persons. This requires equal consideration of the claims of each person affected by the

moral judgement and enjoins prescriptive role taking using procedures designed to ensure

fairness, impartiality, and reversibility in role taking. Kohlberg and his colleagues did not

find many examples of stage 6 reasoning in their studies and eventually dropped this

stage from the developmental sequence (Reed, 1997). Reed constructed the following

moral judgment as an example of stage 6.

A stranger has a right to life just the same as Heinz's wife. Heinz might
have more personal motive for stealing to save his wife's life, but there is
no moral difference between his wife and a stranger. If Heinz were to
consider the situation from the point of view of the stranger, he would see
that there is no moral difference. (p. 72)

Table 2 summarizes the major points of Kohlberg's theory.








Table 2.2

Kohlberg's Theory


Age Levels Stages


Up to 9 years Preconventional Stage 1: Punishment-Obedience Orientation

Stage 2: Personal Reward Orientation

9-20 years Conventional Stage 3: Good Boy/Nice Girl Orientation

Stage 4: Law and Order Orientation

20 + years Postconventional Stage 5: Social Contract Orientation

Stage 6: Universal Ethical Principle Orientation


Relevant Research

Piaget's (1932/1965) and Colby and Kohlberg's (1987a) stage theories have

received substantial attention from the research community and much of this attention has

been directed toward the development of morality in children. Stage theorists have

examined children's moral reasoning as an invariant sequence of developmental stages

that are qualitatively identifiable by their similar structures (e.g., Bussey, 1992; Jones &

Gall, 1995). Others, however, have argued that the developmental process does not define

the structural elements of children's moral judgements. On the contrary, social domain

theorists define the structure of children's moral judgements using three social domains,

the personal, the moral, or the social-conventional (e.g., Laupa & Turiel, 1995). Finally,

social constructivist challenge the universality of stage theories and employ cross-cultural

studies to describe the structural elements of children's moral judgement as a reflection

of cultural specific norms and values (e.g., Brown et al., 1995; Shweder et al., 1987).








Evidence for Stages

Researchers have given much attention to egocentrism in young children and its

dissipation during the early elementary years. According to Piaget (1932/1965),

egocentrism serves as a psychological constraint on young children's reasoning that

diminishes in influence as the child matures, signaling the advent of higher forms of

reasoning. Because of the limits of egocentrism, young children tend to focus on the most

salient quality of the event, the observable consequences--not the actor's intent, when

making moral judgments. Thus, young children judge any event that evokes an adverse

consequence as egregious, regardless of the actor's intention and state a preference for

expiatory punishment over retributive justice in response to moral transgressions. By age

6 or 7, however, egocentrism begins to give way to perspective taking, allowing children

to consider others' motivations. At the same time, children's preferences shift toward

retributive justice instead of expiatory punishment when responding to moral

transgressions.

In typical peers. Recent research employing a stage theory perspective provide

substantial evidence that egocentrism may, indeed, be a defining feature of young

children's thinking that appears to fade in influence as children mature (e.g., Bussey,

1992; Jones & Gall, 1995; Zelazo, Helwiz, & Lau, 1996). To evaluate the level of

egocentrism present in children's thinking at various ages, researchers pose moral

dilemmas and examine children's understanding of others' intentions. Findings suggest

that even when young children can state an actor's intent and can define the differences

between lies and mistakes, they will judge an act as wrong if the consequences are

aversive (Bussey, 1992) or if the actor expends a high degree of effort in committing the








misdeed (Jones & Gall 1995). Furthermore, findings indicate that as children mature,

their ability to understand and use an actor's intent when making moral judgements

seems to increase as they mature (e.g., Barchard & Atkins, 1991; Dixon & Moore, 1990;

Laupa & Turiel, 1995; Zelazo et al., 1996). On the other hand, Barchard and Atkins

found that children continue to prefer expiatory punishment as a response to moral

transgressions, regardless of their level of maturation; therefore, researchers have not

replicated Piaget's developmental effect in children's preference for retributive justice

over expiatory punishment.

Researchers have generally relied on the moral dilemma interview to prompt

children's moral judgements, frequently varying the content of the prompt to provide

richness and depth to the results (e.g. Bussey, 1992; Nucci & Weber, 1995; Turiel,

Killen, & Helwig, 1987). Moreover, researchers generally include children of varying

ages to document maturational changes (e.g., Barchard & Atkins, 1991; Jones & Gall,

1995). For example, Piaget (1965) presented child focused real life moral dilemmas to 5

to 13-year olds, Johnston (1988) presented moral dilemmas in fables to 11 to 15- year

olds, and Colby and Kohlberg (1987a) presented hypothetical moral dilemmas to

individuals across all ages.

Although the use of moral dilemmas has provided valuable information about

how children's moral judgements may vary according to dilemma content, this data

gathering procedure requires specialized training in a labor intensive method of coding

and analysis (see Colby & Kohlberg, 1987a for scoring instructions). In response to this

problem, Rest (1979) developed and validated the Defining Issues Test (DIT), a

standardized, multiple-choice instrument based on Kohlberg's hypothetical dilemmas and







44
scoring criteria. The DIT, however, is written on a 7" grade reading level, limiting its use

to older adolescents and adults. Recently, Narvaez et al. (1999) developed the Moral

Theme Inventory (MTI) specifically for use with children. The purpose of the MTI is to

assess children's ability to understand themes of cooperation in moral stories as the

author intended or as children distort them to match their developmental level of moral

reasoning. Thus, the development of the MTI presents new research opportunities

because it provides child development researchers with a standardized complement to the

moral dilemma interview.

In students with EBD. Hardman & Smith (2001) piloted a study of moral

reasoning in 3rd, 4t, and 5h grade students with EBD (n = 21) and typical peers (n = 21).

Study participants completed the MTI (Narvaez et al., 1999), which measures children's

ability to comprehend themes of cooperation in moral stories. These results were

submitted to a regression analysis accounting for the following independent variables:

SES, gender, ethnicity, grade, reading comprehension, and atypical/typical behavior. This

research design allowed the authors to investigate the relationships among several

cultural variables (i.e., ethnicity, gender, and SES) and the development of moral

judgement. Moreover, by including students with EBD and typical peers, these

researchers sought to further illuminate current understanding of the relationship between

behavior and reasoning. Hardman and Smith also examined the moral reasoning

processes of three students with EBD employing case study research to evaluate their

responses to moral dilemmas. The purpose of the case study research was to elaborate

students' reasoning processes beyond their objectively evaluated abilities to detect moral

themes of cooperation as measured by the MTI (Narvaez et al., 1999).








As a result, Hardman and Smith (2001) found a significant relationships EBD,

SES, and moral theme comprehension, but when accounting for all independent variables

(i.e., SES, gender, ethnicity, grade, reading comprehension, and atypical/typical

behavior) results showed that the effect for EBD was due to the overrepresentation of

children with low SES in the sample of students with EBD. Moreover, case study results

indicated that three students with EBD, who were also low SES, voiced self-focused,

preconventional moral judgements and did not typically voice other-focused moral

themes of cooperation. Instead of cooperation, avoiding punishment, egoistic

consequences, and seeking personal reward appeared to motivate their moral decision-

making.

Hardman and Smith (2001) concluded that the dominance of egocentrism in the

responses of the students with EBD bears preliminary theoretically significant

implications and warrants further documentation. They also caution that because SES and

EBD were confounded, they could not evaluate the relationship between EBD and moral

theme comprehension independent of the significant effect found for SES. These findings

present additional evidence, however, that SES may provide a significant influence on the

development of children's moral reasoning processes. While Colby, Kohlberg, and their

colleagues (1987) found SES to be significantly related to the development of moral

judgment in their longitudinal study of U.S. males, Hardman and Smith are the only

researchers to pursue the nature of this relationship further. In fact, researchers

investigating children's moral reasoning usually conduct their research using purposively

selected samples, thereby seriously limiting the generalizability of the results to the

diverse population that defines public schools.








On the other hand, Hardman and Smith (2001) failed to address the emotional

concomitants of EBD in their analyses. Externalizing disorders characterized by

aggression, disruption, and antisocial behavior are the most prevalent types of EBD

(Kauffman, 1995; Kauffman et al., 1987) and anger is assumed to be the emotional

motivator of these maladaptive behaviors (e.g., Kauffman, 1995; Walker et al., 1995). To

effectively evaluate the relationship between EBD and moral reasoning, the emotional

concomitants of the maladaptive behavior (i.e., anger) should be included in the analyses

(Enright, personal communication).

