 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

 

I DATE DUE DATE DUE DATE DUE
_‘ ,, 6 '1 1?
LAM 81 2,08 @5311 u

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65op.15

ATTITUDES THAT REFLECT RESILIENCE AND THEIR RELATIONSHIP TO
COPING RESPONSES AND PROTECTIVE FACTORS IN AFRICAN AMERICAN
COLLEGE STUDENTS
By

Sign'd Julian Dixon

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY

Department of Counseling, Educational
Psychology, and Special Education

2002

ABSTRACT
ATTITUDES THAT REFLECT RESILIENCE AND THEIR RELATIONSHIP TO
COPING RESPONSES AND PROTECTIVE FACTORS IN AFRICAN AMERICAN
COLLEGE STUDENTS
By

Sigrid Julian Dixon

A sample of 229 African American college students were surveyed in order to
assess attitudes that reﬂect resilience and coping, and to evaluate their relationship. In
addition, religiosity, spirituality, core convictions, and level of acculturation were
assessed. Attitudes that reﬂect resilience were deﬁned as beliefs that one can survive and
make things better for self and others (Biscoe & Harris, 1994).

It was hypothesized that there would be a signiﬁcant positive correlation between
attitudes that reﬂect resilience and coping responses. It was also hypothesized that
religiosity, spirituality, and core convictions would moderate the relationship between
attitudes that reﬂect resilience and coping responses. and that levels of acculturation
would be related to coping responses, core convictions, religiosity, spirituality, and
attitudes that reﬂect resilience. A key focus of the study was to generate evidence
regarding the direction of the relationship between coping responses and attitudes that
reﬂect resilience. Because chains Of relationships were being investigated
simultaneously, and a bi-directional model was being tested, structural equation modeling
(SEM), path analysis, reliability analysis, and exploratory factor analysis (EFA) were

used to analyze the data.

The existence of a relationship between attitudes that reﬂect resilience and coping
responses was supported, but the direction was ambiguous. Additionally, gender-based
differences in coping responses were revealed, and protective religious beliefs were found
to moderate the relationship between speciﬁc coping responses and attitudes that reﬂect
resilience. Culturally salient religious beliefs and practices were directly related to
protective religious beliefs and to spiritual and applied faith coping responses.

Findings from this study further suggest that supportive interventions should
include the teaching of both coping skills and general resiliency attitudes. The results
also point to the importance of religiosity, spirituality, and core convictions as protective
factors. Finally, the results suggest the need for a revision of the items and subscales in
the Resiliency Attitudes Scale, and more careful psychometric analysis of other scales

used in resiliency research.

Copyright by
Sigrid Julian Smith
2002

DEDICATION

To the loving memory of my father, the late James Allen Moses Dixon.

DEDICATION

To my brother, Deric for showing me the true meaning of resilience, persistence, faith,
and love.

vi

 

DEDICATION

TO my mother, Lessie Mabel Brown-Dixon for her love, support, encouragement, and
constant prayers. Thank you for the sacriﬁces you made for dad, my brothers and me.
You are truly the world’s greatest mom.

vii

DEDICATION

To Elmorris for being a great role model and the worlds best big brother.
To Keith for his courage to keep trying, and for the willingness to help.

To Bryan for being my best and closest friend and brother and for the support, help, and
sacriﬁces he made for me.

To Colin for his support and prayers, and for being a responsible and trustworthy brother.

To Christopher for his untiring love, prayers, support, and for being the worlds best little
brother.

And

To my extended family, for your loving support and prayers.

viii

 

ACKNOWLEDGEMENTS

To Dr. Lee N. June, Chairperson of my doctoral committee, I am deeply grateful for his
guidance and continuous support and prayers through this study and throughout the
course of my graduate study. I am especially grateful for his concern, and for his patience
and support of my professional development.

I am deeply grateful and indebted to Dr. Gloria Smith for her guidance, and support as a
member Of my doctoral committee and throughout the course of my graduate study. In
addition, I am grateful to her for the concern, and encouragement she provided in support r
of my professional development.

I am deeply grateful and indebted to Dr. Bill Lawson for his guidance, concern, and
support as a member of my doctoral committee. I am especially grateful for his
encouragement and help during the development of this study and for sharing his time
and friendship.

 

I am grateful and indebted to Dr. Christopher Dunbar for his time, patience, support, and
listening ear as a member of my doctoral committee. Additionally, I am grateful for his
assistance and feedback during the development of the study.

A very special thanks to Dr. Jeannette Gassaway for her concern, support, friendship, and
guidance throughout the duration of the dissertation process. Additionally, I am grateful
for her valued assistance, encouragement and patience during the formulation and
development Of the study.

A special thanks to Dr. Ira Washington for his invaluable assistance in all aspects of the
study, and for his support in preparing the data for analysis.

A special thanks to Mr. Joseph Martineau for his invaluable guidance and technical
assistance in all aspects of the data analysis and interpretation for this study.

Special appreciation to Mrs. Florence Harris, Mrs. Nettavia Curry, Mrs. Betty Sanford,
Mr. Joshua Gillespie, Mr. Murray Edwards and Mr. Rodney Patterson for your support
and assistance.

Special thanks to Mr. Keith Williams for his assistance and kindness.

A special thanks to Mrs. Anna A. Yokoyama for her friendship and support.

A special thanks to Dr. Renee Sanders-Lawson for her support, prayers, and friendship.

A special thanks to Dr. Mary Clearing-Sky, Dr. Charlene Patterson, Dr. Gordon Williams,

Ms. Carmen Gear, Dr. Bill Metcalfe, Dr. DeColius Johnson, Dr. Wendy Singo and all the
staff of the Michigan State University Counseling Center for their kindness, and support.

 

A very special thanks to Dr. Jean-Claude Dutes for his support, encouragement,
mentoring, and friendship.

A special thanks to Dr. Pamela Bellamy for her kind support and encouragement.
A special thanks to Dr. Bonita Pope-Curry for her support and encouragement.

A very special thanks to Ms. Wanda Edwards for her kindness and support.

A very special thanks to Ms. Ann Graves for her friendship, support and prayers.
A very special thanks to Ms. Penny Wali for her prayers, kindness and support.

A very special thanks to Ms. Schelzia English-Ross for her support, prayers, and
friendship.

A special thanks to Briana Glauvitz, Lori Holuta, Barbara McWilliarns, Nicole Hafner,
and Paul Johns for your support and kindness.

Special thanks to Dr. Yevonne Smith for her support and kindness.

A special thanks to Shaun, Donna and Aaron Thomas, Risa and Lee Alexander, Reverend
Eric Washington, Donald and Camille Bryant, Irene Unkefer, Clifford and Dr. Amelia
Wilson, Ms. Cynthia Pugh, Dr. Keith White, Mr. Arnold Ross, Arthur and Janice
Christian, Ms. Dorothy Kurrie, Derrick and Deneice Forbers, David and Debbie Mange
and Carl Allen for your kindness, support, prayers, and friendship.

To Dr. Joseph C. Dunbar, thank you for your listening ears, support, and friendship over
the years.

To Mike Brooks, Kenneth Baker, Jarrod Grandy, and Rudy Serrato, thank you for your
support.

To Ms. Maxine Walton, M.S.W. and my colleagues from the Detroit Institute of Family
Systems Therapy.

To the Friendship Baptist Church family, Reverend Lester D. Stone, Pastor - thank you
for your prayers and support.

To the New St. Mark Baptist Church family, Reverend Larry Smith Pastor — thank you
for your prayers and support.

To LaMaurice and Mother Doris Evans, for your kind support, love, and prayers through
the years.

To Riane, Asha, Tiana, Kialyn, Alfreda and Reginald for their gifts of love and support.

Finally, to my family, Mother Lessie, and brothers Elmorris, Bryan, Colin, Christopher,
Deric and Keith, and the many relatives and friends who showed me love and support, I
express my gratitude.

 

xi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... xv
LIST OF FIGURES ......................................................................................................... xvi
LIST OF APPENDICES ............................................................................................... xvii
CHAPTER I. INTRODUCTION ....................................................................................... 1
Statement of the Problem ............................................................................................... 1
Purpose of the Study ....................................................................................................... 4
Deﬁnition of Terms ........................................................................................................ 5
Overview of the Study .................................................................................................... 8
CHAPTER II. LITERATURE REVIEW ................................................. g ....................... 10
Introduction .................................................................................................................. 10
Overview of the Resilience Literature .......................................................................... 10
Why the Growing Interest in Resilience? ..................................................................... 12
Relevant Theoretical Frameworks that Inform this Study ............................................ 13
Models of Resilience .................................................................................................... 21
Major Criticisms of the Resiliency Literature .............................................................. 26
Resilience and Protective Factors ................................................................................. 30

~ Resiliency in African American College Students ....................................................... 32
Coping, Culture and Theories of Coping ...................................................................... 34
Coping Among African American College Students ................................................... 37
Summary ....................................................................................................................... 40
Research Questions ....................................................................................................... 41

xii

 

CHAPTER III. METHODS ............................................................................................ 43

Introduction .................................................................................................................. 43
Procedures .................................................................................................................... 43
Population and Sample ................................................................................................. 45
Instrumentation ............................................................................................................. 48
Research Hypotheses .................................................................................................... 59
Statistical Methods Used to Analyze the Data ............................................................. 59
CHAPTER IV. ANALYSIS OF DATA ......................................................................... 61
Descriptive Statistics .................................................................................................... 62
Instrument Scaling ........................................................................................................ 62
Model development ...................................................................................................... 63
Summary ....................................................................................................................... 65
CHAPTER V. RESULTS ............................................................................................... 66
Descriptive Statistics of Control Variables .................................................................. 66
Instrument Scaling (Exploratory Factor Analysis) ....................................................... 69
Modeling the Data ........................................................................................................ 88
Summary ..................................................................................................................... 111
CHAPTER VI. SUMMARY AND CONCLUSIONS .................................................. 114
Summary ..................................................................................................................... 114
Conclusions ................................................................................................................ 1 18
Discussion of Results ................................................................................................. 120
Implications of the Study ............................................................................................ 130
Limitations Ofthe Study .............................. 137

xiii

 

APPENDICES ............................................................................................................... 140

REFERENCES .............................................................................................................. 177

xiv

 

LIST OF TABLES

Table 1. Complete sample (N=224) frequencies of categorical variables. ..................... 67
Table 2. Final sample (N=190) frequencies of categorical variables .............................. 67
Table 3. CAAP by program participation frequencies. .................................................... 68
Table 4. Correlations among and descriptive statistics of the six COPE subscales ......... 73
Table 5. Correlations among and descriptive statistics of the three RCAS subscales. ....76

Table 6. Descriptives, reliabilities and intercorrelations among combined COPE and

RCAS factors. ................................................................................................. 79
Table 7. Fit statistics Of the three competing models ....................................................... 90
Table 8. Acronyms used in Tables 9-11 .......................................................................... 96
Table 9. Results of the unidirectional ARR -—* CR model .............................................. 96
Table 10. Results of the unidirectional CR —’ ARR model ............................................ 97

Table 11. Direct, indirect, and total effects in the unidirectional ARR -> CR and CR —+

ARR models. ................................................................................................... 98
Table 12. Factor loading matrix for the COPE. ............................................................. 141
Table 13. Factor loading matrix for the RCAS. ............................................................ 146
Table 14. Complete Combined COPE and RCAS EFA factor loading matrix. ........... 149
Table 15. Factor loading matrix for the RAS EFA. ...................................................... 154

XV

 

LIST OF FIGURES

Figure l. Scree plots of the COPE orthogonal EFA. ...................................................... 70
Figure 2. Scree plot of the RCAS EFA. .......................................................................... 75
Figure 3. Scree plots from the COPE and RCAS combined EF A. ................................. 78
Figure 4. Scree plot of the CCS EFA. ............................................................................. 82
Figure 5. Scree plot of CCS, Religiosity, and Spirituality EFA ...................................... 84
Figure 6. Listwise and pairwise scree plots of the RAS EFA. ........................................ 86
Figure 7. Graphical display of the unidirectional AR -> CR model. ........................... 93
Figure 8. Graphical display of the unidirectional CR -» ARR model. ........................... 94
Figure 9. Spiritual CR as an outcome of ARR, moderated by PRB. ............................ 103
Figure 10. Applied Faith CR as an outcome of ARR, moderated by PRB. .................. 103

xvi

 

LIST OF APPENDICES

APPENDIX A. COPE EFA Technical Appendix ......................................................... 140
APPENDIX B. RCAS EFA Technical Appendix ......................................................... 145
APPENDIX C Combined COPE and RCAS EFA Technical Appendix ...................... 148
APPENDIX D. RAS EFA Technical Appendix ........................................................... 153
APPENDIX E. Consent Form / Instruction Sheet ........................................................ 158

xvii

CHAPTER I. INTRODUCTION

Sgtement of the Problem

Interest in the study of resilience has increased over the past decade in the social
sciences and in education (Glantz & Johnson, 1999). Regarding resilience, Kaplan
(1999) noted that:

Concepts have a life of their own. Sometimes they appear and then suddenly

depart like fashion trends. These early exits are not necessarily premature, at least

when the ideas were insubstantial to begin with. With this in mind, consider our
current fascination with the notion of resilience. Does resilience qualify as an
organizing concept with sufﬁcient logical and emotional resonance to yield
systematic theoretical and research inquiry that will make a lasting contribution?

(p. 18).

The growing interest in resilience has emerged amid important trends. These
trends include: (1) the emergence of the positive psychology movement (Ickovics & Park,
1998); (2) the emergence of multicultural counseling and research imperatives with their
emphasis on developing ethnically and culturally relevant theory that informs
interventions and research with people of color (Pedersen, 1987: Schiele, 1996); and
(3) an increased attention to the role of religion and spirituality in mental and physical
health (Bergin, Stinchﬁeld, Gaskin, Masters & Sullivan, 1988). These trends are
providing new and promising directions for the study of resilience.

Authorities in the ﬁeld have indicated, however, that there is a need for additional

research on resilience (Barbarin, 1993b; Garmezy, 1993; Gannezy & Masten, 1986;

 

Glantz & Johnson, 1999; Werner, 1995). This call for additional research is based on the
consensus that there are yet large gaps in our understanding of this phenomenon
(Barbarin, 1993b; Garmezy, 1993; Garmezy & Masten, 1986; Kagan, 1975; Kinard,
1998; Luthar & Zigler, 1991; Monaghan-Blout, 1996; Mrazek & Mrazek, 1987; Tarter &
Vanyukov, 1999). According to models of resilience posited in the work of Barbarin
(1993b), Flach (1988), Garmezy and Masten (1986), Mrazek and Mrazek (1987), Rutter
(1985), Werner (1995), and Wolin and Wolin (1993), resilient persons cope better with
major life challenges.

Jew, Green and Kroger (1999) however, raise questions about the relationship
between resilience and coping. They hypothesized that “resiliency emerges from a
system of speciﬁc beliefs that interact with environmental stressors to determine an
individual’s coping skills” (p. 77). They further state “This deﬁnition places resiliency,
as a belief system, causally prior to coping, as a set of behaviors based on a belief system”
(p. 7 7). This cognitive behavioral conceptualization of resilience suggests a linear causal
relationship between resilience and coping. The proposed relationship however, has not
yet been empirically supported (Jew, personal communication, October 23, 2000).

In an earlier study, Jew and Green (1998) examined the effects of risk factors on
adolescent resiliency and coping among a population of seventh through twelfth grade
adolescents who reported varying levels of personal stress and adversity. The purpose of
their study was “to assess whether adolescents’ self-reported as high, and low risk
differed in their resiliency and coping scores” (p. 675). Resilience was deﬁned as a set of
beliefs that determine coping behaviors, and coping was measured using the Ways of

Coping Scale (Folkman & Lazarus, 1980).

 

Jew and Green (1998) found that students who reported exposure to a high
number of risk factors (at least ﬁve) had lower levels of resilience (i.e. beliefs that reﬂect
resilience), and students who reported exposure to a lower number of risk factors had
higher levels of resilience. This ﬁnding supported their hypothesis and previous ﬁndings
in the literature on the relationship between resilience and risk factors. A similar pattern
was expected to emerge in coping responses, however, Jew and Green were unable to
detect any signiﬁcant differences in the coping responses among students exposed to
either low or high numbers of risk factors. They therefore concluded that there is a need
to continue to examine the relationship between resiliency and coping.

Although there is a need for additional research on resilience, the literature has
been criticized for the lack of reliable and valid mechanisms for measuring it. This
criticism includes concerns over inadequate operationalizations of resilience (Kinard,
1998). Biscoe and Harris (1994) offer a possible solution to this problem. They
developed the Resiliency Attitudes Scale (RAS). Instead of measuring resilience, this
scale measures attitudes that reﬂect resilience. From such attitudes, one may infer
resilience. This scale is yet under development, but it appears that it may lend itself well
to the empirical study of resilience.

In addition to problems in measuring resilience and in deriving a clear
understanding of the relationship between resilience and coping, resilience has been
inextricably linked with risk and protective factors (Barbarin, 1993a, 1993b; Fine, 1991;
Jew & Green, 1998; Mandleco, 1991; McCubbin, Thompson & McCubbin, 1995; Rutter,
1987; Turner, 1992; Werner, 1995, 1989). Risk factors are acute or enduring stressful life

events and may include characteristics of the individual as well as the individual’s

 

environment. Such factors usually result in increased vulnerability in individuals.
Protective factors on the other hand, are skills, attitudes, personality traits, or
environmental factors that contribute to, or enhance resiliency. While much resilience
research has focused on the study of factors that place children and youth at-risk for
positive outcomes '(e. g., graduation from high school and college), additional work is
needed to identify variables that can offset risk factors or enhance protective mechanisms
(Luthar & Zigler, 1991; Mrazek & Mrazek, 1987). There is also a need to better apply
these concepts to persons of color (Garmezy, 1993).
Purpose of the Stujdy
The overall purposes of this study were (1) to examine the relationship between
attitudes that reﬂect resilience and coping among African American college students, (2)
to examine the moderating role of three factors that are believed to be protective in times
of stress and adversity, and (3) to examine the relationship between level of acculturation
and coping responses, attitudes that reﬂect resilience, core beliefs, religiosity and
spirituality.
The speciﬁc objectives for this study were to:
1.) Broaden our understanding of attitudes that reﬂect resilience in African American
college students.
2.) Better understand the relationship between attitudes that reﬂect resilience, coping
and protective factors (religiosity, spirituality, and core beliefs).
3.) Investigate the relationship between level of acculturation and coping responses,

core beliefs, religiosity and spirituality.

This study is important for two main reasons. First, it attempts to answer several
questions about the relationship between coping and resilience that were raised by Jew
and Green (1998) and Jew, Green and Kroger (1999) who hypothesized a linear causal
relationship between resiliency and coping. A review of the literature and personal
communication with Dr. Cynthia Jew (October 23, 2000) revealed that no empirical
evidence exists either supporting or refuting their hypothesis.

Secondly, information gleaned from this study may be used to help programs that
work with African American students who are both “at risk” and “non-at risk”.

Deﬁnition of Terms
For the purposes of this study, several terms will be operationally deﬁned:
ALBE‘S

Students who were admitted to Michigan State University through the College
Achievement Admissions Program (CAAP) are deﬁned as “at-risk”.

Non-At Risk

Students who were regular admits to the university. In this study, such students
are referred to as non-CAAP.
African American

Refers to members of the American society that are of African ancestry, history,
and culture. Other terms such as Blacks, and Afro-Americans refer to the same

population.

 

Attitudes that Reﬂect Resilience

Attitudes that Reﬂect Resilience are deﬁned as the beliefs that one can survive
and make things better for self and others (Biscoe & Harris, 1994). These attitudes
manifest themselves in an ability to persist in working through difﬁculties (Biscoe &
Harris, 1994). They represent but one aspect of resilience. They shall be operationalized
as a composite score on the General Resiliency Attitudes subscale in the Resiliency
Attitudes Scale (Biscoe & Harris, 1994). The acronym “ARR” will represent Attitudes
that Reﬂect Resilience in this study.
gaping

The “changing cognitive and behavioral efforts to manage psychological stress”
(Tennen, Afﬂeck, Arrneli & Carney, 2000, p. 626). Coping occurs in response to
situational and dispositional stress, and shall be operationalized as subscale scores on the
COPE (Carver, Scheier & Weintraub, 1989). These coping responses shall be designated
in this study using the acronym “CR”.
Religious Coping

The use of faith, religious beliefs, and practices to respond to situational or
dispositional stress, and shall be operationalized as subscale scores on the Religious
Coping Activities Scale (RCAS; Pargament, Ensing, Falgout, Olsen, Reilly, Van Haitsma
& Warren, 1990). These religious coping responses shall be designated in this study

using the acronym “RCAS”.

Religiosig

Perception of the depth of personal religious convictions, and shall be
operationalized as question #10 (“How religious would you say you are?”) of the
Multidimensionality of Religiosity Scale (Levin, Taylor & Chatters, 1995).
Spi 'tuality

Perception of the depth of spiritual convictions, and shall be operationalized as
question #13 (“How spiritual would you say you are?”) and question #14 (“How
important is your spirituality to you?”) of the Multidimensionality of Religiosity Scale
(Levin, Taylor & Chatters, 1995).
Core Convictions

The set Of personal beliefs that reﬂect one’s disposition toward life, and shall be
operationalized as a composite score on a scale developed speciﬁcally for this study and
based on the work of Cooper-Lewter and Mitchell (1996).
Level of Acculturation

The extent of one’s orientation towards African American culture. It shall be
operationalized as a composite score on the African American Acculturation Scale —
Revised (AAAS-R; Klonof & Landrine, 2000). “Studies have found that levels of
acculturation among African Americans are strongly related to type of coping strategy

used to handle stress...” (Klonoff& Landrine, 2000, p. 237).

 

Protective Factors

“The inﬂuences that modify, ameliorate, or alter a person’s response to some
environmental hazard that predisposes to a maladaptive outcome” (Mrazek & Mrazek,
1987, pp. 358-359). Three such factors — core convictions, spirituality and religiosity are
under consideration in this study.
Risk Factors

“The elements in a person, family, community, or culture that may conduce to the
development of maladaptive behaviors” (Garmezy, 1993, p.127), or negative,
unfavorable, or stressful outcomes. Although risk factors are an important element in
resiliency processes, this study will focus on the protective inﬂuences in the relationship
between attitudes that reﬂect resilience and coping responses. Students who were
admitted to Michigan State University under the College Achievement Admission
Program (CAAP) will be considered “at-risk.”
9le

”The patterns of beliefs, values, and commitments, as well as expected behaviors,
resources, and so forth, that shape individual behavior” (Aldwin, 1994, p. 192). Culture
shall be operationalized as the patterns of belief, values, commitments, and behaviors of
the students of color that participated in this study.

Overview of theﬁStudy

Despite the strong interest and increased research, resilience remains a poorly

understood phenomenon. This study will seek to advance our understanding by

investigating attitudes that reﬂect resilience and their relationship to coping among at-risk

and non-at-risk Aﬁican American college students attending a predominantly White mid-
Western institution (PWI).
The research questions that will be the focus of this study are:

(1) Is there a statistically signiﬁcant positive relationship between attitudes that
reﬂect resilience and coping among both “at risk” and “non-at risk” African
American college students?

(2) Do core beliefs, religiosity, and spirituality moderate the relationship between
attitudes that reﬂect resilience and coping?

(3) Is level of acculturation related to core beliefs, religiosity, spirituality, coping
responses and attitudes that reﬂect resilience?

(4) Do attitudes that reﬂect resilience vary by class level, gender, SES, program
participation, and “at risk” status?

(5) Is the relationship between attitudes that reﬂect resilience and coping responses

reciprocal?

CHAPTER II. LITERATURE REVIEW

Introduction

This chapter contains a general overview of the resiliency literature and the
theoretical frameworks that will inform this study, a review of select models of resilience,
and criticisms of the resilience literature. A review of the literature on resiliency in
African American college students will follow. This section will be followed by a review
of the literature on coping among Aﬁican American college students, followed by a
review of the literature on coping and culture. The literature review will end with a
summary Of the ﬁndings that suggests the need for this study and that informs its
methodology.

The main thesis of this study is that there is a signiﬁcant positive correlation
between coping and attitudes that reﬂect resilience among the at-risk and non-at-risk
African American college student population, that religion, spirituality, and core
convictions moderate the relationship between coping and attitudes that reﬂect resilience,
and that level of acculturation is related to coping responses, religiosity, spirituality, and
attitudes that reﬂect resilience.

Overview of the Resilience Literature

Glantz and Johnson (1999) provide a reasonably comprehensive review of the
resilience literature. They indicate that research on resilience is a “formidable” challenge
for anyone to undertake. The greatest challenge in conducting such research is perhaps

integrating and summarizing the highly disparate body of literature.

10

Research in the ﬁelds of prevention and developmental psychopathology
attempted to understand how people cope with and manage stress, the impact of risk and
protective factors, stress resistance, and the development of ego-strength (Fine, 1991;
Garmezy & Masten, 1986; Garmezy, 1993; Luthar & Zigler, 1991; Rutter, 1985, 1987;
Werner, 1989, 1995). This research led to the emergence of the term resilience (Garmezy
& Masten, 1986; Rutter, 1987). Developmental Psychopathologist Norman Garmezy
coined this term in the early seventies (Glantz & Johnson, 1999).

The term “resilience” generally refers to the capacity to spring back or rebound
from setbacks, stress, or trauma, and to successfully cope and adapt in the face of
adversity. It is generally viewed as a multidimensional and contextual phenomenon
(Glantz & Johnson, 1999). It has even been suggested that we replace the term
“resilience” with the phrase “to cope with,” suggesting that resilience is fundamentally a
question Of how one copes with a given situation (Kumpfer, 1999).

In education, the social and behavioral sciences, and medicine, resilience has been
used synonymously with such terms as invulnerability, hardiness, thriving, stress-
resistance, competence, and ego-strength (Garmezy, 1993). In the social science
literature in particular, resilience has been associated with stress management,
adjustment, adaptation, and coping. Contemporary research suggests that resiliency
varies with a number of factors including age, gender, and social environment (Glantz &
Johnson, 1999).

Early studies have tended to treat resilience in individuals as a personality trait,
and focused on it in response to risk as opposed to in response to protective inﬂuences.

Researchers have more recently begun to focus on the idea of resilience as a process

11

(Freitas & Downey, 1998), and on the role of speciﬁc protective inﬂuences and factors
(Rutter, 1987). With the changing paradigms in the social sciences towards a positive,
multi-culturally-informed psychology has come a shift in the direction of research
towards the inﬂuences of culture on resilience.

In general, continued research is needed in order to advance our understanding of
protective inﬂuences on attitudes that reﬂect resilience and how this concept best applies
to African American populations.

Why th_e Growing Interest in Resilience?

Ickovics and Park (1998) conclude that much theory and research in psychology
focuses on mental illness, pathology, and the negative factors that contribute to
pathological processes. Very little theory and research focuses upon personality strength,
mental health, and the factors that contribute to psychosocial and emotional well-being.
The consequence is that we have developed a psychology that is oriented towards
pathology, human weakness, disease, and deﬁciencies. Persons of color have,
particularly been negatively affected by such deﬁcit models.

Within the past three decades, however, the ﬁelds of medicine and the social and
behavioral sciences have experienced a paradigm shift from the disease model of human
health towards a more positive model that emphasizes strengths and assets (Ickovics &
Park, 1998). This shift in paradigms has resulted in new directions for research and
practice in the ﬁeld of psychology. Instead of a strict focus on pathology and human
weakness, psychologists in research and practice are encouraged to make the effort to

identify, focus, and build upon human assets and strengths. The shift towards a, more

12

positive psychology is also reﬂected in the study of resilience and in the literature

regarding services to African American populations.

Relevant Theoretical Frameworks that lnforrn this Study

As already noted, resilience has been conceptualized broadly and inconsistently
throughout the literature (Glantz & Johnson, 1999; Kinard, 1998). It has been
conceptualized most often as a personality trait with an emphasis on risk factors that help
to shape this trait. The more recent literature portrays ecological and cognitive /
behavioral conceptualizations of resilience (Freitas & Downey, 1998). In addition, there
is a plea for greater attention towards identifying and studying the protective factors that
inﬂuence resiliency process (Barbarin, 1993a,1993b; Kumpfer, 1999; Luster & McAdoo,
1994; Rosen, 1982; Rutter, 1987, 1985).

Below, I will discuss the following theoretical frameworks that inform this study:
Cognitive Appraisal Theory of Coping and Resilience, the Wolin and Wolin Theory of
Resilience, Systems Theory, Afrocentric Theory, and Life-Span Developmental Theory.
Cogpg'tive Appraisal Theory of Coping and Resilience

Jew, Green, (and Kroger (1999) developed a theory of resilience that was informed
by the Cognitive Appraisal Theory of Stress and Coping by Lazarus and Folkman (1984).
Their pioneering work may indeed be a signiﬁcant contribution to the study of resilience
as it offers researchers a simple to understand model that lends itself well to empirical
testing.

Lazarus and Folkman (1984) deﬁne stress as a multidimensional phenomenon that

manifests itself variably across situations. There is good stress and there is bad stress.

13

Good stress can be motivational or inspirational, and can be welcomed as a challenge.
The bad stress usually shows up in the form of anxiety, nervousness, and worry. It can
also manifest itself as physical and mental illness. The assessment that a situation is
stressful depends on our appraisal of that situation. Lazarus and Folkman (1984) offer a
framework for conceptualizing our response to stress in the form of a theory called the
Cognitive Appraisal Theory of Stress and Coping.

The Cognitive Appraisal Theory of Stress and Coping was suggested in the work
of Jew, Green, and Kroger (1999) as a framework for explaining the emergence of
resilience. Although Jew, Green and Kroger did not give the details of how this theory
would work in their published studies, they deﬁned resilience as a system of beliefs.
According to cognitive theory in psychology - thoughts, cognitions, and beliefs can
inﬂuence or direct behavior. As such, beliefs or attitudes that reﬂect resilience may help
determine behaviors that reﬂect resilience. These behaviors were identiﬁed as a set of
skills and abilities in Mrazek and Mrazek’s (1987) early work on resilience in maltreated
children.

Jew, Green & Kroger (1999) used the Cognitive Appraisal Theory Of Stress and
Coping (Lazarus & Folkman, 1984) to inform their conceptualization of resilience. This
theory assumes that (1) a person perceives or experiences stress in his or her environment,
(2) appraises the situation, (3) decides upon a coping response (or inaction, depending on
their cognitive set) based on the appraisal, previous learning, values, beliefs, etc., and (4)
reappraises the situation to see if additional action is needed to eliminate or reduce the

stress.

14

This framework says that as an individual experiences stress in his / her
environment, cognitive processes are triggered that start the person appraising their
situation. The appraisal process occurs in two steps: primary appraisal and secondary
appraisal.

During primary appraisal processes, the person makes one of three choices about a
stressful or adverse situation: (1) appraisal of situation or stressor as irrelevant - an
encounter with the environment carries no implication for one’s well-being, (2) appraisal
of the stressor or adverse situation as benign-positive - this happens when the outcome of
an encounter with stress or adversity is positive, as in personal growth, and (3) appraisal
as stressﬁrl, which includes the realization of (a) actual or perceived harm or loss, (b) a
threat or pending negative consequence, or (c) challenge — a situation in which the person
is able to mobilize coping efforts and manage stress with a resulting increase in
knowledge and skills. The chosen response has important implications for adaptation and
adjustment.

Once the person has completed the initial appraisal of the situation and identiﬁed
some possible outcomes, and options, he/she can move to a second level of appraisal in
which the person must decide what is to be, or can be done about the situation. In most
instances, the person realizes that something must be done to manage the situation and
the associated stress / anxiety. What a person chooses to do is determined by a host of
factors such as past learning, available resources, values, beliefs, emotions, etc.

The primary and secondary appraisal processes interact with each other in shaping
the degree Of anxiety experienced by the person and determining the quality of their

reaction or cOping response. Ideally, the older and more mature individual has more life

15

experience, and hence, should also have a broader range of coping responses. Variations
in coping ability may suggest either strength or vulnerability. Circumstances that extend
beyond a persons control as well as individual variations in temperament help determine
the coping responses. Once a response is decided upon, the individual may reappraise the
situation and decide if further action is needed to thwart any further danger or harm to
self.

Cognitive Appraisal Theory appears to be an ideal framework for conceptualizing
stress, coping and cognitive aspects of resiliency processes. However, unless path
analysis is done to assess the direction of these relationships, we cannot make (causal)
inferences about this relationship. According to Jew (personal communication, October
23, 2000) no such analysis has been done on their model of resilience.

_T_'_h_e Wolin and Wolin Theory of Resilience

Wolin and Wolin (1993) developed a theory of resilience based on clinical
interviews with twenty-ﬁve clients from their private practice. According to their theory,
resilience develops over time and is the ability or capacity that emerges from the
individual in the form of skills. These skills become “lasting strengths” and become a
ﬁxed part of the self. These skills are called resiliencies. They include: insight,
independence, relationships, initiative, creativity, humor and morality.