Evidence for Social Domains

Social domain theorists (e.g., Laupa & Turiel, 1995; Turiel, Hildebrandt, &

Waintyb, 1991; Turiel et al., 1987) hypothesized that moral reasoning is defined by three

distinct conceptual domains: (a) the moral, which pertains mainly to concerns with

welfare, justice, and rights; (b) the social-conventional, which pertains to an

understanding of social uniformities and regularities as a necessary part of the smooth

and efficient functioning of social organizations; and (c) the personal, which refers to

those individual prerogatives and entitlements considered to be exempt from social

regulation (Laupa & Turiel). Thus, these researchers challenge the structural integrity of

the Piagetian/Kohlbergian framework and provide evidence that individuals, including

young children, match the content of their reasoning to the domain (i.e., moral, social-

conventional, or personal) in which the transgression applies (e.g., Smetana, Schlageman,

& Adams, 1993; Nucci & Webber, 1995; Tisak, 1996).

In typical peers. Smetana et al. (1993) examined children's responses to

hypothetical and actual conventional and moral transgressions and found that regardless








of transgression type, young children judge moral transgressions to be more serious,

punishable, generally wrong, and independent of rules/authority than social-conventional

transgressions. Similarly, others have found that 3 to 5-year olds distinguish among the

moral, social-conventional, and personal domains (Crane & Tisak, 1995a; 1995b; Nucci

& Weber, 1995) and that mothers (Nucci & Weber) and teachers (Tisak, 1996) respond

differently to children's transgressions according to domain. Therefore, researchers

conclude that the responses of teachers, parents, and peers to different types of

transgressions may significantly influence young children's understanding of events

(Tisak). In support of structural theories, however, social domain researchers have found

that preschoolers and first graders are more likely than third graders to view mixed-

domain acts as only conventional (Smetana et al.), providing evidence that as children

mature, they are increasingly able to recognize the moral components of mixed-domain

events and to incorporate both moral and conventional considerations in their reasoning

(Crane & Tisak, 1995a; 1995b).

In violent children. Astor (1994) examined moral reasoning in 108 nonclinically

referred aggressive and nonaggressive children using vignettes about family and peer

violence. Astor found that violent and nonviolent children condemned unprovoked

violence in all family and peer social contexts. Aggressive children, however, judged

provocation as a transgression that results in serious psychological harm; therefore, it

must be answered with retribution (e.g., hitting back). In contrast, nonagressive children

did not think psychological harm warranted the serious harm that might result with

retribution. Astor concluded that provocation should be examined as a social domain and

suggested further explorations of violent children's reasoning in this domain. Although








Astor's subjects included 54 violent children chosen on the basis of the frequency and

salience of their violent acts during a two-month period, Astor did not include any

measures of anger in his analyses.

In students with EBD. In a follow up study, Astor and Behre (1997) examined the

social context of provocation. Study participants included 17 boys who were enrolled in a

highly restrictive special education day treatment program designed for students with

EBD, their care providers, and a control group that consisted of a matched sample of 17

children and their caretakers. Using Astor's vignettes (1994), Astor and Behre explored

the influence of family and peers on the approval of violence in the family and school

within three types of interpersonal relationships: parent/child; parent/parent; and peer.

Even though Astor and Behre were investigating violence at home and with peers, no

measures of anger were included in the analyses. Using Astor's interpretation of

provocation as a social domain, Astor and Behre concluded that violent children tended

to use moral-only reasoning and nonviolent children more frequently used nonmoral-only

reasoning.

Astor and Behre's (1997) conclusions should be viewed with caution because

they are predicated on the assumption that provocation constitutes a moral domain.

Based on this assumption, Astor and Behre coded nonviolent children's responses that

hitting is wrong under all circumstances as non-moral reasoning and violent children's

responses that hitting is appropriate when one is provoked were coded as moral.

According to Laupa & Turiel's (1995) definition of social domains, provocation that

causes psychological harm falls in the personal domain, which includes individual







49

prerogatives and entitlements considered exempt from social regulation. Therefore, Astor

and Astor and Behre's findings and conclusions may be misleading.

Social Constructivist Challenges

Universal moral principles. Social constructivists also challenge the

conceptualization of moral reasoning as a cross cultural/universal phenomenon that

develops in an invariant sequence of stages. Instead, constructivists posit that moral

development is entirely a social phenomenon, directed by cultural norms and values.

Social constructivists charge that stage theories are singularly focused on justice

reasoning because justice reasoning preferences the individualistically focused values and

norms of Western middle class males, and relegates communitarian ideals, such as care

orientation, to lower developmental stages (e.g., Brown et al., 1995; Gilligan, 1982;

Gilligan, Ward, & Taylor, 1988; Shweder et al., 1987).

In an effort to examine the universality of justice reasoning, Edwards (1987)

conducted a cross cultural ethnographic study of moral development in young Oyugis

(African) children and American preschool children in a Vassar College Nursery School.

Edwards observed that young Oyugis children conform to cultural conventions according

to the salience of the social benefits connected to cooperation with the rules. Thus,

Edwards concluded that the reasons for cooperation with many rules is often

incomprehensible to young children, stating that "Oyugis parents do not need to preach to

children about the rationales underlying any of these rules because the routines of

children's lives amply contain the evidence" (p. 140). In comparison, she found that the

values and norms taught in the Vassar College Nursery School differed from those taught

to Oyugis children, but the method of inculcation was the same. Edwards deduced that








the learning environment, not the child, subdivides morality into separate domains, such

as moral and conventional. She summed up the interplay of reason, culture, and

development saying, "[W]ith increasing age and experience, children apply progressively

more complex and mobile logical schemas to cultural distinctions and categories; they

transform what they are told and what they experience into their own self-organized

realities" (p. 149).

Gilligan (1982) challenged the universality of justice reasoning by examining

gender related cultural differences in moral reasoning. Gilligan and her colleagues (e.g.,

Brown et al., 1995; Gilligan et aL, 1988) present evidence that moral reasoning in males

and females may be different in orientation. These researchers found that males are more

likely to interpret moral dilemmas using a justice/rights orientation and that females are

more likely to prefer a care orientation. Influenced by these findings, Cassidy, Chu, and

Dahlsgaard (1997) investigated children's use of care and justice orientations. Children

were read four moral dilemmas designed to prompt either a care perspective (one

dilemma), a justice perspective (one dilemma), or either justice or care (two dilemmas).

These authors found that regardless of gender, children used justice and care orientations

at equal rates and would accept both orientations to the same dilemma. Barchard and

Atkins (1991) and Smetana (1993) examined the effects of gender in their research;

however, they found no significant differences in children's reasoning based on gender.

A Cognitive-Developmental Approach to Moral Education

Kohlberg envisioned the Just Community as a moral education intervention

intended to provide students with social experiences that would facilitate moral

development by deliberately exposing and using the hidden curriculum to create a liberal,








democratic community (Power et al. 1989). According to Power and his colleagues,

Kohlberg believed that the most fundamental values of our society are the values of

justice and fairness. Therefore, the purpose of moral education is to teach justice by

allowing students to learn the basic valuing processes that underlie their capacity for

moral judgement. As a result, schools transmit the consensual values of a democratic

society and not the values of specific cultures or groups.

Power et al. (1989) stated that public school educators have the responsibility to

teach values, though they do not have the right to impose their own, or any set of values

on their students. These authors agreed with Durkeim's (1961) portrayal of the classroom

as a small society with its own rules, obligations, and sense of social cohesion; therefore,

the procedures of classroom discipline are naturally invested with moral meaning.

Schools cannot eliminate authority from classrooms, but educators can, instead, use their

authority to develop a community that enables children's socialization into a sense of

attachment and obligation to society and its institutions.

Although the social curriculum is a vital component of the school curriculum, this

curricular component has come to been known as the hidden curriculum (Power, et al,

1989). According to Power and his co-authors, the rules that define the hidden curriculum

shape the moral atmosphere, or the context for moral learning, in the school. To

implement a Just Community based on the principles of justice and fairness, these authors

advise educators to employ the hidden curriculum when dealing with everyday rules of

behavior. Therefore, the aim of the Just Community is to expose the hidden curriculum

and transform it into a curriculum of justice, using a different form of school governance







52
where the rights of students and teachers are taken seriously and the values of justice and

fairness take primacy over the value of adult authority.

Kohlberg and his colleagues implemented the Just Community intervention in two

Boston high schools. Results indicated a modest developmental change in student's moral

judgement stage scores. The most dramatic changes, however, were measured in the

judgements of students who reasoned at the preconventional level when the intervention

began. As a result their experiences in the Just Community, students moved from a stage

2: personal reward orientation to stage 3 and stage 4: conventional level of reasoning.