These skills and abilities tend to “cluster” by personality - different personalities
have a different set of resiliencies. Wolin and Wolin (1993) posit that people learn to be
vulnerable to life stress and problems, and that few people have all seven resiliencies.
They believe overcoming vulnerability is a matter of learning to think and behave

differently. According to this theory, you can “uncover” or “discover” your resilient self

16

simply by reading about the success of survivors. Wolin and Wolin credit psychology’s
“preoccupation with pathology and problems” as the culprit in making people vulnerable
to stress. The Wolins believe that people learn and are conditioned not to be resilient.
The Wolins developed the “Damage Model” to help explain the etiology of vulnerability
and its resultant pathology upon individuals. This pathology manifests itself in the form
of pathological thoughts, feelings and behaviors. The “Damage Model” is only
theoretical and is not empirically supported.

The counter to the Damage Model is the “Challenge Model.” This model explains
how people rebound ﬁ'om adversity. It too, is theoretical and lacks empirical support.
The Challenge Model is an attempt to explain the etiology of vulnerability and resilience.
It says that vulnerable people must reframe their view of life if they are to become
resilient. Wolin and Wolin (1993) contend, “By reading this book, you can broaden your
awareness of your strengths and retrieve lost memories of the times that you successfully
rebounded from pain (p. 18).” The key is to change your thinking from victim to
survivor. Wolin and Wolin cite “case reports” as research evidence for the Challenge
Model (p. 18-19). They also contend that a number of experiments on resilience have
added to their understanding of how survivors overcome “formidable stumbling blocks”
(p.18).

Biscoe and Harris (1994) developed a scale that purports to assess attitudes that
reﬂect the seven skills and abilities identiﬁed in the work of Wolin and Wolin (1993).
Items in the six subscales of the Resiliency Attitude Scale (RAS) were written to
represent the seven resiliencies: insight, independence, relationship, initiative,

creativity/humor, and morality (in the RAS, creativity and humor were combined and

17

treated as one resiliency). A seventh subscale was developed by Biscoe and Harris and
included in the RAS as a measure of general resilience. The “General Resilience”
subscale represents skills that reﬂect persistence at working through difficulties, and is
deﬁned as the belief that one can survive and make things better for oneself and for others
(Biscoe & Harris, 1994). The RAS is yet under development. Findings from this study
may help provide additional validity and reliability data for this scale.

Systems Theog

Systems theory (Bowen, 1978; Minuchin, 1974) represents an ideal framework for
understanding resiliency. Systems theory asserts that behavior may be understood in the
context of the dynamic interaction between the person and his or her environment.
Behavior takes on meaning within the context of systems that overlap, converge and
diverge. Although no body of literature has speciﬁcally stated that this framework
should be used to conceptualize resilience, it has been implicated in the work of Kumpfer
(1999) and others who posit ecological models of resiliency.

Freitas and Downey (1998) call for a dynamic, systemic conceptualization of
resilience that helps explain coping in children across the different stages and domains of
development throughout life (e.g. relationships with signiﬁcant others, academics, etc.).
They suggest a shift in focus from striving to identify resilient traits in a person to
identifying systems that foster resilient outcomes. The church, the family, the school,
and the home each represent such systems (Billingsley & Caldwell, 1991; Boyd-Franklin,
1989). This study may inform us about the church and religion as important, resiliency

fostering systems.

18

Afrocentric Theory WM

Much has been written in recent years regarding the need to utilize culture-
speciﬁc approaches in research (Schiele, 1996; Swigonski, 1996). For example, Asante
(1987, 1988) has been the leading proponent of using an Afrocentric approach in research
and theory construction. Afrocentricity (Asante, 1987) may be deﬁned as “placing
African ideals at the center of any analysis that involves African culture and behavior.”
(p.6).

Dana (1993) and Ponterotto and Casas (1991) have also spoken to the issue of
culture-Speciﬁc research and how this approach enhances our understanding of a
particular group. These authors speak of the emic and etic approaches to research. An
emic approach is one that studies a culture from within that cultural perspective. An etic
approach does the opposite and tends to be comparative in nature.

Authorities in the ﬁeld suggest that researchers develop and execute studies, as
well as interpret research results using greater sensitivity to, and consideration for unique
cultural and ethnic differences (Jones, 1991 ). This approach has been used to help build a
Black psychology that includes a value for the oral tradition (the transmission of
knowledge orally), communalism (a valuing of the group more than the individual), affect
(the integration Of feelings with thoughts and actions) and spirituality / religiosity (the
power of God in the matters of this life)(Jones, 1991).

One of the aims of this study was to inform our understanding of the relationship
between attitudes that reﬂect resilience and coping behavior in African American college
students. The moderator variables identiﬁed in this study have been cited in the literature

as important aspects of African American culture and ethos (Cooper-Lewter & Mitchell,

19

 

1996; Donahue & Benson, 1995; Ellison, 1993; Ellison & Day, 1990; Frazier, 1964;
McCall, 1986; Mitchell, 1975; Moore, 2000; Ventis, 1995; Washington, 1964). Level of
acculturation was included in this study to help examine the salience of religion and
spirituality in the sample. Accordingly, Afrocentric Theory may help inform the
investigation of religion, spirituality, and core beliefs as protective inﬂuences in the lives
of African American college students, and may help explain the salience of religion as an
aspect of acculturation.
Life-span Developmental Theory

Glantz & Johnson (1999) Offer a critical review of the resiliency literature and
conclude that perhaps the best model for conceptualizing resilience as a process is a life-
span developmental model. A life-span developmental model of resilience assumes that
resilience manifests itself differently at different stages of development. Life-span
developmental theory appears to offer a reasonable framework for considering some of
the salient issues that college students might experience. Such concerns might include
relationship issues, self-esteem, separation-individuation issues, dependency issues,
gender-identity issues, racial and ethnic identity issues, and career choice. Religion and
related concerns may recur at any stage of development along the life cycle. The way
religious issues manifest themselves and how they are used in coping will also vary with
stage Of development. The ﬁnding that religion and core convictions improve coping,
and resilience in both groups of college students might suggest that religion serves an
important function in early adulthood and late adolescence (Oler, 1997; Cooper-Lewter &

Mitchell, 1996).

20

Models of Resilience

In addition to the theoretical frameworks presented above, several models of
resilience have been identiﬁed and demonstrate that an attempt is being made to expand
our understanding of resilience. However, many models of resilience still equate it with
the mere possession of select protective factors (Glantz & Johnson, 1999). Few existing
models deﬁne resilience as a process and most fail to identify and discuss the possible
implications Of ethnicity and culture in manifestations of resilience processes (Barbarin,
1993b; Kumpfer, 1999). Given this, I will discuss the three models most relevant to this
study - ecological, life cycle, and transactional.
Ecological

In a conceptual article on resilience in child maltreatment victims, Mrazek &
Mrazek (1987) posited an ecological model of resilience. Resiliency was conceptualized
as an outcome of the dynamic interaction between personality traits, protective factors,
and risk factors. They identiﬁed twelve personal characteristics and skills that are
believed to foster resilience. These include:
0 Rapid responsivity to danger
- Precocious maturity
o Disassociation of affect
o Information-seeking
- Formation and utilization of relationships for survival
0 Positive projective anticipation
0 Decisive risk taking

0 The conviction of being loved

21

- Idealization of an aggressor’s competence

0 Cognitive restructuring Of painful experiences
0 Altruism

0 Optimism and hope

(Mrazek & Mrazek, 1987, pp. 359-362)

While it is possible, and even plausible, that these characteristics and skills do
foster resilience, this model fails to explain how that process takes place in the individual.
A major strength of this model is that it offers counselors a repertoire of skills that can
translate into goals to accomplish in therapy or in counseling. Jew & Green (1998) used
this model to develop a scale for measuring resiliency. Their scale is yet under
development.

Life-Cycle
Flach (1988) developed the Life-Cycle model, which conceptualizes resilience as

a process. According to this model, a person progresses through eight stages of the

human life cycle. These are:
0 Birth

0 Childhood

o Adolescence

0 Young, single adulthood
0 Young marriage

0 Parenthood

0 Middle age
0 Aging

22

Each stage in the life-cycle holds special challenges, characteristics,
responsibilities, and opportunities that require a person to change in order to cope
effectively. Movement from one stage to another occurs as one grows and matures. The
actual movement from one stage to another induces some amount of stress that helps
prepare the person for the next level of development.

Between each of the eight life stages there occurs a “bifurcation” - a point where
we respond to the stress and strain of life. It is during these times (i.e., the bifurcation
point) that we might ﬁnd ourselves struggling from within ourselves either mentally, or
physically (or even spiritually) or outwardly via our actions. Our response indicates the
need to change, giving up attitudes and patterns of behavior that are outmoded, or even
harmful. These are replaced by more mature ways of thinking and behaving that are
consistent with growth and development along the lifecycle. Although the actual
conditions per life stage differs from the other stages, there are speciﬁc universal issues
that recur each time the individual passes through a bifurcation point. These include:

o The need to adapt to changing external circumstances.

0 Reconsideration of one’s self-image, and holding on to and restoring self-esteem.
o Forming new human relationships and renewing Old ones.

0 Giving up people and things we love because they die or go away.

0 Balancing our ability to be independent against our need for others.

0 Redeﬁning or reaffirming our purposes in life.

F lach (1988) identiﬁes fourteen traits that he believes contributes to one’s
resiliency. Attitudes that reﬂect resilience are implied in the traits. These traits include:

0 A strong, supple sense of self esteem

23

 

Independence of thought and action, without fear of relying on others or a reluctance
to do so

The ability to give and take in one’s interactions with others

A well-established network of personal friends, including one or more who serve as
conﬁdent

A high level of personal discipline

A sense of responsibility

Recognition and development of one’s special gifts and talents

Open-mindedness and receptivity to new ideas

A willingness to dream

A wide range of interests

A keen sense of humor

Insight into one’s own feelings and those of others, and the ability to communicate
these in an appropriate manner

A high tolerance of distress

Focus - a commitment to life, and a philosophical framework within which personal
experiences can be interpreted with meaning and hope, even at life’s seemingly most

hopeless moments (F lach, 1988, pp. 113-114).

mm.

Glantz & Johnson (1999) posit a transactional model in which resiliency is the

outcome of a dynamic interaction between (1) environmental precursors, commonly

called risk and protective factors, (2) personal characteristics of the resilient person, (3)

reintegration or positive outcome after a negative life experience, and (4) dynamic

24

 

processes that mediate between the person and their environment, and the person and the
outcome (e.g. effective coping). Their model parallels the Flach’s (1988) model, each of
which offer a reasonable framework for exploring the relationship between culture, ethnic
identity, and resilience.

Finally, Kumpfer (1999) organized variables found related to increased resilience
into a dynamic framework that allows for interactions between resilient people and their
high-risk environments. She suggests that resilience may not be a single construct as it is
Often modeled throughout the literature. Instead, it might be a complex of related
processes.

Summm

Cognitive Appraisal Theory, the Wolin and Wolin Theory of Resilience, Systems
Theory, Afrocentric Theory, and Lifespan Developmental Theory may be used to inform
our approach to, and understanding of resilience. In particular, these theories help
explain the importance Of the variables selected as moderators in this study, and they help
provide a framework for understanding the relationships and interactions between these
variables and the relationship between attitudes that reﬂect resilience and coping
responses. The majority of resiliency models suggest, as seen in this review, that
resilience is a complex phenomenon that emerges from the dynamic interaction between
personality, environmental, risk and protective factors. Comparison of protective factors
across models reveals some commonality between models. Each of the models identiﬁes
the importance of family, social and community support and maintaining attitudes that
reﬂect resilience. They also suggest important links between resilience and coping, and

culturally salient factors that may serve to enhance resilience and coping. However, we

25

are yet in need of additional research to ﬁll the gaps in knowledge that these models do
not explain (e. g. the empirical measurement of resilience).
Major Criticisms of the ResiliengLiteraJure

Despite over thirty years of research on this phenomenon, we cannot deﬁnitively
explain resilience. We are not sure if it is a ﬁxed or mutable personality trait or a process.
Although we have some ideas of factors that may contribute to, or detract from the
development and manifestation of resilience, we have not yet identiﬁed a sure mechanism
for fostering it. A critical review Of the literature reveals that the construct resilience has
been operationalized and measured disparately throughout the literature (Glantz &
Johnson, 1999).

In the existing resilience literature, the measurement of resilience is often
accomplished by operationally deﬁning resilience as scores on already existing scales that
were designed to measure different but conceptually or theoretically related constructs.
For example, Chesley-Carter (1998) operationalized resilience as a score on a measure of
social competence. Such an operationalization of resilience merely captures one aspect of
it, and as such, may inadequately represent the phenomenon. In another study of the
relationship between resilience, social competence, coping ability, and gender, resilience
was operationalized as a score on the California Child Q-sort (Mandleco, 1991).

Studies that utilize empirically validated and reliable measures of resilience may
represent a more signiﬁcant contribution to the literature. Several instruments designed
speciﬁcally for quantifying resilience were identiﬁed (Ahn, 1991; Biscoe & Harris, 1994;
Jew, Green & Kroger, 1999; Kelso, 1999). However, these scales are still under

development, and the statistical validity and reliability of these scales are still in question.

26

Although a host of new and promising developments have occurred, a number of
problems yet plague resiliency research. Critique of the resilience literature yet ﬁnds (1)
ambiguity in deﬁnitions of risk and protective factors, (2) the lack of a unifying
conceptual framework throughout the literature, therefore making it nearly impossible to
integrate ﬁndings across studies, (3) the tendency of investigators to use models that are
linear and unidirectional, with static variables as opposed to using more complex,
recursive and bi-directional models, (4) excessive attention to problem behaviors and
negative life conditions as either outcome or predictor, (5) inadequate use of multivariate
approaches in conceptualizing or measuring models, (6) excessive reliance on inferred
measures of resilience, and (7) few if any, attempts to identify important and salient
sociocultural factors that may foster resilience in people Of color. Perspectives in
multicultural counseling and research beg the question of whether resilience in one
culture or race is the same as resilience in another (Genero, 1995; Racklin, 1999).
Below, I will comment on each of these criticisms.

Ambiguitv in deﬁnitions of riskyand protective farctors

A major problem in the research on resiliency is that deﬁnitions of risk and
protective factOrs, as well as deﬁnitions of resiliency, are varied and ambiguous
throughout the literature. This variation and ambiguity makes it difﬁcult to integrate
results across studies. Often, what might be considered a risk factor in one study may be
deﬁned as a protective factor in another.

In addition, little attention is paid to how speciﬁc risk factors might cancel out the
effects of select protective factors. For example, Barbarin (1993b) very poignantly points

out that poverty has the effect of canceling the protective inﬂuence of select protective

27

factors in the lives of some at—risk populations. Often, this type of effect is not taken into
consideration in studies of resilience in at-risk populations.
The Lack of a Unifying Conceptual Framework

It is difﬁcult to ﬁnd two or more studies of resiliency that draw upon a single
(common) psychological theory. Several theoretical frameworks have been used to
conceptualize resilience (Freitas & Downey, 1998; Glantz & Johnson, 1999). Within
some of these framewOrks, resiliency has been modeled as an immutable or ﬁxed trait
(Mrazek & Mrazek, 1987), as a belief system (Jew, Green & Kroger, 1999), and as the
dynamic interaction between risk factors, protective factors, and stress (Rutter, 1987,
1985).
IlraTendency to Use Models that are Linear and Unidirectional

Perhaps due to the relative novelty of this topic, and the potential complexity of
statistically analyzing such models, few studies in the research literature Offer non-linear,
bi-directional models of resiliency. Fortunately, technological advances in computerized
analysis of data may allow for the development of more complex models of resiliency in
the future. Such a model was examined in this study.
Excessive Attention to Problem Behaviors and Negative Life Conditions

Excessive attention to problem behaviors and negative life conditions is a
pervasive feature of much research in psychology. Until recent, the literature has tended
to emphasize risk and resilience as well as resilience amid negative life conditions and
problem behaviors. The shifting paradigms in the social sciences towards health and

wholeness, and the move towards prevention over remediation seem to hold the most

28

promise for developing research studies that emphasize strengths, protective inﬂuences
and resilience.
Inadequate Use of Multivariate Approaches in Conceptuﬂigipggﬂmmgﬂgdeis

Recent advances in technology have made the use of multivariate methods and
data analysis less cumbersome. Our understanding of resiliency can be greatly enhanced
through the use of more complex multivariate models of resiliency. Use of multivariate
analysis is recognized in statistics as the most powerful analysis yet available in research.
As we gain a better understanding of resilience we may witness a greater use of
multivariate methods as a tool for extending studies on resilience. This study employed
Structural Equation Modeling, Path Analysis, and Exploratory Factor Analysis.
Excessive Reliance on Inferred Measures

Many studies of resilience infer resiliency from instruments that were designed to
measure different constructs. A more reliable strategy however, would be to infer
resilience from instruments designed to measure aspects of resilience. The measure used
in this study is an inferred measure of resilience using an instrument designed speciﬁcally
to assess attitudes that reﬂect resilience.
Little Attention to Important and Salient Sociocultural Factors that Foster Re§_i_lie_nc_§
among People of Color

Little resilience research focuses on the role of salient social and cultural factors
in resilience (Barbarin, 1993b; Genero, 1995). The multicultural movement in
psychology and Afrocentric Theory are now urging researchers, counselors, and clinicians
to consider the inﬂuence of culture, ethnicity, social class, gender, and other important

differences on psychological processes and constructs (Pedersen, 1987; Asante, 1987).

29

 

Resilience literature has paid little attention towards exploring the cultural and
ethnic fairness and appropriateness of instruments designed to measure resilience
(McCubbin, Thompson, & McCubbin, 1995). The Resiliency Attitudes Scale (Biscoe &
Harris, 1994) is being piloted on a sample of African American college students in this
study. The results from this study may suggest the appropriateness of using this scale
with this population.

Summary

There is clearly a need for attention to these concerns in resilience research.
Given the advances in our understanding of resilience, and the developments in research
methodology now available, it is conceivable that many of these problems can be
redressed.

Resiliencmd Protective Faptors

As previously mentioned, until recent, much resilience research has tended to
focus on resilience amid risk. The more contemporary trend however, is towards the
study of resilience and protective inﬂuences. Kaplan, Turner, Norman & Stillson (1996)
identify twenty protective factors they believe may contribute to the development of
resilience. These factors are arranged into four categories. They include:

(A) Individual attributes: (1) easygoing temperament or disposition, (2) intellectual
capabilities, especially verbal and communications skills, (3) self-efﬁcacy, (4)
realistic appraisal of the environment, (5) social problem-solving skills, (6) sense
Of direction, or mission, (7) the capacity to understand and respond to other’s

feelings, (8) humor, (9) adaptive distancing;

30

(B) Family Protective Factors: (10) consistent, warm, positive relationships with a
caring adult, (1 1) positive family environment and bonding, (12) high but realistic
parental expectations, (13) family responsibilities and household tasks, (14)
positive parental modeling of resilience and coping skills, (15) extended support
networks, including family and friends,

(C) School Protective Factors: (16) Opportunities for involvement in school decision
making, (17) high but realistic expectations for student academic performance,
(18) caring, supportive school atmosphere, and

(D) Community Protective Factors: (19) positive community norms, and (20)
Community resources (pp. 159-160).

Among these four categories, no speciﬁc mention was made of cultural/ ethnic factors
that might contribute to the development of resilience. The literature suggests that the
Black church is an important institution in the Black community, and that it has played an
important role in the coping behaviors of African American people (Logan, 1980; Oler,
1997; Richardson & June, 1997). The strong sense of faith in God and spirituality that is
fostered in the Black church is believed to have an important protective inﬂuence amid
crisis and adversity. This study will examine the role of religious faith and spirituality in
the relationship between coping and attitudes that reﬂect resilience in African American

college students.

31

Resiliency in AﬁcMmL—Wgtﬂdgta

Few published studies exist that speciﬁcally focus on resiliency in African
American college students. Thompson (1998) and Ford, Kokjie, and Lewis (1996)
represent two such studies. The focus of each Of these studies is on identifying factors
that may have a protective role in coping with stress, and that may contribute to
resilience.

Thompson (1998) conducted a quantitative / qualitative study of predictors of
resilience in African American college students and adults. The quantitative half of the
study was based on survey data taken from African American college students and the
qualitative portion of the study was based on interview data taken from African American
adults. Findings from this study suggest that college students who experienced poverty
during childhood were less likely to consider themselves to be as resilient as other
African Americans their age. This ﬁnding further substantiates what we already know
about the deleterious and pervasive inﬂuence of poverty on attitudes that reﬂect resilience
(Barbarin, 1993b). Further research is needed in order to identify important protective
inﬂuences that may offset the negative effect of similar types of risk factors.

Ford, Kokjie and Lewis (1996) attempted to identify factors that are believed to
contribute towards academic resilience among urban “at-risk” African American male
college students. Academic resilience was deﬁned as “the ability to manifest high
academic competence despite high stress” (p. 19). Participants in this study were African
American males who self-reported exposure to high levels of stress. The psychosocial
variables examined included emotional intelligence quotient, bicultural behaviors, locus

of control, and social support. The sample consisted of 104 African American male

32

 

college students enrolled at a Historically Black college. Ten percent of the sample
(approximately ten participants) was from a predominantly White institution. The
authors found that emotional intelligence quotient was a signiﬁcant contributor to
academic resilience.

In addition to studies that try to identify factors that contribute to the development
of resilience, training programs have been developed that aim to foster resilience. A
regional TRIO association comprised of personnel ﬁ'om the States of Colorado, Montana,
Wyoming, Utah, North Dakota, and South Dakota formed a group named the Association
of Special Programs in the Region Eight (ASPIRE). ASPIRE developed a series of
professional development trainings that target TRIO programs. This group developed a
training program that links retention of high-risk college students with the ability of the
TRIO staff to foster attitudes that reﬂect resilience. They contend that resilience is
related to retention via a demonstrated commitment to students. The focus Of
interventions with students then, is to facilitate the development of the seven resiliencies
cited in Wolin and Wolin (1993). An eighth resiliency — spirituality was added to the list
of attitudes that reﬂect resilience that should be fostered among TRIO students and staff.
The measure that was used in this study to assess attitudes that reﬂect resilience (Biscoe
& Harris, 1994) is derived from the model of resilience posited in Wolin and Wolin
(1993). While not the main focus of this study, data from this study may better inform
TRIO staff and students who are often labeled as “at-risk”, about these resiliencies. In

addition, this study may add support to the notion of spirituality as an eighth resiliency.

33

Coping, Culture and Theories of Coping

As previously mentioned, levels of acculturation among African Americans are
strongly related to the type of coping strategy used to handle stress (Klonoff & Landrine,
2000). This suggests an important link between culture and coping. This section will
focus on coping and culture, and will touch on aspects of coping theory and research that
are relevant to this study.

Coping is recognized as a critical barometer of mental well-being and human
functioning (Aldwin, 1994). As an aspect of normal functioning, coping alone is or can
be the telltale Sign of adjustment difﬁculties. Coping behaviors then, are important
indicators of well-being and health. Coping constitutes a major category of psychiatric
and psychological assessment in many clinical and counseling settings, and it is an
important aspect of assessment in many educational and medical settings (Aldwin, 1994).
It is recognized as the mediator of stress and can serve as either an outcome or a predictor
of an outcome (as in this study). The literature documents quite well the fact that coping
Varies along a developmental and chronological trajectory (Aldwin, 1994). This means
that coping varies with age, personality, stage-of-life, mood, mental state, sex, social-
economic status and many other factors. A major and yet unresolved area of coping
research has been towards understanding the role and importance Of culture and coping
(Aldwin, 1994). There is support in the coping literature for the existence of cultural
variations in coping.

Jenkins (1995) argues that African American people may indeed have a different
way of approaching problems than the majority culture. In addition, a “folk psychology”

seems to have emerged among African American scholars in the social sciences that seem

34

 

to support this idea. The purpose of this folk psychology is to help explain the culturally
different ways African Americans might approach problems, learn, socialize, adjust and
adapt.
Jenkins (1995) states:
Like most other psychologists, I am convinced that the coping characteristics are
what must be developed if the human race is to survive and fully develop its
potential. More speciﬁcally, however, the theoretical trends that emphasize
coping and mastery seem to me to be sorely needed when choosing a

psychological approach to the study of the African American (p. ix).

Jenkins (1995) contends that African American people have their own “folk
theory of success” that may be likened to a “folk theory of coping”. He further states that:
Ogbu (1986) suggests there is another social background inﬂuence on the
educational progress of Black children. He suggests that achievement for any
group is to an important degree a function of those groups’ perceptions of the
opportunity structure based on its history of experiences with that structure.
These experiences lead to a collection of beliefs about ‘status mobility’, about
what one has to do to get ahead. He calls these community beliefs a “folk theory
of success (pp. 126-127).
Aldwin (1994) argues that less attention has been paid in the coping literature to
the more general ways in which culture can inﬂuence the use of coping strategies. She
argues further that culture provides explanations for phenomenon such as death, life,

prosperity, accidents and illness, and that these cultural explanations for the life events

35

 

guide and direct coping responses. This phenomenon can be seen in the African
American community in our explanations of our history.

Aldwin (1994) states “Coping in a non-culturally prescribed manner may result in
greater stress.” (p. 210), and that “bicultural individuals may develop two separate coping
repertoires, depending upon the cultural context” (p. 211). She uses as an example
differences in coping responses between ﬁrst and second-generation international people
(e.g. Japanese and Japanese Americans). Aldwin cites research that supports the ﬁnding
that “coping strategies are inﬂuenced not only by cultural beliefs concerning the most
appropriate means of handling speciﬁc types of problems, but also by social and cultural
institutions for problem-solving and tension reduction.” (Aldwin, 1994, p. 211).

According to Aldwin (1994), the relationship between culture and individual
coping is not unidirectional — rather, individual behaviors can reinforce or change existing
patterns of coping. Aldwin (1994) concludes:

(1) “culture affects individual’s coping behaviors, and coping behaviors

can affect their culture.”

(2) “Psychological research has neglected sociocultural inﬂuences on the
stress and coping process.”

(3) “Cultures do indeed inﬂuence the appraisal process, either through
beliefs and values prevalent within the wider cultural setting or
developed through consensual processes in more speciﬁc social
situations” (p. 215).

In support of examining the role of religion and spirituality in coping and resilience

Aldwin (1994) writes:

36

 

In summary, sociocultural groups appear to generate not only consensual
belief systems concerning the origin and meaning of stressors, but also beliefs
concerning the most appropriate means of both emotion and problem-focused
coping. These beliefs may be situation speciﬁc. Further, trying to cope in ways
that run contrary to the general cultural ethos may increase stress, even though
those same strategies used by members of a different culture may be efficacious in
reducing emotional distress. Even more rarely examined, however, are the more
generalized institutions that cultures provide to individuals in order to help them

cope with problems (p. 211).

Further, cultures may deﬁne certain coping behaviors as more appropriate than
others, and what is deemed appropriate is not uniform within a culture but, rather,
varies as a function of the individual’s position in the cultural mazeway. Finally,
the relationship between the individual and the culture is bi-directional. Not only
do people cope in ways that affect other individuals, but also through the
modiﬁcation or development of new organizations, structures and attitudes, an
individual or group of individuals may affect social change (p. 215).

Cppipg Among African American College Stu_dents

Coping behaviors among African American college students have been examined

in the social sciences literature (see, for example, Jung & Khalsa, 1988; Nero, 1988; Oler,

1997; Rarnbo-Chroniak, 1999; and Turner, 1992). Evidence from this body of literature

shows that religion and church involvement are signiﬁcantly related to coping and

adjustment processes.

37

Jung and Khalsa (1988) found that Black college students reported a higher
frequency of hassles than their White colleagues at predominantly White institutions
(PWI’S). Murphy and Archer (1996) examined stressors facing African American college
students at PWI’s and found that the number and nature Of stressors facing college
students is changing and even increasing. This change has implications for doing
preventive and remedial work. According to Murphy and Archer, if we identify the
coping strategies being used by African American students early enough, as well as the
traits that foster resilience, we may foster them in the students as a way of preventing
mental illness and drop-out.

In a conceptual paper on the coping strategies of resilient African American
adolescents, Turner (1992) hypothesized that African American adolescents achieve
because they possess intrinsic values such as persistence and patience. She suggests that
African American adolescents are socialized and taught that academic achievement is
paramount. She also suggests that the coping behaviors of African American college
students enable them to achieve academically and that most of these strategies are
culture-speciﬁc.

Neighbors, Jackson, Bowman and Gurin (1983) conducted a study examining the
role of religion in coping and found that prayer was an extremely important coping
response used by African Americans, and especially among those earning less than ten
thousand dollars a year, and who are above the age of ﬁfty-ﬁve and female. They also
found that an informal social network was used quite extensively as a means of coping
with problems. African American people tend to use (1) the emergency room, (2) private

physicians, and (3) ministers most frequently when seeking professional help.

38

 

Neville, Heppner, and Wang (1997) investigated the relationship among racial
identity attitudes, perceived stressors, and coping styles in African American college
students attending predominantly White institutions. They found that speciﬁc racial
identity attitudes were statistically signiﬁcant predictors of both general and culture-
speciﬁc stressors, and that the immersion-emersion racial identity status was a unique
predictor of general perceived stressors and problem solving. These results suggest the
importance of racial identity as a critical factor in predicting stress and coping responses
in African American students at predominantly White institutions. These results further
suggest the need to assess the relationship between level of acculturation and coping in
African American college students.

Logan ((1980) examined the role of the Black Baptist Church as an adaptive force
among active African American church members and found that religion serves an
adaptive function among African American church attendees. The Logan study also
found that involvement in the Black church was an important aspect Of (1) African
American culture and (2) coping among African American church attendees.

Oler (1997) investigated the relationship between religious coping and
psychological adjustment among African American Christians. The participants in her
study consisted of church members from several urban African American churches. She
found that spiritually-based coping was positively associated with self-esteem, life-
satisfaction, lowered incidence of depression, and satisfaction with response to negative
life-events. “Religious coping added signiﬁcant variance to general ways of coping in the
prediction of higher self-esteem and lower depression” (p. iii). “Spiritually-based coping

was most strongly and consistently related to positive psychological adjustment and

39

satisfaction with how one responded to a negative event” (p. iii). Most importantly, Oler
found that religious coping activities facilitate psychological adjustment “beyond the
contribution Of general ways of coping” (p. iii).

Giddan (1987) discusses coping and identity development in college students in
general. He asserts that effective coping is part of the psychological, social, and
intellectual development that occurs during late adolescence and early adulthood. Coping
helps college students solve problems that are relatively difﬁcult and challenging, or to
accept unchangeable circumstances; “. .. to cope also means to live with discomfort or
distress connected with problems that cannot be solved” (p. 34). Key coping mechanisms
include access to information, task competence, use of the past, humor, predictability,
control, self-Observation, ﬂexibility, and stability. Some of these coping skills and
abilities are similar to the ones identiﬁed as leading to the development of resilience.

Summary

There is a need to further examine the coping behaviors of African American
college students attending PWI’s. Of particular interest would be the identiﬁcation of
culturally salient patterns of coping. This must be done using instruments that are
sensitive to the different ways of coping used by this population.

Jew & Green (1998) conducted a study of the effects of risk factors on coping and
resilience using a cognitive-behavioral conceptualization of resilience, and a study (Jew,
Green & Kroger, 1999) on the relationship between resilience and coping among
adolescents. They hypothesized in each of these studies that resilience is a belief system,
and as a belief system, it is causally related to coping (i.e. coping emerges from

resilience). However, no data exist to support or refute their hypothesis.

40

 

A major limitation of the Jew, Green and Kroger (1999) study was that like
previous resilience research, it focused upon resilience and risk. Examining the role of
protective inﬂuences and resilience as well as the role of culture in resilience and coping
may extend their work.

Clearly, additional research is needed to either empirically refute or substantiate
the relationship between coping and resilience hypothesized in Jew, Green & Kroger
(1999).

Further, religion, core beliefs, and level of acculturation have been implicated in
the review literature as being important factors in coping and adjustment behaviors
among African Americans, and there is some empirical evidence supporting the notion of
a relationship between resilience and spirituality (Massey, 1999). However, there is need
for additional research to examine these relationships in college students.

Using a correlational design, this study will examine each of the claims above and
seek to provide greater clarity to the unanswered questions.

Research Questions
The research questions that will be the focus of this study are:

(1) Is there a statistically signiﬁcant positive relationship between attitudes that
reﬂect resilience and coping among both “at risk” and “non-at risk” African
American college students?

(2) Do core beliefs, religiosity, and spirituality moderate the relationship between
attitudes that reﬂect resilience and coping?

(3) Is level of acculturation related to core beliefs, religiosity, spirituality, coping

responses and attitudes that reﬂect resilience?

41

(4) Do attitudes that reﬂect resilience vary by class level, gender, SES, program
participation, and “at risk” status?
(5) Is the relationship between attitudes that reﬂect resilience and coping responses

reciprocal?

42

CHAPTER III. METHODS

meat

The purposes of this study were (1) to examine the relationship between attitudes
that reﬂect resilience and coping among at risk and non-at risk African American college
students, (2) to examine the moderating role of three factors that are believed to be
protective in times of stress and adversity and (3) to examine the relationship between
level of acculturation and coping responses, attitudes that reﬂect resilience, core beliefs,
religiosity, and spirituality.

This chapter will provide an overview of the research procedure and
implementation, the characteristics of the population and sample, instrumentation and
scale development, the research hypothesis, and a summary of the statistical tests selected

for the analysis of the data.

Procedures

Several programs exist in the University that recruit and serve large numbers of
students that meet the criteria for inclusion in this study (i.e., African American, both at
risk and non-at risk students). These programs were primarily targeted for participants.

The investigator met with the directors of each of the targeted programs and
explained the purpose of the study. Each of the program directors granted full permission
to approach students for participation, and speciﬁc dates and times were arranged for data
collection. Data collection occurred during regularly scheduled program meetings in
which time was set aside or reserved for the actual administration of the survey packets.

~\ \The data collection sessions were kept uniform and lasted an average of forty to ﬁfty

43

minutes. All survey packets were administered and collected by the investigator. Before
administering the surveys, a brief introduction was given to the student participants by the
director Of each program. Students were informed that their participation was voluntary,
and they were given the option of not participating or withdrawing their participation at
any time during the study without penalty. Each survey packet had an attached consent
form / instruction sheet (see Appendix E).

Student participants were instructed to read silently along with the investigator as
instructions were read aloud prior to any participants ﬁlling out the packets. The consent
form included the names of the investigators, the purpose for the study, notice of
conﬁdentiality, information for contacting investigators for questions, and instructions for
completing the survey packets.

A $100 raffle was included as an incentive for ﬁlling out the survey packets.
Students indicated their voluntary consent to participate in the study by reading the
consent form and signing their name. Students were given the option of participating or
not participating in the rafﬂe via a yes or no response to the questions “Yes, I want to
participate in the $100 Rafﬂe” or “No, I do not want to enter the $100 Rafﬂe.” Students
were told at the start of each data collection session that students wishing to participate in
the $100 rafﬂe must complete all instruments in the survey packet. Participants with
incomplete survey packets were not included in the rafﬂe. Survey packets were collected
upon completion, and participants were fully debriefed via group discussion as to the
purpose of this study. A total of 337 survey packets were collected. However, since the

focus of this study was on African American college students attending a predominantly

44

 

White institution, only data from participants who identiﬁed as African American were
used. The sample size for the African American student participants was 229.

Popliuionﬁ and Sample

Participants were drawn from the general student body and from four pre-existing
programs at Michigan State University that had large populations of African American
students. Participants were included in the subject pool if they were enrolled during the
Spring 2001 and Summer 2001 semesters, and participated in any of the four programs.
These were: (1) MAGIC, (2) SUPER, (3) McNair-SROP and (4) Sports Camp. These
groups were also targeted as they contain large numbers of African American students
who represent different years in college and “at risk” and “non-at risk” status. The
intention was to Obtain responses from all of the African-American students in those four
groups, but a number of non-African-American students also responded. Because the
investigator was interested in comparisons only within race, race was controlled for in
this study, and only data ﬁ'om African American students were used in the analysis (see
Ponterotto & Casas, 1991). Below, I will describe the four groups from which much of
the sample was drawn.

College Achievement Admission Progm (CAAP)

A portion of freshmen students are admitted into the University as CAAP students
and are provided support by the Office of Supportive Services (OSS). The Ofﬁce of
Supportive Services (OSS) is a multifaceted holistic retention initiative designed to
address the needs of CAAP students; either ﬁrst generation, low income or educationally

disadvantaged students. Through the collaborative efforts of MSU faculty and staff, the

45

 

Office of Supportive Services personnel implement an intensive and comprehensive
university-wide service. Students have access to program services from the time they
enter the university, through graduation and placement, into graduate school. The
primary goal of OSS is to increase retention rates, facilitate academic successes and
enhance graduation rates among program participants. The majority of CAAP students
tend to be minority. However, over the years, the population has expanded to include
students with disabilities, students receiving Pell Grants, and/or students who have a 2.50
or lower MSU GPA. Students who participate in these programs are eligible for
academic advising, counseling, tutorial services, access to the computer lab facilities,
skills enrichment courses, and graduate study planning and advising.
The Summer Research Oppprtunities Program and McNair Proggrm (McNair-SROP)

The Summer Research Opportunities Program and McNair Program (McNair-
SROP) are ten-week programs that coincide with Summer semester and are available to
low income, ﬁrst-generation and handicapped students who have achieved junior status
and have a GPA Of 3.0 or better. The goal of the program is to encourage enrollment in
graduate and professional schools. These students are matched with a faculty member
and are required to complete a research project. The student receives a stipend and the
faculty member receives a research allowance that permits both to actively pursue their
research. A portion of these students participated in the CAAP program, but most are
non-CAAP students.
The Summer Universiﬂ Proggrn Encouraging Retention (SUPER)

The Summer University Program Encouraging Retention (SUPER) is a summer

bridge program for a select group of students admitted through the MSU CAAP program.

46

The SUPER program is designed to help underrepresented, low-income, and ﬁrst-
generation college students adjust to the academic and social demands of college. All
SUPER students are CAAP students and are eligible for the programs described earlier.
Maximizing Academic Growth in Collge_(MAGIC) Program

MAGIC is a pre-college program designed to provide newly admitted racial ethnic
minority students with information about the University before they begin classes.
MAGIC students spend one week in residence at the University before the start Of their
ﬁrst semester. The program orients the students to the University and teaches them
strategies that may help them succeed academically. All MAGIC students are freshmen
and students of color.
Sparts Camp Prom

College students serving as Summer Sports Camp Assistants were recruited to
participate in this study. The program trains upper level undergraduate students as
leaders, camp counselors, and house assistants to help manage and supervise junior high
and high school students that visit the campus. The majority of the Sports Camp students
are upper-level college students, with few lower level students (i.e. freshmen and
sophomores). The majority of the Sports Camp Assistants for the Summer 2001 were
students of color.

Survey packages were completed by 337 individuals, 224 of whom were African
American. Since the focus of this study was on the African-American college student

pOpulation, only this sample will be further discussed.

47

Instrumentation

Seven instruments were used to collect data. The instruments were: (1) the COPE
(Carver, Scheier & Weintraub, 1989), (2) the Religious Coping Activities Scale (RCAS;
Pargament, Ensing, Falgout, Olsen, Reilly, Van Haitsma & Warren, 1990), (3) the
Multidimensionality of Religiosity Scale (MRS; Levin, Taylor, & Chatters, 1995), (4) the
Core Convictions Scale (CCS, developed speciﬁcally for this study), (5) the African
American Acculturation Scale - Revised (AAAS-R; Klonof & Landrine, 2000), (6) the
Resiliency Attitudes Scale (RAS; Biscoe & Harris, 1994), and (7) a demographics

questionnaire developed speciﬁcally for this study. Each scale will be discussed below

 

and is presented in its entirety in Appendix B.

All instruments (except the Core Convictions Scale and the Demographics
Survey) were selected if they met the following criterion: (1) they were empirically,
theoretically, or conceptually derived with some evidence of reliability and validity, (2)
they measured a belief, an attitude, or a behavior that might help explain a cognitive-
behavioral conceptualization of resilience and coping among African American college
students, (3) responses could be scaled to be used in correlational analysis and structural
equation modeling (SEM); and (4) their instructions were easy to follow.

The COPE
The COPE (Carver, Scheier & Weintraub, 1989) in its original format, is a 60-

item instrument scored on a four-point Likert-style scale where the responses were:

1 = I usually don’t do this at all

2 = I usually do this a little bit

3 = I usually do this a medium amount
4 = I usually do this a lot

48

The items in the instrument are designed to assess how people generally respond when
they confront difﬁcult or stressful events in their lives. This type of coping may be
considered dispositional. Participants indicate how strongly they agree with such
statements as “I try to come up with a strategy about what to do” and “I restrain myself
from doing anything too quickly.” Low scores on any subscale indicate a less frequently
used way of coping, while high scores on a subscale indicate a more frequently used way
of coping. This instrument was developed speciﬁcally to measure theoretically and
conceptually distinct aspects of problem-focused, emotion-focused, and less-useful
coping responses (Carver, Scheier & Weintraub, 1989).

The COPE (Carver, Scheier & Weintraub, 1989) was designed as an improvement
over the Ways of Coping scale (Folkman & Lazarus, 1980). Folkman and Lazarus (1980)
developed the Ways of Coping Scale to measure:

(1) Problem focused coping - coping efforts aimed at the perceived source of stress.

(2) Emotion-focused coping - efforts to regulate one’s emotions as a way of adapting

to a stressful encounter

(3) Less-useful coping - less useful coping strategies e.g. consume alcohol, behavioral

and mental disengagement.

The COPE by Carver, Scheier, and Weintraub (1989) is based on the Ways of Coping
Scale, but differs in as much as the authors claim it yields (1) ﬁve conceptually distinct
aspects of problem-focused coping, (2) ﬁve conceptually distinct aspects of emotion-
focused coping, (3) three conceptually distinct, less-useful coping responses, and (4) two
subscales were included for research purposes by the authors. Use of the COPE may

reveal a more descript account of coping among Aﬁican American college students.

49

gale Development. The scale authors used the Cognitive Appraisal Theory of
Stress and Coping (Lazarus & Folkman, 1984), and a model of behavioral self-regulation
posited in the work of Carver & Scheier, (1981) as guide for developing scale items.
These theories are congruent with the theoretical basis of this study. The authors selected
and factor analyzed scale items. The authors indicate that items with weak loadings were
either revised, discarded, or new items were written. The authors administered the ﬁnal
version of the scale to a sample of undergraduate students at the University of Miami
(n=978), and the responses were subjected to principal-components factor analysis with
oblique rotation to allow for correlations among factors'. The analysis yielded twelve
factors with Eigenvalues greater than 1.0. The composition of eleven of the twelve
factors was fully in accord with a priori assignment of items to subscales, with the
exception of the active coping and planning items which all loaded on one factor (Carver,
Scheier & Weintraub, 1989). Seeking social support for instrumental reasons and seeking
social support for emotional reasons also loaded together on one factor. The alcohol and
drug use, and the humor subscales were included as exploratory scales only, and were not
included in the published version of the COPE. Hence, there was no internal consistency
reliability or test-retest reliability data reported on these two subscales. They were

however, included in the instrument used in this study for exploratory purposes. The

 

' However, they apparently did not understand that they did not need use oblique rotation to obtain
correlated subscales. The method they used to create factor scores would have allowed correlations among
factors without having used oblique rotation. They allowed each item to load only on a single factor rather
than partitioning the response to each item to contribute to multiple factors. Only when partitioning

responses to each item does one need to worry about deﬁning the factors to be uncorrelated.

50

authors state that the ﬁnal version of the scale contains thirteen conceptually distinct

scales, with two exploratory scales for a total of ﬁfteen.

The scales and their reported internal consistency reliabilities are:

Active coping (.62)
Planning (.80)
Suppression of competing activities (.68)
Restraint coping (.72)
Seeking social support—instrumental reasons (.75)
Seeking social support-emotional reasons (.85)
Positive reinterpretation and growth (.68)
Acceptance (.65)
Turning to religion (.92)
Focus on and venting of emotions (.77)
Denial (.71)
Behavioral disengagement (.63)
Mental disengagement (.45)
Alcohol-drug use (none-reported)
Humor (none reported)

The ﬁrst ﬁve subscales (i.e., active coping, planning, suppression of competing
activities, restraint coping, and seeking social support — instrumental reasons) represent
the ﬁve conceptually distinct aspects of problem-focused coping (Carver, Scheier &
Weintraub, 1989). The next ﬁve subscales (i.e., seeking social support-emotional
reasons, positive reinterpretation and growth, acceptance, turning to religion, and focus
on and venting of emotions) represent the ﬁve conceptually distinct aspects of emotion-
focused coping (Carver et al., 1989). Denial, behavioral disengagement, and mental
disengagement represent conceptually distinct aspects of less-useful coping (Carver et al.,
1989). Alcohol — drug use and humor represent the two exploratory subscales (Carver et
al., 1989).

In general, the reliability values obtained by the authors in the development of this

scale were acceptably high, with only one subscale (mental disengagement) falling below

51

.6. The test-retest reliability data on the COPE suggest that the self-report of coping
strategies measured by the COPE were relatively stable: .62, active coping; .80, planning;
.68, suppression of competing activities; .72, restraint coping; .75, seeking social support
— instrumental; .85, seeking social support — emotional; .68, positive reinterpretation and
growth; .65, acceptance; .92, turning to religion; .77, focus on and venting of emotions;
.71 , denial; .63, behavioral disengagement; .45, mental disengagement; and none reported
for alcohol — drug use. The subscales of the COPE were not strongly intercorrelated (the
sample correlations ranged from -.28 to .69, with an media value of .13. This suggests
that most subscales tap fairly distinct ways of coping. Although not strongly correlated
empirically, the subscales tend to be correlated conceptually (Carver et al., 1989). The
COPE was also able to reveal gender differences in coping. The authors correlated the
COPE with several personality measures (e.g., self-esteem, anxiety, hardiness, locus of
control) and found empirical evidence for both the convergent and discriminant validity
of the COPE. Finally, the authors conclude that the COPE may be used to assess both
dispositional and situational types of coping.
The Religious Coping Activities Scale (RCAS)

The Religious Coping Activities Scale (RCAS; Pargament, Ensing, Falgout,
Olsen, Reilly, Van Haitsma & Warren, 1990) in its original format, is a twenty-nine item

instrument scored on a four-point Likert-style scale:

1 = not at all

2 = somewhat
3 = quite a bit
4 = a great deal

52

Participants were asked to identify a stressful situation or dilemma and respond to items
in the survey based on their handling of the situation or dilemma. This scale measures
situational coping. Participants indicated how strongly they agree with such statements as
“Asked God why it happened” and “My faith showed me different ways to handle the
problems.” Low scores on any of the subscales indicate a less-often used, or less-
preferred form of religious coping. High scores on any subscale indicate a form of
religious coping that is used more often or that is more preferred.

This instrument was designed for use in conjunction with other non-religious
measures of coping to enhance our understanding of coping. This instrument was used in
this study because religious or spiritual coping has been identiﬁed as a preferred way of
coping among African American churchgoing and Christian people (Oler, 1997). Use of
this scale may result in a more complete and accurate understanding of coping among
African American people who attend church and who identify as Christian. In addition,
the theoretical literature on coping (not yet empirically veriﬁed) suggests that religious
coping can affect the ultimate outcomes of negative events (Pargament, Ensing, Falgout,

Olsen, Reilly, Van Haitsma & Warren, 1990).

Scale Development. The sample upon which the authors developed this scale was
intended to be representative of churchgoing people with mainstream Christian beliefs.
However, the sample was 96 percent white, 66 percent female, 38 percent college
educated, and 64 percent married. The RCAS was intended to measure situational
religious coping. A total of 31 items were generated from the empirical literature, written

personal accounts of those who turned to religion in time of stress, and interviews with

53

adults and clergy who report using religion in coping. The authors subjected these items
to principal components factors analysis with Varimax rotation. Using a scree plot, they
selected a ﬁve-factor solution that explained nearly 100 percent of the common variance
in the sample. However, the rule of thumb used by the authors to determine the number of
factors was not stated. A sixth factor (religious avoidance) was created from two of the
original 31 items that did not load on any of the previous ﬁve factors, with an additional
theoretically salient item being added.

The six resulting factors and their reported internal consistency reliabilities were:

(1) Spiritually based (.92)
(2) Good deeds (.82)
(3) Discontent (.68)
(4) Religious support (.78)
(5) Plead[ing] (.61)

(6) Religious avoidance (.61)

Findings from the validity and reliability studies on this scale suggest an
important role of religious beliefs, practices, and motivations among persons who are
religiously involved or who identify as spiritual. Like general coping, research shows that
religious coping is multidimensional (Pargament, Ensing, Falgout, Olsen, Reilly, Van
Haitsma & Warren, 1990).

The Resiliency Attitudes Scale (RAS)

The Resiliency Attitudes Scale (RAS), (Biscoe & Harris, 1994) is currently under
development and is being piloted on an African American college student population via
this study. In its original format, it is a 72-item instrument scored on a ﬁve-point Likert-
style scale with responses ranging from “Strongly disagree” to “Strongly agree.” Half of

the items must be reverse-coded before scoring. The RAS yields seven subscales: ( l)

54

insight, (2) independence, (3) relationships, (4) initiative, (5) creativity / humor, (6)
morality, and (7) general resilience. Examples of items ﬁ'om the general resilience
subscale of the RAS include “No matter what happens, if I keep trying I’ll get through it”
and “Even if bad things happen, I can deal with them.” Participants indicate how strongly
they agree with items in the subscales. Low scores on a subscale indicate a low level of
that resiliency, and a high score on a subscale indicates a high level of that resiliency.
The RAS also gives a composite resiliency score (i.e., the Total Resiliency Score is the
sum of items 1-72).

Saale development. The ﬁrst six subscales identiﬁed in the RAS were developed
based on the conceptualization of resilience posited in the work Of Wolin & Wolin
(1993). Biscoe & Harris (1994) added the general resilience subscale, and deﬁned it as a
belief that one can survive and make things better (Biscoe & Harris, 1994, p.1). These
beliefs fuel persistence in working through difﬁculties. There were no internal
consistency or test-retest reliability studies reported for individual subscales, since this
instrument is still under development. However, the internal consistency reliability of all
items taken together was .87 (Kelso, 1999). Kelso also found that the subscales were
moderately correlated with each other (r = .49 to .69). To date, the items have not yet
been subjected to any kind of rigorous scaling. Kelso provided some divergent validity
evidence in that the RAS was negatively correlated with the Beck Depression Inventory
(BDI). In spite of Kelso’s claim that the RAS was “theory driven, using Wolin and
Wolin’s conceptualization of resiliency” (p.14), a review of the work by Wolin and Wolin

(1993) did not reveal a reasonably developed theory. Their conceptualization of

55

resiliency was apparently derived from anecdotal evidence gathered in interviews with
their clients. No evidence of a rigorous data collection method was found in their work.

The RAS instrument was used in this study for four main reasons: (1) it attempts
to quantify attitudes that reﬂect resiliency, (2) it gives a measure of general resiliency
attitudes, (2) there is a need to further assess the validity and reliability of the RAS scale,
and (3) to help answer the questions —— is there a relationship between attitudes that reﬂect
resilience and coping, and is the relationship between attitudes that reﬂect resilience and
coping a linear causal relationship?

Multidimensionality of Religiosity Scale TMRS)

The Multidimensionality of Religiosity Scale (MRS; Levin, Taylor & Chatters,
1995) in its original format is a twelve-item instrument designed to assess different
aspects of religiosity. Participants indicate the frequency with which they engage in
speciﬁc religious behaviors, the importance of religion in their lives, as well as the extent
of their religiosity. Items include such statements as “How often do you pray”, and “How
religious would you say are you.” The internal consistency reliability for this scale was
not reported in the published version of the instrument. However, the authors indicate
that the scale is a reliable and valid multidimensional measure of religiosity.

An additional four items were added to this scale to expand it as a measure of
spirituality. Examples of these items include “How spiritual would you say you are”, and
“How important is your spirituality to you.” The revised scale was piloted on an African
American population of church-attendees in a study by Martin (2001) and was found to

be a reliable and valid measure of religiosity and spirituality.

56

Only three items were selected from this instrument for use in the actual analysis
of data, since they were most relevant to the research questions. These items were
combined with the items from the Core Convictions Scale and the reliability and validity
were examined. Items selected made conceptual sense to use in this study. The three
items included “(10.) How religious would you say you are”, “(13.) How spiritual would
you say you are”, and “(14.) How important is your spirituality to you.” These three
items were chosen because they represent a general measure of spirituality and religiosity.
Core Convictions Scale (CCS)

The Core Convictions Scale was developed speciﬁcally for this study and is based
on the work of Cooper-Lewter and Mitchell (1996). It is a ten-item instrument containing
statements designed to represent beliefs held by African Americans. Participants indicate
the extent to which they agree or disagree with a belief on a four-point Likert-style scale
for each statement. The instrument is expected to yield a single score. A high score
suggests a strong adherence to this set of convictions. A low score suggests a less intense
adherence or belief in this set of convictions. Items include such statements as “God
works everything out for my good” and “Despite any differences, we are all related or
connected in some way - all of us are brothers and sisters.”

This study was a pilot of this instrument, and the results are discussed in the results

section.

The African America Acculwtion Scale - Revised (AAAS-R)
The African American Acculturation Scale — Revised (AAAS-R; Klonof &

Landrine, 2000) in its original format is a 47-item inventory designed to measure the

57

extent of one’s orientation towards eight theoretically derived dimensions of African
American culture. These dimensions include (1) Religious Beliefs and Practices, (2)
Preferences for African American Things, (3) Interracial Attitudes, (4) Family Practices,
(5) Health Beliefs and Practices, (6) Cultural Superstitions, (7) Racial Segregation, and
(8) Family Values.

This instrument yields eight subscale scores and one composite score. The
development of this instrument was based upon ﬁndings from studies that indicated
levels Of adj ustrnent in a majority-centered society is a major feature of the phenomenon
we call acculturation. High scores on the instrument indicate a high agreement with scale
items, and reﬂect a close identiﬁcation with those things that may be considered uniquely
African American (i.e., a traditional African American cultural orientation). Low scores
indicate disagreement with items, and reﬂect a low identiﬁcation with those things that
are considered traditionally or uniquely African American (i.e., an acculturated
orientation). Participants indicate the degree to which they agree or disagree with each
item based on a seven-point Likert-style scale. Items include such statements as “I like
gospel music”, “When I was young, I was a member of a Black church”, “I don’t trust
most White people”, and “Most Whites are afraid of Blacks.” The internal consistency

reliability of this entire scale was reported as .97 (Klonof and Landrine, 2000), but no

reliability or validity data was reported for the individual subscales.

Demomphics Questionnaire
The Demographics Questionnaire, developed for this study, solicited data on the

age, gender, race/ethnicity, class level, grade point average, ACT score, parent’s income,

58

type Of home, type of high school attended, religious orientation, and reason for coming
to MSU.
Rmh Hypotheses
(1) There is a statistically signiﬁcant positive relationship between attitudes that
reﬂect resilience and coping among both “at risk” and “non-at risk” African
American college students.
(2) Core beliefs, religiosity, and spirituality moderate the relationship between
attitudes that reﬂect resilience and coping responses.
(3) Level of acculturation is related to coping responses, attitudes that reﬂect
resilience, core beliefs, religiosity and spirituality.
(4) Attitudes that reﬂect resilience vary by class level, gender, SES, program
participation, and “at risk” status.
(5) The relationship between attitudes that reﬂect resilience and coping responses is
reciprocal.
Statisticarl—Methods Used to Analyze the Data
The statistical methods used for scaling the data from each instrument were
Principal Components Exploratory Factor Analysis, and Conﬁrrnatory Factor Analysis
(Bryant & Yarnold, 1995); and classical test theory reliability analysis (Crocker & Algina,
1986). AMOS (Version 4.0 WIN95/98/NT, SPSS Inc) was used for the Confmnatory
Factor Analysis, and SPPS (Version 10.0, SPSS Inc) was used for the Exploratory Factor

Analysis and the reliability analysis.

59

The statistical methods used for modeling the resulting scaled data were Path
Analysis (Klem, 1995) and Structural Equation Modeling (Kline, 1998). Both the Path

Analysis and Structural Equation Modeling were carried out using AMOS.

60

CHAPTER IV. ANALYSIS OF DATA

The purposes of this study were to examine the relationship between attitudes that
reﬂect resilience and coping among African American college students, to examine the
moderating role of three possible protective factors, and to examine the relationship
between level of acculturation and coping responses, attitudes that reﬂect resilience, core
beliefs, religiosity and spirituality. This chapter describes the analysis of data collected
from a demographics questionnaire, the COPE (Carver, Scheier & Weintraub, 1989), the
Religious Coping Activities Scale (RCAS; Pargament, Ensing, Falgout, Olsen, Reilly,
Van Haitsma & Warren, 1990), the Multidimensionality of Religiosity Scale (MRS;
Levin, Taylor & Chatter, 1995), the Core Convictions Scale (CCS; developed speciﬁcally
for this study), the Aﬁican American Acculturation Scale-Revised (AAAS-R; Klonof &
Landrine, 2000) and the Resiliency Attitudes Scale (RAS; Biscoe & Harris, 1994).
The data were analyzed in the following manner:
(1) The demographic data was analyzed in order to generate descriptive statistics
for the sample used in the structural equation modeling and in developing the
path models. i

(2) All scales (except the Multidimensionality of Religiosity Scale — MRS, and
the Demographics form) were subjected to Principal Components Exploratory
Factor Analysis for the purposes of scale development and having adequate
measures to use in building the models. The internal consistency reliability

and the test-retest reliability were calculated for the Core Convictions Scale.

61

(3) Structural Equation Modeling and Path Analysis were used to generate the
ﬁnal models of ﬁt.
Descriptive Statistics

Descriptive statistics of the demographic characteristics of the complete sample
(N=224) and the ﬁnal sample (the complete sample minus any participants with missing
data; N=190) were calculated. Descriptive statistics of all variables included in the model
were also calculated. To account for differences in the data of students who had complete
data and those with missing data, these statistics were calculated separately for each
group of students and compared.

Instrument Scaling

From the descriptions of the scales with already described factor structures
(COPE, RAS, and BAS), it appeared that the scales had been adequately analyzed using
Exploratory Factor Analysis (EFA). An attempt was made to perform Conﬁrmatory
Factor Analysis (CFA) on the COPE using the already described factor structures. In this
' case, the CFA statistical procedure was unable to converge. The description of the scale
was scrutinized ﬁrrther, and it was determined that parts of the scale (Humor and
Alcohol/drug'use) were experimental additions to the scale, and had never been subjected
to EFA. The description of the EFA was further scrutinized, and it was found that for
determining the number Of factors in the scale, the authors had used a rule of thumb that
is no longer considered adequate. That is, they cut the number of factors at the number of
Eigenvalues greater than 1. The more accepted method is to examine a Scree plot for

noticeable drops and subsequent leveling.

62

In all cases, it was determined that the pre-described factor structure did not ﬁt
this sample of data. The factor descriptions were either inconsistent with this population,
an out-of.date rule of thumb was used, or there had been no empirical basis for describing
the factor structure. For this reason, attempts at CFA were aborted, and all instruments
were subjected to principal components EF A with Varimax rotation to provide a more
adequate scaling of each instrumentz.

Model development

When the scales were adequately developed, Structural Equation Modeling (SEM)
was used to assess the adequacy of competing models of the data, and to modify the
models to more accurately ﬁt the data. To do this, each competing model was created and

various measures of ﬁt and speciﬁc model inadequacies were Obtained. Using this

 

2 1n EFA, there are multiple ways to create factor scores. Each item can load on more than one factor, since
responses to each item can be dependent on more than one single construct. Factor scores can be created
that allow for each item to load on multiple factors. Another method of creating factor scores is to assume
that each item informs only the factor on which it loads most heavily. In all cases, the authors of the scales
adopted for this study used the second approach. One advantage of the ﬁrst approach is that there is a
slightly more accurate representation of each factor. A disadvantage is in the difficulty of creating the
factor scores and understanding the meaning, since the score on each item must be distributed across
multiple factors. Another disadvantage is that in using orthogonal rotations, factor scores obtained in this
way will be uncorrelated with each other by deﬁnition. Often, this result makes no theoretical sense. The
advantage of the second approach is that each item’s score does not need to be divided between different
factors, and that the resulting factor scores obtained from an orthogonal EFA may be correlated with each
other. The disadvantage is that the representation of the factors is slightly less accurate. The second
method was chosen for all scales for the reasons described in the Results section. The resulting scales and

subscales were then assessed for reliability.

63

information, the models were adjusted to better ﬁt the data. TO account for differences in
the total sample and the sample of students with complete data, the model was estimated
using data from all participants, and data only from participants with complete data. This
was possible because SEM analyzes correlation matrices, either listwise or pairwise.
Listwise correlation matrices are matrices of correlations calculated using only the data
from students who had complete data. Pairwise correlation matrices use all available data
to calculate the correlations between each pair of variables. The advantage of pairwise
correlation matrices is that the correlations in the matrices are more representative of the
population being sampled. The advantage of listwise matrices is that estimation is always
statistically possible when there is no strong multicollinearity of predictors in the model.

The initial models were quite complex because of the large number of outcomes
and predictors, as well as interaction terms included. The models were reduced by
removing non-signiﬁcant interaction terms; removing non-signiﬁcant main effects; and
adding correlations between predictors that were not expected to be causally related, and
outcomes that were not expected to be causally related. Finally, the discrepancy between
the observed correlations between variables included in the model, and the correlation
matrix implied among those variables by the structure of the model were investigated.
Residual cOrrelation with a magnitude larger than 0.10 were investigated to determine
whether adding an additional relationship to the model would improve model ﬁt.

The competing models were then scrutinized to determine which was the best ﬁt
to the data. Various measures of ﬁt, measures of direct, indirect, and total relationships
between predictors and outcomes, and measures of proportion of variance explained in

each outcome were also obtained to evaluate the adequacy of the best-ﬁtting model(s).

Summm

The methods used to test the hypothesis in this study were path analysis and
Structural Equation Modeling. The level of signiﬁcance chosen for all methods of testing
was .05.

The data were analyzed in the following manner:

1. Descriptive statistics for the sample were generated.

2. Instruments were scaled and developed using Exploratory Factor Analysis

(EFA).
3. Final models were developed using Structural Equation Modeling (for the

reciprocal model) and Path Analysis (for the unidirectional models).

65

CHAPTER V. RESULTS

Descriptive Statistics of Control Varia_b_l_e_§
The samples (pairwise, N=224 and listwise, N=190) were described on each
important variable and were inspected listwise and pairwise for differences. Table 1 is a
description of the total sample of African American students who participated in this
study, and it contains a description of the continuous control variables for the complete
sample’. Table 2 is a description of the ﬁnal sample of African American students with

complete data.

 

3 Note that some of the variables in the full sample table are not in the ﬁnal sample table. This was because
they were excluded ﬁ'om the model. Age was excluded because it was highly correlated with class standing,
which was a better control variable. Self-reported family income, self-reported GPA, and self-reported
ACT were excluded from the model because they were self-reported by the students, and probably
unreliable. For example, a number of participants identiﬁed as CAAP students reported very high GPAs,
incomes, and ACT scores. Although these could have been important control variables, because they were
not deemed reliable, any impact they had on the model could not be considered an effect of controlling for

these variables.

66

Table 1. Complete sample (N=224) frepuencies of categorical varia abl.es

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable Category % Variable Category N %
17 58 25.9 Freshman 133 59.4
1 8 80 35 .7 Class Sophomore 12 5.4
19 12 5.4 standing Junior 22 9. 8
20 29 12.9 Senior 53 23. 7
21 23 10.3 Missing 4 1.8
22 13 5.8 Protestant 193 86.2
A e 23 4 1.8 Religious Catholic 20 8.9
g 24 l 0.4 affiliation Muslim 1 0.4
25 1 0.4 Missig 10 4.5
26 1 0.4 Sex Male 73 32.6
29 1 0.4 Female 151 67. 4
35 1 0.4 Regular 13 5. 8
Maria _1_8_.9; Sports camp 37 16.5
_S=D, 2.19 Program McNair-SROP 40 17.9
1: Under $10,000 12 5.4 SUPER 20 8. 9
2: $10,000 - $20,000 18 8 MAGIC 114 50.9
3: $20,000 - $30,000 26 11.6 Student is No 100 44.6
4: $30,000 - $40,000 23 10.3 at risk Yes 120 53.6
5: $40,000 - $50,000 17 7.6 Missing 4 l. 8
_ 6: $50,000 - $60,000 9 4 Single parent 102 45.5
ﬂl’ﬁg‘; 7: $60,000 - $70,000 10 4.5 augment Two parent 113 50.4
8: $70,000 - $80,000 8 3.6 Other relative 9 4
9: $80,000 - $100,000 15 6.7 N 194
10: $100,000 + 16 7.1 Self-repo rte d Mean 19.8
Missing/don't know 70 31.3 ACT SD 3.3
Mean = 5.1 (approximately $41,000) Min 13
SD = 2.83 a roximatel 28 300 Max 29
N 207
Self-reported Mean 2'8
GPA SD 0.5
Min 2
Max 4
Table 2. Final sample gg=l 90) freguencies of categorical variables.
Variable Category Variable CategorL N %
Regular 13 6.8 Freshman 110 57.9
Sports camp 31 16.3 Class Sophomore 11 5. 8
Program McNair-SRO? 36 18.9 standing Junior 19 10.0
SUPER 17 8.9 Senior 50 26.3
MAGIC 93 48.9 Religious Catholic 18 9.5
Home Non-traditional 98 5 1 .6 afﬁliation Protestant 172 90.5
environment Traditional 92 48.4 Student is No 91 47.9
Sex Male 63 33.2 part of CAAP Yes 99 52.1
Female 127 66.8

 

As can be seen in Table 1, more than half of the participants (59.4%) were

freshman, 5.4 percent were sophomore, 9.8 percent were junior, and approximately one

67

fourth (23.7%) were seniors. The percentage by class level did not change appreciably in
the ﬁnal sample (the largest change in percentage for any class level was 2.6, for seniors).

In the complete sample, 45.5 percent of the participants reported coming from a
single parent home; 50.4 percent were from a two-parent home. Just over half of the
participants in the total sample that completed surveys were MAGIC students (50.9%).
53.6 percent of the total sample was made up of participants in the CAAP program.
Again, these percentages did not change appreciably from the complete to the ﬁnal
sample.

The ratio of males to females in the total sample was two females (67.4%) for
every one male (32.6%). Again, the percentages changed only slightly in the ﬁnal
sample. As for religion, most participants (86.2%) identiﬁed as Protestant. In the ﬁnal
sample, 90.5 percent of the participants identiﬁed as Protestant. Participants with missing
data were not included in the ﬁnal sample.