Kohlberg and his colleagues concluded,

The managerial approaches to school administration, with their emphasis
on techno-bureaucratic strategies of problems solving, have undermined
the ideas of moral education. They must give way to a more self-
consciously democratic and communal approach. We are in need of an
educational reform far more significant than any we have know, a reform
that draws its inspiration not from technological advance but from the
moral ideals ofjustice, democracy, and community on which this nation
was founded. (p. 306)

Unfortunately, there have been no efforts to evaluate the effectiveness of the Just

Community approach in elementary schools, where early intervention in children's moral

development might be the most effective.

Review of Seven Independent Variables

The current research agenda appears narrowly focused on the influence of the

developmental process and the universality of this process across cultures and gender,

leaving many important theoretical issues unresolved. For example, cognitive-

developmental researchers have not examined the influences of poverty and ethnicity

within our Western culture (Hardman & Smith, 2001), the relationship between behavior

and reasoning (Hardman & Smith), or the relationship between emotion and moral








judgment. While the study of children's moral reasoning has enlisted a variety of

different perspectives (i.e., stage, social domain, and social constructivist theories),

proponents of these perspectives seem to have missed meaningful opportunities to clarify

the relationship between moral reasoning and behavior, emotion, and cultural issues

specific to America's schools.

Socioeconomic Status (SES)

Colby, Kohlberg, Gibbs, and Lieberman (1987) presented the results of a 20-year

longitudinal study of moral reasoning in 84 U.S. males seeking support for Kohlberg's

hypothesis that moral judgement develops in a invariant sequence of six stages. They

explored the effects of SES, IQ, and achievement assuming that these variables might

provide an important influence on stage development. Results showed that stage 3 was

present at age 10 for middle-class subjects, but not until age 13 for working-class

subjects. Stage 4 appeared at age 16 in middle-class individuals, but not until age 20 in

the lower SES group. They also found that moral judgement stage development was

related to achievement; however, partial correlations showed that differences in rate of

stage development was related to achievement as a reflection of SES differences. As a

result, Colby, Kohlberg et al. surmised that SES influences the development of moral

judgement because students with low SES do not experience the same opportunities to

participate in society as do their peers from middle income families.

Although Colby, Kohlberg, Gibbs, and Lieberman (1987) presented credible

evidence that SES might provide an important influence on the development of children's

moral judgement, SES has been largely ignored in subsequent studies (cf. Bussy, 1992;

Jones & Gall, 1995; Smetana et al., 1993; Nucci & Webber, 1995; Tisak, 1996; Zelazo et








al., 1996). In Hardman and Smith's (2001) study of moral reasoning in students with

EBD, they examined the effects of SES and found it to be significantly related to

children's moral judgement stage development. Moreover, these authors were unable to

interpret the effects of EBD on moral reasoning because EBD and SES were confounded,

indicating that students from low SES families are being identified as emotionally

disturbed in greater numbers than are their middle income peers.

Researchers know that children from poor families appear to be at a cognitive

disadvantage when they enter school and that they are also at risk for social or behavior

problems (Bryant & Maxwell, 1997). Palakow (1998) describes poverty as an endemic

feature of the landscape of America with the largest constituency of poor Americans

being young children. In her ethnographic studies of public school classrooms, Palakow

found

widespread discrimination and prejudice on the part of teachers and school
personnel toward destitute children and their families-where classroom
environments for poor children, particularly difficult and angry children,
become landscapes of condemnation that reveal shared experiences of
exclusion, humiliation, and indifference. (p. 15)

In the case of moral reasoning, Colby, Kohlberg, Gibbs, and Lieberman (1987) and

Hardman Smith (2001) provide evidence that SES provides a significant influence.

Therefore, SES appears to represent an important cultural influence on children's school

experiences and it also seems to represent a significant influence on the development of

children's moral judgement.

Grade

The maturing of children's moral judgement as a function of age is the focal point

of the research literature. Colby, Kohlberg, Gibbs, and Lieberman (1987) found a clear








relationship between age and moral judgement stage development in their longitudinal

study of American males. Using a multiple regression analysis, these authors found that

age accounted for 60% of the variance in stage scores. Moreover, they found that stage

scores at age 10 did not predict adulthood scores, but stage scores at age 13 were much

more highly correlated with later scores. As a result, Colby and Kohlberg (1987a)

conjectured that the most fruitful time period for moral education may occur before the

age of 13. Nevertheless, Kohlberg chose high schools, not elementary schools, as the

context in which to implement his Just Community intervention.

The effect of maturation on the structure of children's moral judgement has been

widely research and documented (e.g., Barchard & Atkins, 1991; Crane & Tisak, 1995a;

1995b; Dixon & Moore, 1990); however, Hardman and Smith (2001) are the only

researchers to employed random sampling. Interestingly, these authors did not find a

developmental effect when measuring respondents' ability to detect themes of

cooperation in moral stories. On the other hand, their study included a small sample

(N=42), a disproportionate number of students with low SES, and a limited age range

(i.e., 3rd, 4th, and 5t grades). For these reasons, Hardman and Smith advised caution when

interpreting these results and recommend continued research on the effects of maturation

on children's moral judgement in samples that represent the diverse nature of public

school populations.

Gender

According to Reed (1997), Kohlberg began his theory building efforts as a

graduate student at the University of Chicago by studying the development of moral

judgment in a sample of American boys. He elected to include only males in his sample








on the advice of his advisor, who suggested that he limit the number of variables to be

include in his analyses (Reed, 1997). This decision led to credible assaults from feminist

researchers on the universality of his theory across gender (e.g., Gilligan, 1982).

Carol Gillligan (1982) was the first to raise the issue of sex bias in Kohlberg's

theory, charging that his justice focused account of the development of moral judgement

was a by-product of sex-bias in sampling procedures. She charged that because the roles

of American males are more likely to include participation in democratic social

institutions, Kohlberg's theory preferences the development of a justice orientation in his

hierarchical scheme of moral stage development. In contrast, Gilligan and her colleagues

(1982, 1988) found that females were more likely to voice a communitarian or care

orientation in their moral judgments, placing value on the importance of maintaining

intimate social relationships. Using Kohlberg's scheme, this kind of reasoning is

relegated to a stage 3: good boy/nice girl orientation; whereas, justice reasoning is

assigned a higher stage 4: law and order orientation (Reed, 1997). As a result, females

were more likely to be evaluated as reasoning at lower stages. Gilligan and her colleagues

concluded that the cultural role of females in American society as maintainers of intimate

social relationships might lead to the development of a different moral orientation, but

not necessarily a less sophisticated one.

Gilligan's (1982) feminist critique of Kohlberg's theory has resulted in the

frequent inclusion of gender in researchers' analyses of children's moral judgments. For

example, Barchard and Atkins (1991) and Smetana et al. (1993) included gender in their

studies of children's moral judgements about naughtiness and punishment and found no

significant effects for this variable. Cassidy et al. (1997) focused on preschool children's







57
ability to use care and/or justice orientations in their moral judgements and reported that

girls and boys used justice and care orientations with equal frequency and could even

accept both orientations to the same dilemma. On the other hand, Johnson (1988)

examined gender differences in adolescents' responses to moral dilemmas presented in

fables and found that girls choose both justice and care orientations more frequently than

boys, who tended as a group to use the rights orientation more exclusively.

Anger

The study of anxiety and depression have dominated researcher interest over the

past 25 years while the study of anger has been relatively ignored (Kassinove &

Sukhodolsky, 1995). Kassinove and Sukhodolsky define anger as

[A] negative phenomenological (or internal) feeling state associated with
specific cognitive and perceptual distortions and deficiencies (e.g.,
misappraisals, errors, and attributions of blame, injustice, preventability,
and/or intentionality), subjective labeling, physiological changes, and
action tendencies to engage in socially constructed and reinforced
organized behavioral scripts. Thus anger refers to a label given to a
constellation of specific uncomfortable subjective experiences and
associated cognitions (i.e., thoughts, beliefs, images, etc.) and may
develop as a result of people's subjective interpretation of an event. (p. 7)

Researchers have learned a great deal about the relationships among depression

and/or anxiety and cognition, finding that emotion and cognition are inextricably related

(e.g., Beck, 1976; Ellis, 1973). On the other hand, little is known about the relationship

between anger and cognition. In fact, Eckhardt and Deffenbacker (1995) state that the

study of anger has been neglected. According to recent findings, angry responses to any

given situation do not appear to be dependent upon actual events, but may, instead, be

dependent on the individuals' subjective interpretations of events (e.g., Dodge & Coie

1987; Levine, 1995), thereby signally the passing of moral judgement (Othof et al.,








1989). Thus, researchers know that the relationship between anger and cognition is

reciprocal (Kassinove, 1995), but the direction of the relationship is not clear. In other

words, feelings of anger may precede judgement (Deffenbacher et al., 1996, Dodge &

Coie; Walker et. al, 1995) or judgment may precede anger (e.g., Downey et al., 1998;

Enright & Fitzgibbons; Fabes & Eisenberg, 1992; Murphy & Eisenberg, 1996). The

former position emphasizes anger as a trait that influences cognition, but the latter

presents trait anger as a justified response to perception of injustice and rejection.