The breakdown for CAAP participation by program is reported in Table 3.
Because of the small numbers in some of the programs, CAAP participation, rather than

program membership (e.g., MAGIC, SUPER, etc.) was used as a control in the models.

Table 3. CAAP by progpam participation freguencies.

 

 

 

 

Program
Type of student Regular Sports camp McNair-SROP SUPER MAGIC
Non-CAAP 8 19 33 0 39
CAAP 5 l8 7 20 71
;r_o_.ta..1N .13 3_7 em 43 L1_0

 

68

Instrument Scaling (Exploratory Factor Analysis)

w

All 60 items of the COPE were subjected to principal components EFA. Since the
authors Carver, Scheier and Weintraub (1989), had originally used an Oblique rotation,
the COPE was analyzed using both an orthogonal and Oblique rotation. In addition, the
EFAs were estimated using both listwise and pairwise correlation matrices. The scree
plots of the orthogonal rotated EFA are presented in Figure 1. The listwise and pairwise

scree plots from the oblique solutions are not presented because they did not differ in any

important way.

69

Pairwise

 

121

 

 

 

 

 

Eigenvalue

17131925313743495

Factor Number

Listwise
12

 

10D

 

 

 

 

 

Eigenvalue

J _
17131925313743495

Factor Number
Figge 1. Scree plots of the COPE orthogonal EFA.

As noted earlier, the preferred method of determining the number of factors

represented in an instrument is to look for noticeable drops and a subsequent leveling of

70

the plot. Figure 1 shows that there are two reasonable cutoffs in these plots. The most
obvious is to cut the number of factors at three. Less obvious, but still reasonable is to set
the cutoff at six. This is especially noticeable in the pairwise plot. These results differed
from the results of the scale authors because of the difference in the rule of thumb used
for determining the number of factors. Figure 1 has a horizontal line placed across the
graphs to denote the cutoff used by the authors. Using this cutoff, the number of factors
extracted would be 16 (one higher than the authors). Note that moving the cutoff point by
a very small value (e.g. 0.1) the number of factors increases or decreases by 2 to 3. This
suggests that the rule of thumb used by the authors is sensitive to random variations in
error. Also note that around this cutoff, there are no noticeable drops with subsequent
leveling.

To determine which was the more reasonable cutoff, both a 3-factor and a 6-factor
solution were obtained for both the orthogonal and Oblique rotations. These four
solutions were scrutinized for conceptual ﬁt. For both the orthogonal and Oblique
rotations, items that loaded on factors in the 3-factor solutions were not linked
conceptually. In these solutions, both the adaptive and non-adaptive coping responses
loaded on the ﬁrst factor, and their factor loadings were all positive, where we would
have expected positive loadings for adaptive strategies, and negative loadings for non-
adaptive strategies. For the 6-factor solutions, there were no substantive differences
between the orthogonal and oblique results. The orthogonal solution was chosen for two
reasons: (1) orthogonal EFA is a less complex process, and (2) the factor loadings of

many factors more closely mirrored those reported by the scale authors.

71

For several reasons, it was determined that each item should load on only a single
factor. First, there were relatively few items that had high cross loadings on multiple
factors. Second, a number of items with relatively strong cross loadings made no
conceptual sense with any factor, and including them in any of the factors actually
decreased the reliability of the factor scores. Those items were dropped from the scales.
Third, even the items with relatively high cross-loadings that did make sense conceptually
ﬁt better with the other items in the factor on which it loaded most heavily. Fourth,
allowing each item to load only on one factor allows the resulting factor scores to be
intercorrelated. Finally, not partitioning each item’s scores amOng multiple factors
facilitates understanding of the scale. Appendix A presents a complete factor-loading
matrix, the labels given to each factor, and the content of each item that loaded on each
factor.

To create the factor scores without losing data because a participant skipped a
small number of items, averaging the responses to each item in each subscale created
factor scores. If participants responded to at least half of the items in a subscale, their
averaged score was kept in the data set. If they responded to less than half the items, it
was determined that their responses were inadequate to represent the construct being
measured by the subscale, and their scores on that subscale were removed from the
database. The same strategy for creating factor scores on this scale was followed for each
factor created using all other scales. The descriptive statistics, internal consistency

reliabilities, and correlations among the six factors are presented in Table 4.

72

Table 4. Correlations among and descriptive statistics of the six COPE subscales.

 

 

 

Correlation with Descriptive statistics

COPE factor 1 2 3 4 5 N Min Max Mean SD Reliability
1 Adaptive 224 1.33 3.86 2.83 0.5 0.89

2 Less useful -0.l7" 224 l 3.77 1.53 0.5 0.85
3 Help-seeking 0.33" 0.06 224 l 4 2.74 0.7 0.83
4Humor 0.18" 0.21" -0.07 224 l 4 2.31 0.9 0.89
5 Religious 0.34" -0.26" 0.20" -0.04 224 1 4 3.28 0.8 0.84
6 Acceptance/restraint 0.41" -0.07 0.07 0.13 0.15“ 224 1 4 2.73 0.7 0.71
‘p<0.05

“p<0.01

As shown in Table 4, the correlations not marked by an asterisk C“) or a double
asterisk (**) may not truly exist between the subscales in this population. As shown,
each subscale has adequate reliability, and the subscales are uncorrelated to moderately
correlated. This provides evidence that distinct but related sets of coping responses are
being measured by each set of items. In addition, Less-useful Coping Responses had the
lowest mean, and Religious Coping Responses had the highest. This indicates that in this
sample, Religious Coping Responses were the most highly used, and Less-useful Coping
Responses were the least used. Again, this pattern was expected in this sample, since the
literature suggests that religious coping responses are a salient coping strategy used by
church-going African Americans (Oler, 1997).

Each correlation in Table 4 can be explained conceptually. The most striking
result is the expected correlations between adaptive and the other coping methods (the
ﬁrst column of numbers). They are all signiﬁcant, and they are all in the expected
direction. Adaptive coping responses are negatively correlated with less-useful coping
responses, but positively correlated with all others. It is to be expected that adaptive and

non-adaptive responses would be negatively related, and that all other coping responses

73

would be positively related to adaptive responses, since they are not necessarily
maladaptive approaches to coping.

The second column of numbers also makes sense conceptually. There is no
reason to expect that Less-useful COping would be related to Help-seeking or
Acceptance/restraint in any particular way, and this was demonstrated in the results.
However, students who tended to use Less-useful coping responses also tended to use
Humor coping responses more often than others. Although Humor is not necessarily a
maladaptive coping response, it can be used maladaptively (e. g. “blowing off” the
problems rather than addressing them). Finally, students who tended to use Less-useful
coping responses also tended to use less Religious coping responses. Because of the
conﬂict between religious belief and many of the Less-useful coping responses (i.e.,
alcohol/drug use), this negative relationship is to be expected.

The third column is also conceptually consistent. There is no reason to believe
that a Help-seeking person would be more likely to use Humor or Acceptance/restraint
responses than any other, and this was borne out in the results. However, Help-seeking
students tended to use more Religious coping responses. It could be said that some of the
Religious coping responses (e.g. praying more than usual, trusting in God, seeking God’s
help) might be another form of Help-seeking.

The fourth and ﬁfth columns are also conceptually consistent. While it may be
expected that people who use Humor coping responses would also be more accepting of
the situation, this relationship is tenuous since Humor may also be an avoidance
mechanism. The relationship between these two coping responses in the table was close

to signiﬁcance, but was not ﬂagged as reliable (or statistically signiﬁcant). There was no

74

reason to believe that people who use humor would also be more religious in their coping
responses, and this was also borne out. Finally, it would be expected that people who had
more Religious coping responses would also be more accepting of the situation, given the
importance of faith that God will address the situation in Religious coping. See Table 12
in Appendix A for the factor-loading matrix for the COPE.
R_C_Aﬁ

The RCAS data was also analyzed using principal components EFA with Varimax
rotation. Again, both listwise and pairwise correlation matrices were analyzed, but the
results were again nearly identical. The same method as for the COPE was used for

determining the number of factors. Figure 2 shows the scree plot for this analysis.

Listwise correlations

 

Eigenvalue

 

 

 

14710131619222528

Component Number
Figure 2. Scree plot of the RCAS EFA.

It was determined that three factors were being measured by the RCAS. Although

the scree plot shows a single very strong factor, the two minor factors explain relatively

75

important variation in the responses, and they made sense conceptually. The complete
factor loading matrix and description of the factors and item contents are presented in
Appendix B. Because of high cross-loadings and conceptual misﬁt, one item was
removed from the analysis.

Factor scores were created in the same way as with the COPE to avoid problems
of missing data. Correlations among the three factors, their reliabilities, and descriptive

statistics are presented in Table 5.

Table 5. Correlations among and descriptive statistics of the three RCAS subscales.

 

 

 

Correlation with Descriptive statistics
RCAS factor 1 2 N Min Max Mean SD Reliability
1 Spiritual CR 220 1 4 3.03 0.77 0.93
2 Applied Faith CR 0.69" 220 1 4 2.36 0.78 0.88
3 Despair/doubt CR 014* 0.20" 220 1 3.67 1.94 0.59 0.60
*p<005
** p < 0.01

As shown, two of the subscales in this three-factor solution have adequate
reliability. The reliability for the Despair/doubt subscale is borderline acceptable. The
subscales are correlated as expected, with the correlations for Despair/doubt Coping
Responses having the lowest correlations with other factors. The scale and item means
indicate that Spiritual Coping Responses are most often used for coping and
Despair/doubt Coping Responses are least used for coping. The pattern ﬁt conceptually
with the idea that Religious coping responses are salient among the Aﬁican American
culture (Oler, 1997).

Again, each correlation in Table 5 can be explained conceptually. Spiritual
Coping Responses have a strong positive correlation with Applied Faith Coping

Responses. This would be reasonable as Spiritual Coping Responses and Applied Faith

76

Coping Responses are closely related, but there is a conceptual difference between the
two that is born out in the correlation that does not suggest that these are exactly the
same. Both Spiritual Coping Responses and Applied Faith Coping Responses have
small but signiﬁcant relationships with Despair/doubt Coping Responses and Spiritual
Coping Responses. This correlation would be expected as items on the Despair/doubt
Coping Responses subscale represent spirituality and faith that is in doubt. It is to be
expected that there must be a certain level of spirituality and faith for those things to be
challenged. See Table 13 in Appendix B for the factor-loading matrix for the RCAS.
Combining the COPml the RCAS

Because one section Of the COPE is directed toward religious coping responses,
and the purpose of the RCAS is to add information to general measures of coping
(Pargament, Ensing, Falgout, Olsen, Reilly, Van Haitsma & Warren, 1990), items ﬁ'om
the COPE and RCAS were analyzed together to create a single multidimensional scale of
coping responses. The 56 remaining items from the COPE (four item were removed from
the analysis and described above) and the remaining 28 items from the RCAS were
submitted together to a principal components factor analysis with Varimax rotation. The

scree plots Of this analysis are presented in Figure 3.

77

Listwise

 

          

 

 

Eigenvalue
'E’

  

1 9 17 25 33 41 49 73
Component Number

Painlvise

 

14‘”

     

 

 

Eigenvalue
'9

1 9 17 25 33 41 49

Component Number
Figge 3. Scree plots from the COPE and RCAS combined EFA.

From the scree plots, it appears that the most reasonable solutions would be

solutions with 4, 6, and 7 factors. These solutions were all investigated, and it was

78

determined that the seven-factor solution made the most sense conceptually, so it was
accepted as the best solution. Again, because of relatively little cross-loading, each item
was allowed to load on only one factor. The complete table of factor loadings, and the
factor contents are presented in Appendix C. Correlations among the three factors, their
reliabilities, and descriptive statistics are presented in Table 6.

Table 6. Descriptives, reliabilities and intercorrelations among combinea COPE aad
RCAS factors.

 

 

 

 

 

Correlation with Descriptive statistics

Factor 1 2 3 4 5 6 N Min Max Mean SD Reliabiliﬁ
lSpiritual CR 220 1 4 3.00 0.72 0.94
2 Adaptive CR 0.30" 228 1.29 3.75 2.82 0.47 0.90
3 Non-adaptive CR -0.06 -0.20"“" 228 1 3.89 1.40 0.50 0.84
4Emotiona1 CR 0.11 0.32" 0.04 228 1 4 2.74 0.66 0.85
5 Humor CR -0.02 0.19" 0.19" -0.07 228 1 4 2.31 0.87 0.89
6 Applied Faith CR 0.63" 0.19" 0.09 0.14“ 0.06 220 1 4 2.17 0.89 0.85
7 Despair/doubt CR 0.02 -0.08 0.40” 0.18" 0.13" 0.04 222 1 3.63 1.86 0.50 0.61
‘p<005

" p < 0.01

As shown in Table 6, the correlations not marked by an asterisk (*) or a double
asterisk (**) may not truly exist between the subscales in this population. As shown,
most subscales have adequate reliability, and the subscales are uncorrelated to strongly
correlated. The only concern is the Despair/doubt Coping Response scale which has a
borderline reliability. It may be necessary to add conceptually consistent items to this
scale, and reﬁne the existing items to more adequately measure this construct.

This provides evidence that distinct but related sets of coping responses are being
measured by each set of items. Again, the pattern of means is consistent with theory
(Oler, 1997). Spiritual Coping Responses are used most often, and Non-adaptive and
Despair/doubt Coping Responses are used least often.

Each correlation in Table 6 can be explained conceptually. As in the COPE

principal component EFA, the most striking result is the expected correlation between

79

Adaptive and the other coping methods, with the exception of Despair/doubt Coping
Responses (see second column). Each correlation is signiﬁcant and in the expected
direction. The ﬁrst column of numbers can be linked conceptually. Spiritual Coping
Responses have a small but signiﬁcant correlation with Adaptive Coping Responses as
expected. They also have a strong signiﬁcant correlation with Applied Faith Coping
Responses as expected. There is no reason to believe that Spiritual Coping Responses
would be related to Non-adaptive Coping Responses, Humor Coping Responses, or
Despair/doubt Coping Responses. Also, the items that loaded on Emotional Coping
Responses focus on human emotional responses to stress, which are very different from
Spiritual coping responses. Hence, one would not expect a relationship between these
two subscales.

The second column of numbers also can be linked together conceptually. Students
that had more Adaptive Coping Responses tended to have less Non-adaptive Coping
Responses. Students that had Adaptive Coping Responses tended to have more
Emotional Coping Responses, Humor Coping Responses, and Applied Faith Coping
Responses. These other Coping Responses can be used in a positive or negative way, and
it is expected that students that had Adaptive Coping Responses will have a range of
coping responses that they use in positive ways. Finally, knowing a student’s tendency to
use Adaptive Coping Responses does not inform how likely it is that that student will also
have Despair/doubt Coping Responses. This may be because the Despair/doubt scale is
composed of competing constructs, where despair and doubt are related, but distinct

constructs, and there was insufﬁcient data to tease them apart in this analysis.

80

The third column is also conceptually consistent. Students that had Non-adaptive
Coping Responses tended to have both Despair/doubt and Humor Coping Responses, but
the tendency is stronger for Despair/doubt Coping Responses. Students that had Non-
adaptive COping Responses tended not to have Emotional Coping Responses, or Applied
Faith Coping Responses any more than any other student.

The items in column four may be linked conceptually. Students that had
Emotional Coping Responses tended to have more Applied Faith Coping Responses, and
more Despair/doubt Coping Responses. This is consistent because important components
of Applied Faith Coping Responses and Despair/doubt Coping Responses can be
considered to be highly emotional. However, students that had Emotional Coping
Responses tended not to use Humor Coping Responses any more than other students.
Although aspects of Humor Coping Responses can be considered to be emotional, the
items that measured Humor Coping Responses were qualitatively different from those
measured by Emotional Coping Responses. The “Humor” items were more dismissive of
the stressor, but the “Emotional” items were more indicative of a response focused on
processing the challenge.

Columns 5 and 6 also ﬁt together conceptually. In these columns the only
signiﬁcant relationship was between Humor Coping Responses and Despair/doubt
Coping Responses. Items from both of these factors indicate a dismissive response to the
challenge of coping. Humor Coping Responses and Applied Faith Coping Responses had
no relationship, nor did Applied Faith Coping Responses and Despair/doubt Coping

Responses. Again, there is no reason to expect that those who applied faith in coping

81

would use more or less Humor or Despair/doubt than others—especially since those who
applied their faith would be less likely to despair or to doubt God.

Finally, note that combining the COPE and RCAS caused some Of the items to
move ﬁom different factors into a combined or related factor when moving from the
individual scales to the combined scale, but the substantive meaning of each factor was
maintained. This also provides some evidence that the combined subscale is a valid
measure of multiple coping responses. See Table 14 in Appendix C for the complete
combined COPE and RCAS EFA factor-loading matrix.

CCS, Religiosity, and Spiritualig

The Core Convictions Scale (CCS) data from the 224 African American students

was subjected to EF A. The resulting scree plot is shown in Figure 4. Because of the

similarity between the pairwise and listwise results, only one graph is shown.

Pairwise correlations

 

 

 

Eigenvalue

 

 

Component Number
Figge 4. Scree plot Of the CCS EFA.

82

Figure 4 shows that there is one major factor represented in this data, with a
possible second minor factor. It was determined that only one factor should be obtained
from this data. The reasoning behind this decision was that this scale is not an important
outcome in the model. For the outcomes (coping responses and resiliency attitudes), it is
desirable to tease out as many different types of the outcomes as possible to more
adequately describe the relationships between the multiple outcomes, even if some of the
factors are relatively minor. Because this scale is included in the model only to control
for core convictions, it is unimportant to include the minor factor as an additional control
variable.

Therefore, the ten items were treated as a single factor. The internal consistency
reliability (Cronbach’s alpha) for the African American group was .92. For all students,
the test-retest reliability was .83, suggesting that these core convictions tended to remain
stable over the course of this study. However, there were two items with low test-retest
correlations. Item 7 (all hmans are created equal) and item 10 (we should persevere apgl
e_ndure until we aucceed, and in car identity and faﬂ) had test-retest correlations of .27
and .45 respectively. This suggests that these items may need to be revised, or that these
particular convictions tended to change over the period of the study. There were 213
Aﬁican American students with data on this factor, with a minimum value of l, a
maximum value of 4, a mean value of 3.68, and standard deviation of 0.52. There were
95 non-African American students with data on this factor.

The intent was also to control for religiosity and spirituality outside what is

measured by the core convictions. As noted above, only three items from the MRS were

83

selected to measure these traits. To determine whether these three items contributed any
information beyond the CCS, an EFA was conducted on the 10 CCS items and the three
MRS items. The scree plot of the listwise EF A is shown in Figure 5. Because of the

similarity of the pairwise results, only one graph is displayed.

Scree Plot

 

 

 

Eigenvalue
—L

112]

i

 

1 3 5

f

\11 EJ
coin

11 13

Component Number
Figge 5. Scree plot of CCS, Religiosity, and Spirituality EF A.

From the scree plot in Figure 5, there appears to be one major factor with a
possibility of two minor factors. Again, because these variables were included in the
model to control for differences in religiosity/spirituality/core convictions, there was no
need to tease out the minor factors from the major factors. Therefore, a single scale was
formed from the 10 CCS items and the 3 MRS items. By collapsing this scale we are
including some additional information about religiosity and spirituality beyond the core

convictions in the scale, in order to control for these things, but the complexity of the

84

analysis is reduced. This scale was termed the W. A total of 213
students had data on this variable, with a minimum value of l, a maximum value of 3.85,
a mean value of 3.44, a standard deviation of 0.47, and a reliability of 0.91.

The Aﬁican; American Acculturation Scale — Revised (AAAS-R)

Only the “religious beliefs and practices” dimension Of the AAAS-R was
conceptually important to this study, but the entire AAAS-R was subjected to an EFA.
The strongest factor in the results of the EFA‘was the “religious beliefs and practices”
subscale, minus item 9. This item (“What goes around comes around”) didn’t ﬁt either
conceptually or statistically, so it was of no concern that this did not contribute to this
factor. This single factor was extracted from the AAAS-R. This subscale was referred to
in this study as the Culturally Salient Religious Beliefs and Practices (CSRBP) scale. The
reliability Of this factor was 0.80. There were 218 students with values on this variable,

with a minimum value of l, a maximum value of 7, a mean of 5.64, and a standard

deviation of 1.23.
The Resiliency AttituAes Scale IRAS)
The 72 items of the RAS were subjected to principal components EFA with

Varimax rotation. The scree plots of both the listwise and pairwise EF A are presented in

Figure 6.

85

Listwise correlations

 

 

 

 

    

 

Eigenvalue
'9

- .-
-.-- '---
u --
.--.--:.--.__
---.-_::-----..-
--.--.---..-----.--
-----.--..-::::0---.
.-_-

 

1 a" '15 22 '29 3543 50 57 F64 71

 

 

 

Component Number
Pairwise correlations
12
105’
81
e.
4.
G) r
2 .
E 2.
S
Li] 0 _“____m_________

 

 

 

 

1 a 15 22 29""56"' 43 50“"57 64 71

Component Number
F 'gge 6. Listwise and pairwise'scree plots of the RAS EFA.

86

As with the COPE, the rule of thumb of “Eigenvalue > 1” is not reasonable for the
RAS. By slightly modifying the nrle thumb, there is a dramatic change in the number of
factors resulting ﬁ'om the analysis. From the pairwise scree plot, the only reasonable
cutoffs are l and 4 factors. From the listwise scree plot, the reasonable cutoffs are l, 4,
and 7 factors. However cutting the number of factors at 7 is questionable because the
drop in the graph to the eighth factor is minor. A four-factor solution was examined for
conceptual integrity. None of the factors in this solution ﬁt together conceptually. The
complete matrix of factor loadings is presented in Appendix D along with an explanation
of why the factors did not ﬁt together conceptually. See Table 15 in Appendix D for the
factor-loading matrix for the RAS EFA.

Because the results of the EFA could be interpreted in a logical manner, the
reliabilities of the proposed seven subscales on this instrument were examined. The
reliabilities of the seven subscales observed in this sample were 0.33, 0.56, 0.47, 0.65,
0.68, 0.60, and 0.7 8 in order. The most reliable proposed subscale was the general
resiliency subscale, which had not ever been subject to scaling analysis. The ﬁrst three
subscales were insufficiently reliable for use as separate scales, however, the next three
could reasonably be used. In addition, the ﬁrst six proposed subscales were conceptually
inconsistent with the deﬁnition of general resilience given by Biscoe and Harris (1994) -
“the belief that one can make things better for self and others” (p 11 of resiliency
manual). However the general resiliency questions tended to be consistent with that
deﬁnition. For these reasons, as well as being unable to obtain a reasonable factor
solution for this instrument, only the questions from the general resiliency subscale were

used.

87

These 10 items (63-72) were evaluated for reliability. The results showed that
items 68 and 69 decreased the reliability of the scale. These items (“68: No matter how
hard I try, I can’t make things right,” and “69: I am willing to go with any approach that
will work”) were also inconsistent with the deﬁnition of resiliency. Item 68 seems to be
measuring hopelessness rather than a willingness to persist, and item 69 seems to be
mixing morality with persistence rather than measuring primarily persistence. These two
items were thus removed from this scale. The reliability of the remaining eight items was
0.82. There were 218 students with values on this variable, with a minimum score Of

2.25, a maximum score of 5, a mean score of 4.07, and a standard deviation of 0.60.

Modeling the Data
Initial Models

Three competing models of the data were developed: a model where the
relationships between coping responses and resiliency attitudes are reciprocal, and two
models where the relationship was unidirectional (one with coping responses causally
prior to attitudes that reﬂect resilience, and one with attitudes causally prior to coping
responses). This was done to answer research question 5, and to provide evidence
concerning Jew, Green and Kroger’s (1999) hypothesis that coping responses emerge
from resilient beliefs. Each of the proposed models is presented below.

Initial reciprocal model. The initial reciprocal model included a causal
relationship from attitudes that reﬂect resilience to each coping response, as well as from
each coping response to attitudes that reﬂect resilience. It also included causal
relationships from protective beliefs, culturally salient religious beliefs and practices,

class standing, sex, and CAAP participation to attitudes that reﬂect resilience and to each

88

coping response. In addition, moderating effects of protective religious beliefs on the
reciprocal relationship between coping responses and attitudes that reﬂect resilience were
also included (i.e., the interaction of protective beliefs with each coping response on
attitudes that reﬂect resilience, and the interaction of protective beliefs with attitudes that
reﬂect resilience on each coping response).

lai_ti_al unidirectionﬂodels. The initial unidirectional models were subsets of the
initial reciprocal models. For the model with coping responses causally prior to attitudes
that reﬂect resilience, the causal relationships from coping responses to attitudes that
reﬂect resilience were removed, and the moderating effects of protective beliefs on those
relationships were also removed (the moderating effects on the reversed causal
relationships remained). For the model with attitudes that reﬂect resilience causally prior
to coping responses, the reverse was done.
Final Models

After modifying the models as described in the 5am section, the models were
inspected for violations of assumptions. All models had violations of the assumption of
multivariate normality, with a relatively large number of outliers. It was determined that
the outliers should not be removed from the model to address this problem because there
were too many. It was also determined that transformations of the offending variables
would not be attempted because the interpretation of the results would be confusing. This
is because the relationships between variables would have to be interpreted in terms of
the transformed variables rather than in terms of the understandable scales. This resulted
in the p-values of statistical tests being somewhat biased. The results of model

development were still guided by statistical tests with a cutoff of 01 = 0.05 with the

89

knowledge that they may be slightly off. This violation of assumptions does not affect
the validity of the decisions, about which models are the best ﬁtting models, nor does it
affect the estimation of effect sizes—it only affects the statistical tests of signiﬁcance Of
the relationships between variables. For this reason, the statistical tests are used as a
guideline for selecting relationships to be evaluated, but the effects sizes are considered
more important to the discussion.

At this point the models were evaluated for best ﬁt. The ﬁt statistics of each

model are presented in Table 7. See Kline (1998) for a description of each of the tests of

 

 

 

 

 

 

 

ﬁt reported in Table 7.
Table 7. Fit statistics of the three comgting models.
Measure of ﬁt Model
Type Statistic Interpretation rule ARR H CR AR -> CR CR —* ARR
Rule-of-thumb AlC Smaller is best ﬁtting model 141.4 148.7 150.2
model-comparison CAIC Smaller is best ﬁtting model 336.8 356.8 358.3
ﬁt measures BCC Smaller is best ﬁtting model 148.8 157.1 158.6
Statistical test 71’ value eater than 0 20 49.4 50.7 52.2
of individual- Df g; x’ lesf’man df ' ’ 45 56 56
model ﬁt p-value 0.302 0.675 0.621
XZ/df Less than 2 = good 111 1.098 0.905 0.931
RMSR Close to zero = good ﬁt 0.026 0.022 0.022
ﬁg‘ifl‘i‘c’ltgl‘f‘ggel GFl Greater than 0.90 = good ﬁt 0.961 0.964 0.963
Wes AGFl Greater than 0.90 = good ﬁt 0.922 0.932 0.931
NFl Greater than 0.90 = good ﬁt 0.936 0.930 0.929
RMSEA Less than 0.05 = goZOd ﬁt 0.023 0.000 0.000

 

As shown in Table 7, all three models adequately ﬁt the data. For the statistical
tests of individual-model ﬁt, each model had a p-value greater than the cutoff of 0.20.
For the rule-of-thumb individual-model ﬁt measures, each model satisﬁed the rule of
thumb for each measure. Finally, on all of these measures, each model appears to be

approximately in the same range of acceptability.

90

Because the competing models were not nested within the reciprocal model, no
exact statistical test of the utility of adding the reciprocal relationships was conducted. In
lieu of a statistical test of signiﬁcance, the rule of thumb of comparing the Akaike
Information Criterion (AIC) and Consistent AIC (CAIC), and the Browne and Cudeck
Criterion (BCC) across models was used (for AIC and CAIC see Kline, 1998; for BCC,
see Brown and Cudeck, 1989). The model with the lower AIC/CAIC/BCC is deemed the
better-ﬁtting model. The AIC, CAIC, and BCC for the three competing models are also
presented in Table 7.

As shown in Table 7, the reciprocal (or bi-directional) model has better ﬁt on all
rule-of-thumb model-comparison ﬁt measures. However, in developing the reciprocal
model, most of the reciprocal relationships dropped out, and the two reciprocal
relationships that remained canceled each other out. That is, there was a positive effect of
Adaptive Coping Responses on Attitudes that Reﬂect Resilience, but there was a negative
effect of Attitudes that Reﬂect Resilience on Adaptive Coping Responses. Furthermore,
there was a negative effect of Non-adaptive Coping Responses on Attitudes that Reﬂect
Resilience, but a positive effect of Attitudes that Reﬂect Resilience on Non-adaptive
Coping Responses. Conceptually, it is expected that the relationships will have the same
sign. For Attitudes that Reﬂect Resilience and Adaptive Coping Responses, both
relationships should be positive, since there is no reason to expect that high levels of
Attitudes that Reﬂect Resilience will cause adaptive coping responses to decrease. For
Attitudes that Reﬂect Resilience and Non-adaptive Coping Responses, both relationships
should be negative, since there is no reason to believe that high Attitudes that Reﬂect

Resilience would cause Non-adaptive Coping Responses to increase. In both cases, this

91

model suggests that both the Coping Responses and Attitudes that Reﬂect Resilience
ﬂuctuate in concert with each other to create some kind of equilibrium. Of course, this
study has no longitudinal data to support such a claim. For these reasons, the reciprocal
model was abandoned, and the task was then to determine which of the unidirectional
models was a better representation of the data.

As shown in Table 7, the model assuming that Attitudes that Reﬂect Resilience
cause some types of Coping Responses has slightly better ﬁt than the model assuming the
Opposite for all three measures of model-comparison ﬁt. However, the differences
between the ﬁt statistics are so small that either model can be considered equally
adequate in explaining the data.

Because presenting the full names of the variables makes these tables
cumbersome, Table 8 presents acronyms used to represent the variables in Tables 9-11.
The ﬁnal unidirectional models are displayed graphically in Figures 7 and 8, and in

tabular form in Tables 9 through 11.

92

 

.15 .53 -.l' .19 .12 -.22 .26

 

.15

 
 
   
 
   
 
   
 
   
 
   

.61 .53
\piriuml
('R .28

\tl.1pli\ k"
(”R .07
\un-zitldplix t-

(‘R .12

lnmtiunul

1 ‘R .02

|111111111
(‘lx’

 

,\ pp I led
1111111
('R

 

 

 

 

 

At risk

 

 

 

Female Culturally Salient Religious

 

.26

Prohctive

Religious
Beliefs

Down 11

11011111
(‘R

‘—

 

.51

 

 

Beliefs and Practices

 

 

 

Figare 7. Graphical display of the unidirectional AR —> CR model.

93

 

.00

 

    
 
   
   

 

 

 

 

 

.28
Protective
.18 Religious
/ Beliefs
.58
.18
\l‘ll'llll;l1 .26
('It '
\11.lpll\ t-
(‘R '02
\1111-;111.1}111\ .‘
1‘11 .09
11111111111:.11
(It .03
111111111: -.19
(‘11 .25
\ppllctl
1.11111 .06

 

(‘R

   

.31

.35

 

Female Culturally Salient Religious
Beliefs and Practices

 

 

 

 

 

 

11011.11:

111111111

1R

 

 

 

F igge 8. Graphical display of the unidirectional CR —r ARR model.

94

.53

In the graphical model, the arrows signify the direction of causality. The numbers
near the lines represent the partial correlation (meaning the correlation between the parts
of the two variables that are unexplained by any other predictors) between the two
variables connected by the lines. Because these are standardized values (or correlations),
these may be considered effect sizes, and may be compared directly with each other to
determine which relationships are the most important in each model. The number just
outside the box of each variable represents the proportion of variation accounted for in
these variables by all of the predictors of that variable. Finally, to clarify the graphical
presentation of the model, the class level of each student was left out of the models. In
both models, class level was a statistically signiﬁcant control of Attitudes that Reﬂect
Resilience and some Coping Response variables, but was considered to be unimportant to
the theory. It was expected that those students who had made it through more years of
school would have more desirable values of Attitudes that Reﬂect Resilience and Coping
Responses.

Table 9 presents the results of the unidirectional ARR -—1 CR model, for direct
effects and moderating effects only. Table 10 presents the same results for the
unidirectional CR —+ ARR model. The effects on the outcomes of the models in Tables 9
and 10 are presented in bold. Table 11 presents all direct, indirect, and total effects of all
predictors in both models. Table 11 adds to the results in Table 9 and 10 by listing the
indirect effects of variables on Attitudes that Reﬂect Resilience and Coping Responses

that are not reported in a traditional single-outcome statistical analysis.

95

Table 8. Acronms used in Tables 9-11.

 

Acronym Full name

ADCR Adaptive Coping Responses

AF CR Applied Faith Coping Responses
ARR Attitude the Reﬂect Resilience
CSRBP Culturally Salient Religious Beliefs and Practices
DDCR Despair/Doubt Coping Responses
EMCR Emotional Coping Responses
HUCR Humor Coping Responses
NACR Non-Adaptive Coping Responses
PRB Protective Religious Beliefs
SPCR Spiritual Coping Responses

 

Table 9. Results of the unidirectional ARR —2 CR model.

 

 

 

 

 

Statistics
Regression weight
Relationship Unstandardized Standardized
Predictor Outcome B sew) B Critical ratio p-value

ARR —’ ADCR 0.44 0.05 0.53 8.76 0.00
AR -> AFCR 0.17 0.10 0.12 1.82 0.07
CSRBP —’ AFCR 0.20 0.05 0.28 3.80 0.00
PRB —» AFCR 0.60 0.16 0.30 3.69 0.00
PRB " ARR -+ AFCR 0.39 0.21 0.13 1.88 0.06
Class —9 ARR 0.11 0.03 0.24 3.47 0.00
PRB —+ ARR 0.35 0.09 0.26 3.85 0.00
ARR -+ DDCR -0.18 0.06 -0.22 -3.08 0.00
At risk -» DDCR 0.13 0.07 0.12 1.75 0.08
Class -—2 DDCR 0.06 0.03 -0.17 -2.38 0.02
ARR —+ EMCR 0.22 0.08 0.19 2.64 0.01
Class —> EMCR -0.08 0.04 -0.16 -2.39 0.02
Female —-> EMCR 0.40 0.09 0.27 4.25 0.00
Female —+ HUCR -0.28 0.13 -0.15 -2.27 0.02
ARR —. NACR -0.17 0.06 -0.19 -2.78 0.01
At risk —» NACR 0.20 0.07 0.19 2.80 0.01
CSRBP —-> PRB 0.19 0.02 0.51 9.16 0.00
ARR -» SPCR 0.18 0.06 0.15 3.22 0.00
CSRBP —+ SPCR 0.19 0.03 0.33 6.13 0.00
Female —v SPCR 0.16 0.06 0.11 2.60 0.01
PRB —+ SPCR 0.83 0.09 0.53 8.79 0.00
PRB ‘ ARR —v SPCR 0.34 0.12 0.15 2.88 0.00

 

96

Table 10. Results of the unidirectional CR —> ARR model.

 

 

 

 

 

Statistics
Raggssion weight
Relationship Unstandardized Standardized Critical
Predictor —v Outcome B sejﬂ) B ratio p-value
Class -» ADCR 0.04 0.02 0.12 1.78 0.08
PRB -» ADCR 0.20 0.07 0.18 2.92 0.00
CSRBP -+ AFCR 0.22 0.05 0.31 4.23 0.00
PRB —+ AFCR 0.52 0.14 0.26 3.61 0.00
ADCR —+ ARR 0.53 0.08 0.43 7.12 0.00
Class —» ARR 0.06 0.03 0.14 2.35 0.02
DDCR —+ ARR -0.22 0.07 -0.19 -3.04 0.00
NACR —» ARR ~0.09 0.08 -0.07 -1.10 0.27
PRB _. ARR 0.00 0.11 0.00 —0.01 1.00
PRB *NACR —> ARR 0.50 0.19 0.18 2.70 0.01
SPCR -» ARR 0.15 0.07 0.18 2.19 0.03
Class -v DDCR -0.10 0.03 -0.25 -3.80 0.00
Class -e EMCR -0.06 0.04 -0.11 -1.66 0.10
Female —o EMCR 0.40 0.10 0.27 4.27 0.00
Female -+ HUCR -0.31 0.13 -0.17 -2.44 0.02
At risk —’ NACR 0.13 0.06 0.13 2.10 0.04
CSRBP —’ PRB 0.19 0.02 0.53 9.08 0.00
CSRBP —e SPCR 0.20 0.03 0.35 6.60 0.00
Female -> SPCR 0.19 0.06 0.12 3.07 0.00
PRB —+ SPCR 0.79 0.09 0.51 9.18 0.00

 

97

Table 11. Direct, indirect, and total effects in the unidirectional ARR —v CR and CR —->

 

 

 

 

ARR models.
Unidirectional ARR - > CR model Unidirectional CR -> ARR model
Relationship Standardized effects Relationship Standardized effects
Predictor Outcome Direct Indirect Total Predictor Outcome Direct Indirect Total
ARR -o ADCR 0.53 0.00 0.53 Class —' ADCR 0.12 0.00 0.12
Class —. ADCR 0.00 0.12 0.12 CSRBP —' ADCR 0.00 0.10 0.10
CSRBP —. ADCR 0.00 0.07 0.07 PRB —. ADCR 0.18 0.00 0.18
PRB -0 ADCR 0.00 0.14 0.14 CSRBP -o AFCR 0.31 0.14 0.45
ARR —~ AFCR 0.12 0.00 0.12 PRB -~ AFCR 0.26 0.00 0.26
Class _. AFCR -0.17 0.03 -0.14 ADCR —o ARR 0.43 0.00 0.43
CSRBP -» AFCR 0.28 0.17 0.45 At risk -+ ARR 0.00 -0.01 -0.01
PRB .. AFCR 0.30 0.03 0.33 Class -o ARR 0.14 0.10 0.24
PRB ‘ ARR —. AFCR 0.13 0.00 0.13 CSRBP _. ARR 0.00 0.15 0.15
Class _. ARR 0.24 0.00 0.24 DDCR —. ARR -0.19 0.00 -0.19
CSRBP _. ARR 0.00 0.13 0.13 Female - ARR 0.00 0.02 0.02
PRB .... ARR 0.26 0.00 0.26 NACR —. ARR -0.07 0.00 ~0.07
ARR —. DDCR -O.22 0.00 -0.22 PRB -. ARR 0.00 0.17 0.17
At risk - DDCR 0.12 0.00 0.12 PRB ‘ NACR -» ARR 0.18 0.00 0.18
Class —. DDCR 0.00 0.05 0.05 SPCR -o ARR 0.18 0.00 0.18
CSRBP -. DDCR 0.00 -0.03 -0.03 Class -o DDCR -0.25 0.00 -0.25
PRB -’ DDCR 0.00 -0.06 -0.06 Class —' EMCR -0.11 0.00 -0.11
ARR _. EMCR 0.19 0.00 0.19 Female —» EMCR 0.27 0.00 0.27
Class -o EMCR -0.16 0.04 -0.12 Female -+ HUCR -0.17 0.00 -0.17
CSRBP —. EMCR 0.00 0.03 0.03 At risk -. NACR 0.13 0.00 0.13
Female _. EMCR 0.27 0.00 0.27 CSRBP —. PRB 0.53 0.00 0.53
PRB —. EMCR 0.00 0.05 0.05 CSRBP _. SPCR 0.35 0.27 0.62
Female —. HUCR -0.15 0.00 ~0.15 Female —. SPCR 0.12 0.00 0.12
ARR —' NACR -0.19 0.00 -0.19 PRB - SPCR 0.51 0.00 0.51
At risk —+ NACR 0.19 0.00 0.19
Class - NACR 0.00 -0.05 -0.05
CSRBP -’ NACR 0.00 -0.03 —0.03
PRB —o NACR 0.00 -0.05 -0.05
CSRBP _. PRB 0.51 0.00 0.51
ARR -. SPCR 0.15 0.00 0.15
Class —+ SPCR 0.00 0.04 0.04
CSRBP _. SPCR 0.33 0.29 0.62
Female _. SPCR 0.11 0.00 0.11
PRB -0 SPCR 0.53 0.04 0.57
PRB‘ARR —oSPCR 0.15 0.00 0.15

 

Results of the AR —’ CR model

The results of the AR —r CR model are presented here as a representation of the

relationships among variables in the data. The competing model is presented in the same

fashion below.

Direct Effecta of Attitudes that Reﬂect Resilience on Coping Resmnses. The

ARR -+ CR model reveals Attitudes that Reﬂect Resilience and Coping Responses are

distinct but related constructs. The results in Table 9 indicate a statistically signiﬁcant

direct effect of Attitudes that Reﬂect Resilience on six out of the seven coping responses

98

assessed in this study. Ordered by decreasing standardized regression weights (size of the
relationships), they are Adaptive Coping Responses (.53), Despair/doubt Coping
Responses (-.22), Emotion Coping Responses (.19), Non-adaptive Coping Responses
(-.19), Spiritual Coping Responses (.15), and Applied Faith Coping Responses (.12).

These standardized regression weights are effect sizes. They represent the amount
of change in the Coping Responses (in terms of standard deviations) expected for a one
standard deviation change in Attitudes that Reﬂect Resilience, controlling for all other
predictors of the individual coping responses being predicted. Where the sign is negative,
an increase in Attitudes that Reﬂect Resilience predicts a decrease in Coping Responses,
but when positive, an increase in Attitudes that Reﬂect Resilience predicts an increase in
Coping Responses. A small effect is indicated by the absolute value of a standardized
regression weight (path coefﬁcient) less than 0.1, a medium effect is indicated by the
absolute value of a standardized regression weight (path coefﬁcient) around 0.3, and a
large effect is indicated by the absolute value of a standardized regression weight (path
coefﬁcient) of 0.5 or larger (see Kline, 1998).

Accordingly, Adaptive Coping Responses are expected to improve by 0.53
standard deviations given an increase of one standard deviation in Attitudes that Reﬂect
Resilience, controlling for any changes in other variables. Results in this study show that
Attitudes that Reﬂect Resilience have a large statistically signiﬁcant relationship
with/effect on Adaptive Coping Responses. Despair/doubt Coping Responses are
expected to decrease by .22 standard deviations given an increase of one standard
deviation in Attitudes that Reﬂect Resilience, controlling for the direct and indirect

effects from other variables in the model. Emotional Coping Responses are expected to

99

increase by .19 standard deviations given an increase of one standard deviation in
Attitudes that Reﬂect Resilience, controlling for the direct and indirect effects from other
variables in the model. Non-adaptive Coping Responses are expected to decrease by .19
standard deviations given one standard deviation increase in Attitudes that Reﬂect
Resilience, controlling for all other direct and indirect effects in the model. And ﬁnally,
Spiritual Coping Responses are expected to increase by .15 standard deviations given one
standard deviation increase in Attitudes that Reﬂect Resilience, controlling for all other
direct and indirect effects in the model. The relationships between Attitudes that Reﬂect
Resilience and the identiﬁed Coping Responses in the model are statistically signiﬁcant
with the exception of Applied Faith Coping Response. The p-value for this relationship
is just above the p < .05 level of signiﬁcance.

The Interpretation of the Relationships. When Attitudes that Reﬂect Resilience
increase, Adaptive Coping Responses, Emotional Coping Responses, Spiritual Coping
Responses, and Applied Faith Coping Responses tend to increase. The Non-adaptive
Coping Responses and Despair/doubt Coping Responses tend to decrease. The p-value of
the relationship between Attitudes that Reﬂect Resilience and Applied Faith Coping
Responses was p < 0.07, slightly above the cutoff of 0.05, but small enough to merit
discussion.

Direct Effects of Protective Religio_us Beliefs on Coping Responses and Attitudes
that Reﬂect Resilience. Protective Religious Beliefs (PRB) had direct, statistically
signiﬁcant effects on Spiritual Coping Responses (0.53), and Applied Faith Coping

Responses (0.30), and Attitudes that Reﬂect Resilience (0.26). This means that students

100

who held Protective Religious Beliefs also tended to use Spiritual Coping Responses,
Applied Faith Coping Responses, and Attitudes that Reﬂect Resilience.

Wat Reﬂect Resilience and Protective Religm
Beliefs on Coping Responﬂ Protective Religious Beliefs had essentially equivalent
moderating effects on the relationship between Attitudes that Reﬂect Resilience and
Spiritual Coping Responses (0.15) and on the relationship between Attitudes that Reﬂect
Resilience and Applied Faith Coping Responses (0.13). Because this is an interaction
effect, the interpretation is difﬁcult. To aid in the interpretation, both relationships are
presented in graphical form in Figures 9 and 10. These ﬁgures show that even in
interaction, the combined effects on Spiritual and Applied Faith Coping Responses on
Attitudes that Reﬂect Resilience and Protective Religious Beliefs are almost identical—
the graphs are very similar. The graphs show that students with high, moderate, and low
use of Protective Religious Beliefs (PRB) who possess low levels of Attitudes that
Reﬂect Resilience (ARR) use about the same amount of Spiritual and Applied Faith
Coping Responses. However, as Attitudes that Reﬂect Resilience increase in strength,
students with high PRB and high ARR tend to use these coping responses more, students
with a moderate level of PRB and high ARR tend to use these coping responses
moderately, and students with low PRB and high ARR tend to use fewer
Spiritual/Applied Faith Coping Responses.

Other Direct Effects on Coping Resmnses. Gender, class-level, Culturally Salient
Religious Beliefs and Practices, and At Risk status demonstrated statistically signiﬁcant
direct effects on coping responses as well. Females tended to use Spiritual Coping

Responses (0.10) and Emotional Coping Responses (0.27) more than males, and tended

101

to use Humor Coping Responses (-0. 15) less than males, with the most important effect
being on the use of Emotional Coping Responses. Females did not show any tendencies
to use any other type of Coping Responses more or less than males.

At Risk students tended to use more Non-adaptive (0.19) and Despair/doubt
(0.12) Coping Responses than non-At Risk students, but showed no differences in the use
of any other Coping Responses. There was no direct or indirect relationship between risk
status and Attitudes that Reﬂect Resilience in this model.

Culturally Salient Religious Beliefs and Practices had a statistically signiﬁcant
positive effect on Spiritual Coping Responses (0.3 3) and Applied Faith Coping
Responses (0.28), meaning that students who identiﬁed with the Culturally Salient
Religious Beliefs and Practices tended to use the faith-based coping responses more than
other students.

Class level had small statistically signiﬁcant effects on Emotional Coping
Responses, Despair/doubt Coping Responses, and on Attitudes the Reﬂect Resilience.
This suggests that these variables vary with class level. However, these cannot be
considered to be effects of class level per se. While it may be that advancing through
school results in increased Attitudes that Reﬂect Resilience and changes in coping
responses, it may also be that having higher Attitudes that Reﬂect Resilience and more
favorable coping response results in being able to advance in school. Therefore, class

level serves only as a control to reduce noise in the model.

102

 

High

 

 

 

 

 

 

 

 

 

m / p
g \ l ....... Low PRB
5 1——Moderate PRB
En 1—High PRB
a, .
1
Low . .
Low Moderate High
AR

1
1

FM 9. Spiritual CR as an outcome of ARR, moderated by PRB.

 

 

 

 

 

 

 

 

 

 

 

1 High
11:
o
g 1 ....... Low PRB
: 1 —— Moderate PRB
é 1—High PRB
a l
O.
<
Low . .
Low Moderate High
1 AR

1

Figm 10. Applied Faith CR as an outcome of ARR, moderated by PRB.

103

 

1
1
1

Indirect effects on coping responag. Culturally Salient Religious Beliefs and
Practices (CSRBP) and Protective Religious Beliefs (PRB) also had important indirect
effects on coping responses. Although it had no direct effect on Adaptive Coping
Responses, Protective Religious Beliefs had an important indirect effect (0.14). This is
because, if we assume the model is correct, Protective Religious Beliefs had a direct
effect on Attitudes that Reﬂect Resilience, which in turn had a direct effect on Adaptive
Coping Responses. Changes in Culturally Salient Religious Beliefs and Practices eﬁ’ect
changes in Protective Religious Beliefs that in turn, effect changes in Attitudes that
Reﬂect Resilience, which effects a change in Adaptive Coping Responses, even though it
does not do so directly.

In addition to the direct effects already described, Culturally Salient Religious
Beliefs and Practices also had an indirect effect through Protective Religious Beliefs on
Spiritual Coping Responses (0.29) and Applied Faith Coping Responses (0.17). Adding
the direct and indirect effects together, the total effects of Culturally Salient Religious
Beliefs and Practices on Spiritual Coping Responses and Applied Faith Coping
Responses were 0.62 and 0.45 respectively. Whereas the direct effects were moderate,
the total effects are large.

Summary of the ARR —» CR model

In the AR -> CR model, higher levels of Adaptive, Emotional, Spiritual, and
Applied Faith Coping Responses are associated with higher levels of Attitudes that
Reﬂect Resilience among both the at-risk and non-at risk Aﬁican American college
students sampled. Use of Non-adaptive Coping Responses and Despair/doubt Coping

Responses was inversely related to Attitudes that Reﬂect Resilience — as attitudes that

104

reﬂect resilience increased, there was a decrease in these two types of coping responses.
While the at-risk students sampled tended to use more harmful coping responses than
other students, they tended not to use any of the non-harmful coping responses any more
or less than students not at risk. There was no direct or indirect relationship between risk
status and Attitudes that Reﬂect Resilience in this model.

For all students, lower levels of attitudes that reﬂect resilience were associated
with use of the harmful coping responses (e. g., Non-adaptive Coping Responses and
Despair/doubt Coping Responses). Finally, females tended to use Spiritual Coping
Responses and Emotional Coping Responses more than males, and they tended to use less
Humor Coping Responses than males.

Protective Religious Beliefs (PRB) also have a large and statistically signiﬁcant
direct effect on Spiritual Coping Responses, medium direct effect on Applied Faith
Coping Responses, and medium direct effect on Attitudes that Reﬂect Resilience.
Culturally Salient Religious Beliefs and Practices (CSRBP) have a large direct effect on
Protective Religious Beliefs (PRB), a medium direct effect on Spiritual Coping
Responses, and a medium direct effect on Applied Faith Coping Responses. It is
important to note that Culturally Salient Religious Beliefs and Practices are distinct but
related to Protective Religious Beliefs.

Protective religious beliefs (PRB) also serve as a statistically signiﬁcant
moderator in the relationship between two Of the outcome variables (Spiritual Coping
Responses and Applied Faith Coping Responses) and the main predictor (Attitudes that
Reﬂect Resilience). This moderating effect takes the following form: where students had

low Attitudes that Reﬂect Resilience they used Spiritual and Applied Faith Coping

105

Responses moderately regardless of their Protective Religious Beliefs, but where students
had high Attitudes that Reﬂect Resilience their use of Spiritual and Applied Faith Coping
Responses depended heavily on their Protective Religious Beliefs. Those with high
Protective Religious Beliefs tended to use Spiritual and Applied Faith Coping Responses
a great deal, but those with low Protective Religious Beliefs tended to use Spiritual and
Applied Faith Coping Responses very little. Hence Protective Religious Beliefs (PRB)
are important moderators in the AR —r CR model.
Raaalja of the CR —’ ARR model

Direct effeﬁ of coping responses on ARR. The results in Table 10 indicate a
statistically signiﬁcant direct effect of four coping responses (CR) on Attitudes that
Reﬂect Resilience (ARR). Ordered by decreasing effect size, they are Adaptive Coping
Responses (.43), Despair/doubt Coping Responses (-.l9), Spiritual Coping Responses
(.18), and Non-Adaptive Coping Responses (-.07). These effect sizes represent the
amount of change in attitudes that reﬂect resilience in terms of standard deviations
expected for one standard deviation change in the coping response predicting attitudes
that reﬂect resilience, controlling for all other predictors of the attitudes that reﬂect
resilience. A negative sign on a standardized regression weight indicates that an increase
in the use of the designated coping response predicts a decrease in attitudes that reﬂect
resilience. A positive standardized regression weight indicates that an increase in the use
of the designated coping response predicts an increase in attitudes that reﬂect resilience.
The deﬁnition of the importance of the effects in this model are the same as in the ARR

—» CR model: a small effect is indicated by an absolute value less than 0.1, a medium

106

effect is indicated by the absolute value around 0.3, and a large effect is indicated by the
absolute value of 0.5 or greater.

Accordingly, Attitudes that Reﬂect Resilience are expected to increase by .43
standard deviations given an increase of one standard deviation in Adaptive Coping
Responses, when Adaptive Coping Response is the only predictor that changes. Attitudes
that Reﬂect Resilience are expected to decrease by .19 standard deviations given an
increase of one standard deviation in Despair/doubt Coping Responses, controlling for
other variables.

Attitudes that Reﬂect Resilience are expected to increase by .18 standard
deviations given an increase of one standard deviation in Spiritual Coping Responses,
controlling for all other variables. And ﬁnally, Attitudes that Reﬂect Resilience are
expected to decrease by .07 standard deviations given an increase of one standard
deviation in Non-adaptive Coping Responses, controlling for all other variables. All of
the relationships in the CR —> ARR model are statistically signiﬁcant with the exception
of Non-Adaptive Coping Responses on Attitudes that Reﬂect Resilience: NACR —> ARR
(p < .27). This relationship was maintained in the model as a main effect in order to
adequately understand the interaction effect to which it contributes.

While at-risk students tend to use more Non-adaptive Coping Responses than
Non-at-risk students, there was no direct relationship, or effect of risk status on Attitudes

that Reﬂect Resilience in this model.

Direct effect of Protective Religious Beliefs on Attitudes that Reﬂect Resilience.

In this model, Protective Religious Beliefs do not have a direct effect on Attitudes that

Reﬂect Resilience (effect size = 0.00, p < 1.00). This relationship was also maintained in

107

the model as a main effect in order to understand the interaction effect to which it
contributes.

Combined effect of Protective Religio_us Beliefs and Non-adapaiye—Copjag
Respgnses on Attitudes that Reﬂect Resilience. As noted above, neither Protective
Religious Beliefs nor Non-adaptive Coping Responses had statistically signiﬁcant main
effects on Attitudes that Reﬂect Resilience, but they did have a statistically signiﬁcant
combined, or interaction effect on Attitudes that Reﬂect Resilience. This suggests that
Protective Religious Beliefs moderate the relationship between Non-adaptive Coping
Responses and Attitudes that Reﬂect Resilience. Figure 11 shows that when students use
Non-adaptive Coping Responses very little, there is no effect of Protective Religious
Beliefs. However, when students use Non-adaptive Coping Responses a great deal,
students with low Protective Religious Beliefs tend to have low Attitudes that Reﬂect
Resilience, but students with high Protective Religious Beliefs tend to have high
Attitudes that Reﬂect Resilience. This suggest that the name “Protective Religious
Beliefs” is a reasonable label for this moderator—having high Protective Religious
Beliefs may protect students from harmful effects of using Non-adaptive Coping
Responses. This is important for the at-risk students because they tend to use Non-

adaptive COping Responses more Often than students not at risk.

108

 

 

 

 

 

 

 

 

 

 

Non-Ada ptlve CR

1

l 2
l A.

1

1 l
1 l
1 1
g l ------- Low PRB

g —Moderate PRB

i \ 1—High lane 11
,2 1
1 1
: Low e .

1 Low Moderate High

1

Figtle ll. ARR as an outcome of Non-adaptive CR moderated by PRB. J

Other direct effects. As in the AR —> CR model, class level had a small
statistically signiﬁcant direct effect (.14) and an indirect effect (.10) on Attitudes that
Reﬂect Resilience, yielding a total effect of class on Attitudes that Reﬂect Resilience of
.24. This suggests that Attitudes that Reﬂect Resilience varies with class level, but as in
the AR —> CR model, this cannot be considered to be an effect of class level. While it
may be that advancing through school results in increased Attitudes that Reﬂect
Resilience, it may also be that having higher levels of Attitudes that Reﬂect Resilience
may result in being able to advance in school. Therefore, this variable serves only as a
control to reduce noise in the model.

In this model, Protective Religious Beliefs have a direct effect on Spiritual Coping
Responses (.51), Applied Faith Coping Responses (.26) and Adaptive Coping Responses

(.18). This means that students that held greater Protective Religious Beliefs also tended

109

to use Spiritual Coping Responses, Applied Coping Responses, and Adaptive Coping
Responses more often than students with lesser Protective Religious Beliefs.

Indirect effects on Attitrales that Reﬂect Resilience. Attitudes that Reﬂect
Resilience were indirectly affected by Protective Religious Beliefs (.17), Culturally
Salient Religious Beliefs and Practices (.15), At-Risk status (.12) in this model. The
route of the effect for Protective Religious Beliefs was through Adaptive Coping
Responses and Spiritual Coping Responses. This does not include the moderation of the
relationship between Non-adaptive Coping Responses and Attitudes that Reﬂect
Resilience explained above. The indirect effect of Culturally Salient Religious Beliefs
and Practices on ARR takes a number of routes: (1) CSRBP-to-Spiritual CR, (2) CSRBP-
to-PRB-to-Adaptive CR, and (3) CSRBP-to-PRB-to-Spiritual CR. Although each Of
these individual indirect effects was small, they combine to create an important effect.
All other possible indirect effects (e. g. At-Risk-to-Non-adaptive CR-to-ARR) were too
small to be considered important.

Summamof the CR -—> ARR model

In the CR -1 ARR model, Attitudes that Reﬂect Resilience and Coping Responses
are separate but related constructs. Students who used Adaptive Coping Responses and
Spiritual Coping Responses more than others also tended to have higher Attitudes that
Reﬂect Resilience. On the contrary, students who used Despair/doubt Coping Responses
and Non-adaptive Coping Responses more than others tended to have lower Attitudes that
Reﬂect Resilience.

Protective Religious Beliefs (PRB) were a statistically signiﬁcant moderator Of

the relationship between Non-adaptive Coping Responses and Attitudes that Reﬂect

110

Resilience. At-Risk students with higher Protective Religious Beliefs tended to have
higher levels of Attitudes that Reﬂect Resilience, even if they use Non-adaptive Coping
Responses; but Protective Religious Beliefs did not predict difference in Attitudes that
Reﬂect Resilience for students who tended not to use Non-adaptive Coping Responses.
Protective Religious Beliefs also have direct positive effects on Adaptive Coping
Responses, Spiritual Coping Responses and Applied Faith Coping Responses.

At-risk students tended to use more Non-adaptive Coping Responses than Non-at-
risk students, but there was no direct relationship, or effect of risk status on Attitudes that
Reﬂect Resilience in this model.

Summm

The overall purpose of this study was to examine the relationship between
attitudes that reﬂect resilience and cOping among at risk and non-at risk African
American college students, and to examine the moderating role of three factors that are
believed to be protective in times of stress and adversity. This summary will be
Organized around the research questions that guided this study.

The methods used to test the hypothesis for this study included Structural
Equation Modeling and Path Analysis. The results from each model may be summarized
in the following manner, in response to the research questions:

1. In both the AR —+ CR and the CR -+ ARR models, Attitudes that Reﬂect
Resilience and Coping Responses are separate but related constructs. There
are statistically signiﬁcant relationships among the Attitudes that Reﬂect
Resilience and select Coping Responses among both At-Risk and Non-At-

Risk African American college students. The nature of these relationships

11]

among Attitudes that Reﬂect Resilience and Coping Responses was the same
for the At Risk and Non-At-Risk students with the exception of Non-adaptive
Coping Responses.

. Core Beliefs, Religiosity, and Spirituality were combined and re-
conceptualized as “Protective Religious Beliefs” and were a statistically
signiﬁcant moderator of the relationships between Attitudes that Reﬂect
Resilience and select Coping Responses. In particular, Protective Religious
Beliefs appeared to enhance the use of Spiritual and Applied Faith Coping
Responses among students with high Attitudes that Reﬂect Resilience, and
High Protective Religious Beliefs appeared to protect students from the
negative effects of using Non-adaptive Coping Responses.

. Level of Acculturation was re-conceptualized as “Culturally Salient Religious
Beliefs and Practices” has a direct positive effect on Protective Religious
Beliefs, both direct and indirect positive effects on select Coping Responses,
and an indirect effect on Attitudes that Reﬂect Resilience.

. Attitudes that Reﬂect Resilience vary by class level, but the validity of that
relationship was highly questionable. It may be that progressing through
school tends to have a positive effect on this outcome, but it may also be that
more favorable values on these outcomes enables students to advance in
school. There were no direct or indirect relationships of At-Risk status with
Attitudes that Reﬂect Resilience in either model. Program participation was
not investigated because the within-group sample size of some groups was too

small, and this was not a speciﬁc purpose of this study. Finally, self-reported

112

measures of SES and achievement (GPA and ACT scores) were not
investigated because the data were deemed unreliable.

. The relationship between Attitudes that Reﬂect Resilience and Coping
Responses may be both unidirectional and reciprocal. This study was unable
to determine in which direction the relationships are stronger, and was unable

to determine whether the relationships are reciprocal.

113

CHAPTER VI. SUMMARY AND CONCLUSIONS

Summary

The primary purposes of this study were (1) to examine the relationship between
attitudes that reﬂect resilience and coping among at risk and non-at risk African
American college students, (2) to examine the moderating role of three factors that are
believed to be protective in times of stress and adversity, and (3) to examine the
relationship between levels of acculturation and coping responses, attitudes that reﬂect
resilience, religiosity, spirituality, and core beliefs.

Prior research assessing the relationship between resilience and coping has

yielded inconclusive results. For example, Jew, Green and Kroger (1999) generated a
hypothesis about the nature of this relationship. Drawing upon the Cognitive Appraisal
Theory of Stress and Coping (Lazarus, & Folkman, 1984), Jew et al. deﬁned resilience as
a system of beliefs and hypothesized that coping emerges from this system of beliefs.
From this hypothesis, they concluded that resilience was causally related to coping, with
coping arising from resilience. Jew et a]. were able to identify differences in resilience
between adolescents who reported being exposed to a high number of risk factors, and
adolescents having few or no risk factors, however, they were unable to identify any
differences in coping that could be associated with high or low levels of resilience. They
concluded that there is a need for additional research on the relationship between coping
and resilience.

The current study attempted to further advance our understanding of the

relationship between resilience, coping, and protective factors by examining attitudes that

114

reﬂect resilience in African American college students attending a predominantly White
institution. It differs from previous studies of resilience in that it emphasized attitudes
that reﬂect resilience rather than attempt to measure resilience directly.

The population for this study consisted of 229 African American college students
enrolled at Michigan State University during the Spring and Summer 2001 semesters.
Four student groups were sampled to create most of the subject pool. They included
SUPER, MAGIC, McNair-SROP and the Sports Camp Program. Approximately 13
students from the general population were included in this pool of participants. More
than 95 percent of the sample used in this study identiﬁed with some religion, primarily
Christianity, and approximately half of the participants met the operational deﬁnition of
“at-risk”.

Seven instruments were used to collect data. The instruments were: (1) the COPE
(Carver, Scheier & Weintraub, 1989), (2) the Religious Coping Activities Scale (RCAS;
Pargament, Ensing, Falgout, Olsen, Reilly, Van Haitsma & Warren, 1990), (3) the
Multidimensionality of Religiosity Scale (MRS; Levin, Taylor, & Chatters, 1995), (4) the
Core Convictions Scale (CCS, developed speciﬁcally for this study), (5) the African
American Acculturation Scale - Revised (AAAS-R; Klonof & Landrine, 2000), (6) the
Resiliency Attitudes Scale (RAS; Biscoe & Harris, 1994), and (7) a demographics
questionnaire developed speciﬁcally for this study.

Instruments (except the Core Convictions Scale and the Demographics Survey)
were selected if they met the following criterion: (1) they were empirically, theoretically,
or conceptually derived with some evidence of reliability and validity, (2) they measured

a belief, an attitude, or a behavior that might help explain a cognitive-behavioral

115

conceptualization of resilience and coping among African American college students, (3)
responses could be scaled to be used in correlational analysis and structural equation
modeling (SEM); and (4) their instructions were easy to follow.

To gather demographic information, an ll-item demographics questionnaire was
developed and used. The information requested included age, gender, race/ethnicity,
class level, GPA, ACT score, parental annual income, home environment, reason for
coming to MSU, religious preference, and type of high school attended.

The second instrument used was the COPE (Carver, Scheier & Weintraub, 1989).
The original 60-item instrument was adapted for this study to identify conceptually
distinct coping responses. The items from this scale were combined with items from the
Religious Coping Activities Scale (RCAS) and factor analyzed to simplify the analysis,
and to create a representative measure of coping for the participants in this study. Coping
Responses and Attitudes the Reﬂect Resilience served as (1) predictor variable and (2)
outcome variable in the models generated in this study.

The third instrument used was the Religious Coping Activities Scale (RCAS;
Pargament, Ensing, F algout, Olsen, Reilly, Van Haitsma & Warren, 1990). The original
29-item scale was adapted for this study to identify three conceptually related but distinct
religious coping responses. As previously mentioned, these three Coping Responses were
combined with an adaptation of the COPE to simplify the analysis, and to form a
representative measure of coping for the participants in this study.

The fourth instrument used was the Resiliency Attitudes Scale (RAS; Biscoe &

Harris, 1994). The original 72-item instrument was adapted for this study to identify

116

attitudes that reﬂect resilience. The instrument was scaled down to an 8-item scale
consisting of eight items from the general resiliency subscale of the RAS.

The ﬁfth instrument used was the Core Convictions Scale (CCS). This scale was
developed speciﬁcally for this study, and consisted often items that reﬂect beliefs that are
held by African American people about God, faith, and life.

The sixth instrument used in this study was the Multidimensionality of Religiosity
Scale (MRS; Levin, Taylor, & Chatters, 1995). Three questions were extracted from the
original 16-item instrument and were combined with the ten items from the Core
Convictions Scale to create the variable called “Protective Religious Beliefs (PRB).”

The seventh instrument used in this study was the African American
Acculturation Scale - Revised (AAAS-R; Klonof & Landrine, 2000). The AAAS-R is a
47-item assessment of how closely an individual identiﬁes with African American culture
and race. Nine items from the Religious Beliefs and Practices subscale of this instrument
were used to represent Culturally Salient Religious Beliefs and Practices (CSRBP) in this
study.