According to Kassinove and Sukhodolsky (1995), the typical instigation of anger

begins with a value judgment. Therefore, anger represents an attribution of blame that is

dependent upon individuals' perceptions of their social relationships. The emerging

literature on children's anger presents evidence that there may be a significant

relationships among social competence, rejection, anger, and moral judgement. For

example, Murphy and Eisenberg (1996) found that social competency predicts anger

intensity and Downey, and her colleagues (1998) found that poor and minority children

who angrily expect social rejection behavior more aggressively, experience increased

interpersonal difficulties, and decline in academic learning over time. In addition, Fabes

and Eisenberg (1992) found that children who were low in social status and social

competence seemed to invite aggressive conflict. Nevertheless, research on children's

anger seems to be in an incipient stage (e.g., Boekaerts, 1993; Fabes & Eisenberg;

Murphy & Eisenberg), leaving many important questions about the relationship between

anger and moral judgment unexplored and unanswered.








Emotional/Behavioral Disorders (EBD)

An emotional/behavioral disorder is a complex, multifaceted disability that is

defined by the presence of disordered behavior, emotion, and cognition (Kauffman,

1995). According to Kauffman the maladaptive behavior of students with EBD can be

represented in externalizing disorders characterized by aggression, disruption, and

antisocial behavior or it may be represented in internalizing disorders such as withdrawal

and anxiety. Externalizing disorders seem to represent the most common form of EBD

(Kauffman et al, 1987; Walker, 1995); however, this may be the case because

externalizing disorders also represent the most commonly cited reason for referral to

special education (Kauffman & Wong, 1991). Because of their maladaptive behavior,

students with EBD often face rejection and rebuke from peers and teachers, resulting in

their removal from the general education classroom and likely placement in special

classes, day schools, and alternative schools (Kauffman).

While the federal definition of EBD refers specifically to unhappiness and

depression as the underlying emotional concomitants of EBD, researchers suggest that

anger may also contribute to the development of EBD, especially when maladaptive

behavior is represented in aggression, disruption, and antisocial behavior (Walker et. al.,

1995). Walker and his co-authors proposed (1995) a seven phase acting-out cycle that

includes the following phases: calm; trigger; agitation; acceleration; peak; de-escalation;

and recovery. These authors also identified numerous school-based (e.g., conflicts,

changes in routine, provocations, pressure, and ineffective problem solving) and

nonschool-based "triggers" (dysfunctional homes, health problems, nutrition, substance

abuse, and gangs) that instigate aggressive behavior. Kassinove and Sukhodolsky (1995)








point out, however, that aggression is not always instigated by anger and suggest that the

intent of aggression can be either instrumental or emotional. Instrumental aggression is

carried out for an extrinsic purpose, while emotional or hostile aggression derives from

an urge to attack someone when one feels angry, even though one may not profit from

aggression. One may even be willing to pay a price for it. According to Kassinove and

Sukhodolsky, this type of aggression frequently occurs in response to a perceived

injustice.

An emotional/behavioral disorder is also defined by cognitive limitations other

than emotional disorders. According to the federal definition, (IDEA, 1997) students with

EBD exhibit "an inability to learn that cannot be explained by intellectual, sensory, or

health factors." Thus, the definition specifically identifies learning deficiencies as an

important element of the disability in addition to emotional and behavioral disorders.

Researchers generally point to a population tendency toward a low average IQ to provide

evidence that learning deficiencies may accompany EBD (Kauffman, 1995). Therefore,

the emotional/behavioral/cognitive complexity of EBD seems to present child

development researchers with a unique opportunity to explore the relationships among

behavior, emotion, and cognition and children's moral judgement. Yet, students with

EBD have not been included in studies examining children's moral reasoning processes

(cf. Brown et al., 1995; Bussey, 1992; Colby & Kohlberg, 1987a; Jones & Gall, 1995;

Laupa & Turiel, 1995; Piaget, 1932/1965; Shweder et al. 1987). Therefore, the study of

students' thinking processes has been limited to identifying cognitive distortions/deficits

that lead to aggression, disruption, and antisocial behavior and the reinforcement








contingencies that maintain cognition, emotion, and behavior (e.g., Kaplan & Carter,

1995).

Ethnicity

According to Colby, Kohlberg, and Nisan (1987), both the stages and the

sequence of development for moral judgement are universal, or culturally invariant.

[M]oral judgement represents underlying thought organization rather than
specific responses; its development results from a process of interaction
between organismic structuring tendencies and universal features of social
experience, rather than from "transmission" through genetics or direct
shaping; and the direction of development is toward greater equalibration
in the organism-environmental interaction and reciprocity between the self
and other. (p. 119)

To provide support for this proposition, Colby and Kohlberg (1987) reported the results

of two longitudinal cross-cultural studies, a study of moral judgment in Turkish males

and a study of Israeli kibbutz adolescents.

In the study of Turkish males, Colby, Kohlberg, and Nisan (1987) found that the

moral judgements of Turkish study participants fit the stage structures and sequence

proposed by Kohlberg's theory. The sequence observed, however, was limited to the first

four stages in Kohlberg's scheme, prompting the authors to suggest a continued

examination of the universality of stages 5 and 6 across cultures. These authors suggested

that it is possible that other cultures hold moral principles that are different from

American males, thus their moral reasoning may not fit the higher stage structures

described by Kohlberg.

Colby, Kohlberg, Snarey, and Reimer (1987) reported the findings of Kohlberg's

longitudinal study of 92 kibbutz adolescents. They found support for Kohlberg's theory

observing that the development of moral judgement in kibbutzniks followed the patterns








reported for samples in the U.S. and other cultures. Moreover, they observed that the

maturation of moral judgement in this population seemed to be accelerated when

compared to the other cultures studied, including the U.S. As a result, the kibbutz culture

inspired Kohlberg to model his idea for a moral education intervention, The Just

Community, after the democratically organized community exemplified in the Israeli

kibbutz (Power et al., 1989).

Others have also conducted cross-cultural studies frequently offering findings that

challenge Kohlberg's claim of universality (e.g., Edwards, 1987). Few, however, have

studied the relationship between moral reasoning and identification with an ethnic

minority within American society. For example, Narvaez and her co-authors (1999)

examined the relationship between ethnicity and moral theme comprehension by

evaluating participants' ratings of"ingroup" themes found in the moral message choices

of the MTI (e.g., Helping strangers instead of your friends can cause trouble. If you think

of others first instead of your family, your family may suffer.). Because there were too

few members for each minority group, they collapsed all ethnic categories into a "non-

white" category and compared it with "white". Even though the participant sample

included university students, only the 3rdand 5th grade children's scores were included in

this analysis because there was only one "non-white" university participant. The test was

significant meaning that "non-white" children rated the ingroupp" themes higher (i.e. "the

same" or "very much the same") than "white" children rated these same themes.

Hardman and Smith (2001) also used the MTI (Narvaez et al. 1999) to assess participants

ability to comprehend moral themes of cooperation; however, they did not analyze







63
respondents' ratings of"ingroup" themes. They did include ethnicity in their correlation

and regression analyses and found no significant effects for ethnicity.

Reading Comprehension

In 1993, William Bennett proclaimed that reading moral stories ("Book of

Virtues") to children would improve their moral literacy. Narvaez and her co-authors

(1999) challenged Bennett's prescription stating that little research had been done about

the influence of reading moral stories on the development of moral literacy. They

proposed that moral theme comprehension is an ability that is related to but separate from

reading comprehension. These authors developed the MTI as a way to assess children's

ability to comprehend moral themes in stories. Using the MTI children were asked

whether or not they understood the moral lessons from four stories and they were asked

to respond to reading comprehension questions for each story.

As a result, Narvaez et al. (1999) reported that developmental differences in moral

theme comprehension were evident even after reading comprehension was taken into

account. These authors found that the study participants assimilated the moral messages

of the stories to match their level of moral thinking or moral schema development.

Furthermore, results indicated that even when standardized reading comprehension and

vocabulary scores were used as covariates along with the reading comprehension items,

there were still significant differences between the 3rd and 5th grade students on moral

theme comprehension task. They argued that while these results are preliminary, they

support the proposition that moral text comprehension requires something beyond

reading comprehension. Moreover, they concluded that it is possible to use an objective

method to peer into the moral mind of the child and recommended continued research








examining such things as the differences between moral and non-moral theme

comprehension, the relation between moral theme comprehension and scores on moral

judgment interviews, and whether certain kinds of moral themes are understood

developmentally sooner than other moral themes.