Statistical methods used in this study included (1) Reliability and Exploratory
Factor Analysis for sealing instruments, and (2) structural equation modeling and path
analysis to derive the ﬁnal models. Four hypothesis were developed to examine the
research questions in this study. A .05 level of signiﬁcance was used as a guide for all
hypothesis testing, but effect sizes were used to interpret the relationships in terms of

importance.

”7

Conclusions

Attitudes that Reﬂect Resilience were deﬁned as beliefs that one can survive and
make things better for self and others. This study found a relationship between these
attitudes and Adaptive, Spiritual, Applied Faith, Emotional, Non-adaptive, and
Despair/doubt coping responses. There was however, no relationship between these
attitudes and Humor Coping Responses, and no direct or indirect relationship between
Attitudes that Reﬂect Resilience and risk status. In addition, Protective Religious Beliefs
were found to be an important moderator of the relationships between Attitudes that
Reﬂect Resilience and select Coping Responses. Protective Religious Beliefs may
enhance the use of Spiritual and Applied Faith Coping Responses among students with
high Attitudes that Reﬂect Resilience, and High Protective Religious Beliefs may protect
students from the negative effects of using Non-adaptive Coping Responses

The results of this study inform the research questions as follows:

1. In both the AR -> CR and the CR —» ARR models, Attitudes that Reﬂect
Resilience and Coping Responses were separate but related constructs. There
were statistically signiﬁcant relationships among the Attitudes that Reﬂect
Resilience and select Coping Responses among both At-Risk and Non-At-
Risk Aﬁican American college students. The nature of these relationships
among Attitudes that Reﬂect Resilience and Coping Responses was the same
for the At Risk and Non-At-Risk students with the exception on Non-adaptive
Coping Responses.

2. Core Beliefs, Religiosity, and Spirituality were combined and re-

conceptualized as “Protective Religious Beliefs” and were a statistically

118

signiﬁcant moderator of the relationships between Attitudes that Reﬂect
Resilience and select Coping Responses. In particular, Protective Religious
Beliefs may enhance the use of Spiritual and Applied Faith Coping Responses
among students with high Attitudes that Reﬂect Resilience, and High
Protective Religious Beliefs may be associated with low or no use of Non-
adaptive Coping Responses.

. Level of Acculturation was reoconceptualized as “Culturally Salient Religious
Beliefs and Practices” had a direct positive effect on Protective Religious
Beliefs, both direct and indirect positive effects on select Coping Responses,
and an indirect effect on Attitudes that Reﬂect Resilience.

. Attitudes that Reﬂect Resilience varied by class level, but the validity of that
relationship was highly questionable. It may be that progressing through
school tends to have a positive effect on this outcome, but it may also be that
more favorable values on these outcomes enables students to advance in
school. There were no direct or indirect relationships of At-Risk status with
Attitudes that Reﬂect Resilience in either model. Program participation was
not investigated because the within-group sample size of some groups was too
small. Finally, self-reported measures of SES and achievement (GPA and
ACT scores) were not investigated because the data were deemed unreliable.

. The relationship between Attitudes that Reﬂect Resilience and Coping
Responses may be both unidirectional and reciprocal. This study was unable
to determine in which direction the relationships were stronger, and was

unable to determine whether the relationships were reciprocal.

119

Discuasion of Results

This study contributes to the ﬁeld of research on resilience in a number of ways.
First, this study further informs the ﬁeld on the scaling of some instruments considered to
be useful in the study of resilience. Second, it provides additional evidence regarding the
relationships between Attitudes that Reﬂect Resilience and Coping Responses
particularly as they relate to African Americans. Third, it provides support for Cognitive
Appraisal Theory of Stress, Coping and Resilience. Fourth, it raises questions about
labeling students as “at-risk”. Finally, it adds empirical evidence for the protective
inﬂuences of religious beliefs. Each of these will be discussed below.
Instrument scaling

Path analysis requires the use of measures with relatively good psychometric
qualities and a clear and substantive rationale for specifying the direction of relationships
indicated in the model (Kline, 1998). The use of valid and reliable instruments improves
the interpretation of the models. The instruments that were used in this study were
scrutinized for their validity and reliability, and data from each instrument were carefully
entered and coded for interpretation. Overall, each of the instruments had good reliability
and validity indices. The Resiliency Attitudes Scale, the combined COPE — Religious
Coping Activities Scale, and the Core Convictions Scale were piloted in this study.

Given concern about the adequacy of the scaling methods used by prior
researchers, ﬁve of the seven instruments used in this study were subjected to principal
components exploratory factor analysis: the COPE (Carver, Scheier & Weintraub, 1989),
the Religious Coping Activities Scale (RCAS; Pargament, Ensing, Falgout, Olsen, Reilly,

Van Haitsma & Warren, 1990), the Resiliency Attitudes Scale (RAS; Biscoe & Harris,

120

1994), the Aﬁican American Acculturation Scale - Revised (AAAS-R; Klonof &
Landrine, 2000), and the Core Convictions Scale (CCS, developed speciﬁcally for this
study). The demographics survey did not require EF A, and since only three questions
from the Multidimensionality of Religiosity Scale (MRS; Levin, Taylor, & Chatters,
1995) were used, it did not require EFA.

Surprisingly, after Exploratory Factor Analysis (EFA), the number of factors for
each instrument decreased from the number of expected factors with the exception of the
Core Convictions Scale. The reduction in factors may be due to the fact that an outdated
rule-of-thumb was used for determining the number of factors when these instruments
were created, but this study used a more current rule of thumb that is less susceptible to
random variations in error. This reduction in factors assured that in modeling, there was
no over-interpretation of the data since there were no empirically unsupported scales
represented as important outcomes or predictors.

To ﬁt with existing theory and to redress concerns about being able to measure
resilience directly, instruments designed to be a direct measure of resilience were
rejected, and a more ﬁtting construct was identiﬁed that would still allow the author to
investigate the relationships of interest. Hence, a measure of attitudes that reﬂect
resilience (the RAS) was identiﬁed and used in the study.

There was considerable concern about the reliability of six of the subscales in the
Resiliency Attitudes Scale (RAS; Biscoe & Harris, 1994) and the concepts that they
represent. The reliability on three of the subscales (Insight, Independence, and
Relationships) was poor, and on three out of the four remaining subscales (Initiative,

Creativity / Humor, and Morality), it was marginal. It is difﬁcult to explain the relatively

l2l

low reliabilities on these subscales other than to suggest that they could possibly have
been inﬂuenced by signiﬁcant differences between the sample upon which the scale was
normed and the sample in this study.

The other major concern with these six subscales is that they did not appear to ﬁt
well conceptually with the general resilience subscale. General resilience was deﬁned as
the belief that one can survive and make things better for self and others (Biscoe and
Harris, 1994). The items that make up this subscale ﬁt well conceptually with this
deﬁnition — “NO matter what happens, if I keep trying I’ll get through it”, and “Even if
bad things happen, I can deal with them.”

Items from the remaining six subscales were intended to measure six distinct but
related “resiliencies.” Even with reverse scoring, many of these items did not ﬁt well
conceptually with the deﬁnition of resilience given by Biscoe and Harris (1994).

A rescaling of this instrument was attempted, but no solution was found that made
any conceptual sense. It was determined that only the experimental general resiliency
subscale was useful to this study. The low reliabilities in the subscales of the Resiliency
Attitudes Scale, and the conceptual inconsistency of other possible scaling results suggest
the need to examine whether these other “resiliencies” exist at all, and if so, the items
measuring them may need to be reworked completely.

Since coping was an important variable in this study, two scales were selected to
measure it. Each scale was a valid and reliable measure of coping for the sample. Each
instrument was factor analyzed, and the resulting solutions had fewer factors.

Even though the number of factors in the instruments decreased, the instruments

sustained their conceptual integrity and reliability. For instance, the Adaptive Coping

122

items loaded on the correct factor in this sample. The ease of explaining the relationships
between subscales in the restructured COPE, for example, provides some strong construct
validity evidence for the subscales as used in this sample, and validates the claims of the
authors that the instrument measures conceptually distinct sub-constructs of coping.

To make the path analysis less cumbersome, however, items from the two
instruments were combined and factor analyzed. These two instruments were
successfully merged without changing the meaning of factors, and provided further
evidence of its validity. The combined scale contributes to the literature a reliable and
valid instrument that allows for the study of much more speciﬁc coping behaviors than
possible with the instrument used in previous studies of resilience and coping. The
combined instrument may be used with populations having similar characteristics to the
one used in this study.

Evidence of a Causal Relationship between Attitudes that Reﬂect Resilience and Coping?

This investigation resulted in the development of two competing models that
illustrate the relationship between coping responses and attitudes that reﬂect resilience.
Regarding interpretation of results from path analysis, Kline (1998) warns

Thus, the interpretation that direct effects in a path model that was not rejected

must correspond to causal relations in the real world is typically unwarranted. It

is usually only with the accumulation of the following types of evidence that
researchers can even begin to think that the results of a path analysis may indicate

causality: (1) replication of the model across independent samples; (2)

corroborating evidence from experimental studies of variables in the model that

are manipulable; and (3) the accurate prediction of the effects of interventions.

123

Without such evidence, structural models, whether of observed or latent variables,

are best seen as “as if” models of causality (pp. 142-143).

While important patterns of relationships were demonstrated in the resultant
competing models, this study was unable to deﬁnitively determine the direction of the
relationships or conﬁrm a causal relationship. It is therefore imperative that researchers
continue to investigate the relationship between attitudes that reﬂect resilience and coping
responses. Additional studies may include a larger sample of participants, continue to
gather psychometric information using the Resiliency Attitudes Scale (Biscoe & Harris,
1994), and should examine other measures of risk (e.g., social economic status, measures
of both dispositional and situational stress, etc.).

Several conclusions about the sample however can be drawn ﬁom each model.
Both models show that attitudes that reﬂect resilience and coping responses are related
but distinct constructs. This ﬁnding is important since it has been suggested in the
literature that we do away with the term resilience and instead, use the term “coping with”
(Kumpfer, 1999). In addition, both models illustrate that there are gender-based
differences in coping responses and a tendency for the at-risk students to use more
negative coping responses (i.e., non-adaptive coping and despair/doubt coping) than other
students in the study. It is important to note that there were no direct or indirect gender-
based differences in attitudes that reﬂect resilience.

The CR —+ ARR model shows that Protective Religious Beliefs enhance the
relationship between Non-adaptive Coping Responses and Attitudes that Reﬂect
Resilience. The Protective Religious Beliefs (Core Convictions, Religiosity, and

Spirituality) have direct and indirect positive inﬂuences on coping behaviors and

124

Attitudes that Reﬂect Resilience. The importance of this relationship is even more salient
in the CR —-> ARR model, where Protective Religious Beliefs moderate the relationship
between Non-adaptive Coping Responses and Attitudes that Reﬂect Resilience. And
ﬁnally, both models show that Culturally Salient Religious Beliefs and Practices have a
signiﬁcant direct relationship with Protective Religious Beliefs and some Coping
Responses.

The lack of a relationship between Attitudes that Reﬂect Resilience and Humor
Coping Responses in each model would be expected as the humorous coping responses
often include a form of behavioral and mental disengagement, and dismissal of, or
ignoring the seriousness of a real problem. This subscale reﬂected a similar factor
structure as in the original COPE (Carver, Scheier & Weintraub, 1989) and consists of the
following items: “I laugh about the situation”, “I make jokes about it”, “I kid around
about it”, and “I make fun of the situation.” It makes sense that these items are not
related to the belief that one can survive and make things better for self and others.

In the model in which attitudes predict coping responses (ARR —» CR), the
relationships between Attitudes that Reﬂect Resilience and Coping Responses were
signiﬁcant for Six out of the seven possible coping responses. The p-value for the
relationship between Attitudes that Reﬂect Resilience and Applied Faith Coping
Responses was .07. This p-value is just above the acceptable limit for signiﬁcance
(p<.05). This barely non-signiﬁcant p-value could reﬂect the inﬂuence of outliers that
were not removed from the data. All other direct and indirect relationships in this model
were signiﬁcant with the exception of the At Risk —-1 Despair/doubt Coping Responses

relationship (p = 0.08). Here again, outliers that were not removed from the data might

125

have inﬂuenced the level of signiﬁcance. Also, because of the violation of multivariate
normality in this sample, the p-values may be slightly biased.

The mean scores for the coping subscales (see Table 6.) also suggest that the
participants in this study tend to use Spiritual, Adaptive, Emotional, Humor and Applied
Faith Coping Responses in times of stress and adversity, and that they tend to use Non-
adaptive and Despair/doubt Coping Responses least. Of particular interest is the ﬁnding
that the participants use Spiritual Coping more than any of the other coping responses.
This ﬁnding corroborates the ﬁndings of Oler (1997) and Logan (1980) regarding the role
of religious coping among church-going Aﬁican Americans, and it further substantiates
the idea that religiosity and spirituality are important and even salient aspects of life
among Aﬁican American college students attending a PWI, and who identify as
Christian.

In the model in which coping responses predict attitudes that reﬂect resilience
(CS —> ARR), the relationships between three coping responses - (1) Adaptive Coping
Responses, (2) Despair/doubt Coping Responses, (3) Spiritual Coping Responses, and
Attitudes the Reﬂect Resilience were signiﬁcant. The relationship between Non-adaptive
Coping Responses and Attitudes that Reﬂect Resilience was non-signiﬁcant (p = 0.27).

Three variables were identiﬁed as possible moderators of the relationship between
coping and attitudes that reﬂect resilience. These included core convictions, religiosity,
spirituality, and level of acculturation. Factor analysis of the Core Convictions Scale
revealed it to be a valid and reliable measure. Three questions were extracted from the
Multidimensionality of Religiosity Scale and were used as an indicator of the depth of

one’s religious and spiritual conviction. To further simplify the model and the analysis,

126

these items were combined with the Core Convictions Scale to create a single measure
that was renamed the Protective Religious Beliefs Scale. When items from the CCS were
combined with items from the MRS scale, and factor analyzed, the resulting scale turned
out to have good validity and reliability. Thus, this study contributes to the literature by
offering a reliable and valid scale for the measurement of Protective Religious Beliefs.
This adapted scale was developed for use with African American populations. Results
indicate that the beliefs identiﬁed in this scale are really protective. This ﬁnding
represents another major contribution to the literature.

Level of acculturation was included in this study in an attempt to explain variance
that was not accounted for by the other variables in this model. However, factor analysis
of the acculturation scale resulted in a six-factor solution that did not make sense
conceptually. The validity of the solution was questionable. Therefore, all items were
dropped from the scale with the exception of nine items from the Religious Beliefs and
Practices subscale of the AAAS-R. This subscale was factor analyzed and it yielded a
valid and reliable scale. Because the items in the scale were written to represent beliefs
and practices that are important in African American culture, it was renamed the
Culturally Salient Beliefs and Practices Scale (CSRBP). This adaptation from the
AAAS-R scale represents another contribution to the literature as the results Show that it
has important direct and indirect effects on coping responses, protective religious beliefs,

and attitudes that reﬂect resilience for this population.

Support for Cognitive Appraisal Theory
This study offers support for the Cognitive Appraisal Theory of Stress, Coping

and Resilience (Jew, Green & Kroger, 1999). Although the two ﬁnal models differ in

127

important ways, each could be explained using Cognitive Appraisal Theory of stress and
coping. In the ARR -—> CR model, Attitudes the Reﬂect Resilience predict six of the
seven Coping Responses whereas in the CR —i ARR model, only four of the Coping
Responses predicted Attitudes that Reﬂect Resilience. In each case, Non-adaptive and
Despair/doubt Coping Responses have an inverse relationship with Attitudes that Reﬂect
Resilience.

Model-ﬁtting data indicated that each model was equally probable. It is
conceivable and even plausible then, that primary and secondary appraisal processes (1)
ﬁrst follow the paths in the AR -+ CR model, and (2) once a coping response is
determined, during the reappraisal stage of coping, Emotional and Applied Faith Coping
Responses drop out, and the individual’s coping responses follow the paths in the CR —->
ARR model - Spiritual, Adaptive, Despair/doubt and Non-adaptive Coping Responses
shape or inﬂuence Attitudes that Reﬂect Resilience.

Re-conceptualization of Risk

Since there was no direct or indirect relationship between risk status and Attitudes
that Reﬂect Resilience, it was not possible to say that one group was more or less resilient
than the other. However, the data did show that the average response on the Resiliency
Attitudes Scale (containing the eight items from the General Resiliency subscale in the
original RAS) was “agree” to “strongly agree.” This suggests that at-risk and non-at-risk
students may be more similar than we think. This observation may have important
implications for how Universities deﬁne risk and the assumptions made about students

that present with academic proﬁles that deﬁne them as being at-risk.

I28

Membership as an “at risk” student is sometimes interpreted as meaning that one
has deﬁciencies in ones’ academic readiness to pursue a college education. This often
leads to the assumption that an “at risk” student lacks the tenacity to survive the
challenges associated with this pursuit. Results from this study show however, that there
was no direct or indirect relationship between risk status and attitudes that reﬂect
resilience. The majority of students (n=218) who participated in this study tended to
hold beliefs that they can survive and make things better for themselves and others.

Given the results of this study, regarding students who are considered at-risk, and
the fact that there is a stigma associated with such a label, greater attention should be
placed on reconsidering this concept and how it may affect student performance. The
work of Steele (1997) may be helpful in this regard. Steele (1997) suggests that students
who are labeled “at-risk” may be subject to a phenomenon he calls “stereotype threat”.
This type of threat places such students at risk for conforming to negative stereotypes.
Evidence for the Protective Inﬂuence of Religious Beliefs

The models that were developed in this study clearly suggest that Protective
Religious Beliefs may truly be “protective.” These beliefs are thought to be protective in
the sense that they may afford some degree of calm or assurance that even in the midst of
adversity or trouble there is hope. Instructions for the Religious Coping Activities Scale
asked participants to think about and describe in one sentence, the most serious negative
event they have experienced in the past year. Participants identiﬁed a host of serious and
negative events. These events were primarily situational. Jenkins (1995) suggests that
African American people have a characteristic way of approaching problems or negative

life experiences that reﬂects their ethos, values, beliefs and practices. This position is

129

corroborated in the work of Aldwin (1994) who suggests that African American people
cope in a culturally-prescribed manner. This way of coping may reﬂect ethos, and beliefs
that are unique to the particular culture or ethnic/social group. These beliefs and ethos
may be used as a framework for deciding how to best respond to stressful or adverse
conditions.

The models developed from the data generated in this study clearly demonstrate
that Protective Religious Beliefs were involved in the coping responses associated with
the negative events they identiﬁed, and in the attitudes the students developed in response
to these situations. In both models that relationship was medium-tO-strong.

In the model showing Coping Responses as predictors of Attitudes that Reﬂect
Resilience, the Protective Religious Beliefs tended to moderate the small inverse
relationship between Non-adaptive Coping Responses and Attitudes that Reﬂect
Resilience. The effect size for the moderation of this relationship was larger than the
effect size of the direct relationship between Non-adaptive Coping Responses and
Attitudes that Reﬂect Resilience. This suggests that even when the students used Non-
Adaptive Coping Responses, the Protective Religious Beliefs cancelled the negative

effect of the Non-adaptive Coping Responses.

Mlications of the Study
This study was not able to conﬁrm a causal relationship between resilience and
coping as hypothesized in the work of Jew, Green and Kroger (1999). However, it does
offer preliminary data that could be used to inform and develop future studies of coping

and resilience in Aﬁican American college students. Perhaps the most important

I30

implication of this study is for researchers to continue to examine the relationship
between coping and resilience as there are yet many unanswered questions regarding the
nature of this relationship. This study has other important research, theoretical, and
practical implications, which will be discussed next.

Reseaach Implications

As previously mentioned, there is a need for continued research on the “causal”
relationship between Attitudes that Reﬂect Resilience and Coping Responses. This
includes path analysis of the unidirectional and reciprocal models. To aid in such efforts,
additional work is needed for improving the reliability and validity of direct and inferred
measures of resilience as well as instruments used to measure related constructs. Most
importantly, investigators must exercise more care in making sure that resilience is
appropriately operationalized, and that the operationalization ﬁts the design of the study.
Otherwise, as Glantz & Johnson (1999) pointed out, we will continue to have difficulty
integrating research across studies. Future research efforts addressing causality should be
longitudinal, and should include at least three groups of students—those being trained for
coping responses, those being trained for Attitudes that Reﬂect Resilience, and those
receiving no training (control group).

This study was unable to answer whether attitudes that reﬂect resilience varied by
class level, gender, SES, program participation, and risk status. To simplify the analysis,
a decision was made early in the study to broadly categorize students as At Risk and Non-
At Risk. It was later realized that there were insufﬁcient and uneven numbers of
participants from various programs to do a between program analysis of attitudes that

reﬂect resilience or coping responses. In this way, program sample size served as

131

somewhat of a limitation in this study. Therefore, future research should include a larger
sample size, a re-conceptualization of risk since the conceptualization of risk used in this
study had no effect on the relationships between coping responses and attitudes that
reﬂect resilience, and an emphasis on differences by class level. This would include more
reliable measures of SES, and a larger sample of participants from each program.

In addition, this study looked at both dispositional coping responses (the COPE) and
situational coping responses (RCAS), but did not separate them in the analysis: they were
joined into a general coping response framework. There was adequate empirical evidence
that they worked well this way, but it may be important to make this distinction.

This study investigated only dispositional attitudes that reﬂect resilience. It may
also be important to distinguish between dispositional and situational attitudes that reﬂect
resilience. It may be that the relationships between coping responses and attitudes that
reﬂect resilience depend upon the combination of situational versus dispositional coping
response and attitudes that reﬂect resilience. Finally, it may not be important to make this
distinction at all for either attitudes that reﬂect resilience or for coping skills. Figure 12.
shows with the solid lines what was done in this study, and with the dashed lines what

remains to be investigated.

132

 

 

ARR ARR

 

Dispositional + 4 Situational

 

 

 

 

 

 

 

 

 

 

 

 

 

.3
y" "a
Dispositional Situational
CR CR

 

 

 

 

Figure 12. Schema of relationships investigated (solid lines) and not investigated (dashed
lines) in this study.

Finally, a brief comment needs to be made regarding the use of the Resiliency
Attitudes Scale (RAS; Biscoe & Harris, 1994) rather than the Resilience Scale (Jew,
Green & Kroger, 1999). The RAS was used rather than the scale developed by Jew,
Green and Kroger(1999) because it was less costly to do so, and because of its potential
for informing an intervention currently being utilized with a group of students who
present as “at-risk”. The results from this study suggest a shift away from fostering the
seven Resiliencies described in Wolin and Wolin (1993) towards fostering the general

resiliencies identiﬁed by Biscoe and Harris (1994).

I33

Theoretical Mplicagins

Though widely considered a ﬁxed personality trait and / or dimension,
contemporary thinking in the social sciences describes resilience as a multidimensional
and mutable process (Kumpfer, 1999; Glantz & Johnson, 1999; Rutter, 1987). After
reviewing the literature and conducting this study, I would propose that resilience
manifests itself in one’s personality and that it is a process. I therefore would deﬁne it as
the process wherein a person or persons utilize certain skills and abilities to regain and
sustain normal or near-normal and competent psychological, emotional, social, and
physical functioning during or after exposure to chronic or acute stressors or adversity.

The Attitudes that Reﬂect Resilience construct consisted of eight items from the
General Resilience subscale in the original Resilience Attitudes Scale (Biscoe & Harris,
1994). The remaining items were dropped from the scale because they did not ﬁt
conceptually with this subscale and the factor loadings for the solution generated from
factor analysis of the RAS was inconsistent with the loadings described in the manual.
The results from this factor analysis are difﬁcult to explain. One possible explanation is
that the original RAS was developed for, and normed upon a substance abuse and prison
population. It was piloted on a non-substance abusing, non-prison population for the ﬁrst
time in this study. The results in this study do not necessarily mean that Wolin and
Wolin’s (1993) theory of The Seven Resiliencies is in error, but may also mean that the
items in the scale developed by Biscoe and Harris (1994) may require re-working or
revision in order to apply to the general population. It may also be that this theory of

resiliencies is inappropriate outside the population on which the instrument was normed.

134

The General Resilience subscale was added to the RAS as an eighth resiliency by
Biscoe and Harris (1994). The results from the analysis of the General Resilience
subscale in this study suggest that it has good validity and reliability. The ASPIRE
Incorporated TRIO training team included spirituality as an addition to the list of
Resiliencies identiﬁed in Wolin and Wolin (1993). Results from this study cannot
conﬁrm that spiritualin should be considered a reﬂection of resiliency, but it does suggest
that certain religious beliefs can have a protective or enhancing effect on attitudes that
reﬂect resilience. This protective or enhancing effect provides a rationale for studying, if
not evidence of, spirituality as a reﬂection of resiliency.

Pmtical Implicam

Feagin, Vera and Imani (1996) identify a number of factors that serve as serious
challenges for African American students attending White institutions. These challenges
are believed to have a direct effect upon attrition rates among this population. Research is
therefore needed to further identify and study the factors that offer the greatest amount of
protection from the deleterious effects of stress among African American college students
attending White institutions.

Richardson and June (1997) indicate that African American college students
attending predominantly White schools tend to seek out the Black Church for spiritual
ediﬁcation, social and emotional support, guidance, and other important needs. They
indicate that identifying with the church can appropriately reinforce an individual’s moral
and ethical belief system. More than 95 percent of the participants in this study report
some involvement in religion. This study found that students who identiﬁed more with

religion or spirituality tended also to use more adaptive coping responses and tended to

135

have greater attitudes that reﬂect resilience. Together with the ﬁndings from this study
regarding the protective inﬂuences of religion, spirituality and core convictions, these
ﬁndings suggests opportunities for partnerships with local African American churches
and religious institutions in providing the needed types of social, emotional, and spiritual
support to African American college students attending predominantly White institutions.

In addition to fostering attitudes that reﬂect resilience as a strategy for preventing
drop out, ﬁndings from this study suggest that using academic support programs to teach
coping skills may be equally or even more effective. The results were not deﬁnitive in
this regard. Additional efforts should be made towards investigating any relationship
between coping responses (such as self-agency, self-efﬁcacy, persistence, and others), and
attitudes that reﬂect resilience. Use of longitudinal studies may aide in assessing the
effects of teaching coping responses on resilience as well as on the effects of teaching
attitudes that reﬂect resilience on coping responses.

This study found that students operationally deﬁned as at-risk tended to use more
of the negative coping behaviors (i.e. Non-adaptive and Despair/doubt Coping
Responses) than other students. These behaviors included such things as consumption of
alcohol and denial that a problem exists. This ﬁnding provides valuable information for
counselors and educators by suggesting that their interventions with this population
should include greater drug and alcohol use prevention education as well as assertiveness
training.

While not speciﬁcally addressed in this study, certain assumptions come with
identifying a student as “at-risk”. Typically, when we think of a person as being “at-risk”,

we tend to think of them to some extent, as being deﬁcient, or challenged. In this study,

136

the designation “at risk” was given to those who were academically disadvantaged, ﬁrst
generation college students, and/or low-income. The data show that more than ninety
percent of the students in this study agree to strongly agree with most of the attitudes that
reﬂect resilience. By inference, this suggests that there is not much difference between
the at-risk students and the non-at-risk students in terms of attitudes that reﬂect resilience.
This result raises questions about how we think about at-risk students. It may be that at-
risk status would be more adequately measured by attitudes that reﬂect resilience, by
typical coping responses, or by factors that predict drop out.

Finally, there is much work to done in trying to construct a link between practice
and research. The results from this study suggest the need for additional work towards
understanding how to integrate psychology and religion. These ﬁndings suggest that
programs that educate and train counselors and psychologists pay greater attention to and
exercise sensitivity towards the role of religion, spirituality, and core convictions in the
work done with African American college students.

Limitations of the Study

Several of the scales in this study were piloted for the ﬁrst time with an Aﬁican
American college student population. As such, ﬁndings from this study may not
generalize to other populations. There is therefore a need for researchers to use these
scales with other African American populations so that we may add to the validity,
reliability, and generalizability of ﬁndings from these scales in this population.

The Core Convictions Scale and the Levels of Acculturation Scale were
developed for use with an Aﬁ'ican American population speciﬁcally. As such, it may not

be possible to replicate this study with non-African American populations. New versions

137

or revisions of these scales would need to be developed to allow for replicating this study
with different populations.

Cross-sectional resilience research has been criticized in the literature; especially
since it has been hypothesized that resilience may actually be a multidimensional,
longitudinal process that has meaning only in the context of a stressful or adverse
situation or condition. This study attempted to redress this concern by using an inferred
measure of resilience that was designed to assess attitudes that reﬂect resilience. The
process of resilience can only be analyzed in a longitudinal context.

As shown in Figure 12, the nature of the relationships between coping responses
and attitudes that reﬂect resilience may change depending on the combination of
dispositional versus situational responses and attitudes being investigated. This study
informs only three pieces of the puzzle (the solid lines in Figure 12), but leaves
unscrutinized some important remaining pieces (the dashed lines in Figure 12).
Additionally, it is not yet clear whether there is a need to separate attitudes that reﬂect
resilience and coping responses into situational versus dispositional types.

Another implication of this study is the importance of researchers exercising care
in the construction of ethnically and culturally relevant psychometric instruments. This is
an especially important concern in survey research. The interpretability of the responses
may be diminished or even made impossible when the ethos, values and beliefs of the
persons responding to survey questions are not represented in the responses. This Often
results in forced choices and perhaps worthless data.

Finally, because of violations of the assumption of multivariate normality in the

data, the p-values for the relationships in the models may be biased. Removing all

138

offending outliers would have biased the data too much to make the results useful.
Therefore, effect sizes were used for interpreting the models with no strict understanding

of which relationships were truly statistically signiﬁcant.

I39

APPENDICES

APPENDIX A. COPE EFA Technical Appandix

140

Appendix A: COPE EFA Technical Apmndix

Table 12. contains the factor-loading matrix resulting from the six-factor,
orthogonally rotated solution. The individual factors are denoted by boxes around the
items forming them, and important cross-loadings are presented in boldface. Note that
there is relatively little important cross-loading of items on multiple factors. This
supports the decision to assume that responses to each item are inﬂuenced by only a

single factor.

Table 12. Factor loading matrix for the COPE.

 

 

 

 

 

Factor

Item 1 2 3 4 5 6
32 0.72 -0.19 0.17 -0.06 0.06 0.06
56 0.70 -0.08 0.18 -0.0l -0.06 0.12
19 0.67 ~0.18 0.18 0.07 0.09 -0. 12
42 0.65 0.10 -0.08 -0.02 0.08 0.1 1
58 0.64 -0.20 0.10 -0.16 0.14 0.13
39 0.64 -0.26 0.09 0.01 0.05 0.23
29 0.55 0.04 0.00 0.08 0.28 0.16
5 0.53 -0.26 0.16 0.01 0.05 -0.06
25 0.52 -0.07 0.04 0.09 -0.05 0.18
30 0.52 -0.13 0.37 0.10 0.04 -0.01
41 0.51 0.01 —0.15 0.01 0.21 0.10
38 0.48 -0. 12 -0.01 0.20 0.30 0.30
49 0.48 -0.02 0.08 0.12 ' 0.17 0.31
14 0.47 -0.22 0.39 0.13 0.05 -0.04
15 0.46 0.09 0.04 0.02 -0.01 0.04
59 0.45 -0.32 0.24 0.02 0.28 0.23
1 0.43 -0.32 -0.06 0.1 l 0.27 0.09
47 0.38 0.06 0.17 0.01 -0.01 0.33
33 0.37 0.1 1 0.10 0.1 1 0.01 0.09
10 0.37 0.07 -0.24 0.10 0.18 -0.04
55 0.36 0.16 0.15 0.12 -0.19 0.01
35 0.1 l 0.83 -0.06 -0.04 -0.1 l -0.16
53 0.10 0.83 -0.07 -0.06 -0.12 -0.12
26 0.08 0.75 -0.07 -0.06 -0.1 l -0.08
12 -0.04 0.71 -0.1 1 -0.1 l -0.22 -0.17
27 -0.04 0.67 0.20 0.15 0.07 -0.10
40 -0.08 0.64 0.05 0.26 -0.08 0.04
24 -0.12 0.57 -0.02 0.05 -0.1 l 0.05
9 -0. l 7 0.56 0.06 0.07 -0.09 0.16
37 -0.20 0.48 0.08 0.06 —0.04 0.29
57 -0.06 0.48 -0.02 0.21 -0.04 0.16

 

 

 

 

141

 

 

 

 

 

 

 

 

 

 

6 ~0.01 0.43 0.13 0.06 0.04 0.02
51 -0.15 0.40 0.07 0.20 0.14 0.24
43 -0.10 0.30 0.23 0.12 0.05 0.21
52 0.28 -0.08 0.73 0.02 0.15 0.00
28 0.04 0.01 0.69 -0.19 0.00 -0.10
3 -0. 15 0.23 0.67 -0. l 3 -0.04 0.02
23 0.17 -0.08 0.66 0.04 0.13 0.02
l l 0.26 -0.07 0.66 -0.05 0.06 -0.10
46 0.05 0.19 0.63 -0. l 3 -0.03 -0.03
4 0.20 -0. 12 0.62 0.02 0.09 -0.09
34 0.16 0.10 0.61 -0.02 0.20 0.08
45 0.41 -0.08 0.53 0.02 -0.06 0.1 l
17 -0.06 0.09 0.51 0.03 -0.01 0.15
3 l -0. 12 0.22 0.41 0.25 -0.08 -0.02
16 -0.01 0.07 0.27 0.21 -0.04 0.17
50 0.13 0.15 -0.04 0.84 -0.06 0.06
36 0.08 0.19 0.01 0.84 0.01 0.10
20 0.16 0.06 -0.07 0.84 -0.03 0.01
8 0.15 0.07 -0. 12 0.77 0.02 0.00
18 0.14 -0.21 0.12 -0.01 0.81 0.03
48 0.20 -0.05 0.02 -0. 14 0.80 0.03
60 0.13 -0.01 0.17 -0.04 0.75 -0.01
7 0.10 -0.33 0.05 0.09 0.74 0.00
54 0.21 -0.04 0.03 0.02 0.03 0.68
21 0.07 0.1 l -0. 14 -0.04 0.02 0.65
13 0.31 0.02 -0.01 0.07 -0.01 0.60
44 0.32 -0.36 0.04 -0.06 0.10 0.56
22 0.14 0.1 1 -0.01 0.09 -0.06 0.36
2 0.1 l -0. 12 0.20 0.04 0.08 0.25

 

 

 

There were a few items that did not have strong factor loadings for any factor, or
that had cross-loadings nearly as strong as the primary loading. These items were
investigated individually to determine whether the item was conceptually congruent with
any factors, and if so, which factor the item should be a part of. The items contents of all
items loading on a single factor were inspected for shared meaning to create labels for
each factor. The labels created and the item contents of each item contributing to each
factor are presented below, along with the decisions made concerning items that did not
ﬁt well into any factor.

COPE#1 - Adaptive Coping Responses (01 = 0.89)

32 — I try to come up with a strategy about what to do.
56 — I think hard about what steps to take.

142

l9 — I make a plan of action.
42 - I try hard to prevent other things from interfering w/ my efforts at dealing w/ this.
58 - I do what has to be doe, one step at a time.

39 — I think about howl might best handle the problem.

29 - I try to see it in a different light, to make it seem more positive.

05 - I concentrate my efforts on doing something about it.

25 - I take additional action to try to get rid of the problem.

30 — I talk to someone who could do something concrete.

41 — I make sure not to make matters worse by acting too soon.

38 — I look for something good in what is happening.

49 — I force myself to wait for the right time to do something.

14 — I talk to someone to ﬁnd out more about the situation.

15 — I keep myself from getting distracted by other thoughts or activities.

59 — I learn something from the experience.

01 - I try to grow as a person as a result of the experience.

47 — I take direct action to get around the problem.

33 — I focus on dealing with this problem, and if necessary, let other things slide.
10 — I restrain myself from doing anything too quickly.

55 - I put aside other activities in order to concentrate on this.

COPE#2 - Less Useful Coping Responses (01 = 0.83)

35 — I drink alcohol or take drugs, in order to think about it less.

53 - I use alcohol or drugs to help me get through it.

26 — I try to lose myself for a while by drinking alcohol or taking drugs.
12 - I use alcohol or drugs to make myself feel better.

27 -— I refuse to believe that it has happened.

40 — I pretend that it hasn’t happened.

24 -— I just give up trying to reach my goal.

09 — I admit to myself that I can’t make myself feel better.

37 - I give up the attempt to get what I want.

06 - I say to myself “this isn’t real.”

51 — I reduce the amount of effort I’m putting into solving the problem.
43 — I go to the movies or watch TV, to think about it less.

(Item 5 7 was deleted from this factor)

5 7 — I act as though it hasn ’I even happened (Denial/E17 C)

COPE#3 - Help-Seeking Coping Responses (01 = 0.85)

52 - I talk to someone about how I feel.

28 — I let my feelings out.

03 — I get upset and let my emotions out.

23 — I try to get emotional support from friends or relatives.

11 — I discuss my feelings with someone.

46 — I feel a lot of emotional distress and I ﬁnd myself expressing these feelings a lot.
04 — I try to get advice from someone about what to do.

34 —1 get sympathy and understanding from someone.

143

45 — I ask people who have had similar experiences what they did.

17 - I get upset, and am really aware of it.

(Items 16 and 31 were deleted from this factor)

I 6 — I daydream about things other than this. (Mental Disengagement /Less Useﬁil Cope)
31 - I sleep more than usual. (Mental Disengagement /Less Useﬁil Cope)

COPE#4 -Humor Coping Responses (01 = 0.89)
50 - I make fun of the situation.

36 - I kid around about it.

20 - I make jokes about it.

08 - I laugh about the situation.

COPE#5 -Religion Coping Responses (01 = 0.84)
18 - I seek God’s help.

48 - I try to ﬁnd comfort in my religion.

60 — I pray more than usual.

07 - I put my trust in God

COPE#6 - Acceptance / Restraint Coping Responses (01 = 0.71)

54 — I learn to live with it.

21 — I accept that this happened and that it cannot be changed.

13 — I get used to the idea that it happened.

44 — I accept the reality of the fact that it happened.

(Items 2 and 22 were deleted from this factor)

2 — 1 turn to work or other substitute activities to take my mind 017 things (Mental Dis.)
22 — 1 hold of doing anything about it until the situation permits. (Restraint Cope)

APPENDIX B. RCAS EFA Technical Apmndix

I45

Appendix B: RCAS EF A Technical Appendix
Table 13. contains the factor-loading matrix resulting from the three-factor,
orthogonally rotated solution. The individual factors are denoted by boxes around the

items forming them, and important cross-loadings are presented in boldface.

Table 13. Factor loading matrix for the RCAS.

 

 

 

 

 

 

 

 

Cormonent ,

Item 1 2 3

27 0.84 0.26 0.02
28 0.78 0.29 -0.03
16 0.78 0.27 -0.13
19 0.73 0.16 -0.12
22 0.73 0.26 -0.06
26 0.72 0.10 0.21
10 0.70 0.34 -0.02
12 0.69 0.29 0.07
21 0.68 0.42 -0.08
4 0.66 0.18 -0.07
15 0.65 0.23 —0.16
14 0.58 0.21 0.32
29 0.49 0.45 0.22
17 0.45 0.38 -0.01
20 0.18 0.82 0.03
8 0.17 0.80 0.05
11 0.24 0.77 0.01
5 0.33 0.70 -0.02
1 0.19 0.63 0.08
6 0.22 0.62 0.00
18 0.51 0.59 -0.04
13 0.41 0.43 0.34
3 0.18 0.26 0.08
25 . 0.23 -0.18 0.71
24 -0.24 0.05 0.62
9 0.17 0.08 0.59
23 -0.16 0.19 0.54
2 -0.28 -0.08 0.51
7 0.33 0.36 0.39

 

 

 

Item 7 (Asked fog miracle) was removed from the scale because it had approximately
equal loadings on all three factors, and did not ﬁt neatly into any of the three factors.

I46

Items that Loaded on Factors 1-3 in the RCAS Scale:

RCAS#1 - Spiritual Coping Responses ((1 = .93)

04 - Trusted that God would not let anything terrible happen to me.
10 — Experienced God’s love and care.

12 — Realized God was trying to strengthen me.

14 - Confessed my sins.

15 — I let God solve my problems for me.

16 — In dealing with the problem, I was guided by God.

17 — Realized that I did not have to suffer since Jesus suffered for me.
19 - Took control over what I could, and gave the rest up to God.

21 — My faith showed me different ways to handle the problems.

22 — Accepted that the situation was not in my hands, but in the hands of God.
26 - Found the lesson from God in the event.

27 — God showed me how to deal with the situation.

28 — Used my faith to help me decide how to cope with the situation.
29 — Prayed or read the Bible to keep my mind off of my problems.

RCAS#2 - Applied Faith Coping Responses ((1 = .88)

01 — Received support from clergy.

O3 — Focused on the world-to-come rather than the problems of this world.
05 — Attended religious services or participated in religious rituals.

06 - Led a more loving life.

08 — Participated in church groups.

11 — Received support from other members of the church.

13 - Tried to be less sinful.

18 — Used Christ as an example of how I should live.

20 - Provided help to other church members.

RCAS#3 — Despair/Doubt Coping Responses ((1 = .60)

02 — Felt angry wit or distant from God.

09 — Asked God why it happened.

23 — Felt angry with, or distant from the members of the church.
24 — Questioned my religious beliefs and faith.

25 - Bargained with God to make things better.

147

APPENDIX C. Combined COPE and RCAS EFA Technical Apmndix

l48

Appendix C: Combined COPE apd RCAS EFA Technical Appendix
Table 14 contains the factor-loading matrix resulting from the seven-factor,
orthogonally rotated solution for the combined COPE/RCAS. Boxes around the items
forming them denote the individual factors, and important cross-loadings are presented in
boldface. Note that there is relatively little important cross-loading of items on multiple
factors. This supports the decision to assume that responses to each item are inﬂuenced
by only a single factor.

Table 14. Complete Combined COPE and RCAS EFA factor loading matrix.
Component Component

Item 1 2 3 4 5 6 7 Item 1 2 3 4 5

6 7

 

 

COPE07
COPE18
COPE48
COPE6O
RCASO3
RCASO4
RCASOS
RCAS 10
RCAS 12
RCAS l 3
RCAS l 4
RCAS l 5
RCASl 6
RCAS 1 7
RCAS l 8
RCAS l 9
RCAS2 l
RCAS22
RCASZ6
RCA827
RCASZS
RCASZ9
COPEOl
COPEOS
COPE 1 0
COPE l 3
COPE l 4
COPE l 5
COPE 1 9
COPE25
COPE29
COPE30
COPE32

 

0.63
0.65
0.65
0.60
0.26
0.63
0.58
0.75
0.75
0.48
0.57
0.71
0.77
0.54
0.66
0.70
0.77
0.79
0.65
0.83
0.80
0.59

0.08
0.16
0.16
0.09
0.19
0.05
0.10
0.14
0.05
0.09
0.10
0.10
0.14
0.02
0.05
0.09
0.25
-0.03
0.1 1
0.17
0.09
-0.07

 

0. l 8
0.16
0.13
0.1 l
-0.03
0.01
0.08
0.00
0.24
0.13
0.07

 

 

0.47
0.55
0.32
0.46
0.45
0.43
0.64
0.60
0.54
0.50
0.75

 

-0.44
-0.34
-0.26
-0.17
0.21
0.05
0.02
0.05
0.07
0.01
0.03
~0.01
-0.03
0.10
0.04
-0.03
-0.01
0.01

0.12
~0.01
-0.01
-0.01
-0.01
0.06

-0.06
0.12
0.07
-0.12
-0.02
-0.01
0.05
0.04
0.00 -0.03
-0.03 0.06
-0.09 0.04
-0.03 -0.01
-0.34 0.00
-0.08 0.23
0.07 -0.l7
0.12 0.06
~0.20 0.39
0.09 0.06
-0.13 0.12
0.02 0.08
-0.12 0.01
-0.12 0.35
-0.14 0.08

0.01
0.21
0.06
-0.06
0.02
-0.06

-0. 16 -0.07

0.00
0.08
-0.04
0.07
-0.02
0.01

-0.01
-0.03
-0.03
-0.04
-0.06
~0.03
0.09
-0.05
0.13

0.10
0.15

0.14
0.06
-0.01
0.18

0.12

-0. 12

0.08 0.09 -0.05
0.18 0.02 -0.0l
0.04 -0.09 0.11

0.1 l

0.13

0.01

0.47
0.08
0.05
0.36
0.08
~0.04
0.04
0.13
0.37
-0.04
0.17
-0.07
-0.13
0.01

0.07
0.33
0.13

0.04
0.05

-0.43
0.00
0.03

0.18

-0.1 1
0.06
0.07
0.14

0.06
0.07
0.09
0.04
0.06
-0.10
0.03
-0.01
0.06
0.30
0.23
-0. 16
-0.20
0.05
-0.04
-0. 12
-0.08
-0.07
0.16
0.00
-0.07
0.17
-0.08
'0.13
-0.20
0.05
-0.13
0.07
-0.06
0.07
0.02
-0.06
0.06

COPEO6
COPE09
COPE 1 2
COPE24
COPE26
COPE27
COPE35
COPE40
COPE53
COPE03
COPE04
COPE] 1
COPE l 7
COPE23
COPE28
COPE34
COPE45
COPE46
COPESZ
COPE08
COPE20
COPE36
COPESO
COPE57
COPEZ 1
RCASOI
RCASO6
RCASOB
RCAS] l
RCAS2O
COPE37
COPE43
COPES 1

0.05 -0.10
-0.07 ~0.11
-0.04 ~0.16
-0.09 -0.11
0.00 0.01
0.06 -0.09
0.03 -0.02
-0.04 -0.08
0.04 0.00
-0.06 -0.13
0.02 0.13
0.05 0.23
-0.02 0.01
0.07 0.15
-0.02 0.04
0.22 0.08
-0.04 0.46
'0.07 0.04
0.13 0.24
0.02 0.17
0.00 0.15
-0.04 0.13
-0.02 0.15
-0.17 0.01
0.05 0.25
0.35 0.05
0.35 0.23
0.44 0.09
0.48 0.09
0.47 0.11
-0.03 ~0.ll
0.05 0.05
0.04 -0.07

 

0.33
0.46
0.83
0.55
0.75
0.49
0.84
0.52
0.87

0.12
0.03
-0.05
0.07
-0. 12
0.22
-0.06
0.12
-0.08

 

0.18
-0. l 8
-0.09
0.09
-0.09
0.03
0.13
-0.1 l
0.08
-0.07
0.00
0.08
0.15
0.17
0.29
0.02
-0.05
0.00
0.07
-0.02
0.16
0.34
0.17
0.26

 

0.66
0.62
0.70
0.45
0.67
0.69
0.64
0.47
0.62
0.78

0.18
0.15
-0.05
0.13

0.02
0.21

0.02
0.33
-0.05
-O.15
0.06
-0.01
0.00
0.1 l

-0.23
0.07
0.06
.0. l 3
0.06

 

-0.08
0.02
0.00
-0.07
0.00
~0.10

 

0.76
0.83
0.82
0.83
0.29

0.17 0.22
-0.06 0.33
-0.01 -0.04
-0.08 0.01
-0.01 0.11
0.10 0.40
0.08 0.13
-0.03 0.18
0.06 0.07
0.02 0.17
0.13 0.06
0.07 -0.15
-0.07 0.22
-0.01 -0.07
0.10 0.02
-0.07 0.01
0.05 0.09
-0.03 0.23
0.03 -0.12
-0.01 0.08
0.00 0.01
-0.05 0.12
-0.04 0.00
0.17 0.11

 

0.11

0.11 -0.01

-0.01
0.02
0.10
0.04
0.06
0.03
-0.04

0.09
0.00
0.09
0.12
0.04
0.07
0.21

 

-0.44 0.07
0.49 0.07
0.48 -0.04
0.62 0.02
0.58 0.05
0.59 -0.01

 

-O.12 0.34
-0.09 0.32
-0.02 0.40

 

 

149

 

 

COPE33 -0.03 0.41 -0.01 -0.08 0.13 0.00 0.11 RCASOZ -0.40 -0.04 -0.02 0.00 0.07 0.04 0.41
COPE38 0.29 0.53 -0.19 0.00 0.27 0.01 0.03 RCASO9 0.11 0.07 0.05 0.13 -0.10 0.10 0.52
COPE39 0.12 0.70 -0.15 0.12 0.00 0.01 -0.20 RCASZ3 -0.02 0.02 0.12 0.03 0.12 0.11 0.57
COPE4l 0.23 0.45 0.08 -0.11 0.11 0.00 -0.14 RCASZ4 -0.18 0.03 0.09 -0.05 0.06 0.07 0.70
COPE42 0.06 0.67 -0.02 -0.07 0.03 0.11 0.11 RCASZS 0.14 -0.11 0.00 0.08 -0.09 -0.20 0.59

 

 

 

COPE44 0.21 0.51 -0.32 0.02 -0.10 -O.22 -0.16
COPE47 0.05 0.45 0.04 0.14 -0.02 -0.12 0.20
COPE49 0.20 0.51 0.02 0.04 0.12 -0.23 0.06
COPE54 0.11 0.42 -0.06 0.02 0.04 -0.33 0.10
COPESS —0.09 0.36 0.19 0.11 0.15 0.05 -0.01
COPE56 0.06 0.70 0.02 0.18 -0.09 0.00 -0.04
COPE58 0.15 0.67 -0.14 0.07 -0.17 0.00 -0.10
COPE59 0.28 0.51 -0.38 0.21 0.02 -0.02 0.00

 

 

 

Items that Loaded on Factors 1-7 in the Integrated COPE and RCAS Scale:

Factor #1 - Religious Coping:

C07 - I put my trust in God

C18 — I seek God’s help.

C48 — I try to ﬁnd comfort in my religion.

C60 - I pray more than usual.

R03 — Focused on the world-to-come rather than the problems of this world.
R04 — Trusted that God would not let anything terrible happen to me.
R05 — Attended religious services or participated in religious rituals.
R10 — Experienced God’s love and care.

R12 - Realized God was trying to strengthen me.

R13 — Tried to be less sinful.

R14 - Confessed my sins.

R15 — I let God solve my problems for me.

R16 - In dealing with the problem, I was guided by God.

R17 - Realized that I did not have to suffer since Jesus suffered for me.
R18 — Used Christ as an example of how I should live.

R19 - Took control over what I could, and gave the rest up to God.
R21 — My faith showed me different ways to handle the problems.

R22 - Accepted that the situation was not in my hands, but in the hands of God.
R26 - Found the lesson from God in the event.

R27 — God showed me how to deal with the situation.

R28 — Used my faith to help me decide how to cope with the situation.
R29 — Prayed or read the Bible to keep my mind off of my problems.

Factor #2 — Adaptive Coping:

C01 — I try to grow as a person as a result of the experience.

C05 — I concentrate my efforts on doing something about it.

C10 - I restrain myself from doing anything too quickly.

C 13 — I get used to the idea that it happened.

C14 — I talk to someone to ﬁnd out more about the situation.

C15 —- I keep myself from getting distracted by other thoughts or activities.

150

C19 -1 make a plan of action.

C25 -— I take additional action to try to get rid of the problem.

C29 — I try to see it in a different light, to make it seem more positive.

C30 — I talk to someone who could do something concrete.

C32 —— I try to come up with a strategy about what to do.

C33 — I focus on dealing with tis problem, and if necessary, let other things slide.
C38 — I look for something good in what is happening.

C39 — I think about how I might best handle the problem.

C41 — I make sure not to make matters worse by acting too soon.

C42 — I try hard to prevent other things from interfering w/ my efforts at dealing w/ this.
C44 -— I accept the reality of the fact that it happened.

C47 — I take direct action to get around the problem.

C49 — I force myself to wait for the right time to do something.

C54 - I learn to live with it.

C55 - I put aside other activities in order to concentrate on this.

C56 -- I think hard about what steps to take.

C58 - I do what has to be doe, one step at a time.

C59 — I learn something from the experience.

Factor #3 - Non-adaptive Coping:

C06 — I say to myself “this isn’t real.”

C09 — I admit to myself that I can’t make myself feel better.

C12 — I use alcohol or drugs to make myself feel better.

C24 — I just give up trying to reach my goal.

C26 — I try to lose myself for a while by drinking alcohol or taking drugs.
C27 — I refuse to believe that it has happened.

C35 — I drink alcohol or take drugs, in order to think about it less.

C40 — I pretend that it hasn’t happened.

C53 — I use alcohol or drugs to help me get through it.

Factor #4 - Emotional Coping:

C03 — I get upset and let my emotions out.

CO4 - I try to get advice from someone about what to do.

C11 — I discuss my feelings with someone.

C17 - I get upset, and am really aware of it.

C23 - I try to get emotional support from friends or relatives.

C28 - I let my feelings out.

C34 - I get sympathy and understanding from someone.

C45 — I ask people who have had similar experiences what they did.
C46 — I feel a lot of emotional distress and I find myself expressing these feelings a lot.
C52 — I talk to someone about how I feel.

Factor #5 -— Humor: (Item C57 was omitted, since it did not ﬁt conceptually)
C08 — I laugh about the situation.
C20 — I make jokes about it.

151

 

C36 — I kid around about it.
C50 — I make fun of the situation.

Factor #6 — Agentive Coping (Focus on Agency):

C21 — I accept that this happened and that it cannot be changed.
R01 - Received support from clergy.

R06 - Led a more loving life.

R08 — Participated in church groups.

R11 — Received support from other members of the church.
R20 — Provided help to other church members.

Factor #7 — Non-Agentive Coping (Shift Responsibility):
C37 - I give up the attempt to get what I want.

. C43 — I go to the movies or watch TV, to think about it less.
C51 — I reduce the amount of effort I’m putting into solving the problem.
R02 - Felt angry wit or distant from God.

R09 - Asked God why it happened.

R23 — Felt angry with, or distant from the members of the church.
R24 — Questioned my religious beliefs and faith.

R25 — Bargained with God to make things better.

152

APPENDIX D. RAS EFA Technical ADDCILdg

153

Apmndix D: RAS EFA Technical Amndix

Table 15 contains the factor-loading matrix resulting from the four-factor,
orthogonally rotated solution. The individual factors are denoted by boxes around the
items forming them, and important cross-loadings are presented in boldface. Note that
there is relatively little important cross-loading of items on multiple factors. However,

the factor structure is not consistent with the eight resiliencies identiﬁed by Biscoe and

 

 

 

 

 

 

 

 

 

Harris (1994).
Table 15. Factor loading matrix for the RAS EFA.

Factor Factor
Item 1 2 3 4 Item 1 2 3 4
63 0.75 0.10 0.04 0.00 l l 0.13 0.59 0.04 -0.05
33 0.68 0.08 0.11 0.01 35 -0.1 l 0.53 0.10 0.16
67 0.67 0.00 0.10 -0.17 19 0.05 0.50 -0.04 -0.21
64 0.64 0.10 -0.06 0.00 20 0.04 0.49 -0.07 0.14
55 0.64 0.08 -0.08 0.17 18 -0.04 0.46 0.15 0.23
54 0.58 -0.02 0.02 0.02 56 0.00 0.46 0.07 -0.02
30 0.57 0.05 -0.06 0.06 59 0.41 0.44 -0.1 1 0.07
32 0.56 -0.02 0.19 -0.09 26 0.24 0.41 -0.08 0.02
62 0.54 0.09 0.23 0.14 31 0.1 1 0.38 0.24 0.10
58 0.53 -0.02 0.21 -0.15 15 0.07 0.30 0.18 -0.07
23 0.51 0.07 0.19 -0.1 l 21 -0.06 0.29 0.04 0.1 l
65 0.50 0.37 0.18 -0. 13 10 0.13 -0.29 0.23 0.12
53 0.50 0.24 0.02 0.10 14 0.24 0.25 -0.03 0.02
43 0.47 0.15 0.46 0.03 12 0.04 0.21 0.00 0.07
40 0.46 0.02 0.18 0.30 6 0.16 0.20 -0.04 -0. l 7
72 0.45 0.14 0.14 -0.18 16 0.1 l 0.14 0.59 0.00
66 0.45 0.34 0.30 —0.01 70 0.27 0.26 0.57 0.04
4 0.43 -0.07 0.04 -0.10 22 0.13 -0.01 0.50 -0.08
42 0.42 0.12 0.00 0.34 60 0.1 1 -0.22 0.42 0.01
57 0.42 0.04 0.16 -O. l 7 28 -0.05 0.28 0.41 0.28
68 0.41 0.25 -0.06 0.32 3 0.27 0.17 -0.39 0.07
50 0.41 0.18 -0. l 7 0.37 71 0.35 0.22 0.38 0.02
36 0.38 0.32 0.33 -O. 10 8 -0.02 0.06 0.36 0.28
41 0.38 .022 0.12 0.23 25 -0. 14 0.02 0.35 0.25
37 0.38 0.10 0.14 0.17 13 0.16 0.19 -0.31 0.17
39 0.37 0.32 -0. 10 0.20 49 0.17 -0.01 0.30 -0.02
29 0.37 0.08 -0.05 0.04 27 0.03 -0.04 0.26 0.18
24 0.35 0.01 0.31 0.1 l 9 0.13 0.13 0.13 0.00
61 -0.34 0.1 l -0.08 0.03 48 0.01 -0.02 0.02 0.55
46 0.34 0.07 0.08 0.16 2 —0.01 0.13 -0.08 0.46
7 0.33 0.22 0.27 -0.07 44 0.34 -0.13 -0.04 0.45
34 0.33 -0.1 1 0.33 O. 14 52 0.05 0.34 -0. 12 0.43
5 0.25 0.18 0.18 0.16 47 0.15 0.10 0.16 0.41
45 0.24 -0. 16 0.03 0.08 38 0.07 0.30 0.16 0.39

 

 

 

 

154

 

 

17 0.31 0.06 -0.05 -O.38
69 0.10 -0.14 0.25 0.38
51 0.15 0.18 0.28 -0.37
l -0.01 0.02 0.16 0.20

 

 

 

Because the factors could not be interpreted, the reliabilities of the factors were not
investigated. Items in this solution were examined for conceptual consistency, and no
factor was found to have reasonable conceptual integrity. Hence, a decision was made to
drop all items from the scale with the exception of items 62-72 from the General
Resiliency subscale. Also, no label was given for the four factors since there was no
conceptual integrity per factor.

Items that loaded on factors 1-4 of the RAS.

RAS#1 - No label given

04 I try to notice signals from other people that spell trouble.

05 It doesn't do any good to try and ﬁgure out why things happen.

07 I am willing to ask myself tough questions and answer them honestly.
23 I try to ﬁgure out why a relationship was not healthy and avoid repeating it.
24 I am good at starting relationships with other people.

29 I am good at keeping relationships going.

30 I am able to love others and be loved by them.

32 I oﬁen talk myself through a problem.

33 I can learn from the past and use that information to make the future better.
34 I have hobbies or other activities that I take seriously.

36 I am successful in taking care of my physical and emotional needs.

37 I don't like to try to ﬁnd out how things work.

39 I do enough to get by, but not much more.

40 I enjoy getting involved in constructive activities.

41 Sometimes I forget my problems when I'm pursuing creative activities.
42 I don't think that I'm creative.

43 I'm good at ﬁnding new ways to look at things.

45 The positive feelings I get from creating help make up for the pain of my past.
46 Using my imagination doesn't help to solve problems.

50 Most problems have only one solution.

53 I can't help repeating the mistakes that my parents made.

54 I like to help other peOple.

55 There's no way I could make a difference in other people's lives.

57 I stand up to people when I see them being dishonest, petty or cruel.
58 I am willing to take risks for the sake of doing what I think is right.

61 I like to help others even if they are not willing to help themselves.

62 I am involved in things that will make people's lives better.

63 No matter what happens, if I keep trying I'll get through it.

64 There are things that I can do to make my life better.

65 Sometimes it's hard, but I don't let things keep me down.

66 Even if bad things happen, I can deal with them.

155

67 It's not that hand you are dealt, but how you play it.
68 No matter how hard I try, I can't make things right.
72 Failure is something you learn from rather than feel guilty about.

RAS#2 — No label given

06 Often I ﬁnd myself taking responsibility for other people's problems.

10 I try to ﬁgure out why people act the way they do.

11 I will often stay with someone, even though I know that person is bad for me.

12 I am able to step back from troubled family members and see myself as OK.

14 I can't help acting like a child around my parents.

15 I am able to recognize when I'm in a bad relationship and end it.

18 It's hard for me to stay calm when someone I care about is being unreasonable.

19 If I love someone, I can put up with that person hurting me.

20 I often ﬁnd myself around people who aren't well adjusted.

21 There are few people who I can really count on.

26 It's hard for me to believe that I'll ever ﬁnd a good relationship.

31 It's beyond me how most things work.

35 I often get really frustrated when dealing with problems and can't ﬁgure out what
to do.

56 I don't always do what I know is right.

59 Sometimes I feel like I'm just drifting along with no purpose in life.

RAS#3 - No label given

03 When others think badly of me, there's probably a good reason for it.

08 I have a hard time telling what someone new is like until I get to know the person
well.

09 I can ﬁx hurts from my past that could keep me from letting people get close to me.

13 If you care about someone, you should try to do what the person wants, even if it
seems unreasonable.

16 I can stay calm around troubled people because I understand why they act the way
they do.

22 I am good at sizing up people.

25 I can't do anything about whether people like me or not.

27 I'm shy around people I don't know.

28 I can't really tell if a relationship is going to be good until I try it.

49 I am good at using humor to reduce tension between others and myself.
60 I almost always stand up for underdogs.

70 I'm good at making the most of a bad situation.

71 When life gives me lemons, I make lemonade.

RAS#4 — No label given

01 I usually can't predict what other people will do.

02 I avoid accepting responsibility for other people's problems.

17 I realize that I can't change other people; they have to change for themselves.
38 There are few things that I am good at doing.

156

 

44 One way to express my feelings is through my artwork, dance, music or writing.
47 It's hard for me to see the humor in a bad situation.

48 One has to take life very seriously to get by.

51 I ﬁnd it easy to choose between right and wrong.

52 It's a dog eat dog world where one has to do what it takes to get by.

69 I am willing to go with any approach that will work.

157

APPENDIX E. Consent Form / In_struction Sheet

158

Appendix E - Consent Form / Instruction Sheet

Thank you for considering participating in this study. Your participation is voluntary and you
may choose not to participate, or to withdraw participation at any time during the study without
penalty. Mr. Sigrid Dixon, a PhD. candidate in the Department of Counseling, Educational
Psychology and Special Education at Michigan State University, under the supervision of
Professor Lee N. June, is conducting this study.

The purpose of this study is to further investigate the relationship between various personal and
demographic factors such as coping, religiosity, and resilience in college students. As a
participant, you will be asked to complete seven questionnaires. It will take approximately 40
minutes to complete them.

All information collected will be kept strictly conﬁdential, and your conﬁdentiality will be
maintained to the maximum extent allowable by law. You will be fully debriefed regarding this
study after completing the survey packet.

Any participant desiring results from this study may contact Sigrid Dixon by e-mail at
dixonsig@msu.edu with your request. In your e-mail request, please include a return address for
the purpose of mailing results to you.

With your consent, your name will be assigned a number and entered into a rafﬂe. The prize is
$100, and the rafﬂe will be drawn after all data is collected. Your name, and the number that
will be assigned to you will be used only for the purpose of distributing the prize. Otherwise, all
identifying information is held strictly conﬁdential.

You must complete each survey in the survey packet before your name can be assigned a number

and entered into the raffle. Withdrawal from participating in the study will disqualify you for the
rafﬂe.

If you have any questions or concerns, you may contact Sigrid J. Dixon, M.A. at 14C Student
Services Building, telephone: (517) 355-2264. If you have questions about your rights as a
participant, you may contact Dr. David Wright, at (517) 355-2180.

You indicate your voluntary consent to participate in this study by reading this consent form and
signing your name below.
Name (Please print):
Signature:
Date:

 

 

 

 

If you are interested in participating in the raffle, please include the following contact
information: an e-mail address and telephone number where you may be reached in the event you
win the raffle.

_Yes, I want to participate in the $100 Rafﬂe
_ No, I do not want to enter the $100 Raffle

E-Mail address:
Telephone Number:
This participant is assigned number

 

 

 

159

 

 

Demographic Information

9‘1".“pr

 

What is your age? _

What is your gender? (circle one) Male / Female

Class level (circle one): Freshman Sophomore Junior Senior
Grade Point Average:

ACT Score:

 

What is your parent’s annual income (Please check one):
a.) Under $10,000 _
b.) $10,000 - $20,000
0.) $20,000 - $30,000
(1.) $30,000 - $40,000
e.) $40,000 - $50,000
f.) $50,000 - $60,000
g.) $60,000 - $70,000 __
h.) $70,000 - $80,000 _
i.) $80,000 - $100,000 __
j.) $100,000 and above _
k.) Don’t know _

7. Were you reared in:
l.) A single parent home __
m.) A traditional two parent home __
n.) Reared by relative other than parent(s) _
0.) Foster Parents __
p.) Other (please describe brieﬂy)

 

q.) Why did you come to Michigan State University?

 

 

r.) Do you identify as
5.) Christian (Protestant) __
t.) Christian (Catholic)—
u.) Muslim__
v.) Jewish—
w.) Other (Please indicate your religious
preference)

x.) What type of High School did you attend:
y.) Predominantly White, suburban
z.) Predominantly Black, inner-city
a.) Mixed or Integrated, suburban
bb.) Mixed or Integrated, inner-city
cc.) Public

160

dd.) Private
ee.) Other (please describe)

 

l6l

COPE

We are interested in how people respond when they confront difﬁcult or stressful events
in their lives. There are lots of ways to try to deal with stress. This questionnaire asks
you to indicate what you generally do and feel, when you experience stressful events.
Obviously, different events bring out somewhat different responses, but think about what
you usually do when you are under a lot of stress.

Please respond to each of the following items by circling one number in the choice
options given at the end of each statement. Please try to respond to each item separately
in your mind ﬁom each other item. Choose your answers thoughtfully, and make your
answers as true FOR YOU as you can. Please answer every item. There are no “ri t” or
“wrong” answers, so choose the most accurate answer for YOU - not what you think
“most people” would say or do. Indicate what YOU usually do when YOU experience a
stressful event.