Summary

The cognitive developmental perspective of the development of morality in

children is grounded in the assumption that children create knowledge as a result of a bi-

directional relationship between thinking and experience. Reasoning is conceptualized as

an intentional endeavor to discover specific connections between the things children do

and the consequences that ensue (Dewey, 1916/1944). The development ofjudgement,

however, is couched in a developmental process that may constrain or facilitate higher

stages of thinking (Piaget, 1932/1965).

In 1932, Piaget conducted a study of children's moral reasoning, employing a

cognitive-developmental approach to describe and explain the phenomenon. Assuming

that children construct their moral realities as a result of their mental and physical

actions, Piaget examined how nature and experience influence children's understanding

of morality. As a result, Piaget authored the first stage theory describing the development

of morality as a cognitively structured, maturational process. Piaget identified four

developmental stages that define children's practice of rules. These four stages are

sensorimotor, egocentric stage, cooperation stage, codification of rules. Through the

practice of rules (experience), children also develop a consciousness of rules (thinking),

in three stages: nonmoral; heteronomous morality; autonomous morality.








In 1987, Colby and Kohlberg authored what might be one of the finest examples

of a stage theory (Higgins, 1995). Their theoretical framework was based on deontic

reasoning; that is, questions involving judgements of rightness, duties, and rights. These

authors defined moral judgements as (a) judgements of value, not of fact; (b) social

judgements, involving people; and (c) prescriptive or normative judgements, judgements

of duty, or rights and responsibilities rather than value judgements of liking or

preference. Moreover, moral judgements are prescriptive because they command or

oblige a prescribed course of action and are derived from some rule or principle of action

that the speaker takes as binding. For example, moral judgements are different from

social-conventional judgements (i.e., judgements about dress, manners) because

conventional judgements are relative to a particular situation; whereas, moral judgements

are unalterable and universal. Colby and Kohlberg (1987a) identified six stages of moral

reasoning, grouped into three levels: the preconventional level which includes

punishment obedience and personal reward orientations; the conventional level which

includes good boy/nice girl and law and order orientations; and the principled or

postconventional level which includes social contract and universal ethical principle

orientations.

The structural theories of moral development posed by Piaget (1932/1965) and

Colby and Kohlberg (1987a) have received substantial attention from child development

researchers investigating the development of morality in children. In this regard,

researchers have focused their efforts primarily on the development of moral reasoning as

an invariant sequence of developmental stages (e.g., Bussey, 1992; Jones & Gall, 1995),

as a function of social domains (e.g., Laupa & Turiel, 1995), and as a reflection of







66
cultural norms and values (e.g., Brown et al., 1995; Shweder et al., 1987). Nevertheless,

child development researchers have missed opportunities to examine the relationship

between moral reasoning and cultural issues relevant to our Western culture (i.e., poverty,

ethnicity), behavior, and emotion.

Since Piaget's (1932/1965) seminal study, researchers have generally relied on

the moral dilemma interview to elicit children's moral judgements (e.g. Bussey, 1992;

Nucci & Weber, 1995; Turiel et al., 1987). Recently, however, Narvaez et al. (1999)

developed the Moral Theme Inventory (MTI) which presents new opportunities for

research on the development of morality in children. The MTI was developed for use

with children and provides child development researchers with a standardized

complement to the moral dilemma interview.

Only three studies have examined moral reasoning in children with EBD, with

one of those being a pilot of the present study. None have examined the relationship

between moral reasoning and emotion. In this study, Hardman and Smith (2001) reported

that study participants with EBD demonstrated by their responses to the MTI (Narvaez et

al., 1999) and moral dilemmas presented by a researcher that they did not understand

moral themes of cooperation and that cooperation as a moral motivator was not a part of

their experience. Instead, the participants with EBD perceived morality as dictated by

external forces such as law and adult authority and not obligated from within. Moreover,

their moral reasoning appeared to be motivated by the avoidance of punishment and

trouble and seeking personal reward. Hardman and Smith concluded that the dominance

ofegocentrism in the responses of the students with EBD bears preliminary theoretically

significant implications and warrants further documentation. These researchers also








caution that SES and EBD were confounded; therefore, the relationship between EBD

and moral reasoning could not be evaluated independent of the significant effect found

for the variable SES.

The other two studies were both couched in a social domain perspective. Astor

(1994) and Astor and Behre (1997) found that violent, antisocial children view

provocation as causing such severe psychological harm that inflicting physical harm to

the perpetrator (hitting) is a morally appropriate response. These authors used this

interpretation of violent and non violent children's moral judgements to reach the

conclusion that violent, antisocial children reason more frequently in the moral domain

when compared with typical peers.

Of the seven independent variables targeted in the present study, only grade (or

age) and gender have received attention in the research literature. The influence of SES

on moral reasoning has not been examined since Colby, Kohlberg, and their colleagues

(1987) first identified this variable as one that is significantly related the development of

moral orientation. The relationship between EBD and moral reasoning has not been

addressed until recently (Astor & Behre, 1997, Hardman & Smith, 2001) and ethnicity,

too, has received little attention from researchers. Narvaez et al. (1999) have been the

only researchers to examine the dependence of moral theme comprehension on reading

comprehension and none of the moral development researchers has examined the

relationship between anger and moral reasoning, even though there is ample evidence to

suggest that emotion and cognition are related.













CHAPTER III
METHOD

In Chapter III, I present the research methods and procedures of the study. For the

purposes of presentation, the chapter has been divided into five sections. Because

researchers have not targeted moral reasoning in elementary students with EBD using

stage theory, I found it necessary to begin my inquiry with a pilot study. The methods,

procedures, and results of the pilot is followed by an introduction to the present study,

which includes a statement of research hypothesis and a description of the research

design. The final three sections provide a description of the setting and participants,

instrumentation, and research procedures.

Introduction to the Pilot Study

The purpose of the pilot was to explore the moral reasoning of 3rd, 4t, and 5th

grade students with EBD. Previously, researchers have not determined whether moral

reasoning in elementary students with EBD is different from their typical peers; however,

they have found that moral reasoning may be influenced by gender (e.g., Brown et al.,

1995; Gilligan 1982), culture (e.g., Edwards, 1987; Shweder et al., 1987), socioeconomic

status, and age (e.g., Colby & Kohlberg, 1987a; Piaget, 1932/1965). I targeted moral

reasoning in elementary-aged students because the number of students with EBD seems

to increase rapidly during the elementary years (OSEP, 1998), 3" 5th grade present the

best opportunity for intervention (e.g., Walker et al., 1995), and mid to late elementary

years provide a context for rapid growth in the development of moral judgement (Piaget).








Because a disproportionate number of African American males and students with SES

characterize the population of students with EBD (Kauffman, 1995), I also examined the

effects of SES, ethnicity, and gender on moral reasoning. Moreover, I included reading

comprehension in my analyses because researchers have found it to be strongly related to

moral theme comprehension (Narvaez et al., 1999).

Statement of Research Hypothesis

Specifically, I collected and analyzed data to address the following hypothesis:

When accounting for gender, ethnicity, reading comprehension, SES, atypical/typical

group, and grade, there will be a significant positive relationship between reading

comprehension and atypical/typical group and moral theme comprehension.

Research Design

I employed both qualitative and quantitative research methods to address the

hypothesis. I used causal comparative methods to explore the relationship between EBD

and moral reasoning in typical and atypical samples and case study research methods to

explore themes and relationships in students' thinking at the case level, thereby adding

depth and richness to the quantitative results. Ex post facto criterion-group research is a

causal comparative design used when the researcher arrives at the scene after the

treatment has been administered (Shavelson, 1996). According to Gall, Borg, and Gall

(1996) the primary objective is to discover possible causes and effects of a behavior

pattern or personal characteristic by comparing individuals in whom it is present with

individuals in whom it is absent, or present to a lesser degree. The causal comparative

design allows the determination of difference and, at the same time, an analysis of the

relationships several variables. Thus, causal comparative methodology is particularly








useful when cause-and-effect relationships are not amenable to experimental

manipulation. A disadvantage of causal-comparative research, however, is that the task of

establishing causality becomes more difficult (Gall et al.). Therefore, the researcher must

use both theory and data to eliminate rival hypotheses and justify causal relations within

the proposed model (Shavelson).