1 = I usually don’t do this at all
2 = I usually do this a little bit

3 = I usually do this a medium amount
4 =1 usually do this a lot

I try to grow as a person as a result of the experience. 1 2 3 4

I turn to work or other substitute activities to take my mind oﬁ‘ things. 1 2 3 4
I get upset and let my emotions out. 1 2 3 4

I try to get advice from someone about what to do. 1 2
I concentrate my efforts on doing something about it. 1
I say to myself “this isn’t real.” 1 2 3 4
IputmytrustinGod.12 3 4

I laugh about the situation. 1 2 3 4

. I admit to myself that I can’t deal with it, and quit trying. 1 2 3 4

10. I restrain myself from doing anything too quickly. 1 2 3 4

1 1. I discuss my feelings with someone. I 2 3 4

12. I use alcohol or drugs to make myself feel better. 1 2 3 4

13. I get used to the idea that it happened. 1 2 3 4

14. I talk to someone to ﬁnd out more about the situation. 1 2 3 4

15. I keep myself from getting distracted by other thoughts or activities. 1 2 3 4
16. I daydream about things other than this. 1 2 3 4

17. I get upset, and am really aware of it. I 2 3 4

18. I seek God’s help. 1 2 3 4

19. I make a plan of action. 1 2 3 4

20. I make jokes about it. 1 2 3 4

21. I accept that this has happened and that it can’t be changed. 1 2 3 4

22. I hold off doing anything about it until the situation permits. 1 2 3 4

23. I try to get emotional support from ﬁiends or relatives. 1 2 3 4

34
234

09°89‘P'P‘P‘Nt"

162

24

25.
26.
27.
28.
29.
30.
31.
32.
33.

34.
35.
36.
37.
38.
39.
40.
41.
42.

43.
44.
45.
46.

47.

48

56
57
58

59.
60.

I just give up trying to reach my goal. 1 2 3 4

I take additional action to try to get rid of the problem. 1 2 3 4

I try to lose myself for a while by drinking alcohol or taking drugs. 1 2 3 4

I refuse to believe that it has happened. 1 2 3 4

I let my feelings out. 1 2 3 4

I try to see it in a different light, to make it seem more positive. 1 2 3 4

I talk to someone who could do something concrete about the problem. 1 2 3 4

I sleep more than usual. 1 2 3 4

I try to come up with a strategy about what to do. 1 2 3 4

I focus on dealing with this problem, and if necessary let other things slide a little.
1 2 3 4

I get sympathy and understanding from someone. I 2 3 4

I drink alcohol or take drugs, in order to think about it less. 1 2 3 4

I kid around about it. 1 2 3 4

I give up the attempt to get what I want. 1 2 3 4

I look for something good in what is happening. 1 2 3 4
I think about how I might best handle the problem. 1 2 3
I pretend that it hasn’t really happened. 1 2 3 4

I make sure not to make matters worse by acting too soon. 1 2 3 4

I try hard to prevent other things from interfering with my efforts at dealing with
this. 1 2 3 4

I go to movies or watch TV, to think about it less. 1 2 3 4

I accept the reality of the fact that it happened. 1 2 3 4

I ask people who have had similar experiences what they did. 1 2 3 4

I feel a lot of emotional distress and I ﬁnd myself expressing those feelings a lot. 1
2 3 4

I take direct action to get around the problem. 1 2 3 4

4

. I try to ﬁnd comfort in my religion. 1 2 3 4
49.
50.
51.
52.
53.
54.
55.

I force myself to wait for the right time to do something. 1 2 3 4
I make fun of the situation. 1 2 3 4

I reduce the amount of effort I’m putting into solving the problem. 1 2 3 4
I talk to someone about how I feel. 1 2 3 4

I use alcohol or drugs to help me get through it. 1 2 3 4

I learn to live with it. 1 2 3 4

I put aside other activities in order to concentrate on this. 1 2 3 4
I think hard about what steps to take. 1 2 3 4

I act as though it hasn’t even happened. 1 2 3 4

I do what has to be done, one step at a time. 1 2 3 4

I learn something ﬁom the experience. I 2 3 4

Ipray more than usual. 1 2 3 4

163

Most Serious Negative Event:
Please think about the most serious negative event you have experienced in the past year.

In one sentence, please describe this event:

 

 

When did this event occur?

 

164

Religious Coping Activities Scale

To what extent did you use the following behaviors in coping with the event you
described? Please circle one response per item.

 

Not at all Somewhat Quite a bit A great deal

 

a b c d
1. Received support from clergy. a b c d
2. Felt angry with or distant from God. a b c d
3. Focused on the world -to-come rather than the problems of this world. a b c d
4. Trusted that God would not let anything terrible happen to me. a b c d
5. Attended religious services or participated in religious rituals
(e.g. Holy Communion). a b c d
6. Led a more loving life. a b c d
7. Asked for a miracle. a b c d
8. Participated in church groups (e. g., support groups,
prayer groups, Bible-study groups). a b c d
9. Asked God why it happened. a b c d
10. Experienced God’s love and care. a b c d
11. Received support from other members of the church. a b c d
12. Realized God was trying to strengthen me. a b c d
13. Tried to be less sinﬁrl. a b c d
14. Confessed my sins. a b c d
15. I let God solve my problems for me. a b c d
16. In dealing with the problem I was guided by God. a b c d
17. Realized that I did not have to suffer since Jesus suffered for me. a b c d

18. Used Christ as an example of how I should live. a b c d

19. Took control over what I could, and gave the rest up to God. a b c d

20. Provided help to other church members. a b c d

21. My faith showed me different ways to handle the problems. a b c d

22. Accepted that the situation was not in my hands, but in the hands of God. a b c d
23. Felt angry with or distant from the members of the church. a b c d

24. Questioned my religious beliefs and faith. a b c d

25. Bargained with God to make things better. a b c d

26. Found the lesson from God in the event. a b c d

27 . God showed me how to deal with the situation. a b c d

28. Used my faith to help me decide how to cope with the situation. a b
c d

29. Prayed or read the Bible to keep my mind off of my problems. a b c d

165

Resilient Attitude Scale (Biscoe & Harris, 1994)

We are interested in how you view yourself. Please be as honest as possible when rating
each of the statements below. There are no right or wrong answers. In the space to the
right of each statement below, please circle the response that best describes how you feel
about that statement. Please read each item carefully and rate how strongly you agree or
disagree with it using the scale below:

Disagree Disagree Neutral Agree Agree

Strongly Strongly
1 2 3 4 5
l. I usually can’t predict what other people will do. 1 2 3 4 5
2. I avoid accepting responsibility for other people’s problems. 1 2 3 4 5
3. When others think badly of me, there’s probably a good reason for it. 1 2 3 4 5
4. I try to notice signals from other people that spell trouble. 1 2 3 4 5
5. It doesn’t do any good to try and ﬁgure out why things happen. 1 2 3 4 5
6. Often I ﬁnd myself taking responsibility for other people’s problems. 1 2 3 4 5
7. I am willing to ask myself tough questions and answer them honestly. I 2 3 4 5
8. I have a hard time telling what someone new is like until I get to know the person

well. 1 2 3 4 5

I can ﬁx hurts from my past that could keep me from letting people get close to

me. 1 2 3 4 5

10. I try to ﬁgure out why people act the way they do. 1 2 3 4 5

11. I will often stay with someone, even though I know that person
is badforme. l 2 3 4 5

12. I am able to step back from troubled family members
and see myself as OK. 1 2 3 4 5

13. If you care about someone, you should try to do what the person wants, even if it
seems unreasonable. 1 2 3 4 5

14. I can’t help acting like a child around my parents. 1 2 3 4 5

15. I am able to recognize when I’m in a bad relationship and end it. 1 2 3 4 5

16. I can stay calm around troubled people because I understand why they act the way
they do. 1 2 3 4 5

17. I realize that I can’t change other people; they have to change
for themselves. 1 2 3 4 5

18. It’s hard for me to stay calm when someone I care about is
being unreasonable. 1 2 3 4 5

19. If I love someone, I can put up with that person hurting me.

20. I often ﬁnd myself around people who aren’t well adjusted.

21. There are few people who I can really count on. 1 2 3 4 5

22. I am good at sizing up people. 1 2 3 4 5

>0

1345
1345

2
2

166

 

23.

24.
25.
26.
27.
28.
29.
30.
31.
32.
33.

34.
35.

36.
37.
38.
39.
40.
41.

42.
43
44.

45.

46.
47.
48.
49.

50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.

I try to ﬁgure out why a relationship was not healthy

and avoid repeating it. 1 2 3 4 5

I am good at starting relationships with other people. 1 2 3 4 5

I can’t do anything about whether people like me or not. 1 2 3 4 5

It’s hard for me to believe that I’ll ever ﬁnd a good relationship. 1 2 3 4 5
I’m shy around people I don’t know. 1 2 3 4 5

I can’t really tell if a relationship is going to be good until I try it. 1 2 3 4 5

I am good at keeping relationships going. 1 2 3 4 5

I am able to love others and be loved by them. 1 2 3 4 5

It’s beyond me how most things work. 1 2 3 4 5

I often talk myself through a problem. 1 2 3 4 5

I can learn from the past and use that information to make

the future better. 1 2 3 4 5

I have hobbies or other activities that I take seriously. 1 2 3 4 5

I often get really frustrated when dealing with problems and can’t ﬁgure out what
to do. 1 2 3 4 5

I am successful in taking care of my physical and emotional needs. 1 2 3 4 5
I don’t like to try to ﬁnd out how things work. 1 2 3 4 5

There are few things that I am good at doing. 1 2 3 4 5
I do enough to get by, but not much more. 1 2 3 4 5
I enjoy getting involved in constructive activities. 1 2 3 4 5

Sometimes I forget my problems when I’m pursuing
creative activities. 1 2 3 4 5
Idon’t think that I’m creative. 1 2 3 4 5

. I’m good at ﬁnding new ways to look at things. 1 2 3 4 5

One way to express my feelings is through my artwork,
dance, music or writing. 1 2 3 4 5

The positive feelings I get from creating help make up for
the pain ofmy past. 1 2 3 4 5

Using my imagination doesn’t help to solve problems. I 2
It’s hard for me to see the humor in a bad situation. 1 2 3
One has to take life very seriously to get by. 1 2 3 4 5

I am good at using humor to reduce tension between

others and myself. 1 2 3 4 5

Most problems have only one solution. 1 2 3 4 5

I ﬁnd it easy to choose between right and wrong. 1 2 3 4 5

It’s a dog eat dog world where one has to do what it takes to get by. l 2 3 4 5
I can’t help repeating the mistakes that my parents made. 1 2 3 4 5

I like to help other people. 1 2 3 4 5

There’s no way I could make a difference in other people’s lives. 1 2 3 4 5

I don’t always do what I know is right. 1 2 3 4 5

I stand up to people when I see them being dishonest, petty or cruel. 1 2 3 4 5
I am willing to take risks for the sake of doing what I think is right. 1 2 3 4 5
Sometimes I feel like I’m just drifting along with no purpose in life. 1 2 3 4 S

I almost always stand up for underdogs. 1 2 3 4 5

345
45

167

61. I like to help others even if they are not willing to help themselves. 1 2 3 4 5
62. I am involved in things that will make people’s lives better. 1 2 3 4 5

63. No matter what happens, if I keep trying I’ll get through it. 1 2 3 4 5

64. There are things that I can do to make my life better. 1 2 3 4
65. Sometimes it’s hard, but I don’t let things keep me down. 1 2
66. Even if bad things happen, I can deal with them. 1 2 3 4 5
67. It’s not that hand you are dealt, but how you play it. 1 4 5
68. No matter how hard I try, I can’t make things right. 1 4 5
69. I am willing to go with any approach that will work. 1
70. I’m good at making the most of a bad situation. 1 2 3
71. When life gives me lemons, I make lemonade. 1 2 3 4 5

72. Failure is something you learn from rather than feel guilty about. 1 2 3 4 5

5
345

23

23
2345
45

I68

 

Multidimensionality of Religiosity Scale

Directions: Next, we’d like to ask some more questions concerning religion.

1. How often do you usually attend religious services?

1. Less than once a year

2. A few times a year

3. A few times a month (1 - 3x)

4. At least once a week (1- 3x)

5. Nearly everyday (4 or more times a week)

2. Are you an ofﬁcial member of a church or other place of worship?

1.) Yes
2.) No

3. How many church clubs or organizations do you belong to or participate in?

1.) None
2.) 1-3
3.) 4 or more

4. Besides regular service, how often do you take part in other activities at your place of
worship?

1.) Never

2.) A few times a year

3.) A few times a month (l-3x)

4.) At least once a week (l-3x)

5.) Nearly everyday (4 or more times a week)

5. Do you hold any positions in your church or place of worship?

1.) Yes
2.) No

6. How often do you read religious books or other religious materials?
1.) Less than once a year
2.) A few times a year

3.) A few times a month (l-3x)
4.) At least once a week (l-3x)

169

5.) Nearly every day (4 or more times a week)
7. How often do you watch or listen to religious programs on TV or radio?

1.) Less than once a year

2.) A few times a year

3.) A few times a month (1-3x)

4.) At least once a week (1-3x)

5.) Nearly everyday (4 or more times a week)

8. How often do you pray?

1.) Less than once a year

2.) A few times a year

3.) A few times a month (l-3x)

4.) At least once a week (1-3x)

5.) Nearly everyday (4 or more times a week)

9. How often do you ask someone to pray for you?

1.) Less than once a year

2.) A few times a year

3.) A few times a month (1-3x)

4.) At least once a week (1-3x)

5.) Nearly everyday (4 or more times a week)

10. How religious would you say you are?

1.) Not religious at all
2.) Not too religious
3.) Fairly religious
4.) Very religious

1 1. How important was religion in your home when you were growing up?

1.) Not important at all
2.) Not too important
3.) Fairly important
4.) Very important

12. How important is it for Black parents to send or take their children to religious
services?

1.) Not important at all

2.) Not too important
3.) Fairly important

170

4.) Very important

13. How spiritual would you say you are?
1.) Not spiritual at all
2.) Somewhat spiritual
3.) Very spiritual

14. How important is your spirituality to you?
1.) Not important
2.) Somewhat important
3.) Very important

15. Looking back at your spiritual beliefs since you were six years old, are these beliefs
the same, less, or more?
1.) Less than they were
2.) About the same
3.) More committed than they were

16. How much help is your church/religious community to you? Would you say:

1.) A lot of help
2.) Some help

3.) A little help
4.) No help

5.) No attendance

l7l

 

Beliefs and Attitudes Survey (AAAS-R)

Below are some beliefs and attitudes about religion, families, racism, Black people,
White people, and health. Please tell us how much you personally agree or disagree with
these beliefs and attitudes by circling the number. There are no right or wrong answers;
we simply want to know your views and your beliefs.

 

I Totally Sort of I Strongly
Disagree Agree Agree
Not True Sort of Absolutely
at all True True
01. I believe in the Holy Ghost. 1 2 3 4 5 6 7
02. I like gospel music. 1 2 3 4 5 6 7
03. I believe in heaven and hell. 1 2 3 4 5 6 7
04. The church is the heart of the Black
community 1 2 3 4 5 6 7
05. I have seen people “get the spirit” or
speak in tongues. 1 2 3 4 5 6 7
06. I am currently a member of a Black
church. 1 2 3 4 5 6 7
07. When I was young, I was a member of a
Black church. 1 2 3 4 5 6 7
08. Prayer can cure disease. 1 2 3 4 5 6 7
O9.What goes around, comes around. 1 2 3 4 5 6 7
10. I used to sing in the church choir. 1 2 3 4 5 6 7
11. Most of the music I listen to is by
Black artists. 1 2 3 4 5 6 7
12. I like Black music more than White
music. 1 2 3 4 5 6 7
13. I listen to Black radio stations. 1 2 3 4 5 6 7

14. I try to watch all the Black shows on TV. 1 2 3 4 5 6 7

15. The person I admire the most is Black. 1 2 3 4 5 6 7

172

 

16. I feel more comfortable around Blacks
than around Whites.

17. When I pass a Black person (a stranger)
on the street, I always say hello or nod
to them.

18. Most of my friends are Black.

19. I read (or used to read) Essence or
Ebony magazine.

20. I don’t trust most White people.

21. IQ tests were set up purposefully to
discriminate against Black people.

22. Most Whites are afraid of Blacks.

23. Deep in their hearts, most White people
are racist.

24. Whites don’t understand Blacks.

25. Most tests (like the SAT’s and tests to
get a job) are set up to make sure that
Blacks don’t get high scores on them.

26. Some members of my family hate or
distrust White people.

27. When I was young, I shared a bed at
night with my sister, brother, or some
other relative.

28. When I was young, my parent(s) sent me
to stay with a relative (aunt, uncle, grand-
mother) for a few days or weeks, and then
I went back home again.

29. When I was young, my cousin, aunt,
grandmother, or other relative lived
with me and my family for a while.

30. When I was young, I took a bath with

173

 

31.

32.
33.
34.
35.

36.

37.
38.
39.
4o.
41.
42.

43.

44.

my sister, brother, or some other relative.
Some people in my family use Epsom
salt.

Illnesses can be classiﬁed as natural types

and unnatural types.

Some old Black women / ladies know
how to cure diseases.

Some older Black women know a lot
about pregnancy and childbirth.

I was taught that you shouldn’t take a
bath and then go outside.

I avoid splitting a pole

When the palm of your hand itches,
you’ll receive some money.

There’s some truth to many old
superstitions.

I eat black-eyed peas on New Year’s
Eve.

I grew up in a mostly Black neighbor-
hood.

I went to (or go to) a mostly Black high
school.

I went to a mostly Black elementary
school.

I currently live in a mostly Black
neighborhood.

It’s better to try to move your whole
family ahead in this world than it is to

174

 

 

be out for only yourself.
45. Old people are wise.

46. I often lend money or give other types
of support to members of my family.

47. A child should not be allowed to call a
grown woman be her ﬁrst name, “Alice.”
The child should be taught to call her

“Ms. or Mrs. Alice.”

175

The Core Convictions Scale
Instructions:
Below are statements conceming personal beliefs. Please read each statement and circle

the responses that reﬂects the extent to which you agree with each item.

Strongly disagree Disagree Agree Strongly agree
a b c d

1.) God works everything out for my good. a b c d
2.) God is caring, just, fair, and impartial. a b c d
3.) God is all-powerful and in control of every situation. a b c d
4.) God knows everything. a b c d
5.) God is good and everything that God has created is good. a b c d
6.) God is merciful and gracious. a b c d
7.) All humans are created equal. a b c d
8.) Every human being is unique and worthy of respect. a b c d
9.) Despite any differences, we are all related or connected in
some way. We are brothers and sisters. a b c d
10.) We should persevere and endure until we

succeed, and in our identity and faith. a b c d

176

 

REFERENCES

Ahn, R. L. (1991). Development and validation of the Washington resiliency
scale. Disse_rtation Abstracts International. 52 (08B), 4457.

Aldwin, C. M. (1994). Stress, coping. and development: An integﬁive
mrsmctive. New York: The Guilford Press.

Asante, M. K. (1987). The afrocentric idea. Philadelphia: Temple University
Press.

Asante, M. K. (1988). Afrocentricig. Trenton: Aﬁica World Press.

Baer, H. A. (1981). PrOphets and advisors in Black spiritual churches: therapy,
palliative, or opiate? Culture. Medicine. and Psvchi_atrv. 5. 145-170.

Barbarin, O. A. (1993a). Emotional and social development of African American
children. Joumpl of Black Psychology, 19 (14), 381-390.

Barbarin, O. A. (1993b). Coping and resilience: Exploring the inner lives of
African American children. Journal of M Psychology. 19(14), 478-492.

Bergin, A. E., Stinchﬁeld, R. D., Gaskin, T. A., Masters, K. S. & Sullivan, C. E.
(1988). Religious lifestyles and mental health: An exploratory study. Journal of
Counseling Psychology. 35 (1), 91-98.

Biscoe, B. & Harris, B. (1994). RAS: Resiliency attitudes manual. Oklahoma:
Eagle Ridge Institute, Inc.

Billingsley, A., & Caldwell. C. H. (1991). The church, the family, and the school

in the Aﬁican American community. J oumal of Negro Education, 60 (3), 427-439.

177

Boyd-Franklin, N. (1989). Fivekey factors in the treatment of Black families:
Minorities and family therapy. New Jersey: Haworth Press, Inc.

Bowen, M. (1978). Family therapy in clinical practice. New York: Aronson.

Bryant, F. B. & Yarnold, P. R. (1995). Principal-components analysis and
exploratory and conﬁrmatory factor analysis. In L. G. Grimm & P. R. Yarnold (Eds),
Mg And Understanding Multivariate Statistics. Washington, DC: American
Psychological Association.

Browne, M. W. & Cudeck, R. (1989). Single sample cross validation indices for

covariance structures. Multivariate Behavioral Research 24 445-455.

 

Carver, C. S. & Scheier, M. F. (1981). Attention and self-regglation: A control-
theory approach in human behavior. New York: Springer-Verlag.

Carver, C. S., Scheier, M. F. & Weintraub, J. K. (1989). Assessing coping
strategies: A theoretically based approach. Journal of Personalig and Social Psychology.
3 (2), 267-283.

Chesley-Carter, J. (1998). The relationship between at-risk students’ resilience

and dropout behavior. Dissertation Abstracts International 59 (07A).

 

Cooper-Lewter, N. & Mitchell, H. H. (1996). Soul theology. Nashville, TN:
Abingdon Press.

Crocker, L. M. & Algina, J. (1986). Introdﬂtion to classical and modern test
M. New York: Holt, Rinehart, and Winston.

Dana, R. H. (1993). Multicultural assessment perspectives for professional

psychology. Boston: Allyn & Bacon

178

Donahue, M. J. & Benson, P. L. (1995). Religion and well-being of adolescents.
Journal of Social Issues. 51 (2), 145-160.

Ellison, C. G. (1993). Religious involvement and self-perception among Black
Americans. Social Forces, 71 (4), 1027-1055.

Ellison, C. G. & Gay, D. A. (1990). Religion, religious commitment, and life
satisfaction among Black Americans. The Sociology Quarterly, 31(1), 123-147.

Feagin, J. R., Vera, H. & Imani, N. (1996). The agony of education: Black
students at White colleges and universities. New York: Routledge.

Fine, S. B. (1991). Resilience and human adaptability: Who rises above
adversity? American Journal of Occupational Therapy, 45 (6), 493-503.

F lach, F. (1988). Re_silience: di§covering a new strengm at times of stress. New

York: Ballantine Books.
Folkman, S. & Lazarus, R. S. (1980). An analysis of coping in a middle-aged

community sample. Journal of Heali and Social Behavior 21, 219-239.

 

Ford, C. A., Kokjie, F. A. & Lewis, M. K. (1996). Factors that contribute to
academic resilience among urban “at risk” African American male college students.
Challenge: A Journal of Res:earch on Aﬁican American Men. 7 (3), 17-29.

Frazier, E. F. (1964). The Negro church in America. New York: Schocken
Books

Freitas, A. L. & Downey, G. (1998). Resilience: A dynamic perspective.
International Jourpal Of Behavioral Development, 22 (2), 263-285.

Garmezy, N. (1993). Children in poverty: Resilience despite risk. Special issue:

children and violence. Psychiatry Interpersonal & Biological Processes. 56(1), 127-136.

179

Garmezy, N. & Masten, A. S. (1986). Stress, competence, and resilience:
common frontiers for therapists and psychopathologists. Behavior Therapy, 17 (5), 500-
521.

Genero, N. P. (1995). Culture, resiliency, and mutual psychological development.
In H. I. McCubbin, E. A. Thompson, A. 1. Thompson, & J. A. Futrell (Eds), Resiliency in
African-American Farah. California: Sage Publications.

Giddan, N. S. (1987). Coping and identity development in college students.
Jouraal of Colleg Student Psychotherapyg, 33-58.

Glantz, M. D. & Johnson, J. L. (1999). Resilience and development: Positive life
adaptations. New York: Kluwer Academic/Plenum Publishers.

Ickovics, J. R. & Park, C. L. (1998). Paradigm shift: Why a focus on health is

important. Journal of Social Issues 54(2), 227-244.

 

Jenkins, A. H. (1995). Mngomers: The psychology of Aﬁican Americans.
Massachusetts: Allyn & Bacon.

Jew, C. L., Green, K. E., & Kroger, J. (1999). Development and validation of a
measure of resiliency. Mutement And Evaluation In Counseling And Development,
2, 75-89.

Jew, C. L. & Green, K. E. (1998). Effects of risk factors on adolescents’
resiliency and coping. Psychological Repprts, 82. 675-678.

Jones, R. L. (1991). Black Psychology. Berkeley, CA: Cobb & Henry Publishers.

Jung, J. & Khalsa, H. K. (1988). The relationship of daily hassles, social support,
and coping to depression in Black and White students. Journal of General Psychology.

m (4), 407-417.

180

Kagan, J. (1975). Resilience in cognitive development. Ethos.3 (2), 231-247.

 

Kaplan, H. B. (1999). Toward an understanding of resilience: A critical review of
deﬁnitions and models. In M. D. Glantz & J. L. Johnson (Ed.), Resilience and
Development: Pasitive Life Adaptations. New York: Kluwer Academic/Plenum

Publishers.

Kaplan, C. R, Turner, 8., Norman, E., & Stillson, K. (1996). Promoting

 

resilience strategies: A modiﬁed consultation model. Social Work In Education 18 (3),
158-168.

Kelso, T. W. (1999). A study to establish the validity of a measure of resiliency.
Dissertation Abstracts International. Thesis (M.A.) - Humboldt State University.

Kinard, E. M. (1998). Methodological issues in assessing resilience in maltreated
children. Child Abuse and Neglect. 22 (7), 669-680.

Klem, L. (1995). Path analysis. In L. G. Grimm & P. R. Yarnold (Ed.), Rﬂdﬂg
And Understanding Multivariate Statistics. Washington, DC: American Psychological

Association.

Kline, R. B. (1998). PrincipleL and practices of structural equation modeling.

New York: The Guilford Press.
Klonoff, E. A. & Landrine, H. (2000). Revising and improving the African
American acculturation scale. Journal of Bleﬁc Psychology. 26 (2), 235-261.
Kumpfer, K. L. (1999). Factors and processes contributing to resilience: the
resilience framework. In M. D. Glantz, & J. L. Johnson (Eds.), Resilience and
development: msitive life adaptations (pp. 179-224). New York: Kluwer

Academic/Plenum Publishers.

181

 

Lazarus, R. S. & Folkman, S. (1984). Stress. appraisal. aad_g_op_ipg. New York:
Springer Publishing Company.

Levin, J. S., Taylor, R. J ., & Chatters, L. M. (1995). A multidimensional
measure of religious involvement for African Americans. The Sociological Quarterly. 36
(1), 157-173.

Logan, S. M. L. (1980). The Black Baptist church: A social psychological study
in coping and growth. Dissertation Abstracts International. 41 (06A), 2769.

Luster, T. & McAdoo, H. P. (1994). Factors related to the achievement and
adjustment of young African American children. Child Development, 65, 1080-1094.

Luthar, S. S. & Zigler, E. (1991). Vulnerability and competence: A review of
research on resilience in childhood. American Journal of Orthopsychiagy. 61 (1), 6-22.

Mandleco, B. L. H. (1991). The relationship between resilience, social
competence, coping ability, and gender. Dissgtation Abstracts International. 52 (038),
1749.

Martin, P. P. (2001). The Aﬁican American church and African American
parents: Examining relationships between racial socialization practices and racial identity
attitudes. Dissertation Abstracts International. 62 (03B), 1560.

Massey, S. D. (1999). A study of the relationship between resilience and
spirituality among high risk youth (at risk, academic competence, social competence,

behavioral competence, high school students). Dissertation Abstracts International 60

 

(03A), 0697 (1-141).

McCall, E. L. (1986). Black church lifestyles. Nashville, TN: Broadman Press

182

McCubbin, H. 1., Thompson, A. I., & McCubbin, M. A. (1995). falily
W.
Madison: University of Wisconsin Publishers.

Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard
University Press.

Mitchell, H. H. (1975). m belief: folk beliefs and 81W
West Aﬁica. New York: Harper & Row Publishers.

Monaghan-Blout, S. (1996). Re-exarnining assumptions about trauma and
resilience: implications for intervention. Psychotherapy in Private Practice. 15 (4), 45-68.

Moore, D. W. (2000). Two of three Americans feel religion can answer most of
today’s problems. _Ciallop News Service; The Gallop Organization; Princeton, NJ.

Mrazek, P. J. & Mrazek, D. A. (1987). Resilience in child maltreatment victims:
a conceptual exploration. Child Abuse & Neglecg, 11. 357-366.

Murphy, M. C. & Archer, J. A. (1996). Stressors on the college campus: a
comparison of 1985 and 1993. Journal of College Student Development. 37. (1). 20-28.

Neighbors, H. W., Jackson, J. S., Bowman, P. J. & Gurin, G. (1983). Stress,
coping, and Black mental health: preliminary ﬁndings from a national study. Innovations
in Prevention. 5-29.

Nero, R. L. (1988). Ways of coping among alumni and students of a historically

Black church related college. Dissertation Abstracts International 49 (12A), 3629.

 

Neville, H. L., Heppner, P. P., & Wang, L. (1997). Relations among racial
identity attitudes, perceived stressors, and coping styles in African America college

students. Journal of Counselig & Development. 75. 303-311.

183

Oler, G. A. (1997). Religious coping and psychological adjustment among
African American Christians (Stress, Depression, Self-Esteem). Dissertation Abstragta
International. 58 (l 13), 6242.

Pargament, K. I., Ensing, D. S., Falgout, K., Olsen, H., Reilly, B., Van Haitsma,
K., & Warren, R. (1990). God help me: (I): religious coping efforts as predictors of the
outcomes to signiﬁcant negative life events. American J ournal_ﬁC_o_mm_ty_uiPsyc_hQQgL
18 (6), 793-824.

Pedersen, P. (1987). Ten frequent assumptions of cultural bias in counseling.
qu_m_al of Malticultuaal Counseling and Developmeat, 15 (1), 16-24.

Ponterotto, J. G. & Casas, J. M. (1991). Handbook of racial / ethnic minorig
counseling research. Springﬁeld, Illinois: Charles C. Thomas Publisher.

Rarnbo-Chroniak, K. M. (1999). Coping and adjustment in the freshman year
transition. Dissertation Abstracts International Section A: Human and Social Sciences.
ﬂ (12-A), 4378.

Racklin, J. M. (1999). The roles of sense of coherence, spirituality, and race in
response to trauma. mum Abstracts International: Section B: The Science &
Epgineering. 59. (9-B). 5106.

Retention Through Resilience (RTR) MM. (August, 1999). Training Seminar:
ASPIRE Training Institute Professional Development Seminar, Michigan State

University, East Lansing, MI.

184

Richardson, B. L. & June, L. N. (1997). Utilizing and maximizing the resources
of the Aﬁican American Church: strategies and tools for counseling professionals. In C.
C. Lee (Eds), Multicultural Issues in Counaeling: New Approaches to Diversity (2“ ec_l,,
pp. 155-170). Alexandria, VA: American Counseling Association.

Rosen, C. E. (1982). Ethnic differences among impoverished rural elderly in use
of religion as a coping machinism. Journal of Rural Communig Psychology, 3 (2), 27-
34.

Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American
Journal of Orthop§vchi_atrv. 57 (3), 316-331.

Rutter, M. (1985). Resilience in the face of adversity: protective factors and
resistance to psychiatric disorder. British Journal of Pachiagy, 147, 598-611.

Schiele, J. H. (1996). Afrocentricity: and emerging paradigm in social work
practice. Social Worl_<.;41(3), 284-294.

Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual
identity and performance. American Psychologist, 52 (6), 613-629.

Swigonski, M. E. (1996). Challenging privilege through Africentric social work
practice. Social Work. 41(2), 153-161.

Tarter, R. E., & Vanyukov, M. (1999). Re-visiting the validity of the construct of
resilience. In M. D. Glantz , & J. L. Johnson (Ed.), Resilience and development: Positive
life adaptations (pp.85-100). New York: Kluwer Academic/Plenum Publishers.

Tennen, H., Afﬂeck, G., Armeli, S. & Carney, M. A. (2000). A daily process
approach to coping; linking theory, research, and practice. American Psychologist, 55

(6), 626—636.

185

 

 

Thompson, G. L. (1998). Predictors of resilience in African American adults.
Dissertation Abstracts International Section A: Humap_ and Socig Sciences. 59 (4-A),
1082.

Turner, L. (1992). Coping strategies of resilient African American adolescents.

Paper Presented a the Annual Convention of the American Psycholcgical Association

 

(100th, Washington, DC, August 14-18, 1992).

Ventis, W. L. (1995). The relationship between religion and mental health.
Journal of Social Issues. 51 (2), 33-48.

Washington, J. R. (1964). BM religion: the Neggo and Christianig in the
United States. Boston: Beacon Press.

Werner, E. E. (1995). Resilience in development. Current Directions in
Psychological Science. 4(3), 81-85.

Werner, E. E. (1989). High-risk children in young adulthood: A longitudinal
study from birth to 32 years. American Journal of Orthopsychiat_ry. 59(1), 72-81.

Wolin, S. J. & Wolin, S. (1993). The re§ilient self: How survivors of troubled

gmilies rise above adversity. New York: Villard.

186

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1111:11lllzllﬂlllllcllzlﬂlllzlllwlﬂlrHI