Case study is a form of qualitative research that allows the researcher to study a

phenomenon in its natural context by focusing on specific instances or cases, thereby

conducting an in-depth analysis of each case from an emic (Gall et al., 1996). For

example, cognitive-developmental researchers have employed case study research to

examine children's moral reasoning by posing moral dilemmas or problems to elicit their

reasoning processes. Children's responses to these moral prompts are, then, submitted to

an analysis in which the researcher seeks a window into the basic motives, desires, and

intent of the informant's reasoning (Gubrium & Holstein, 1997). In some instances, talk

may be viewed as descriptive; that is, informants are treated as reporters on experience.

In case study research, however, talk is considered constituitive. The informant becomes

the author of experience (Gubrium & Holstein).

Setting and Participants

I conducted the pilot in one moderately sized school district with approximately

30,000 students in rural, Central Florida. I petitioned 23 elementary schools and one

special day school for students with serious emotional disturbance (SED) to participate in

the project. The principals often elementary schools and the special day school agreed to

participate. From the 11 participating schools, I selected a computer generated random

sample of 40 students from the 3rd, 40, and 5th grade general education population and 40








students from the 3r"d, 4 and 5t grade special education population designated as

emotionally handicapped (EH) or SED (n = 40). Parental consent letters were sent home

with the 80 randomly selected students and 42 students subsequently returned signed

consents. Interestingly, the informed consent process produced an equal number of study

participants for each group. One elementary school had only one student selected and that

student did not return a parental consent form, reducing the number of participating

schools to 10. Results of a one-sample chi-square test indicated that the final participant

sample (N = 42) did not differ proportionately from the random sample (N= 80) on any

of the variables of interest: grade, ethnicity, SES, gender, or behavior type (typical versus

atypical).

When the magnitude of the treatment effect is not directly under the researcher's

control, the researcher should investigate the treatment effect by pilot testing the study

(Shavelson, 1996). According to Shavelson, estimating the effect size is an important step

in planning a study because the greater the treatment effect, the greater the power of the

test to detect statistical difference and the probability of making a Type II error is

reduced. Because previous researchers have not included elementary students with EBD

in their samples, a pilot was necessary to estimate the magnitude of the treatment effect.

Shavelson (1996) identified three other factors that may also influence the power

of the statistical test: (a) variability in the population, (b) sample size; and (c) level of

significance. According to Shavelson, the power of the statistical test can be improved by

sampling subjects from a homogeneous population. Sample homogeneity, however, may

also reduce estimates of reliability, thereby limiting statistical power. Narvaez et al.

(1999) estimated the reliability of the MTI using a sample that included a wider age range








of respondents than those targeted in the pilot, creating additional concerns about

statistical power. Moreover, because the pilot was exploratory, sample sizes were small,

but reducing the sample size also limits the power of the test.

Finally, Shavelson (1996) stated that by decreasing the level of statistical

significance, the power of the test can be increased and the probability of making a Type

II error can be reduced. Usually researchers are conservative when setting the level of

significance (e.g., .05 or .01); however, since this was a pilot study, the probability of

making a Type II error increased in importance. Therefore, I set alpha at .10 for all

analyses to increase statistical power because there was no information about the

expected magnitude of effect sizes from previous research, participant samples were

small, and the reliability of the MTI using a limited age range was not known.

A chi-square analysis comparing the group of students with EBD (n = 21) and

their typical peers (n = 21) on each variable of interest (i.e., gender, ethnicity, SES,

grade) was significant (a = .10) for SES, x2 (1, N = 21) = 9.528, p = .002 and ethnicity,

X2 (1, N = 21) = 3.055, p = .081, indicating that the participants with EBD included

proportionately more minority children and more children from low SES families when

compared with typical peers. All other proportions were within chance expectations (i.e.,

gender, grade).

I purposively selected three students from the participant sample of students with

EBD to participate in moral dilemma interviews. The sample included two female

students, one 4th grader and one 5th grader, and one 3rd grade male student. The 3r" and

4 graders attended the same school, but were not in the same classroom. All three

participants attended special classes for students with EBD.








Pilot Study Instrumentation

Moral Theme Inventory (MTI)

The purpose of the MTI (Narvaez et al., 1999) is to assess students' ability to

understand the themes of cooperation in moral stories as the author intended or as they

distort them to match their developmental level of moral reasoning. The MTI requires

respondents to listen to four stories and respond to five tasks for each story. It is given in

two 50-minute periods and yields two composite scores: reading comprehension and

moral theme comprehension. The reading comprehension task includes 10 true-false

reading comprehension questions for each of the four stories. The moral theme

comprehension composite is a 12-item test that includes four tasks: (a) rate how well four

vignettes match the story's theme on a five point Lickert scale, (b) select the vignette that

best matches the story theme, (c) select two theme choices that best match the story

theme, and (d) rate how well 7 or 8 theme choices match the story theme on a five-point

Lickert scale. All stories and response choices are on audiotape.

Reliability and validity. Narvaez and her co-authors (1999) used the responses of

two pilot groups to generate the moral themes and distortions included in the theme rating

and theme selection tasks. The MTI was, then, administered to a sample that included 50-

3rd graders and 54 5h graders selected from one public elementary school and 28 adults

enrolled in an educational psychology class at a public university. As a result of this

administration of the MTI, Narvaez et al. (1999) reported a Cronbach alpha reliability of

.89 for the moral theme comprehension composite and an alpha of.81 for the reading

comprehension composite. Using pilot results, I calculated a Cronbach alpha reliability of







74
.74 for the moral theme composite and a Cronbach alpha reliability of.83 for the reading

comprehension task.

Moral Dilemma Interview

I used three types of moral dilemmas to elicit informants' moral judgements:

Kohlbergian (see Colby & Kohlberg, 1987a); Piagetian (see Piaget, 1932/1965); and

literature based using the characters and setting found in the children's book, The Boxcar

Children by Gertrude Chandler Warner. The Boxcar Children is a story about the

adventures of four children who have lost both parents and are trying to live on their own.

I varied the dilemma type to examine the consistency of children's reasoning across

dilemmas, and I created The Boxcar Children dilemmas to provide an increased focus on

the issue of obedience to an external authority, a dominant theme of children's reasoning

(Colby & Kohlberg; Piaget).

Over the course of three 30 to 45-minute interviews, informants listened to a total

of eight hypothetical dilemmas. Each dilemma presented a conflict between two moral

issues (e.g., obedience to authority versus obedience to conscience), requiring informants

to make a choice as to which issue should take priority in the hypothetical situation.

Immediately following the reading of each dilemma, I asked informants to demonstrate

an adequate understanding of the pertinent facts and issues presented in the dilemma,

providing reviews when necessary. After informants demonstrated an adequate level of

comprehension, I asked them to make an issue choice. Following a statement of the

informants' issue choice, I probed the informant's reasoning about their choice to elicit

moral judgements and I also probed their reasoning about the issue that was not favored








to generate additional information about the consistency of informants' reasoning

processes.

Reliability and validity. Colby and Kohlberg (1987a) evaluated construct validity

for the Moral Judgment Interview by how well the data match the hypothesized

theoretical framework for the development of moral reasoning. Thus, evidence of

construct validity is represented in a good fit between the patterns discovered and theory.

To the extent that individuals' responses to hypothetical dilemmas predict real-life moral

judgements, Colby and Kohlberg concluded that this element of validity is limited by the

absence of techniques for scoring real-life moral judgements.

Data Collection

Moral Theme Inventory

I created an SPSS Base 10.0 file that included a list of student numbers,

participant numbers, and extant data (i.e., atypical/typical group, SES, ethnicity, gender,

and grade). Every participant (42) completed the Moral Theme Inventory (MT) (Narvaez

et al., 1999) at their schools and at a time and place arranged by the school counselor.

The MTI was given in two 50-minute sessions and in each session, students listened to

two stories and responded to five tasks for each story (i.e., reading comprehension task

and four moral theme comprehension tasks). A research assistant scored participants'

responses to the MTI and entered them into the SPSS Base 10.0 file. The scores of four

participants with EBD were eliminated from the data analysis procedures because of

missing data on some of the moral theme comprehension tasks.

I interviewed each of the 3 informants three times at their schools with each

interview lasting approximately 30-45 minutes. The time and date for each interview was








arranged with the informant's teacher. In the event of absences, interview appointments

were rescheduled. All interviews were audiotaped and upon the completion of Interview

III, I gave each participant a copy The Boxcar Children. Interviewing began on May 1,

2000 and was completed on June 2, 2000.

As a result of this data gathering procedure, I conducted a total of 9 interviews (3

students x 3 interviews). A student assistant transcribed all interviews into a word

processing file and I entered them into The Ethnograph (v5.0) for coding and analysis.

Each informant was given a code name, using the names of the three older children in

The Boxcar Children Warner, 1977). The 3rd grade male participant was given the name

Henry, the 4h grade female participant was given the name Violet, and the 5th grade

female participant was assigned the name Jessie. The 9 interviews produced yielded 130

pages of interview protocol.

Moral Dilemma Interview

I used a modified form of standard issue scoring (see Colby & Kohlberg, 1987a)

to interpret informants' responses to eight moral dilemmas. Each dilemma presented a

conflict between two moral issues (e.g., obedience to authority versus obedience to

conscience), requiring informants to make a choice as to which issue should take priority

in the hypothetical situation. Table 3.1 lists the stories by title and the conflict in values

posed by each dilemma and the definitions for each issue choice are presented in Table

3.2.

Comprehension. I initiated the interpretive process by first evaluating informants'

comprehension of the moral dilemmas. Immediately following the telling of each

dilemma, I asked the informants' a few comprehension questions to generate data about








their basic understanding of the facts and issues presented in the dilemma. Whenever

informants responded to these questions with confusion about any pertinent issues or

Table 3.1

Issues Represented in Eight Dilemmas


Dilemma Interview Issue


Heinz's Dilemma

Part 1

Part 2

Part 3

Louise's Dilemma

Father's Pen

Mother's Scissors

Doing Chores

The Foolish Brother

The Long Walk

The Boxcar Children

Part 1

Part 2

Part 3


Life versus Law

Conscience versus Law

Conscience versus Punishment

Authority versus Contract

Conscience versus Punishment

Punishment versus Fairness

Authority versus Contract

Authority versus Affiliation

Fairness: Equity versus Equality



Affiliation versus Authority

Affiliation versus Law

Affiliation versus Life


facts, I reviewed the story and continued asking comprehension questions until the child

demonstrated an adequate understanding. When the informant and I were both

comfortable with the level of comprehension, I then proceeded with questions designed






78

to probe the informants reasoning processes. Therefore, the first interpretive task was to

examine all data coded comprehension to evaluate informants' understanding of the

dilemmas. As a result of this analysis, I determined that every informant seemed to have

an adequate understanding of the issues and facts presented in the dilemmas. Thus, I

proceeded with the analysis of informants reasoning processes by segmenting their

responses into two broad categories, issue choice (IC) and moral judgement (MJ).

Issue choice. The purpose of all eight dilemmas was to present the informant with

a conflict between two moral issues, so after listening to a dilemma and reviewing the

basic issues and facts presented, I asked each informant to choose the issue that should

take priority. Informants usually made clear issue choices following the comprehension

probes; however, occasionally they had to work through their justifications for each

choice and then make a decision after examining the pros and cons indicated by each

choice. Ultimately, informants made clear issue choices in all instances, resulting in a

total of 36 issue choices (12 choices x 3 informants).

Moral judgement. The next interpretive task was to examine informants'

justifications for their issue choices as expressed in their moral judgements. A segment of

text was categorized as a moral judgement (MJ) when the justification was prescriptive

(i.e., included words like "should" and "ought") and the informant identified a valued

norm (e.g., authority) and a moral element (e.g., obey to avoid punishment) that provided

a reason for valuing the norm (Colby & Kohlberg, 1987a). This analysis produced 247

MJs that were submitted to further analysis to identify valued norms and moral elements.

Norm. The norm is the component of the MJ that represents the moral values) or

objects) of concern that is being brought to bear by the informant in justifying the issue








choice in the dilemma (Colby & Kohlberg, 1987a). Thus, the norms informants valued in

supporting their issue choices could include norms other than those presented in the issue

choice. For example, in "Heinz's Dilemma", informants were presented with a conflict

between the issues of stealing a drug to save the life of one's wife (IC life) or upholding

the law and not stealing (IC law). In this example, informants might choose the law issue

(not steal), but could justify the IC law by valuing the norms property, law, affiliation,

and/or conscience. An informant could decide to "not steal" because it is against the law

or they could decide to "not steal" because it's not right to take someone else's proDerty

without asking. I also probed their reasoning on issues not favored. The purpose of those

probes was to generate additional information about the informants' reasoning processes

and the consistency of their reasoning across issues. Table 3.2 provides a list of the norms

informants valued to justify their issue choices.

Moral element. To complete the moral judgement, the informant must give a

reason for endowing the norm with value. This segment of an MJ is called the moral

element and is categorized as either a modal or value element. (Colby & Kohlberg,

1987a). According to Colby and Kohlberg, a model element expresses the mood or

modality of the moral language and included words such as should, must, deserves, and

has a right. Sometimes informants offered only modal elements as terminal values or

justifications. For example, when asked why stealing is wrong, an informant might

simply reply, "It's just wrong", failing to provide a terminal value element in explaining

why stealing is wrong. This kind of response indicates that the informant is speaking

from a normative order orientation in which maintaining the norm is an end in itself. In








Table 3.2

Norms Valued and Their Definitions


Norm Definition


affiliation maintaining relationship

authority obligated by an authority figure

conscience obligated by good intentions

contract keeping promises

fairness issuing justice with equity or equality

law obligated by law or rules

life maintaining quality or quantity of life

property respecting property rights

punishment obligated by fear of retribution


contrast, value elements represent ultimate ends, values, or reasons set forth by the

informant as the terminal value in the MJ. For example, when an informant says that

stealing is wrong because "you might go to jail", the informant is expressing the

avoidance of punishment as the terminal value. Table 3.3 lists the modal and value

elements found within informants' MJs in response to the moral dilemmas posed.








Table 3.3

The Moral Elements


Modal Elements

Upholding normative order

1. Obeying/consulting persons or deity. Should obey, get consent (should consult,

persuade.

2. Blaming/approving. Should be blamed for, disapproved (should be approved).

3. Retributing/exonerating. Should retribute against (should exonerate).

4. Having a right/having no right.

5. Having a duty/having no duty.


Egoistic consequences

1. Seeking reward/avoiding punishment

2. Avoiding trouble for oneself


ue Elements

Ideal or harmony-serving consequences

1. Serving a social ideal or harmony

2. Fairness: equality versus equity

3. Balancing perspectives or role taking

4. Good/bad consequences for others


Pilot Study Results

The results of a simultaneous multiple regression analysis indicated that the

coefficients for SES (B = 4.009, p = .080) and reading comprehension (B = .516, p =

.001) were significant. The coefficient for typical/atypical group was not significant;

however, correlation results showed that EBD and SES were confounded. Therefore, the







82
significance of the relationship between EBD and moral theme comprehension could not

be determined with these data.

An analysis of interview data suggested that 3r grade Henry and 5h grade Jessie's

moral judgements were dominated by a Punishment/Obedience Orientation, but also

contained frequent references to an individualistic perspective consistent with a Personal

Reward Orientation. In contrast, 4't grade Violet's moral orientation seemed to be almost

entirely consistent with a Personal Reward Orientation. A synthesis of the quantitative

and qualitative results provided evidence that students in the EBD sample may rely on

preconventional reasoning to guide their moral judgement. These results, however,

should be viewed with caution since EBD and SES were confounded in both data

sources. Preconventional reasoning is characterized by a morality that is obligated by

external forces, such as adults, law, and punishment, and is motivated by egoistic

concerns. The theoretical and practical implications of these findings bear significant

import and warrant further investigation using a larger sample in which SES is controlled.

Introduction To The Present Study

Cognitive developmental researchers portray the development of morality as a

maturational progression through a series of qualitatively different stages (e.g., Colby &

Kohlberg, 1987a; Piaget, 1932/1965). Typically, researchers use the moral dilemma

interview to gather data about children's reasoning processes (e.g., Colby & Kohlberg;

Piaget), however, the recent development of the Moral Theme Inventory (Narvaez, et al.,

1999) presents a unique opportunity for child development researchers to integrate

quantitative and qualitative research methods to facilitate theory building and increase

explanatory power. In the present study I integrated qualitative and quantitative research







83
methods to extend the pilot findings by adapting sampling procedures to control for SES,

a variable known to be confounded with EBD, and by examining the influence of trait

anger on the development of moral reasoning. The most prevalent types of EBD are

externalizing disorders that include maladaptive behavior in the form of aggression,

disruption, and antisocial behavior (Kauffman et al., 1987). Thus, the variable trait anger

was added because it is commonly recognized as the underlying affective component of

externalizing disorders (Walker et al., 1995). Therefore, the purpose of the present study

is to explore the relationships among behavior, trait anger (emotion), and moral reasoning

(cognition) in a sample of elementary students with EBD and their typical peers while

controlling for SES.

Statement of Research Hypothesis

The present study was designed to employ a cognitive developmental approach to

examine the relationships among moral reasoning, trait anger, and EBD. I included

elementary aged students because the number of students with EBD seems to increase

rapidly during the elementary years (OSEP, 1998), 3r 5t grade present the best

opportunity for intervention (e.g., Walker et al., 1995), and mid to late elementary years

provide a context for rapid growth in the development of moral judgement (Piaget,

1932/1965). Because the population of students with EBD is characterized by a

disproportionate number of African American males and students with low SES

(Kauffman, 1995), I included gender, SES, and ethnicity in my analyses. I also included

reading comprehension as a cognitive variable, trait anger as an emotional variable, and

the student's grade in school to gauge developmental change.









Specifically, I collected and analyzed data to address the following research

hypothesis: When accounting for the variables gender, ethnicity, trait anger, reading

comprehension, SES, atypical/typical group, and grade, there will be a significant

positive relationship between reading comprehension and moral theme comprehension

and a significant negative relationship between trait anger and moral theme

comprehension.

Design

To implement the present study, I replicated the same research design described

in the pilot. I integrated quantitative and qualitative research methods using an ex post

facto criterion group design and case study research to facilitate an in-depth study of

informants' moral reasoning processes. An ex post facto criterion group design prescribes

a type of quantitative methodology grounded in the assumption that features of the social

environment constitute an independent reality and are relatively constant across time and

settings. According to Gall et al. (1996), the objective is to discover possible causes and

effects of a behavior pattern or personal characteristic by comparing individuals in whom

it is present with individuals in whom it is absent, or present to a lesser degree. The

researcher collects data on observable behaviors of samples and then subjects these data

to numerical analysis and interpretations that can be generalized to populations.

In contrast, qualitative methodology implied by case study research requires the

study of phenomena in natural settings as a subjective reality (Gall et al., 1996). In case

study research, researchers interpret phenomena in terms of the meanings people bring to

them (Guba & Lincoln, 1994). Thus, the objective is to obtain intricate details about

phenomena such as feelings, thought processes, and emotions that are difficult to extract








or learn through more conventional research methods (Strauss & Corbin, 1998). The

researcher treats social facts as things and interprets them with the aim of describing and

explaining relationships among concepts (Gubrium & Holstein, 1997). Therefore, the

qualitative researcher is concerned primarily with process (Creswell, 1994), and the

interpretive activities persons undertake, moment by moment, to construct, manage, and

sustain the sense that their social worlds exist as factual and objectively "there" (Gubrium

& Holstein).

According to Gall et al. (1996), when the researcher uses a multiple-case design,

the generalizability of constructs and themes across cases can be checked by noting

whether a particular theme observed in one case also is present in the other cases. The

generalizability of case study results to individuals or situations beyond those included in

a particular study is problematic. Therefore, case study results are limited in

generalizability and cannot be inferred to the situations and persons beyond those

included in the study.

Sampling

Design requirements. When researchers want to draw inferences from a sample to

a population, the statistical model requires a random selection of participants (Shavelson,

1996). A random sample allows each member of the population an equal and independent

chance of being included in the sample. When the assumption of random selection is

satisfied, sample findings can be generalized to a known population (Shavelson).

Stratified random sampling is a sampling technique researchers use to achieve balanced

representation on a variable of interest when the researcher wants to be certain that

subgroups in the population are adequately represented (Gall et al., 1996). Since






86
socioeconomic status (SES) is a variable known to influence moral reasoning (e.g., Colby

& Kohlberg, 1987a; Hardman & Smith, 2001) and is also significantly related to EBD

(Hardman & Smith), I employed stratified random sampling techniques to equalize the

atypical and typical samples on the variable SES.

Sampling procedures. I initiated sampling procedures by submitting an

application to conduct research in 43 elementary schools and three special education

centers for students with severe emotional disturbance (SED) in two moderately sized,

rural school districts located in Central Florida. Only 19 of the 24 schools in district one

agreed to participate in the study, so the sample selected in district one represents a

subpopulation within that district. Every school in district two agreed to participate,

thereby increasing the external validity of these results.

The sample selection process was different for each district, so I describe the

selection process separately for each one. I selected 100 students from district one and 60

students from district two. I divided the sample in this way because district one was

larger and was able to provide a computer generated sample. The sample selection

process in district two had to be done by hand, requiring more time and effort.

School district one. I submitted an application to conduct research to the

principals of 23 elementary schools and one special day school for students with EBD.

Only 19 of the 24 applications were returned with 12 elementary schools and the school

for students with SED agreeing to participate and six elementary schools declining. To

stratify the sample on the variable SES, I requested a computer-generated list of all 3rd,

4t, and 5th grade students with EBD in participating schools who were not eligible for

free/reduced lunch. This request produced a list of 13 students, so I purposively selected







87
those 13 children and requested a computer-generated random selection of an additional

37 3d, 4h, and 5th grade students with EBD, bringing the number of students in the

EBD sample to 50. I then selected a random sample of 50 3rd, 4h, and 5t grade students

from the general population in participating schools using a 13:37 ratio of free/reduced

lunch to full pay students. As a result, sampling procedures in district one yielded a total

sample of 100 students selected from 12 elementary schools and one school for students

with EBD.

School district two. All 19 elementary schools and the 2 special day schools for

students with EBD agreed to participate in the study. I requested a list of student numbers

for all 3rd, 4, and 5th grade students with EBD and their typical peers. Using a random

number table, I selected 30 3", 4t, and 5h grade students from the general education

population, observing that 11 were eligible for free/reduced lunch and 19 were full pay

students. To stratify the special education sample to match the general education sample

on SES, I separated the list of special education student numbers into two lists,

free/reduced lunch (94 students) and full pay (21 students). I, then, randomly selected 11

free/reduced lunch students and 19 full pay students to participate in the study. As a

result, sampling procedures in district two yielded a sample of 60 students representing

19 elementary schools and two special day schools for students with SED. Sampling

efforts in both districts yielded a final sample of 80 3r, 4t, and 5' grade students with

EBD and 80 3", 4th, and 5" grade typical peers (N = 160) selected from 31 elementary

schools and three special day schools for students with SED.

After completing the informed consent process, I purposively selected an

informant sample of 12 students from the participant sample that would represent an








equal distribution on the variables group, grade, and SES. I selected six students with

EBD and six typical peers. Each group, EBD and typical peers, included two students

from each grade (3d, 4th, and 5th) and one student from each group and grade was chosen

from the free/reduced lunch participants and one student from each group and grade was

chosen from the full pay participants.

Setting and Participants

Setting

The present study was conducted in two moderately sized school districts

(approximately 30,000 students each) in rural, Central Florida. One district had 24

elementary schools and one school for students with serious emotional disturbance (SED)

and the other had 19 elementary schools and two schools for students with SED.

Participants

Teachers in both school districts sent home informed consent letters for all

students included in the random sample (N=160) and subsequently obtained consent from

the parent/guardians of 42 students with EBD and 40 typical peers. Each parent/guardian

received two letters, one giving permission to take the MTI (Narvaez et al., 1999) and the

Feelings Ouestionnaire (Jacobs & Blumer, 1984) and another giving permission for

participation in moral dilemma interviews. Every consenting parent, however, returned

both letters. Students in four elementary schools in district two failed to return any letters,

reducing the number of participating elementary schools to 27. Appendix A contains

copies of both informed consents.

Before data collection began, two participants with EBD and two typical peers

participants moved. One student with EBD refused to participate during data collection,








reducing the final participant sample to 77. Results of a one-sample chi-square test (a =

.05) indicated that the final participant sample (77) did not differ proportionately from the

random sample (160) on any of the demographic variables of interest; that is, grade,

ethnicity, SES, gender, or atypical/typical group.

Table 3.4 presents the results of a chi-square analysis of group differences

between the group of students with EBD (atypical group) (n = 39) and their typical peers

(n = 38) on each demographic variable of interest (i.e., gender, ethnicity, SES, or grade)

using a one-sample chi-square test. The results of the test were significant (a = .05) for

gender, x2 (1, N = 39) = 9.231, p= .002, grade, X2 (2, N = 39) = 35.250, p< .001, and

ethnicity X2 (1, N = 39) = 7.385, p = .007. Consistent with previous findings, the

participants with EBD included proportionately more minority students and males when

compared with typical peers. The significant result for grade indicated that the group of

students with EBD was older than the group of typical peers as measured by grade in

school. The chi square result for SES was non-significant, suggesting that the stratified

Table 3.4

One-Sample Chi-Square Test of Atypical Sample


df Z2 Significance


Gender 1 9.231 .002

Grade 2 35.250 <.001

Ethnicity 1 7.385 .007

SES 1 .506 .477

a =.0




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EB3SPQOWG_ELI1L6 INGEST_TIME 2013-03-27T14:50:51Z PACKAGE AA00013630_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES