~

-

m .

m

m

w

m
I

 

 

 

 

 

 

_ .
. .
.
.
. .
.. .
.‘
, .l
. .
. ..
.
. . ..
q .
. .
_
. . .
. .
. .
l .
ll

 

 

 

 

 

 

 

 

f .. ;,.. . .
L'r . -‘- I 'uuu‘ ~ ..

 

HWHWWlezbbéfibU'
l m .le :l l l H] l: an . h.

‘l
Ill) rllll ll

3 1293 00563 5317

 

LIBRARY
Michigan Stat.

L ’ University 4

 

 

This is to certify that the

dissertation entitled

AN INVESTIGATION OF THE RELATIONSHIPS BETWEEN
ADULT RELIGIOUS ORIENTATION AND ERIKSONIAN
PSYCHOSOCIAL DEVELOPMENT

presented by

Jay Murray Terbush, IV

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Counseling Psychology

‘4.) Limin— ’jquzAL

 

 

Major professor

Ma 18 1989
Date y ’

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

.—- ~__,

fi'“ _
“-vfiF-

 

 

 

MSU

 

 

 

RETURNING MATERIALS:
Place in book drop to

 

LJBRARJES remove this checkout from
-_1-—. your record. FINES will
be charged if book is
returned after the date
, stamped below.
% a"?

 

EMU \u‘k .0 ::—)('16

3.?

FEB 2 4 H
.4 ’99 x
0:) ,5 '7 "
APR 2 4 19 ‘PI.

l
l .

t“

x

"‘ ’ ill:

diam

 

Dray c f WM?
' I "

[12:3

APRJh 4 1995

\

JCS? .2 9 1991 jj -.

4 3-“

 

JAN l7 l9???

0

 

AN INVESTIGATION OF THE RELATIONSHIPS
BETWEEN ADULT RELIGIOUS ORIENTATION

AND ERIKSONIAN PSYCHOSOCIAL DEVELOPMENT

by

Jay M. Terbush

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

1989

ABSTRACT

AN INVESTIGATION OF THE RELATIONSHIPS
BETWEEN ADULT RELIGIOUS ORIENTATION
AND ERIKSONIAN PSYCHOSOCIAL DEVELOPMENT

by

Jay M. Terbush

This descriptive study investigated the relationship between adult
intrinsic (I) and extrinsic (E) religious orientations and the eight
psychosocial developmental tasks of Eriksonian theory. Previous research
and theory with the religious orientations suggested that I and E were
related to other aspects of personality development and developmental
processes.

A randomly identified sample of 184 Protestant, religiously-
affiliated adults completed the Religious Orientation Scales (ROS); the
Assessment of Adult Adjustment Patterns (AAAP), which measures mastery of
the eight Eriksonian stage tasks; and a demographic sheet. Subjects'
scores on I and E were correlated with scores on the eight AAAP stages.
Subjects were divided into four groups: Intrinsic (High I, Low E), Non-
religious (Low 1, Low E), Extrinsic (Low I, High E), and Indiscriminately
Religious (High 1, High E). Group means for the eight stages were

compared utilizing ANOVAs. Presence or absence of self-reported

Jay M. Terbush

religious conversion, disillusionment with religious, and experience
profoundly affecting one's life were studied in relation to scores on I
and E and AAAP stages.

Results support the conclusion that extrinsic religious orientation
is a less-psychologically healthy way of being religious than intrinsic
orientation. I is positively correlated to Stage 8 (wisdom). E is
negatively correlated to Stage 2 (autonomy), Stage 5 (identity), Stage 6
(intimacy), and Stage 8 (wisdom). Intrinsic and Non-religious persons
are higher on autonomous functioning (Stage 2) than Extrinsic and
Indiscriminately Religious persons. Using a two-group analysis, the Low
E group scored better on all eight stage tasks than the High E group.
Extrinsic religiousness is related to less-healthy overall psychological
or personality development and psychological functioning. The evidence
suggests that Intrinsic and Non-religious persons enjoy better
psychological functioning across all of the psychosocial characteristics
studied. Conversion experience relates to increased I and lowered E.
Both a period of disillusionment with one's religious faith and report of
an experience profoundly affecting one's view of self, others, and life
are related to higher I, lower E, higher Stage 8, and to a generally
healthier functioning personality. Reflected-on experiences related to

committed intrinsic religion and meaningful, purposeful living.

DEDICATION

To Dad

He would be very proud

iv

ACKNOWLEDGMENTS

A project of this magnitude could neither be undertaken nor
successfully completed without the support and assistance of other
people. There are many people to whom I am grateful; without their help,
the dream would not have become a reality.

Linda Forrest, as chairperson of my committee, you helped me to
discern when I was ready to focus my energies on this research. Your
balance of support of me (even from a phone booth in a national park!)
and challenge of me to do my best, matched what I needed to complete each
phase of this study. And you convinced me to enjoy the process because
it was my research and to actually have "fun" doing this.

John Schneider, your affirmation of my personal growth early in this
program and lately your encouragement to me through your interest in and
enthusiasm about this research have been appreciated.

John Powell, your incisive analysis of theoretical issues is always
stimulating and your professionalism has been an example to me of a
psychologist who is a committed Christian.

Gloria Smith, your accessibility for consultations, particularly in

the early phases of narrowing the topic and focusing the methodology,

helped me to develop a research proposal both relevant to me agd
manageable.

The people of First Congregational Church, you gave me your love and
support in granting me a leave of absence to pursue my dissertation. And
though to some it may not be clear why I had to do this, your commitment
to me and prayers for me have been a source of strength. I promise you
now my best efforts in service to our beloved church.

Jim, you helped me to find my way back to the dream when I had almost
given up in despair of ever finding it. You believed in me and showed me
how to affirm myself through a sometimes lonely process of producing
something "good enough."

Rich, our regular late night Big Boy consultations helped me to keep
this all in perspective. Your friendship is important to me.

Barbara, you kept your sense of humor (and helped me keep mine!)
while typing my nearly illegible handwriting--often under tight
deadlines. McDonald's breakfast biscuits will remind me of one "all
nighter" and a 6:00 a.m. word processing session!

To those anonymous subjects of this research, I say a special thank
you for your time, your willingness to participate, and your honesty in
responses. You have helped, not only this Ph.D., but to advance our
understanding of religious adults' development. Mae, Cathy, Jim, Carol,
Doug, Rich, and Diane, thank you for participating in the informal pilot

studies which helped me decide on this particular study as ”do-able."

vi

Bill, your generosity provided me a convenient, quiet office space in
which to spread out my materials and to think and write; and Norma, you
expedited my first post—deadline financial aid application, which
encouraged me to decide I can fulfill the dream, even if it was with
borrowed money!

Judge Jay M. Terbush and Geraldine Terbush, my parents: you gave me
the precious gift of life and nurtured me to adulthood, and I thank you
both for that. You also taught me to do the best that I can in whatever
task I undertake, and you showed me, by the way you lived, that the
Christian values of giving and service to others are the best way to a
meaningful life, helping, thereby, to set me on a career path that
intertwines ministry and counseling.

Marlene, your faithful, loving support of me as my life partner has
made my achievements possible. You have stuck with me, believed in me,
and put up with me through an educational process that took much longer
than we ever anticipated when we said "for better or worse, for richer or
poorer." I don't know how I would have done this without your
encouragement and love.

Finally, Katharine Rose, Carolyn Jeanne, Jessica Anne, and Jonathan
David Christian, my beloved children: your dad has missed much of this
important year in your lives. Each of you has brought me so much
joy--you girls learning to say "dissertation” at two, and Jon at 19

months saying, "Good luck, daddy," as I left for my oral defense! Your

vii

antics and the miracle of your unfolding personalities have been a
wonderful distraction for your daddy from the seriousness of scientific
research. It was partly for you that I dreamed the dream. Someday,

you'll understand.

viii

TABLE OF CONTENTS

List of Tables .

List of Figures

CHAPTER I: THE PROBLEM

Purpose of This Study

Need for This Study

Overview of Theory . . .
Religion and Mental Health . . .
Allport' 3 Mature Religious Sentiment .
Religious Orientations .

Religious Orientation in Developmental Perspective .

Summary .
Statement of Hypotheses in Broad Research Form .
Overview of the Dissertation .

CHAPTER II: LITERATURE REVIEW .

Introduction . . . . . . . . . . . . . .
Religious Orientation: Bipolar or Orthogonal? .
Religious Orientations and Psychological Health
Correlational Studies of I and E .
ROS and the Fourfold Typology
Summary .
Religious Orientations as Developmental Phenomena
Summary

CHAPTER III: METHODOLOGY

Sample and Data Collection Procedures
Instrumentation .
Demographic Sheet .
Religious Orientation Scales .
Assessment of Adult Adjustment Patterns

ix

. xiii

H
“\INUlb

17
22
28
30
31

33

33
33
37
37
43
47
48
53

54

54
64
64
65
72

Design of the Study
Hypotheses in Testable Form
Data Analysis Procedures .
Summary

CHAPTER IV: RESULTS .

Results for the Original Hypotheses
Hypotheses 1-4 . . . . . .
Hypotheses 5-6 .

Post Hoc Analyses
AAAP . . .
Religious Orientation Scales .

Conversion, Disillusionment, and Profound Experiences

Four- fold Religious Orientation Typology .
Summary . . . . .

CHAPTER V: SUMMARY AND CONCLUSIONS

Summary of the Study . .
Limitations of the Study .
Limitations of the AAAP
Other Limitations
Discussion .
Conclusions .
Implications for Practitioners . .
Implications for Future Research .

Appendix .

Bibliography .

79
79
81
83

84

84
84
89
93
93
98
100
106
108

110

115
115
115
120
124
129
134
136

140

155

.10

.ll

.12

LIST OF TABLES

Components Associated with Intrinsic and Extrinsic
Religious Orientations .

Hypothesized Relationships of Models of Development

Erikson's Psychosocial Stages and Faith Dimensions .

Age of Subjects
Educational Level of Subjects
Personal Theological Orientation of Subjects .

Frequency of Self-Reported Attendance at a Church
Worship Service

Subjects' Report of Religious Conversion Experiences .

Age of Religious Conversion (for Those Who Indicated
Such an Experience) . . . . . . . . . . . . . .

Subjects' Disillusionment with Religion
Age of Disillusionment (for Those Who Indicated Such
an Experience) . . . . . . . . . . . . . . . .

Experience Profoundly Affecting One' 8 View of Self,
Others, and Life . . . . . . . . . . . .

Age of Experience Profoundly Affecting One' 3 View of
Self, Others, and Life . . . . . . . . . . . .

Internal Reliability Coefficients for I and E

Subscale Statistics for Extrinsic and Intrinsic
Religious Orientations .

xi

21

29

3O

58

58

59

59

61

61

62

62

63

63

66

68

 

.13

.14

.15

Fourfold Split of the Sample with the Number of
Subjects in Each Category . . . . . .

Means, Standard Deviations, and Reliabilities for the
Eight Stages of the AAAP Across Several Studies

Factors Emerging from the AAAP .

Pearson Correlation Coefficients for Relationships
Among ROS Subscale Scores and AAAP Stage Scores and
Highest Stage Mastered in Order . . . . .

Multivariate Analyses of Variance Utilizing Four
Religious Orientation Groups and Mean Scores for Eight
Stages of the AAAP .

Univariate Analyses of Variance for Mean Stage Scores
by Religious Orientation Groups

Mean Scores on Stage 2 and Stage 4 by Religious
Orientation Group . . . . . .

Rank Order of the Three Age Groups for Each of the
Eight Stages and the Highest Stage Mastered in Order .

Percentage of Subjects Who Mastered Each Stage
Regardless of Chronological Order and Percentage of
Subjects Who Mastered up to Each Stage in Order

Rank Order of the Mean Stage Scores for Each of the
Four Religious Orientation Groups

Means, Standard Deviations, T- values, and Significance

Levels for the Eight AAAP Stages by the High Extrinsic
and Low Extrinsic Groups . . . . . . . .

xii

72

75

76

85

90

91

91

95

96

107

109

LIST OF FIGURES

Hypothesized process of development within
religious orientation dimensions .

xiii

CHAPTER I

THE PROBLEM

Throughout the lOO-year history of psychology as a science, theorists
have held diverse attitudes about religion and religious people. Some
psychologists and psychiatrists have regarded religion as generally
unhealthy for people. Sigmund Freud, in his classic philosophical piece,
Ihg_£u§uzg_g£_ég_lllgsign (1927), argues that religion is an illusion
rising out of human beings' instincts, anxieties, wishes and feelings of
helplessness. Freud contended that science and rationality were
antithetical to and superior to religion. Interestingly, Albert Ellis
(1980) has maintained much the same position, arguing that religiosity is
opposed to rationality and emotional health. Ellis' perspective is that
religious people are neurotic and he writes forcefully:

Religiosity is in many respects equivalent to
irrational thinking and emotional disturbance .

. The elegant therapeutic solution to emotional
problems is to be quite unreligious . . . the less
religious they are, the more emotionally healthy
they will be. (p. 637)

In contrast to these theorists, Carl Jung had an essentially positive
view of religion and spiritual growth broadly defined. Jung (1933) at
one point wrote that of his patients over 35 years old ". . . there has

not been one whose problem in the last resort was not that of finding a

religious outlook on life.” (p. 229). Victor Frankl (1962) developed

his concept of logotherapy from observations of survivors of the
Holocaust. His own experience in a concentration camp led him to
conclude that a will to meaning in life was basic to every person. Frankl
believes that persons of faith often not only survive but grow and
transcend the difficult, even devastating conditions of life.

Whether religion is healthy or unhealthy is not only a theoretical
issue, but an empirical question as well. When one examines the
published research which might bear on the broad question of the
relationship between religiosity and mental health, the findings, not
surprisingly, are mixed. Several reviews of the literature in the area
of the psychology of religion have been completed in the last 20 years
(Sanua, 1969; Becker, 1971; Stark, 1971; Argyle and Beit-Hallahmi, 1975;
Bergin, 1983).

In a recent attempt to clarify the ambiguities present in many of the
published studies, Bergin (1983) quantitatively combined the data across
samples and studies through a meta-analytic procedure. He combined the
results of all the studies he could locate that used at least one measure
of religiousness and one measure of psychopathology or adjustment.
Overall, his findings were statistically insignificant with only a
marginal trend toward the positive effect of religion. When considering
the general question of the relationship of religion to mental health,
one can only give a highly equivocal answer.

In both the theoretical and empirical literature in the psychology of
religion, there has been an increasing tendency to consider religion, not
as a single factor, but as a phenomenon made up of several different

factors. As Bergin (1983) has suggested, this is similar to the

conceptualization of intelligence as involving a general (G) factor and
several specific (S) factors. Religion and its functions in an
individual's life are complex and multidimensional. Various researchers
have considered different ways of being religious, including religious
beliefs, attitudes, practices, rituals, experiences, and affiliations.

By considering various ways of being religious, it is possible to then
ask if a particular way of being religious is more healthy than any other
way.

A variety of conceptual models and research studies have attempted to
factor out dimensions of religiousness and to relate those dimensions to
measures of psychological health or pathology. One of the most promising
frameworks for research is Gordon Allport‘s (1967) concept of an
intrinsic (I) and extrinsic (E) orientation to religion. Meadow and Kahoe
(1984), in a recent textbook entitled Ihg_£§yghglggy_gf_ggligigg, state
that no other theoretical approach to religion in an individual's life
has had a greater impact on the scientific study of the psychology of
religion. Allport's Religious Orientation Scale (ROS), based on this I-E
orientation distinction, is one of the most frequently used instruments
for measuring religiousness in empirical studies (Donahue, 1985).

The use of this intrinsic and extrinsic distinction has generated a
body of research (of. Donahue, 1985) that, in general, supports the idea
that religious persons who are extrinsically oriented are less
psychologically healthy (i.e., more anxious, fearful, prejudiced) and
that intrinsically oriented persons are more healthy (i.e., internal

locus of control, sense of purpose in life). The research making use of

the Religious Orientation Scale will be discussed in more detail in
Chapter 2.

In addition to approaching religion and attempting to factor out
healthy and unhealthy ways of being religious, religion has also been
studied as a developmental phenomenon (Allport, 1950; Goldman, 1964;
Wilcox, 1980; Droege, 1978; Fowler, 1981; Bergin et a1., 1988). Although
the authors cited come from a variety of disciplines and theoretical
perspectives, each would support the idea that a healthier religiousness
is related to healthier personality development. Droege (1978) and
Fowler (1981) specifically relate their theory of religious development
to Erik Erikson's (1963, 1968) theory of healthy psychosocial development
across the lifespan.

One interesting proposal made by Kahoe and Meadow (1981) is that the
concept of extrinsic and intrinsic religious orientation is itself a
developmental phenomenon. In their article, they state they specifically
hope to stimulate research utilizing their developmental conceptual
framework. Their model is consistent with Allport's (1950) original
ideas about a mature and an immature religious sentiment. Allport and his
associates later refined this theoretical idea into the intrinsic
(mature) and extrinsic (immature) religious orientation (Allport & Ross,

1967).

W

There are two primary purposes of this research project. The first
purpose is to study the construct of religious orientation as a

developmental phenomenon. This study will attempt to find evidence in

adult religiously affiliated persons that an intrinsic orientation is
related to successful sequential mastery of Erikson's psychosocial
developmental tasks. Theories of religious development would lead one to
expect that a person with an intrinsic orientation would be more highly
psychologically developed than a person with either an extrinsic
religious orientation or an indiscriminately religious person.

The second purpose of this study is to consider the relationship
between religious orientation and psychological health as measured by the
level of mastery of each of the individual Eriksonian psychosocial
tasks. Both proposals made in the theoretical literature and the results
of previous research with the Religious Orientation Scale lead to an
expectation that the more a person's religious orientation is intrinsic
(I) as opposed to extrinsic (E), the more likely that person is more
psychologically healthy. The more successfully an individual resolves
the individual Eriksonian developmental tasks, the healthier that person

functions both intrapsychically and interpersonally.

W

In a most general sense, this research project will build on and
extend previous research utilizing the Religious Orientation Scale.
Several recent psychology of religion textbooks have reviewed and
summarized the research concerning the I-E orientation and have
encouraged on-going research utilizing this instrument and conceptual
framework (cf. Batson and Ventis, 1982; Donahue, 1985; Meadow and Kahoe,
1984). Donahue (1985), in summarizing his review of the I-E literature,

writes that ". . . although this research is at an initial stage, the

findings currently available bode well for the potential of the I-E
framework as a powerful explanatory tool in personality-social
psychology” (p. 416).

This research will contribute to the small but growing body of
knowledge in the psychological study of religion and of religious
orientation specifically. It will contribute to an understanding of the
differences in religious orientation.

A concern of many theorists in the study of religion has been to
discriminate between "good" and "bad” religion (Bergin, 1983), that is,
psychologically healthy and unhealthy religious faith, life, and
practice. By employing a measure of religious orientation, this study
will, it is hoped, further explicate the concept of healthy
religiousness. This distinction has important implications for
personality theory and research, for counselors and other mental health
professionals working with religious clients, for religious educators,
and for religious professionals (particularly ministers) who influence
the directions of growth for religious people.

This study seeks to consider some aspects of the relationship between
religion and developmental psychology. Although there have been several
theoretical frameworks for religious development and its relationship to
other aspects of personality development, there have been very few
empirical studies of this hypothesized relationship. The small number of
studies of religious adults and psychological developmental phenomena
indicate a need both for refinement of theory and additional empirical

study.

The concept of religious orientation has been hypothesized to be a
developmental phenomena (Kahoe and Meadow, 1981; Chirban, 1980; Gorsuch &
Venable, 1983), but to date the systematic study of this hypothesis has
been scant (Venable, 1984). By relating religious orientation to a
measure of healthy psychosocial development, this study will provide

evidence related to this largely untested intuitive notion.

W
W331

At some point in his or her career, nearly every psychotherapist will
treat a psychologically unhealthy, very religious person, for example,
some one with a paranoid psychosis manifesting itself in part through
grandiose religious delusions, or someone with a strongly restrictive
family background who struggles with excessive and inappropriate guilt
and intrapunitive thoughts, feelings, or behaviors, or someone who is
anxious and rigid with constricted attitudes toward self, others, and
life, lacking in spontaneity and an appreciation for the paradoxical in
life.

Each of these clinical examples are manifestations of dysfunctional
religion, a religious life that is clearly not healthy. William James
(1961), in his classic work ya11gt1gs_gf;3g11gigu§_fixpg;1gngg, first
published in 1902, described this type of religiousness as a "sick
soul." Since James, many theorists have provided conceptual frameworks
for describing the 'sick soul"; that is, the religion which is
psychologically unhealthy. Paul Pruyser's (1977) article entitled ”The

Seamy Side of Current Religious Beliefs,” discusses from a psychodynamic

perspective five ways in which religious beliefs gag (though not
necessarily must) promote neurotic constriction. Pruyser argues that
excessive religiosity can (1) distort reality testing; (2) trigger (or
maintain) regression from rational cognitive functioning to fixation on
archaic thought patterns and defense mechanisms; (3) lead to inadequate
and inappropriate coping with aggression and anger; (4) condone infantile
wishes; and (5) encourage helplessness and dependency by surrender of
higher order ego functioning and autonomy to authoritarian demands. An
experienced clinician could perhaps add to this list some of the
following as characteristics of unhealthy religiousness: an overactive
conscience or superego; excessive guilt; magical thinking that abrogates
personal responsibility; and intolerance for ambiguity, which is
symptomatic of narrowness, prejudice, and lack of creativity.

Clearly, pathology in religious beliefs, experiences or practices is
related to pathology elsewhere in the life and personality of the
individual (cf. Martin & Nichols, 1962; Stark, 1971). This observed
relationship between religiousness and personality is one of correlation
and not necessarily of causation. Different theoretical formulations
might argue a direction of cause and effect. For example, a pathological
religion causes pathology in the personality or behavior of the
individual, or on the other hand, that unhealthy religion is symptomatic
of an unhealthy personality. To put this in question form: Is the person
sick because his or her religion is sick? Or is the person's religion
sick because the person is? Developmentalists following James (Allport,
1950; Meadow & Kahoe, 1984; Darling, 1976; Aden, 1976) would probably

argue the latter perspective, while acknowledging the interaction of

various factors making these questions extraordinarily complex. Though
there are various ways to describe it theoretically and empirically, it
does seem that there is a "sick soul”, that is, an unhealthy way (or
ways) of being religious. Later in this chapter and the next, many of
these descriptions of unhealthy religiousness will come to be related to
extrinsic religious orientation or an indiscriminately pro-religious
orientation.

James (1961) also described a more satisfactory way of being
religious. One that promoted a general sense of happiness and wellbeing,
which he called the "religion of healthy-mindedness' (p. 85). As
discussed earlier, some psychological theorists would question whether
there is such a construct as a ”healthy religion" or that very religious
persons could also be mentally/psychologically healthy (Rokeach, 1960;
Ellis, 1980). Of course, this issue can be addressed both theoretically
and empirically by psychologists and others asking whether or not there
is such a thing as healthy religiousness and, if so, what it looks like?

Certainly, the great religious teachers, organized religious groups,
and religious people have answered questions about religion and mental
health affirmatively. All religions claim to provide their followers
with the good life. Even those religions which teach of a life after
death, also claim that the benefits to believers are part of ghis life,
not just the after life. In general, religious leaders claim to be
teaching the best way to live, and imply, if they do not overtly state,
that adherence to their beliefs, practices and lifestyle will promote

life satisfaction and, more importantly, mental health.

10

There are several ways in which one might intuitively expect that
religion could enhance personal growth and mental health and could,
therefore, be considered as healthy ways of being religious. For
example, first of all, religion can provide a philosophy of life,
informing life events with a context of meaning and purpose. Religious
faith is frequently concerned with ultimate issues of life and death,
good and evil, and can address in a comprehensive fashion questions
like: Who am I? and Why am I here? Religion can help to make sense of
life. Second, most religious philosophies encourage the development of a
system of values, ethics and morality, and can provide a framework of
principles for living and for human relationships. The concepts of love
for others, forgiveness, justice, fairness and equity in human relations
are basic to most religions of the world. The teachings of religion
address fundamental needs of individuals and societies. Organized
religions and informal religious groups, could, it would seem, be healthy
in a third way - by fostering a sense of belong and community. These
groups could provide a place for the development of relationships with
significant other people. The support and security of a religious group
and belief in a benevolent god could lead to optimism (hope) and personal
peace of mind, particularly in the face of personal suffering or
tragedy. Fourth, at least some religious activities and behaviors might
be therapeutic and promote psychological health; for example: prayer,
meditation and contemplation; the discipline of self-reflection and
introspection; and the activity of meeting with others for worship in a

quiet, beautiful, and inspirational setting.

11

Given the claims of religion and both its teachers and adherents, it
makes sense to consider scientifically the relationship between being
religious and being mentally healthy. This relationship between
religious variables and variables of psychological health or pathology is
one of the overarching research questions and methodological challenges
in the broad subfield called the psychology of religion.

During the decades of the 1930's through the 1950‘s, it was a common
research paradigm to compare religious versus non-religious people on
specific psychological measures. Diverse measures of religiousness were
used, as were various measures of mental functioning. Reviews of these
studies, interestingly, were often contradictory and, at best,
inconclusive. Martin and Nichols (1962) summarized a dozen or so studies
and conclude that religious believers are conforming, unintelligent,
defensive, and emotionally distressed. Davis (1965) concluded that there
was fairly consistent support for the claim of better mental health among
those who were religious. Sanua (1969), in a review of empirical
studies, concluded that most studies show no relationship between
religiousness and mental health, and Wright (1972), after a comprehensive
analysis of the literature on religion and mental health, felt that it
was difficult to find any coherent or meaningful pattern in the data.
Treating religiousness as a single dimension or variable produced few, if
any, significant results between those who were "religious" and those who
were not. Bergin's (1983) meta-analysis of empirical findings relating
psychopathology and religiousness was statistically insignificant with
only a trend toward “marginal support for a positive effect of religion"

(p. 176).

12

A great many of the studies summarized by these researchers had one
element in common: they tended to treat religiousness as a
unidimensional concept, as a single unified factor or trait of a person
(Bergin, 1983). Often the group being studied was a religiously
conservative, Protestant one, but, in general, a person was considered
religious or not-religious. In research on human behavior, religiousness
does emerge as a single factor, if one studies a large heterogeneous
population and limits the measurement of this factor to obvious indices
of conformity to institutionalized religious traditions. The early
research by Rokeach (1960) on prejudice correlating with very religiously
conservative people is an example of the results of this approach to
religion.

If, however, one chooses a more homogeneous sample of "religious"
persons and includes measures of detailed variations in religiosity,
several factors become apparent and the concept of ”religion" becomes
multi-dimensional. Religious persons (even those of a similar church
affiliation or theological persuasion) differ on many aspects of
religiousness: participation in organized religious activities, personal
religious experiences, the functions of beliefs in one's life, etc.
Spilka (1977), for example, found that over half of those persons who
said that religion was very important to them indicated that religious
beliefs had no real effect on their ideas or their conduct of their
everyday lives. Several researchers have attempted to parcel the
religious factor in various ways (cf. Spilka, Hood & Gorsuch, 1985;
Meadow & Kahoe, 1984). King and Hunt (1972), in recognition of the

multidimensional aspects of religious beliefs, commitments and practices,

13

have developed and validated scales for measuring such aspects of
religiousness as creedal assent, church attendance, financial support of
a church, religious knowledge, devotionalism, orientation to religion and
behavioral and cognitive salience. Later research in the area of
prejudice differentiated subtypes of very religious people and found that
religious people vary widely among themselves on measures of prejudice
(Feagin, 1964; Allport & Ross, 1967).

It is against the background of the multidimensionality of
religiousness that some theorists and researchers have attempted to
formulate theoretical notions of psychologically healthy religion and to

generate research to validate the concept.

WWW;

Although several psychological theorists have written about a healthy
religiousness (cf. Fromm, 1950; Clark, 1958; Allen & Spilka, 1967;
McConahay & Hough, 1973; Droege, 1978; Fowler, 1981; Batson & Ventis,
1982), this writer has found the work of Allport particularly appealing
as a starting point. Allport's (1950) concept of what he originally
called the mature religious sentiment, and later labeled the intrinsic
religious orientation, became a focus of a significant body of research
looking at two distinct ways of being religious, one of which was posited
to be more psychologically healthy then the other. A considerable body of
research, some of which will be reviewed in Chapter 2, has both supported
this theoretical distinction and modified it.

Allport's seminal work, Ihe_1ndLgigg§1_§ng_31§_fig11gigg, was
published in 1950. In the introduction to this work, he set forth his

purpose and his perspective on the subject matter:

14

I am seeking to trace the full course of religious
development in the normally mature and productive
personality. I am dealing with the psychology,
not with the psychopathology, of religion. The
neurotic function of religious belief, its aid as
an 'escape from freedom,’ is indeed commonly
encountered, so commonly that opponents of
religion see only this function and declare it to
dominate any life that harbors a religious
sentiment. With this view I disagree. Many
personalities attain a religious view of life
without suffering arrested development and without
self deception. Indeed it is by virtue of their
religious outlook upon life -- expanding as
experience expands -- that they are able to build
and maintain a mature and well-integrated edifice
of personality. (p. xiii)

Allport sets his conceptualization of the mature religious sentiment
in the context of both personality theory and developmental psychology.
His theory is cognizant that all the human psychological processes are
involved (or at least can be involved) in the religiousness of a person.
He treats religiousness as a developmental phenomenon which changes as
the person changes, and places the mature religious sentiment at the apex
of normal personality developmental processes, which begin in early
childhood and culminate in adulthood.

Allport maintains that most criticisms of religion are directed at
what he calls its "immature forms” (p. 61). Immature religion narrows
the interest and experiences of the person. It is either impulsive and
self-gratifying or constrictive, repressive and guilt producing. The
immature religious sentiment is 'unreflective” (p. 62) and provides no
context of meaning in which to locate the self. Finally, for Allport, an
unhealthy religion, the immature religious sentiment '. . . is not
unifying in its effect upon the personality. Excluding, as it does,

whole regions of experience, it is spasmodic, segmented, and even when

15

fanatic in intensity, it is but partially integrative of the personality"
(p. 62).

In contrast to this immature religion which has in a sense not grown
up, Allport delineated six characteristics of the mature religious
sentiment. First, the mature religious sentiment is gell_d1f£gzgnt1a§eg
and well thought out. Maturity of religion requires processes of
reflection, doubt, honesty, and balance, and represents many successive
discriminations and reorganizations. It implies the presence of critical
thinking and abstraction which Piaget termed formal operations. There is
an openness to new experiences and contradictory evidences as
opportunities for reevaluation. In contrast, undifferentiated immature
religion accepts uncritically with no reflection what is taught by
parents, political ideologues, or religious institutions or leaders. "In
compulsive religion there is a defensive ruling out of disturbing
evidence" (p. 73).

A second characteristic of the mature religious sentiment is that it
is dynamig in spite of its derivative nature. By derivative, Allport
means that religiosity arises developmentally from a complex of
intrapsychic and interpersonal factors including biological drives,
infantile needs, patterns of learning, and socialization. Both James
(1961) and Maslow (1964) speak of the "instinctual” bases of
religiousness in life. Immature religion, in a sense, gets stuck at this
level of basic needs and egocentrism. The focus is on concern for the
comfort of self, self-justification, and magical thinking in impulse and
wish fulfillment. The religion that is dynamic moves beyond this to

become "functionally autonomous" (p. 72); that is, it functions largely

16

independently of its origins. In a sense, the religious sentiment
becomes the master motive, exercising influence over all of the life of
the individual and able to direct it toward goals which are no longer
wholly self-interest.

The third attribute of mature religion is that it produces a
c2ns1s§gn§_ngzg11§y. The relationship between religion and morals is
often independent, and for Allport this would characterize an immature
religiosity. For him, religion should steadily and persistently
influence conduct. Not that there will never be moral conflict or
difficult choices or morally ambiguous and confusing situations, but the
mature religion is able to tolerate the ambiguity without becoming
sporadic or impulsive. Though some would not agree with him, Allport
believes that ”ethical standards are difficult to sustain without
idealism; and idealism is difficult to sustain without a myth of Being"
(p. 75).

Mature religion is also ggmpxghgnsixg: it brings order to all of
life. The religion provides a philosophy which covers all the aspects of
existence which are important to the person. Mature religion serves an
integrative function as it conveys intelligibility and direction to
behavior.

The demand that one's religious sentiment be
comprehensive makes for tolerance. One knows that
one's life alone does not contain all possible
values or all facets of meaning. Other people too
have their stake in truth. The religion of
maturity makes the affirmation ”God is," but only
the religion of immaturity will insist, "God is
precisely what I say He is." (p. 78)

A fifth characteristic closely related to its comprehensive nature is

what Allport discusses as the ingggral nature of the mature religious

17

sentiment. This is a life-long process of evolving a harmonious pattern
making sense of life experiences. Particularly troublesome is the
problem of evil and suffering in human experience, and developing a
consistent approach to this problem opens one to greater toleration for
the ambiguous and paradoxical in life.

The final characteristic of mature religion is that it is
fundamentally_hggzig§1g. Religious faith is held tentatively and subject
to revision, but it serves as the person's "working hypothesis" (p. 81).
Religion helps the person find answers to life's tough questions, but
ultimately those who have faith are risk takers. In mature religion,
doubt is possible concerning the tenets of the faith and certainty is

impossible. Certitude is sufficient for guiding one's life.

Religigus Orientatigns
Following the publication of WW (1950).

one of Allport's primary research interests was the relationship between
religion and ethnic prejudice (Allport, 1954, 1959, 1963, 1966; Allport &
Ross, 1967). This interest encouraged Allport and his associates to
refine the theoretical constructs of mature and immature religion and to
attempt to operationalize them for measurement purposes. Although in
earlier formulations Allport (1954) discusses ”two kinds of religion,”
which Allport labeled 'interiorized' and ”institutionalized”, and their
different ways of manifesting prejudice, a sharpening of these constructs
occurred toward seeing two ways of being religious or two different
religious outlooks (Allport, 1959, 1966, 1967). The labels intrinsic (I)

and extrinsic (E) were first introduced in 1959. In two articles

18

published shortly before his death, "The Religious Context of Prejudice"
(1966) and ”Personal Religious Orientation and Prejudice” (Allport &
Ross, 1967), Allport gives the most succinct discussion of these two
different "orientations” or ”motivations” (p. 434). What was being
described is not “religion” per se, or certain behaviors that could be
classified as religious, but rather the motivation associated with a
person's religious practices or beliefs.

Allport conceptualized I and E as opposite ends of a bipolar
continuum. Intrinsic and extrinsic religious orientations are "two polar
types of religious affiliation" (Allport, 1966, p. 451). Perhaps, in
part, because of the earlier developmental framework, it was hypothesized
that "most people, if they profess religion at all, fall upon a continuum
between these two poles" (Allport & Ross, 1967, p. 434). The simplest
and most often quoted way to characterize this intrinsic and extrinsic
distinction is that "the extrinsically motivated person Eggs his
religion, whereas the intrinsically motivated liggs his religion"
(Allport & Ross, 1967, p. 434).

Several extended quotations will serve to define these two religious
orientations. Regarding the extrinsic orientation, Allport (1959)
writes,

. religion is not the master motive in the
life. It plays an instrumental role only. It
serves and rationalizes assorted forms of
self-interest. In such a life, the full creed and
teaching of religion are not adopted. The person
does not serve his religion; it is subordinated to

serve him (p. 264).

About the intrinsically oriented religious life, he writes

l9

. . dogma is tempered with humility . . . . A
religious sentiment of this sort floods the whole
life with motivation and meaning. It is no longer
limited to single segments of self-interest. And
only in such a widened religious sentiment does
the teaching of brotherhood take root. (1959, p.
265)

Allport (1966) explained that he borrowed from axiology the concepts
of gagginsig and ingzinsig values because the distinction helped to
separate religiously affiliated people whose church membership "supports
and serves other non-religious ends“ from those people "for whom religion
is an end in itself -- a final, not instrumental, good" (p. 454). In
contrast to the intrinsic religious person who is oriented to faith as a
supreme, integrating value in life, which leads to unification of being
and transcendence of self, the extrinsic type have

. no true association with the religious
function of the church . . . they feel no
obligation to attend church regularly nor to
integrate religion into their way of life .
most extrinsics are casual and peripheral
churchgoers . . . a type of religion that is
strictly utilitarian: useful for the self in
granting safety, social standing, solace, and
endorsement of one's chosen way of life.
(Allport, 1966, p. 454-5)

Allport and Ross (1967) give the following as formal definitions of

these two orientations:

W: Persons with this

”orientation are disposed to use religion for their
own ends. The term is borrowed from axiology, to
designate an interest that is held because it
serves other, more ultimate interests. Extrinsic
values are always instrumental and utilitarian.
Persons with this orientation may find religion
useful in a variety of ways -- to provide security
and solace, sociability and distraction, status
and self-justification. The embraced creed is
lightly held or else selectively shaped to fit
more primary needs. In theological terms, the
extrinsic type turns to God, but without turning
away from self.

20

1W: Persons with this

orientation find their master motive in religion.
Other needs, strong as they may be, are regarded
as of less ultimate significance, and they are, so
far as possible brought into harmony with the
religious beliefs and prescriptions. Having
embraced a creed the individual endeavors to
internalize it and follow it fully. It is in this
sense that he ljxgg his religion. (p. 434)

These definitional ideas about religious orientation, while clearly
related to the earlier construct of mature-immature religion, have been
refined, and of necessity, simplified. Table 1.1 lists some of the key
concepts related to each of these orientations. Consistent throughout
the development of this religious orientation was an understanding that
the intrinsic religious orientation was the preferable orientation, the
way of being religious that was also more psychologically healthy
(Donahue, 1985a).

In research with religious orientation and prejudice, Allport and
others developed scales to measure ”religious orientation" (Wilson, 1960;
Feagin, 1964; Allport & Ross, 1967). They had noted repeated evidence in
the research on prejudice that some of the very religiously conservative
people were consistently free of prejudice, open, liberal socially and
democratic as opposed to authoritarian. They hypothesized that
differences in religious orientation were being confounded in studies
treating religiousness as a simple single factor. In a series of studies
using scales to measure prejudice and religious orientation, these
researchers demonstrated significant differences in prejudice for the two
religious orientations. Intrinsically religious persons were

significantly less prejudiced than extrinsically religious persons, and

the extrinsically religious persons accounted for most of the variance

21

Table 1.1
Components Agsociated with Intrinsic and Extrinsic Religious
Orientations

 

Intrinsis W
l. Relates to all of life, floods l. Compartmentalized, not
life with meaning, other needs integrated into life.

brought into harmony with
religious beliefs.

2. Ultimate, master motive, 2. Instrumental, utilitarian,
meaning endowing, integrative, religion as means,
unifying, religion as "end", self-serving.

self- transcending.

3. Unprejudiced, tolerant, 3. Prejudiced, exclusionary of
universal. those unlike oneself.

4. Mature, makes for mental 4. Immature, dependent, seeks
health. comfort, security, defensive.

5. Regular church attendance, 5. Irregular church attendance,
involvement for fellowship and affiliation for sociability
deeper values of faith. and status.

 

*
Chart based on Hunt and King (1971) pp. 342-343; and Donahue (1985a),
p. 401.

 

 

between religious persons and non-religious persons (cf. Feagin, 1964;
Allport & Ross, 1967). Of interest was the finding that the I and E
orientations, as measured by a refined Religious Orientation Scale, were
not bipolar but were separate and interactive (Feagin, 1964). Allport had
expected religious people to be relatively consistently either extrinsic
or intrinsic, but found some subjects 'provokingly inconsistent. They
persist in endorsing any or all items that to them seem favorable to

religion in any sense" (Allport & Ross, 1967, p. 437). He called these

22

persons “indiscriminately religious", and they were the most prejudiced
of all.

This partitioning of religious people into intrinsic, extrinsic and
indiscriminately religious categories and the concept of greater or
lesser degrees of the intrinsic and extrinsic religious orientations has
generated a large body of research that, in general, supports the
original ideal of Allport that one way of being religious is more healthy
than the other. Support for this statement will be presented in the
literature review in Chapter Two. Suffice it at this point to quote the
conclusion of Meadow and Kahoe (1984) after a review of many of the
empirical studies utilizing religious orientation as a variable.

The disorder-health distinction is evident among
correlates of intrinsic and extrinsic religious
orientations. In general, the intrinsic
orientation has been positively related to
variables that psychologists have deemed to
indicate positive personality attributes (internal
control, responsibility, achievement). Similarly,
it has either been independent of or negatively
related to most undesirable psychological traits
(prejudice, dogmatism, extrinsic motivation).
Likewise, extrinsic religion tends to be related
to such negative personality characteristics as
authoritarianism, dogmatism, external control, and

lower levels of education and academic aptitude.
(p. 298)

WWW

Allport himself never explicitly stated a developmental relationship
between extrinsic and intrinsic religious orientation. However, the
evolution of the I-E distinctions from the earlier ”mature-immature
religious” developmental scheme would imply that extrinsic is a less
mature and intrinsic a more mature, highly developed orientation toward

religious life. Several researchers have noted "the developmental

23

sequence implied in Allport's writings” on the I-E distinction (Hood,
1985, p. 415; cf. Brown, 1964; Kahoe & Meadow, 1981; Gorsuch & Venable,
1983; Spilka et al., 1985). Gorsuch and Venable (1983), noting that the
sequencing of I after E developmentally is ”implied in much of the I-E
literature" (p. 186), revised the Religious Orientation Scale to allow
its use with children and adolescents. According to Hood (1985), Allport
repeatedly emphasized ”intrinsic faith in a hard won process of maturity

. achieved relatively late in life, as a final overcoming . . . of
earlier selfish, extrinsic commitments" (p. 416).

In its simplest formulation, an extrinsic orientation would precede
an intrinsic orientation, with the intrinsic orientation the.
developmental achievement of a mature personality (Hood, 1985; Gorsuch &
Venable, 1983). One would then posit either a chronological age
relationship between E and I (with I increasing and E decreasing as a
function of age) or else a relationship between E and I and other
developmental processes and constructs. In this latter understanding, an
intrinsic religious faith would grow out of and be related to more
healthy psychological developmental experiences, while an extrinsic
religious orientation would be related to less successful (less healthy,
more dysfunctional) psychosocial developmental processes.

Kahoe and Meadow (1981) have proposed a model which combines the idea
of religious orientation as discussed here with some related
conceptualizations that have been prOposed as alternatives or
modifications to the intrinsic-extrinsic religious orientations (Allen &
Spilka, 1967; Fleck, 1976; Batson, 1976). These various

conceptualizations have been combined by Kahoe and Meadow into a

24

two-dimensional developmental model. Their model is reproduced in Figure
1.

This model is actually a revision of a two-dimensional model of
religious faith proposed by Brown (1964). One dimension of the Brown
model used the intrinsic and extrinsic religious orientations of Allport,
whereas the other dimension contrasted an outward orientation to
institutional religion and an inward orientation to individual conscience
and judgment. In their discussion of this revised model, Kahoeand
Meadow (1981) attempt to show how persons would be characterized at any
particular point on the model in terms of their religious development.
They hypothesize that all persons move through (or have the potential to
move through) the two dimensional space of the model in a specific
order: extrinsic religiousness ---> observance religiousness --->
intrinsic religiousness ---> autonomous religiousness.

Kahoe and Meadow believe, as did Allport, that while a mature
religion may become intrinsically motivated by its own inner drives,
religious faith and practices are extrinsic in their origins. They
recall Allport's (1950) characterization of mature religion as
"derivative yet dynamic" (p. 71). Religion in the life of the individual
and for humankind is derived from basic fears and anxieties, whether
physical, psychological, social or existential. Religion for the child
(or the adult who has never grown up), in addition to its habitual
elements of belief and practice, is essentially egocentric and self
serving.

Because of involvement with an institution and its religious system,

most religious people will develop an observance (or institutional,

25

 

 

ABEQDQE!
(individualism,
quest)
6 \
\
Extrinsic , \ Intrinsic
l
(defensive, \ (committed,
self-serving) \ / self-giving)
.\ //
\ __ ,/
Qbaemanse
(institutional,
consensual)

 

Figure l: Hypothesized process of development within
religious orientation dimensions (source:
Kahoe and Meadow, 1981, p. 10)

*
----- > indicates the direction of movement.

 

 

consensual) orientation. At this level, involvement with the religious
institution and its activities and identification with this social group
and its defining religious system become important. Loyalty to the
institution or group is fostered by the individual's needs being met ‘
within that context (i.e., loneliness, anxiety about death, etc.). Kahoe
and Meadow distinguish the social or affiliative activities of observance

religion, which are closer to the extrinsic orientation, from the

conformity to the doctrinal or belief system, which is more closely

26

aligned with the intrinsic orientation. The social aspect (at least as an
end in itself) they regard as more primitive psychologically. The
relative strength of either of these aspects "may be influenced by both
the psychological characteristics of the individual and the emphases of
the religious institution. It may also be affected by the individual's
stage of religious development" (Meadow 6 Kahoe, 1981, p. 10). The shift
to an observance religion means that the person must move away from the
purely self-serving extrinsic orientation.

Some people, of course, will not continue through the developmental
paradigm, but for those who do, there is a further turning away from
self, as beliefs and values become more internalized and lived out. All
of the higher religions advocate that their adherents give themselves in
devotion to religious ideals and causes. However, many factors, according
to Kahoe and Meadow, mitigate against a person developing an intrinsic
orientation. Some religious groups or leaders may not encourage this
growth process, while ”individual differences“ (p. 11) may make it less
likely that a particular person becomes intrinsically oriented. Kahoe
and Meadow (p. 11) state, ”A general disposition toward intrinsic
motivation may be a precondition to intrinsic religiousness." The
implication is that religious orientation is strongly related to other
personality or psychosocial developmental variables.

The final step in this developmental scheme is autonomous
religiousness, an orientation that few people reach, in part because this
orientation tends to be antagonistic to the interests of organized
religion. This is particularly the case when the institutional religion

requires conformity to rules and regulations and emphasizes externalized

27

authority. An intense personal religious experience, greater abstract
reasoning ability, higher education and training in critical thinking and
self-reflective disciplines may be related to development of this
autonomous religiousness. Persons who need the security of and
identification with the religious group and/or its beliefs and practices,
will be unlikely to move toward this most independent, functionally
autonomous of religious orientations.

Kahoe and Meadow (1981) see religious development as a continuum with
specific ”psychometrically identifiable and 'characteristic' orientations
along the way” (p. 15). They avoid utilizing the label ”stages" because
of the implication that religious experience functions uniquely at each
stage. Though acknowledging the difficulty of depicting this, they
believe that each new level in this model encompasses all of the previous
levels, suggesting that a three-dimensional model with an upward spiral
would be better at illustrating this encompassing of earlier levels. In
this regard, this model is similar to Erikson's epigenetic theory of
psychosocial development in which higher stages build on and in some ways
recapitulate the developmental tasks of earlier stages.

Although specifically stating that they hope their model will
stimulate research, these authors do not present empirical evidence in
support of the model. They hypothesize that this development schema does
have “an intuitive validity" (p. 12) in part because of its convergence
with several other developmental theorists. Both in the original article
(1981) and in their textbook (1984), Kahoe and Meadow briefly discuss
conceptual parallels between their theory of religious development and

other theorists, including Loevinger, Rohlberg, Maslow, Erikson, and

28

Fowler. In Table 1.2, an attempt is made to diagrammatically represent
the relationships between their model of religious development and these
other developmental frameworks. Table 1.3 presents the eight stages of
Erikson's theory, the task of each stage, the virtue or strength which
results from a positive resolution of the task, and a dimension of
religious faith related to each of the eight stages. After a
consideration of some of these hypothesized relationships, Meadow and
Kahoe (1984) state, ”The developmental levels of our model almost surely
interact with a religious person's general personality development --
especially in cognitive stages, moral judgment, ego development,

motivational style and socialization." (p. 327)

Samara

The concept of religious orientation has been discussed with
attention to two separate but related emphases in the theoretical
literature. The first emphasis is the relative psychological health of
an intrinsic and an extrinsic religious orientation. The theory posits
that an intrinsic religious orientation is more healthy and an extrinsic
religious orientation less psychologically healthy. The second emphasis
in the preceding theoretical discussion was on a developmental sequencing
for the two religious orientations with an extrinsic orientation being an
earlier developmental phenomenon and an intrinsic orientation being a
later, more mature developmental phenomenon. This theoretical overview
has provided a basis for hypothesizing the relationships between
intrinsic and extrinsic religious orientations and normal (or healthy)

psychosocial developmental tasks.

table 1.2

29

kmthesized Relationship; of Models of Develmt

 

 

Ag: grikson Fowler Kohlberg oevi er Maslow Kahoeaeadow
Psychosocial Faith Moral Ego Needs Religious
Development Develomnnt Developnent Develqsnent hierarchy Development

0-2 I. Irust Urdifferen- Syebiotic Physiologi- Extrinsic

tiated Faith cal/Safety
leads

2-6 II. Autonomy 1) Intuitive- Precomen- Iswlsive Love and

projective tional level Self- lelonging
faith Protective
III. Initiative
7-12 IV. Irnntry 2. Hythic- I. heterono- Conformist Self weervance
literal m Esteem
faith nrality (other)
(pmishnent
and obedience
2. Iratrlnental
Exchange
(hedonistic)

13-20 V. Identity 3. mthetic- Conventional Self Aware Self Intrinsic
Commtional level Esteeu
faith (self)

3. Mutual
interpersonal
relations
(“good WI
“nice girl“)

20-35 VI. Intimacy 6. Individaa- 4. Social Conscien- Self-
tive systonand tious Actualiza-
faith cormcience tion

(law and
order)
35-60 VII. Gmera- 5. Conjmc- Post cemen- lndividu- Autonwy
tivity tive timal alistic
faith principled

level

5. Social Autonomus Nets Needs
Contract]
individual
ridits

600 VIII. Wisdon 6. lhiversal- 6. Universal Integrated

izing faith ethical
principles!
loyalty to
being

 

'sourm: m s rm (19w. pp. 55, 325; and fowler noon, pp. 52, m.

30

Table 1.3
Erikson's Psychosocial Stages and Faith Dimensions

 

Trust or Virtue or
was Cosme; mm Fem
1. Trust or Trust own per- Hope Trust
mistrust ceptions and

other goodness

2. Autonomy or Gain impulse Will/power Courage
shame/doubt control
3. Initiative Self-guidance Purpose/ Obedience
or guilt and self— direction
punishment
4. Industry or Acquire rudi- Competence, Assent
inferiority ments of ' skill
technology
5. Identity or Integrate all Devotion, Identity
role con- one's personal fidelity
fusion identifica-

tions into one

6. Intimacy or Make commit- Love, Self-
isolation ments, accept bonding surrender
obligations
7. Generativity Become teacher Care, Uncondi-
or stagnation and guide of production tional
next generation caring
8. Integrity or Acquire post- Wisdom Uncondi-
despair narcissistic renunciation tional
love of the accept-
human ego ance

 

Sources: Aden (1976), Erikson (1963, 1968), and Meadow & Kahoe (1984)

 

 

St e t o e d ese c m
The following are the major hypotheses to be tested in this study. At

present, these hypotheses will be stated in a broad, general way, to give

31

the reader an idea of the direction the research will take and to guide

the reader through the material in the early chapters of this

dissertation. In Chapter 3 the research hypotheses will be restated in a

more detailed and testable form.

I.

II.

III.

IV.

The Intrinsic Religious Orientation will be positively related to
an adult's mastery of the Eriksonian psychosocial developmental
tasks. The Extrinsic Religious Orientation will be negatively
related to an adult's mastery of the Eriksonian psychosocial
developmental tasks.

The Intrinsic Religious Orientation will be positively correlated
with an adult's resolution of the individual Eriksonian
psychosocial developmental tasks. The Extrinsic Religious
orientation will be negatively correlated to an adult's
resolution of the Eriksonian psychosocial developmental tasks.
Persons who are Intrinsically Religious will master in order more
of the Eriksonian psychosocial developmental tasks than persons
who are Extrinsically Religious or Indiscriminately Religious.
Persons who are Intrinsically Religious will demonstrate better
resolution of each of the individual Eriksonian psychosocial
developmental tasks than persons who are Extrinsically Religious

or Indiscriminately Religious.

W

In Chapter 2, there is a review of the empirical literature involving

religious orientations with a focus on studies that utilized the

Religious Orientation Scales and other measures of psychological health

32

or development. Because of its bearing on this study, the research will
be summarized that relates to correlates of I and E, as well as
correlates using a four-fold typology based on scores on the two scales.

Chapter 3 will present the design of this study including sampling
procedures and a description of the demographic sheet and the two
instruments to be used (the Religious Orientation Scales and the
Assessment of Adult Adjustment Patterns). The hypotheses will be
restated in testable form and a specification of the data analysis
procedures will be given.

In Chapter 4, the results of the analysis of the data will be
presented. Each hypothesis will be stated with a statement of acceptance
or rejection of the null hypothesis.

Chapter 5 will discuss the results of this study, drawing conclusions

and implications for future research.

CHAPTER II

LITERATURE REVIEW

Introduction
This chapter will review the empirical studies utilizing the

Religious Orientation Scale (ROS). It will begin with a discussion of
the orthogonal nature of the two subscales I and E and two different
analytical procedures that have been used by researchers: correlational
studies and typology studies. Research will then be summarized under two
broad headings relating religious orientations to psychological health

and to developmental phenomena.

e1 io s ie t ' O t 0 al

Allport's original theoretical conceptions of intrinsic and extrinsic
religious orientations had placed these two constructs as opposite ends
of a continuum (1950, 1954, 1959, 1963). The early empirical studies
utilizing this conceptualization initially attempted to measure these
unidimensionally (Wilson, 1960; Feagin, 1964) to differentiate between
persons who were “religious.” This attempt was based on Allport‘s (1967)
assertion, "To know a person is in some sense 'religious' is not as
important as to know the role religions plays in the economy of his life"

(p. 442).

33

34

The earliest research, however, casts doubt on the construct validity
of the bipolar conception of I and E. Feagin (1964), in a factor
analysis of his religious orientation items, reported loadings on two
separate, orthogonal scales. He concluded that Allport's hypothesis of
different ways of being religious was ”substantiated" but that the I-E
orientations were separate dimensions, not unidimensional, and that the
extrinsic subscale was somewhat more useful than the intrinsic scale at
making the differentiation. Virtually all of the studies since (cf.
Allport & Ross, 1967; Hunt 6 King, 1971; Hood, 1970) have corroborated
this finding that I and E are separate orientations. Interestingly,
Allport himself began to notice a subgroup of religious people he called
"muddleheads" because they "refuse to conform to our neat religious
logic” (1966, p. 6). Despite Allport's attempt to construct a scale to
represent polar opposites, these persons agreed with items on both ends
of the continuum.

It had been expected that I and E would be strongly negatively
correlated and, therefore, bipolar; but, in fact, Allport and Ross (1967)
found I and E only somewhat negatively correlated. Donahue (1985a), in a
meta-analysis of reported correlations between I and E, calculated an
insignificant correlated of -.06 (although the range is from .24 to
-.58). Further, Donahue presented data suggestive that the population
value of the E-I correlation is close to zero (p. 404). Apparently when
the sample consists of evangelical or very conservative religious persons
the correlation becomes strongly negative (cf. Dodrill et al., 1973;
Bolt, 1977; Shoemaker & Bolt, 1977) but when the sample is more broad

based the observed I-E correlation is low (cf. Hoge & Carroll, 1973;

Patrick, 1979).

35

If I and E are uncorrelated in the general population, then I and E are
representing orthogonal constructs.

In response to his empirical findings, Allport modified his bipolar
approach making it a fourfold typology (Allport & Ross, 1967; Hood,
1970). In this approach, I and E are treated as two separate unipolar
dimensions of religious orientation and can be considered simultaneously
to generate a fourfold classification: Intrinsics (high I, low E),
Extrinsics (low I, high E), Indiscriminately Religious (high I, high E),
and Indiscriminately Anti- religious (henceforth "Non-religious”) (low I,

low E). Diagrammatically, this typology is:

l I
i
V Extrinsic , Indiscriminately

l

| Religious

l

i

i

i

l

I

I

Score on E ------------ . -----------------

a

I

l

l

Non-religious : Intrinsic

I

i

i

i

l

' >

 

 

Score on I

36

Hood (1970) proposed forming the above categories on the basis of
median splits from the research sample. This is the procedure most
researches have used who have incorporated this typology into this
studies. As will be discussed, this distinction has produced some
interesting results, although for this researcher the category that has
been labeled ”indiscriminately anti-religious” is both conceptually and
empirically mislabeled. Better and probably less effectively laden is
”non-religious" or perhaps ”less-interested-in-religion.” Unless the
sample is very broad, it is inaccurate to label those persons who score
relatively lower on both I and E as "anti-religious.” Bergin (1987)
noted that the fourfold typology made no sense on his sample of very
religiously active students who scored very high on I and very low on E.
Though much of the research that is summarized here will report a
"non-religious” category, this caution must be kept in mind when
interpreting the findings.

Research using the Religious Orientation Scales has provided those
interested in the psychology of religion with a useful tool for
discriminating between differing ways of being religious. Some
researchers have chosen to subdivided their sample into two or more of
the subgroups of the Hoochllport fourfold typology. Others have used
the I and E subscales and correlated these measures with other measures.

Still other studies have utilized the ROS in both ways.

37

Walsh
Wm

Two recent correlational studies illustrate the value of separately
relating the I and E scales to other psychological constructs and also
indicate some of the weaknesses of much of the I-E research literature.
Baker and Gorsuch (1982) considered the question of the relationship
between religious orientation and trait anxiety as measured by the IPAT
Anxiety Scale and five of its subscales. They cited confusion in the
literature over the relationship between religion and anxiety, with some
studies showing a positive relationship between religious persons and
anxiety and other studies showing no relationship. They argued that the
discriminant power of the ROS and previous research would lead to
prediction of significant differences between the two religious
orientations and trait anxiety. Their results provided strong support
for their prediction. On the total scale trait anxiety was significantly
negatively correlated with intrinsic (-.33) and significantly positively
correlated with extrinsic (.35). Three of the five subscales repeated
these statistically significant findings, while the other two had trends
in the same directions. A measure of state anxiety showed no significant
results. Baker and Gorsuch argued that trait anxiety manifesting itself
in suspiciousness of others, frustration intolerance, guilt and a
weakened ego unable to balance emotional forces within oneself is an
indication of less psychological adjustment for those with a more
extrinsic orientation. Intrinsicness, they argued, is apparently

associated with greater ego strength, less insecurity, and less anxiety.

38

There are several weaknesses in the report of this study. There is
no clear description of the sample. Subjects are described as "taken
from a religious wilderness camping organization . . . which offers
stressful backpacking trips” (p. 120). Age, denominational affiliation,
gender, and any other relevant demographic data are omitted, despite the
fact that these may have an influence on the results. A more serious
weakness is that Baker and Gorsuch do not report the scores on the I and
E subscales. Though they indicate the two scales are significantly
negatively correlated (-.32), they do not provide mean, median, or
standard deviation for either scale. This omission (common to many of
the published studies using ROS) make comparisons across studies much
less possible. Given the strong negative correlation of I and E they
report, it is likely that I was fairly high and E fairly low, meaning
that the correlations performed were with a very range-restricted
measure. This latter criticism is also true for the next study.

Bergin et a1. (1987) also approached their study against the backdrop
of previous research suggestions that, in general, I is related to better
psychological functioning and E is negatively related to better
psychological functioning. They studied undergraduate psychology
students at a Mormon college giving them, in addition to the ROS, the
California Personality Inventory, the Beck Depression Inventory, the
Ellis Irrational Beliefs Test, the Manifest Anxiety Scale, and the Self
Control Schedule. Different classes of students were given somewhat
different combinations of these measures, so that sample sizes vary
across the instruments. A detailed discussion of all the correlational

' results is beyond the scope of this summary. However, in general, I was

39

negatively correlated with anxiety and positively correlated with self
control and such other characteristics as sense of well being,
responsibility, tolerance, and intellectual efficiency (subscales of the
CPI). E was positively correlated with anxiety and negatively correlated
with psychologically more healthy traits such as sociability, sense of
well being, responsibility, tolerance, achievement, and intellectual
efficiency. A striking result as one examines the non-significant trends
is that most of them consistently show I tending to correlate with
positive personality traits, whereas E shows the opposite tendency.

This study has several weaknesses that temper the generalizability of
the results. For several of the subgroups, a small sample size of 32 was
used. In addition, the use of undergraduates and the sampling of a very
religiously committed and conservative group moderate any broad
conclusions drawn from this study. Bergin et a1. do report the mean I
(38) and E (24) scores, but without standard deviations on either scale.
These means are very high for I and very low for E, indicating a strong
negative correlation between them and a very restricted range of scores
on the ROS. In discussing this sample, Bergin et al. are careful to
apply the results only to an intrinsically religious orientation, but
their assumption "that the magnitude of the reported correlations was
underestimated due to the restricted range of scores on the I and E
subscales” (p. 199), while possible, is one that requires empirical
validation. Bergin et a1. had originally intended to also classify their
samples using the fourfold typology. However, using the midpoints of the
two subscales as the dividing point, which is the procedure Donahue

(1985a) recommends, yielded 98.6% of their sample being classified as

40

intrinsic. Further, because their subjects' scores were so skewed on
both I and E, the use of the observed medians of this sample to form the
fourfold classification would have tended to “grossly distort the meaning
of the classifications“ (p. 199). This writer has argued above that the
labels of the four quadrants, especially "non-religious," are applicable
only when the sample is very broad based.

A detailed summary of all of the other studies which considered
correlations between I and E as separate, unipolar dimensions is beyond
the scope of this literature review. However, some generalizations can
be made that are relevant to the present study. When the relationship
between other measures of religiousness and I and E are made, a
consistent pattern is observed. Whether the variable is self-reported
importance of religion (cf. Batson, 1976) or religious beliefs,
practices, and experiences (cf. Dodrlll et al., 1978; Spilka et al.,
1968; Hood, 1978), I is significantly more positively correlated than is
E. This is what would be expected on the basis of the conceptual and
empirical definitions of E as a type of religious commitment that treats
religion as only one of the many influences in life and seeks religious
identification as social support or for its own comfort.

Allport's original research interest was the construct of prejudice
and its relation to religious people. A number of studies have
considered this relationship. Donahue (1985a), in summarizing all of the
available studies on prejudice, found mean correlations with I of -.05
and with E of .34 (p. 406). For the extrinsic orientation, this is
supportive of Allport's (1966) conceptualization that E would be related

to prejudice. But I is essentially uncorrelated with measures of

41

prejudice, while Allport's writings expected a negative correlation. A
similar prediction based on Allport's writings would be made for
dogmatism and its relationship to I and E. The implication would be that
I would be openminded and E relative more closeminded, dogmatic, and
authoritarian. As with prejudice research, Donahue (1985a) has computed
mean correlations across five studies utilizing a measure of dogmatism or
authoritarianism as the dependent variable and found I essentially
uncorrelated (mean correlation - .06) and E rather strongly positively
correlated (.36). These studies represent the use of ROS and a variety
of measure of prejudice and dogmatism, and their generalizations appear
rather clearly established in the literature: E is related to being more
rigid, closeminded, suspicious, and fearful of other people, more
defensive and prejudiced; whereas I is not related to such variables in
any meaningful way. Put another way, E is a less socially,
psychologically desirable trait, while no specific conclusions can be
drawn about I on the basis of these groups of studies. When this review
turns to a consideration of studies utilizing the fourfold typology, the
relationship between I and prejudice and dogmatism will be clarified
further.

If an intrinsic religious orientation is one that serves as a
pervasive motivator in life, it would be expected that I would be able to
provide assistance in persons dealing with broad existential questions
about life and death, while E would not serve in this way. There is
considerable support for these predictions across several different ROS
studies. I has been shown to be related to a sense of meaning and

purpose in life (Bolt, 1975; Crandall & Rassmussen, 1975) as measured by

42

Crumbaugh's Purpose in Life test. The facing of death raises the most
profound questions for human beings, not only about an after life, but
about reasons for existence, loneliness, and generalized anxiety about
life ending. Relating ROS scores to a variety of fear of death measures
has been another fruitful line of study for several researchers (cf.
Spilka et al., 1968; Bolt, 1977; Patrick, 1979). The pattern emerging
from these studies indicates that E is positively correlated to fear of
death, anxiety about life ending, and the losses associated with dying,
while I appears to be negatively correlated with these same constructs.
In one recent study involving terminally ill cancer patients, Acklin et
a1. (1983) found a positive relationship between intrinsic religious
orientation and life meaning attribution and that I was associated with
lower levels of despair, social isolation, and anger-hostility.

/ Other sets of constructs that have received considerable attention in
the ROS literature are those of internal-external locus of control and
intrinsic-extrinsic motivations (cf. Strickland & Schaffer, 1971; Kahoe,
1974, 1975, 1985; Morris & Hood, 1981). The positive relationships
between I and internal locus of control and intrinsic motivation and
between E and extrinsic motivation are so strong that Kahoe (1975, 1985)
has argued that the religious orientations may have pervasive personality
bases that emerge in the course of psychosocial development.

Worth mentioning as questions not satisfactorily addressed by the
empirical research is the relationship of I and E to gender and social
desirability. Despite the fact that gender differences in religiousness
are frequently reported in the literature, only a very few published

studies report I and E scores broken down by gender (cf. Strickland &

43

Schaffer, 1971; Alker & Gavin, 1978). There is some evidence that women
tend to score higher on I than men, with no gender differences on E
(Donahue, 1985a). With regard to social desirability, the evidence is
slim but consistent. Though Donahue makes a case that social
desirability is not related to religious orientation, I appears to be
somewhat positively related to measures of social desirability (Batson et
al., 1978; Watson et al., 1984). The confounding effects, if any, of the
relationships of I to gender and social desirability have not been

explored in the literature.

RO nd t e ou d

As has been discussed above, when I and E are considered
simultaneously, a fourfold classification system is formed of extrinsics,
intrinsics, indiscriminately religious, and non-religious persons. The
use of this typology by researchers has increased the explanatory power
of the construct of religious orientation. When considering I and E
simultaneously, the two categories of intrinsic and extrinsic are
relatively pure, whereas the other two quadrants are mixtures of these
two variables. This approach, though not without its conceptual and
methodological problems as discussed elsewhere, does clarify some of the
results which are obtained from the correlational research involving I
and E. Intrinsic religiousness, it will be shown, may fail to gozrglgte
with some non-religious variables because it confounds two distinct
types. Persons who score high on I may be either pure intrinsics (low in

E) or they may be indiscriminately religious (high on E).

44

Allport and Ross (1967) were the first to propose this typology and
apply it to research in prejudice. The scoring procedure they used only
produced extrinsic, intrinsic, and indiscriminately pro-religious
groups. Their data show highly significant differences, however, among
these three groups with the intrinsics less prejudiced than the
extrinsics who, in turn, are less prejudiced than the indiscriminates.
This pattern of relationships will be seen in other research involving
the ROS and many other psychological constructs.

In passing, it may be interesting to note the pattern that emerges
when the dependent variable is some other religious measure and the ROS
is used to split respondents into the four groups. When this is done,
the result is a main effect for I, no effect for E, and no interaction
(Cf. Hood, 1970; Tate & Miller, 1973; Dodrill et al., 1973). The results
of these studies show that whether the dependent variable is religious
experiences, beliefs, practices, or values, the intrinsics and the
indiscriminates are equivalent and both score significantly higher than
the extrinsics or the non-religious group.

The pattern is different, however, when the dependent variable is
non-religious and it resembles that first observed by Allport and Ross
(1967). A study by Thompson (1974) is illustrative of the discriminant
power of this typology and of the general pattern of relationship that
emerges when the fourfold classification is used. Thompson surveyed a
large (n-532) sample of high school students and their parents, giving
them the ROS and Rokeach's Dogmatism Scale. In three separate ANOVAs for
adolescents, their mothers and fathers, the same pattern emerged. Those

who were classified as indiscriminately religious were the most dogmatic,

45

followed by the extrinsics. The least dogmatic were the intrinsics and
non-religious (Thompson called this group ”indiscriminately
anti-religious"), and there was no differences between these two groups.
The pattern that emerges is intrinsic - non-religious < extrinsic
< indiscriminately pro-religious. Joe et a1. (1977) report a similar
pattern. A one way ANOVA showed the intrinsics significantly less likely
to devalue a rape victim than the other three groups. The ordering of
mean scores was intrinsic < non-religious < extrinsic - indiscriminates.
Rice (1971) designed a study specifically to consider the
relationship of religious orientation to mental health, following
Allport's hypothesis that an intrinsic orientation is facilitative of
psychological health and an extrinsic orientation is not. He
administered the ROS, Barron's ego-strength scale, and a measure of
psychosical adjustment to a sample of 151 men from six religious bodies
(churches). Using the fourfold typology, Rice hypothesized that the
ordering of mental health (as defined by ego strength and adjustment)
would be intrinsic > extrinsic > indiscriminately pro-religious -
indiscriminately anti- religious. His initial analysis of the data
showed the relationship to be indiscriminately anti-religious the
healthiest with the intrinsics similar although scoring lower on most of
the dependent variables and the extrinsics and indiscriminately
pro-religious least healthy. Rice questioned the utility of the ROS for
classifying into the focus categories as Allport labeled them,
particularly with his sample of religiously affiliated persons. He chose
to reclassify his subjects, eliminating the indiscriminately

anti-religious category. With this reclassification Rice found the

46

intrinsics the most psychologically healthy on both measures and the
order was intrinsic > extrinsic Z indiscriminately pro-religious.

Some studies have not chosen to utilize all four of the types in
their comparison. For example, McClain (1978) compared intrinsically
religious students with non-religious students and found the intrinsic
respondents higher on factors of personal and social adequacy, self
control, and stereotyped femininity and lower on egocentric sexuality and
restlessness. Alker and Gavin (1978) found intrinsics healthier than
extrinsics on measures of psychological well being. Wiebe and Fleck
(1980) found that intrinsic personality profiles on the 16 PF were
significantly different than those of extrinsic and non-religious
subjects which were similar to each other. Intrinsic subjects tended to
be more conscientious, responsible, sensitive, dependent, empathic, and
open to their emotions. They also tended to be more conservative and
traditional which can be indicative of rigidity and neurosis. Extrinsic
and non-religious subjects tended to reflect greater self-indulgence,
undependabilty, and skepticism, but also tended to be more flexible,
self-reliant, innovative, and analytic. Extrinsics and non-religious
subjects tended to be less rigid. This study tends to moderate the
results of some of the other studies summarized here, as Wiebe and Fleck
comment that “the question of the relationship between pathology and
religiosity remains difficult to solve” (p. 187).

Two studies attempting to utilize the fourfold typology reported that
the typology was not able to distinguish respondents on the dependent
variables being studied. Gibbs and Achterberg-Lewlis (1978) found "no

statistical relationships of import obtained for the religious

47

orientation variable” (p. 566) when studying fear of death in terminally
ill cancer patients, though other measures of religiousness were related
to positive coping with terminal illness. The lack of significance
within the fourfold typology is probably explained in part because 94% of
their sample was high on I (50% intrinsic, 44% indiscriminately
religious) and only 6% were extrinsics. This skewed sample distribution
supports the perspective argued elsewhere in this proposal that the
fourfold typology is perhaps most appropriate when scores on I and E are
more broadly distributed. Kraft et al. (1986) also report that the
fourfold typology did not differentiate levels of assertiveness. One
possible explanation for this is that the median split reported for the I
subscale was 22, a value that is lower than the theoretical mean score
for I (27) and is much lower than the median reported on all other
studies examined by this writer. This study by Kraft et al. of
undergraduate psychology students may be in general less intrinsically
religious than most studies (median split on E was 31, which compares
favorably to other studies), so that the fourfold typology may be skewed
toward the less religious end. This analysis is speculative at this
point, though it is consistent with the analysis of previous studies

using the fourfold typology.

Samar:

In general, as one considers the body of research with R08, a clear
pattern emerges. Across most studies, I is related to better
psychological health than E. Whether studies use correlations with I-E

scores or the fourfold typology, this generalization holds true. There

48

are some indications that I and E may be related to other pervasive
personality variables, which may be a function of developmental
processes. The explanatory power of the ROS argue for its utility in
continuing research with religious people. This study sought to build on
this empirical literature, for example, by utilizing a measure of
psychological health, by sampling to get a broad based adult subject pool
on I and E, by collecting careful demographic information, and by

utilizing a measure that includes socially desirable responding.

O tt V

As discussed in the theoretical overview in Chapter I, a
developmental sequencing of intrinsic after extrinsic, with an intrinsic
religious orientation emerging out of an earlier extrinsic orientation
has been hypothesized (Hood, 1985). Gorsuch and Venable (1983) comment
that this developmental sequence of I after E is "implied in much of the
I-E literature” (p. 186). This sequencing was formalized in a model of
religious orientation development proposed by Kahoe and Meadow (1981).
Though this hypothesized relationship has been a part of theoretical
conceptualizations of religious orientation since Allport's seminal
discussions of the mature religious sentiment (1950) and the
intrinsic/extrinsic religious orientations (1966, 1967), few empirical
studies have directly addressed this issue.

Two studies (Chirban, 1980; Ernsberger & Manaster, 1981) have
considered intrinsic and extrinsic religious orientation and their
relations to two other sets of developmental constructs: faith

developmental stage and moral development. These studies provide

49

interesting suggestive support for the idea that E and I are related to
other aspects of personal development, which is the thrust of this
research project.

Chirban (1980) studied the relationship between intrinsic and
extrinsic religious motivation and the stages of Fowler's (1981) faith
development paradigm. Fowler has proposed a structural developmental
theory of faith development which explicitly parallels the work of
Erikson in psychosocial development, Kohlberg in moral development, and
Piaget in cognitive development. Fowler proposes that there is a
universal, invariant, hierarchical sequence of stages through which
persons develop throughout the life cycle as they seek to structure their
lives in relationship to the ultimately meaningful (that is, ”faith”).
The six stages and their conceptual relationship to other developmental
theories can be seen in Table 1.2 in Chapter 1. Assignment of a person
to a particular stage is made on the basis of a very lengthy structured
research interview called the Faith Development Interview (FDI).

Chirban discerned in Fowler's work (and in the developmental theories
which it parallels) a shift from a more extrinsic religiousness
(security-seeking, protective, affiliative, utilitarian, defensive,
dependent) to a more and more intrinsic religiousness (inner directed,
personal, committed, autonomous). After reviewing the conceptual idea of
I and E and some of the empirical literature measuring I and E through
the Religious Orientation Scale, Chirban concluded that an alternative to
the ROS subscales would be worth developing. He developed a scoring
procedure for rating verbal responses from an interview as extrinsic or

intrinsic in their religious orientation and trained raters in the

50

scoring procedure. The raters then analyzed previously completed Faith
Developmental Interviews for I and E responses. He found that I
positively correlated with Faith Stage and that E negatively correlated
with Faith Stage. Intrinsic religious responses were much more
characteristic in the three highest Fowler stages which are the stages
which correspond to adulthood. Chirban also found both I and E present
at all stages and comments that these two constructs are ”structurally
different at different stages" (p. 80). An obvious weakness of this
study as it relates to the present study (or any other research utilizing
religious orientation) is that Chirban's operationalization of I and E
may or may not have construct and criterion validity with I and E as
measured by the ROS. Nevertheless, this study is suggestive of some
developmental correlates for intrinsic and extrinsic religiousness.

In a study which primarily focused on Kohlberg's levels of moral
development and effect of a particular church's denominational moral
teaching, Ernsberger and Manaster (1981) include some evidence that bears
on the issue of I and E as developmental phenomena. In a theologically
moderate church, intrinsic religious orientation was found to
significantly correlate with the presence of the higher levels of moral
reasoning. However, in two theologically conservative churches, this
relationship was not observed. The author's explanation for this
difference was that moral reasoning at the conventional level was
strongly socially reinforced through the teaching of these religiously
conservative churches.

Another way to study I and E in a developmental sequence is to

hypothesize changes in relative strength of I and E as a function of

51

chronological age. A study by Thompson (1974) offered some support to
this idea. His study was focused on the constructs of dogmatism and

openmindedness. He sampled a large number of Catholic adolescents and
their parents and reports on median scores on I and E for these groups,

as follows:

Intrinsic Extrinsis
Adolescents 29.5 36.7
Parents 32.2 29.5

He reports the differences on E are statistically significant. Although
other interpretations of these data are possible, he suggests that E may
decrease as a function of age.

In order to test the hypothesis of a change in I and E as a function
of age, Gorsuch and Venable (1983) developed an ”age-universal" I-E scale
by taking the Religious Orientation Scale (ROS) and re-writing the items
in simpler language. They studied the reliability and validity of the
use of this instrument with children in fifth grade and older. After
demonstrating its strong psychometric properties which are similar to
those of the original ROS, Venable (1984) administered the I-E scale to
groups of children in fifth, seventh, ninth, and twelfth grades,
expecting to find that relative score on I increased and E decreased as
grade level increased. His results did not support this hypothesis.
Rather the mean scores for I and E were not statistically different
across the ages. Spilka et a1. (1985), in commenting on these results

and the proposal that I and E are developmental phenomena, write that "a

52

social psychological model may be preferable to a cognitive developmental
model” (p. 72).

This comment is consistent with the suggestion made by Hunt and King
(1971) that these two religious orientations have pervasive personality
bases. Evidence supporting this idea comes from a study of job
motivation by Kahoe (1974). In his study, Kahoe has shown that intrinsic
religious orientation is strongly correlated with intrinsic job
motivation, while extrinsic religious orientation correlates with
extrinsic job motivation. In discussing these results, Kahoe (1985)
suggests that "intrinsicness and extrinsicness probably have different
sources within the developing personality” (p. 411).

These studies and their conclusions suggest a different way of
approaching the issue of religious orientation as a developmental
phenomenon. Rather than linking E and I to specific chronological age,
these results would imply that extrinsic and intrinsic religious
orientations may develop separately in the personality. Instead of an E
then I sequence, in which these two orientations are located on a
unidimensional developmental continuum, it may be that different
personality and developmental processes within the individual produce
either a more extrinsic or intrinsic religious orientation. This would
support the idea developed elsewhere in this chapter that separate
personality correlates of I and E may be expected, as well as expecting
differences in subjects classified as intrinsic, extrinsic, and
indiscriminately religious on measures of other personality or

developmental variables.

53

Emu

The paucity of studies directly relating to religious orientation and
developmental measures as dependent variables, despite the presence of
theoretical suggestions of a relationship, make the study proposed here
an important one in examining the relationship between religious
orientation and psychosocial development. In addition, further research
is warranted to understand the relationship of religous orientation to

other measures of psychological health.

CHAPTER III
METHODOLOGY

In this chapter, the process for obtaining a sample of subjects is
described, and the instruments used in the study are discussed. The
design of the study and the method of analysis is presented. The

hypotheses are restated in testable form.

C t e

The population under consideration for this study is adults (over 18)
who are church affiliated. Several churches in the greater Lansing area
agreed to allow their members to be approached to participate in this
study. The ministers' support of this research was, in each case,
confirmed by a vote of an elected lay leadership committee from the
church. It is likely that this official "endorsement" of participation
increased the response rate one might otherwise have expected from an
anonymous, lengthy, mailed survey. The churches that agreed to
participate are: First Congregational Church of St. Johns; Christ
Episcopal Church of Owosso; Plymouth Church of Lansing; Congregational
Christian Church of Maple Rapids; University Baptist Church, East

Lansing; and River Terrace Christian Reformed Church, East Lansing.

54

55

The researcher was allowed to utilize each church's mailing list from
which the sample was obtained. A random sampling technique identified
the persons from the mailing lists to be approached for participation in
this research. It is expected that a sample obtained in this way will be
representative of the church's population (membership).

A mailing was done to each person who had been randomly identified
from each participating church's mailing list. A packet of materials was
mailed containing the following items:

1. Cover letter explaining the research and
enlisting participation of the recipient,
signed by a church official as well as this
researcher. The cover letter served to inform
respondents of the nature of the research and
their participation. It indicated that by
filling out the questionnaires and returning
them, the respondent was giving consent to
participate.

2. Demographic information sheet.

3. Religious Orientation Scales.

4. Assessment of Adult Adjustment Patterns.

5. Stamped, addressed envelope for returning the
materials.

Copies of the materials are in Appendix A.

The recipients of these packets were asked to complete all of the
materials and return them in the envelope provided. Because of the
length of the instruments, participants were instructed that they could
complete the forms in a couple of sittings. It was anticipated that
approximately an hour to an hour and a half would be needed by the
average person to complete these materials. Participants who chose not

to fill out the instruments were asked to mail the blank materials back,

56

so that they could be reused. Participants were assured that they were
free to not participate.

The anonymous nature of the responses to the questionnaires was
stressed in the cover letter. To ensure anonymity, no list of names of
people who were sent materials was kept by the researcher, nor were the
forms coded in any way. All respondents received the exact same
materials. Respondents were specifically asked not to put their names on
any of the questionnaires or answer sheets when they were returned to the
researcher.

Publicity concerning the on-going research was handled through
announcements in the church newsletter, the Sunday bulletin, and the
announcement section of the Sunday worship hour. These announcements
encouraged participation, mentioned that the researcher had been
supported by an official board/committee of the church and the minister,
and stressed the anonymous nature of responses to the questionnaires.

Two weeks after the original mailing of materials, a follow-up post card
was sent to all persons who received the initial mailing of
questionnaires. Because the researcher had no way to know who had and
who had not returned the questionnaires, the follow-up post card served
two purposes: (a) to thank people who had returned their questionnaires,
and (b) to encourage people who had not responded to do so as soon as
possible. The researcher had prepared a second follow-up letter; but,
due to the high response rate, this second reminder was not mailed.

The initial mailing of material went to 373 persons, of which 184
completed all of the items and returned the questionnaires to this

researcher. This represents a response rate of over 49%--a very high

57

return rate for this type of survey, probably facilitated by the
publicity about the research in each of the participating churches. The
original mailing went to 206 women (55%) and 166 men (45%). Of the
completed and returned materials, 118 (64%) were from women and 66 (36%)
were from men. Women were more likely to return these materials. One
can only speculate about the reasons for this difference: a greater
interest in religion, a greater willingness to take the time to
participate in this research or more available time to complete the
materials, or, perhaps, a greater willingness to be self-disclosing about
-a variety of personal issues. The sample was 99.5% Caucasian. One
subject marked ”other" for race.

The mean age of the sample obtained was 49.8 years, with a range from
24 to 86 years. The median age for the sample was 45 years. The sample
represents a broad distribution of ages, as can be seen in Table 3.1

Table 3.2 summarizes the highest educational level completed by the
subjects, which ranged from only completing grade school (1%) through
doctoral-level education (7%).

Several questions were asked about the subjects' religiousness,
including personal theological orientation, degree of interest in
religion, self-reported religiousness, and frequency of church
attendance. In general, the subjects' responses indicate a fairly
religious sample as was expected, given the population from which the
potential subjects were drawn. On a seven-point Likert scale, on the
question ”How religious are you?” in which '1" was "not religious," ”7"
was ”very religious" and '4” was a midpoint between those two

self-ratings, the mean, median, and mode were all ”5." More than 92%

58

Table 3.1
Age of Subjects

 

 

 

Ass Ems! c an e
24-29 12 7
30-39 47 26
40-49 45 24
50-59 23 13
60-69 31 17
70-79 21 11
80-86 5 . 3
Table 3.2

Educational Level of Subjects

 

Laid PM: W
Grade School 2 . 1
High School 32 17
Some College 57 31
Bachelor's Degree 51 28
Master's Degree 29 16
Doctorate/Professional 13 7

Degree

 

 

(170) of the subjects described themselves as "moderately" or "very"
interested in religion. The subjects adhered to a wide variety of

personal beliefs and religious practices from fundamentalist to liberal,

59

as can be seen in Table 3.3, and were diverse in their report of the

frequency of attending a church worship service, as seen in Table 3.4.

Table 3.3
Personal Theological Orientation of Subjects

 

 

 

e0 0 ic ta ugggg; Pgrgent of Samplg

Fundamentalist 7 4
Evangelical 18 10
Conservative 46 25
Moderate 73 40
Liberal 28 15
Other 12 6
Table 3.4

Frequency of Self-Reported Attendance at a Church Worship Service

 

e e c Ngmbg; er Sam le
Rarely 7 a
Infrequently 8 4
Occasionally 13 7
Once a Month 20 11
Two or Three Times a 47 26
Month
Once a Week 71 39

More than Once a Week 18 10

 

 

60

Three of the questions on the demographic sheet were utilized in a
variety of post hoc analyses, because of interest in their possible
explanatory power in the results. Subjects were asked whether they had
ever had (a) a religious conversion experience, (b) a period of
disillusionment with their religious faith, and (c) an experience that
profoundly affected their view of self, others, and life. If they
answered "yes” to any of these questions, the subjects were asked to give
the age at which the experience occurred and to briefly describe what
happened and how it affected them. In tabulating subjects' responses to
these three questions, four possible responses were observed and
recorded. Subjects were asked to check "yes” or ”no" and then to explain
any "yes" answers. However, some subjects did not answer the question at
all, and others who checked "yes” offered no explanation. The researcher
chose to separate these two responses from those of subjects who checked
"no” or who checked ”yes” and provided some specific details about the
nature of the experience and its affect on them. The four possible
responses, then, to these questions were (a) no answer to the question,
(b) ”no," (c) ”yes,” and (d) ”yes” with explanation or detail.

In response to the question about having a religious conversion
experience, one-third of the sample indicated they had had such an
experience (see Table 3.5). The mean age of conversion was 26.4 years,
and for 50% of this sample their religious conversion occurred before age
21. However, conversion experiences were reported from ages 8 to 73 (see
Table 3.6).

Subjects were also asked to state whether they had ever gone through

a period of being disillusioned or disenchanted with their religious

61

Table 3.5
Subjects' Report of Religious Conversion Experiences

 

 

 

ADEEEI RENEE: Sam e
No Answer 0 0
No 132 67
Yes 39 21
Yes with Details 22 12
Table 3.6

Age of Religious Conversion (for those who indicated such an experience)

 

e 0 ve Egmbg; e e f S e
Under Age 10 3 5
Age 11 - 19 24 41
Age 20 - 29 13 22
Age 30 - 39 7 12
Age 40 - 49 9 15
Age 50 - 59 0 0
Age 60+ 3 5

 

 

faith and practice. Of those who answered this question, 78 persons (43%
of the sample) indicated they had been disillusioned about their personal
faith at some time in their lives. The average age for this
disillusionment was 28.3 years., Interestingly, the mean and median ages
(both 25 years) for this experience were a few years older than for
conversion experiences. Tables 3.7 and 3.8 summarize the report of

disillusionment and the age for the reported experience.

62

Table 3 . 7
Subjects' Disillusionment with Religion

 

 

 

52m: when 12W
No Answer 8 4
No 20 53
Yes 49 27
Yes with Details 29 16
Table 3.8

Age of Disillusionment (for those who indicated such an experience)

 

e of is 10 t E2222: e of e
Under Age 10 O 0
Age 11 - 19 20 27
Age 20 - 29 27 37
Age 30 - 39 l4 19
Age 40 - 49 5 7
Age 50 - 59 6 8
Age 60+ 2 4

 

 

The course of a person's life and development is often dramatically
influenced by a particular event. Subjects were asked to indicate if
they had ever had an experience of whatever nature that profoundly
affected their views of self, others, or life. Nearly half of the sample
(90 persons) reported that they had, in fact, had such a profound,

life-changing experience. The nature of these profound experiences

63

varied from adoption at age three to the positive influence of a music

teacher to the death of a spouse and to the loss of physical health.
These experiences tended to occur on average at a slightly older age
the disillusionment with religion experiences (mean - 29.4, median -

30). Tables 3.9 and 3.10 summarize the subjects' responses and ages

the profound experiences.

Table 3.9

Experience Profoundly Affecting One's View of Self, Others, and Life

than

of

 

 

 

 

Aaner Number W
No Answer 20 11
No 74 40
Yes 59 32
Yes with Details 31 17
Table 3.10
Age of Experience Profoundly Affecting One's View of Self, Others, and
Life
Wm limbs:
Under Age 10 6 8
Age 11 - l9 6 8
Age 20 - 29 25 32
Age 30 - 39 29 37
Age 40 - 49 8 10
Age 50 - 59 l 1
Age 60+ 4 5

 

 

64

122W

The ”data sheet" which was developed for this research project asked
respondents for such standard information as gender, age, race, and level
of education. In addition it asked for information which may be either
(a) confounding variables in the analysis of the results or (b) relevant
variables on which specific comparisons can be made.

In the latter category are identifiers like theological orientation
and interest in religion which have been utilized in other research with
ROS and found in some cases to produce differential comparisons in
subsamples (cf. Spilka et al., 1985; Batson & Ventis, 1982). The
question regarding recent major loss was included to assess the effect,
if any, on the pattern of scores on ROS for those persons adjusting to a
significant loss.

Much of the literature of the psychology of religion focuses on the
experience of conversion and its antecedents as well as its
consequences. Though the evidence is mixed and the theoretical
perspectives varied on the relationship between conversion experiences
and mental health (cf. James, 1961, chapters 8-10; Hiltner, 1966; Conn,
1986), there is some recent evidence that those persons with a continuous
religious development were more healthy than those who had a more
discontinuous religious development (Bergin et al., 1988). Bergin et a1.
define a continuous religious development as one in which the young adult
identifies with the denomination of his or her parents, was educated in
their religious traditions as a child, has not had a conversion

experience leading to a change of religious identification, and reports

65

mild rather than intense religious experiences. It was felt that some of
these variables might confound the results obtained on the ROS or the
AAAP, so brief questions were asked on the data sheet regarding religious
conversion or disillusionment experiences, religious education, and

profound experiences affecting one's view of self, others, and life.

W

More than 70 published studies have used the Religious Orientation
Scales (ROS) to address questions of the relationships between religion
and other personality and behavioral constructs, and it has been called
one of the most frequently used psychometric measures of religiousness
(Donahue, 1985a). The 20-item ROS was developed by Gordon Allport and
some of his students to measure orientation or motivation toward personal
religion. Allport's R08 is a subset of a 21-item religious orientation
scale first used by Feagin (1964). There are two subscales to the ROS:
an extrinsic religious orientation scale of 11 items and an intrinsic
religious orientation scale of 9 items. There have been somewhat
different scoring procedures used by researchers with these items
including a four-choice response, a true-false format, and a reverse
scoring for the intrinsic subscale, so that the lower the score, the more
highly intrinsic the response (see Meadow and Kahoe, 1984; Donahue,
1985a). This has made comparisons of results across studies somewhat
difficult, but recently there has been a trend toward standardization of
the scoring procedure (Donahue, 1985a) with each item having a five point

response (from "I definitely disagree" to “I definitely agree”) and both

66

subscales scored so that the greater the sub-scale score, the greater the
presence of that orientation.

Interestingly, neither Allport and Ross (1967) nor Feagin (1964),
whose original set of religious orientation items was used by Allport
(with the exception of one item) report internal reliability
coefficients. Several later researchers have reported their scale

reliabilities which are summarized in Table 3.11.

Table 3.11 '
Internal Reliability Coefficients for I and E

 

 

Batson Spilka et a1. Griffin 8 Thompson
(1976) (1977) (1983)
Intrinsic
Scale .76 .91 .81 - .93
Extrinsic
Scale .70 .85 .69 - .82

 

 

The item to total scale correlations have not been as good, ranging
from .28 to .58 for the Intrinsic scale and .18 to .50 for the Extrinsic
scale (Hunt 8 King, 1971; Robinson 8 Shaver, 1973). Several researchers
have found that by eliminating a few items the reliabilities of the
scales can be improved (McConahay 8 Hough, 1973; Patrick, 1979; Donahue,
1985b). Feagin's (1964) early study factor analyzed a subset of an
original 21 items he used and generated two six-item subscales. Feagin's
shorter scales have better item-to-total scale correlations (.54 to .71
for I, and .48 to .68 for E). Most researchers have chosen to use the

Allport and Ross (1967) 20-item Religious Orientation Scale (cf. Meadow 8

67

Kahoe, 1984). Donahue (1985b) in his recent report on the status of
research with R08 strongly recommends use of the ROS plus the one item of
Feagin's not used by Allport and Ross. He then urges researchers to
score both the longer and the shorter versions and decided which version
has a higher reliability with their sample. This study included all 21
of Feagin's items, and before comparisons were made with other measures,
an analysis of the scale reliabilities was done to determine which
version (the shorter or longer version) would used in the analyses.

For the present sample, intrinsic and extrinsic sub scales were
computed in the two ways described. El and Il designate the six item
shorter versions, and E2 and I2 designate the longer Allport and Ross
versions most commonly used in research. Table 3.12 summarizes the scale
reliabilities for each of these subscales. Based on an evaluation of
these scale statistics, the researcher chose to utilize E2 and 12, the
longer versions. Although the E2 item to total scale correlations were
lower than for E1, the standardized item alphas for both E2 and 12 were
higher than for El and 11. Of particular significance was the fact that
the internal reliability coefficients for E2 and 12 were higher than for
El and Il. Alphas of .69 for E2 and .82 for 12 compare favorably with
those of other researchers, as summarized in Table 3.11. The item to
total scale correlations for E1 and 11 for this sample are lower than
those reported by Feagin (1964), whereas the item to total scale
correlations for the longer versions (E2 and 12) compare with or are
better than those reported by other researchers cited above.

It may be of interest to note in passing that the correlations

between the two methods of computing extrinsic and intrinsic for this

68

Table 3.12
subscale Statistics for Extrinsic and Intrinsic Religious Orientations

 

Range of Correlations

Internal Reliability Standardized for Items to Total
WHILE. 1mm Stale
El .6763 .6780 .29 - .54
E2 .6887 .6910 .23 - .45
Il .7502 .7615 .34 - .59
12 .8162 .8212 .37 - .66

 

 

sample are very high. For El and E2, r - .74 (p - .000); and for 11 and
12, r - .96 (p - .000). The relationship between these two different
orientations has been of interest, as detailed in Chapter Two. For this
sample, 12 and E2 have a negative correlation (-.25) which is similar to
the original report by Allport and Ross (1967). Based on the above
statistical evidence for the two methods of computing intrinsic and
extrinsic scales, this researcher chose to use the longer version (E2 and
12), hereafter labeled simply E and I.

The validity of the intrinsic and extrinsic orientations, although
not unchallenged (Hunt 8 King, 1971), does seem to have support from a
variety of sources. Probably the most important evidence to determine the
type of religiousness that E and I are measuring is to look at their
relationship to other religious measures. Across a variety of studies
using various measures of religiousness, including church involvement,
financial support, knowledge of church teachings, personal growth in
faith, salience of beliefs, etc., I and E exhibit significantly different
patterns of relationships. In a summary of six studies (Batson, 1976;

Batson 8 Ventis, 1982; Dodrill et al., 1973; Hoge 8 Carroll, 1973; King 8

69

Hunt, 1972; Spilka et al., 1968), Donahue (1985a) reported intrinsic
religious orientation correlated .39 with such measures (.59 if one study
was eliminated), and the extrinsic religious orientation correlated .16.
The theory underlying these two ways of being religious would predict
these empirically observed differences. Even more striking evidence of
the concurrent validity of I and discriminant validity for E comes from
four studies in which the relationship between these two scales and the
respondents' rating of the importance of religion or religious commitment
in their lives (Batson, 1976; Batson 8 Ventis, 1982; Jackson, 1981;
Spilka et a1, 1968). I correlated .76 with rating of the importance of
religion and for E the correlation was .03.

Additional support for the validity of these two constructs comes
from a study by Hoge (1972) who sought to validate the distinction
between religion as ultimate (intrinsic) versus religion as instrumental
(extrinsic) and to refine a measurement instrument for this conceptual
distinction. Hoge utilized persons nominated for his studies by
ministers who were asked to identify persons from their congregations who
were either intrinsically or extrinsically oriented after being given a
summary of these religious orientations. Using all of the ROS items plus
many new items, Hoge sought to develop a more reliable and valid measure
of intrinsic religiousness by using ministers' judgments of parishioners'
orientations as the concurrent validity standard. His Intrinsic
Religious Motivation (IRM) scale had better item-to-total scale
correlations than the ROS ranging from .48 to .80 and a scale reliability
of .90. Interestingly, though this scale appeared to have somewhat

better psychometric properties than the ROS, this researcher could find

70

no study, since Hoge's was published 16 years ago, to use this IRM
scale. Perhaps, in part, this is due to the very high correlations that
the IRM exhibited with the Allport R08 (.87 total ROS; .86 intrinsic
subscale; .71 extrinsic subscale). Hoge's study does provide evidence
that ministers were able to select, with some degree of reliability,
those who were extrinsic or intrinsic. In part, Hoge was responding to
criticisms by Hunt and King (1971) that the definition of intrinsic had
included several theoretical components and that scale construction had
involved several factors rather than a single I factor. These criticisms
of I have been considerably moderated since Hunt and King's paper was
published. In part, this has been because of the evidence presented in
Chapter II, in which a large number of studies have found results in the
direction predicted. As Meadow and Kahoe (1984, p. 294) comment, "We
evaluate the religious orientations by how they work.”

Recent reviews of the ROS, while taking into account the difficulties
discussed above, encourage the use of this measure of orientations or
motivations toward religion (Batson 8 Ventis, 1982; Meadow 8 Kahoe, 1984;
Donahue, 1985a). This research involved use of the scores on the two
subscales in both ways that previous researchers have used them. First,
the separate subscale scores for I and E were correlated with the scores
on the psychosocial developmental measure (AAAP). Second, the scores for
both scales were used to form the four-fold typology which was discussed
in Chapter II. The sampling procedure was designed to produce subjects
with a range of scores on both I and E, so that four groups could be
formed: Intrinsics (high 1, low E), Extrinsics (low I, high E),

Indiscriminately Religious (high 1, high E), and Non-religious (low I,

71

low E). The typology was formed in two ways: first, by splitting the
sample using the theoretical midpoint of the I and E subscales as Donahue
(1985a) suggests. Then, for the sake of comparison, the typology was
also formed using the sample medians, which is the standard practice of
most previous research using the ROS.

For the E subscale, scores could theoretically range from 11 to 55,
with a midpoint of 33. The sample in this study had a range of scores
from 11 to 42, with a mean of 26.0, a standard deviation of 6.1, and a
median of 26. Crosstabs for the fourfold typology using the theoretical
midpoint would be 11-33 and 34-55 and using the sample median were 11-26
and 27-42.

For the I subscale, scores could theoretically range from 9 to 45,
with a midpoint of 27. The sample in the study had a range of scores
from 16 to 45, with a mean of 33.8, a standard deviation of 6.7, and a
median of 35. Crosstabs for the fourfold typology using the theoretical
midpoint would be 9-27 and 28-45 and using the sample median were 9-34
and 35-45.

The fourfold split of the sample was constructed using the
theoretical midpoints and the sample medians for both E and I. Results
of these procedures are summarized in Table 3.13. It was decided that
the use of the theoretical midpoint of the subscale to form the four
religious orientation types (at least for this sample) would make further
comparison and statistical analyses invalid. The small cell size for
Extrinsic (6) and Indiscriminately Religious (17) make the results of
statistical procedures questionable. The sample median split, however,

generated ample numbers in each of the four cells for valid statistical

72

Table 3.13
Fourfold Split of the Sample with the Number of Subjects in Each Category

 

Religious Orientation Type

Indiscriminately
Intrinsic Religious Extrinsic

Non-Religious

High I High I Low I Low I

Low E High E High E Low E

Median Split
of Sample 59 35 55 35
Theoretical _
Midpoint of 126 17 6 35
ROS Scales

 

 

analyses. The four religious orientation subtypes referred to in

Chapters Four and Five are based, then, on median splits of this sample.

W

The AAAP is a recently developed instrument designed to objectively
measure an individual's degree of mastery of each of the eight
psychosocial developmental tasks postulated by Erik Erikson (1963,
1968). It is an objective, self-report measure consisting of 320 items
which includes items for each of the eight Eriksonian developmental
tasks, as well as two validity scales. One of the validity scales is a
Likert form of the Crowne-Marlow (1960) social desirability index. The
other validity scale is an unusual response scale patterned and named
after the F-scale on the MMPI. Each item consists of a statement on
which respondents are asked to rate themselves using a four-point Likert

scale as follows:

73

H
I

Definitely true of me

N
I

True of me
3 - Not true of me

4

Definitely not true of me

If an item does not apply to a respondent, the instructions direct that
the item be left blank on the scoring sheet. Items are presented either
"positively", meaning that endorsing the item as ”True of me" indicates
mastery of the stage to which the item corresponds, or "negatively",
meaning that non-endorsement of the item (”Not true of me") is an
indication of mastery of that stage. The test is scored by giving the
respondent four points for an item when it is answered in the direction
most supportive of mastery of the stage, three points if the respondent
has endorsed the response next most indicative of mastery and two points
or one point for the least supportive statements. If a respondent leaves
an item blank (meaning "Does not apply to me"), the item is scored with
no points. Individual item scores which pertain to a given stage are
then summed to yield a stage score.

Mastery levels for each of the stages have been established so that
the eight stages exhibit the characteristics of Guttman scales, that is,
that the stages are mastered in hierarchical order of difficulty,
beginning with Stage 1 and continuing in order through Stage 8, the last
one to be mastered. If the score on a given stage is above the cutoff
point, it indicates that the respondent has answered a substantial
proportion of that stage's items in the direction of mastery. This being
the case, the individual is presumed to have resolved the psychosocial

task of that particular stage. The necessary mastery levels are: Stage

74

1 - 67%; Stage 2 - 78%; Stage 3 - 69%; Stage 4 - 78%; Stage 5 - 85%;
Stage 6 - 80%; Stage 7 - 73%; Stage 8 - 84% (Azar, 1982). Several
studies have utilized the AAAP and the mastery level indicated
(Picciotto, 1987; Arulpragasam, 1986). Other studies have chosen to
ignore the stage mastery level and compare subjects' relative scores on
each stage (Valdez, 1984; Scabbo, 1984). Presumably, the higher the
individual's score on a particular stage, the more complete or successful
the mastery of that developmental task. The eight developmental stages
of the AAAP have relatively high internal consistency as measured by
Cronbach's alpha. Table 3.14 summarizes the means, standard deviations,
and reliability data for several studies, including the present one.
Originally normed (Azar, 1982) on a population of university staff and
faculty who ranged in age from 19-69, the AAAP has been used with college
students (Valdez, 1984), alcoholics (Athy, 1986), normal adults
(Arulpragasam, 1986; Kalikow, 1987), cocaine and heroin abusers (LeBlanc,
1988), psychiatric inpatients (Azar, 1982), and handicappers (Scabbo,
1984). In general, its results have been in the anticipated direction for
these diverse populations.

The modified Crowns-Marlow social desirability scale has been found
to correlate highly with the original Crowne-Marlow when administered at
the same time (Farquhar et al., 1983). Factor analysis of the instrument
identified twenty-three factors that are consistent with Erikson's theory
and are shown to be moderately reliable (see Table 3.15). Construct
validity was established originally by comparing a normal and a
psychiatric population (Azar, 1982). The means for each stage were

significantly higher for the normal population than for the psychiatric

75

.08.. .Uooooca.a< “08.. .53: 38— .30..) ”3958

 

an. m0.

m». 50.

8.8
9 0w.» 8 o
05.:
on 00 S «5.0 on 2.
3......
8. 8.2. 3.:
50.2.
co. 8.: 5a....
5.»...
no. mm. —~ 8.0.. 3.5.
3.0
50. 9.5 5.5.0
3.2.
8. 55.3 no.0
55.0
8. . mud 3.5 3.5
”qua manna muudq» _x;mq
-2332

mamwum.»mm.uummammw

~0.o..

05.2.

~5.n..

5.5

55.2

9.0

5.5.8

55.05

50.05
05.55 9.05

8.8
5%..» No.9

N060
3.0.0 3.50

3.3..
50.85

S. .6—
50.9..

3.55
3.5.5 “.05

55.8—
50.50..

00.05
8.55 55.~5

3.35

N0. .65

3.8..

3.05 35.53.... ....>

3.3

3.8

3.3..

3.3..

00.05

and:

3.05

manna “mama” snug aqua

6.5.3.2

mummy

su.>.u
.utacoa .__>

it... ..>

su.ucun_ .>

tag. .>.

9).»

.035:— .2.

585‘ .:

ugh. .—

uqquwuuuuum

 

1.6:: acooocn 05 .55 8.333.. 2.» 2e 80in. 5o 8.. gone :38 tobacco 33 2.» £8...
3:58 .293 39.3 g 05 5o moo-um :33 65 .35 3.3.33.8. .8- !B.ua.>oa Pix-3m .28:

«En 0.95

76

Table 3.15
Factors Emerging from the AAAP

 

Cronbach's

W W Alpha
1. Trust Basic Trust .88
II. Autonomy Will to be oneself .89
Solitude .82
Holding on, letting go .82
III. Initiative Self-punishment, guilt .86
Anticipation of roles .81
by parents
IV. Industry Apply self to task .92
V. Identity Trust in peers .86
Ideological thought .79
Molding identity .84
Fidelity tests .79
VI. Intimacy Commitment to affiliation .90
Fusion with another .68
VII. Generativity Establishing and guiding the .88
next generation
Charity .84
VIII. Integrity Order and meaning .83
Accepting one's life cycle .80

* Source: Azar, 1982.

 

 

population with the exception of Stage 6. Additional construct validity
comes from Athy's (1986) study in which means for several of the stages
for an alcoholic sample were lower than for a normal comparison group.
Arulpragasam (1986), as part of her study of the construct of "hope",
planned a criterion validation hypothesis on the AAAP. As expected by
Erikson's theory, persons who mastered through Stage 7 were significantly

older than persons who mastered Stage 4 but not Stage 7. She found

77

further that for her sample there was no relationship between education
level and mastery of Eriksonian stages.

Because the AAAP is a new instrument, additional studies need to be
completed to further establish its criterion validity. The construct
validity of Stage 6 is questionable, given its lack of discrimination
between a normal and psychiatric population (Azar, 1982). There is no
test-retest reliability data. However, given the good internal
consistency of the stage scales and the evidences of construct validity,
the use of the AAAP as a measure of psychosocial development was
supported for this study. The AAAP appeared to be useful as a measure of
psychological health in adults and as a developmental measure, which were
the types of comparisons the present study was undertaking. In addition,
the AAAP scores were utilized in both ways that previous researchers have
found helpful. First, the individual respondent's scale score for each
of the eight stages were used in some analyses. Second, the highest
stage mastered in order (based on the mastery levels established by Azar,
1982) were used for other comparisons.

The use of the AAAP in the present study was also based on
consideration of two other available measures of Eriksonian psychosocial
development. These are the only other instruments in the literature
which attempt to access all eight of Erikson's stages. Many researchers
(Constantinople, 1969; Rosenthal et al., 1981) have attempted to develop
instruments which measure some of the Eriksonian developmental tasks.
Typically, as with the studies just cited, the interest is in
psychosocial development of young adults and focuses on identity (Stage

5) and intimacy (Stage 6). Boyd and Koskela (1970) developed the Self

78

Description questionnaire as an objective format, self-report
instrument. Consisting of 160 items, it asked subjects to respond on two
Likert scales for each item: "Like me - unlike me” and ”of concern - of
no concern to me”. Despite its interest in measuring all eight of
Erikson's stages, the normative sample was college undergraduate and
graduate students with a restricted age range and which was inappropriate
for the two highest stages. The researchers concluded that the
instrument did provide support for Erikson's theoretical notion that the
stages are ordered chronologically. Relatively high test-retest
reliability and internal consistency scores for this instrument were
reported. However, no control for social desirability was performed and
the validity of this instrument is somewhat uncertain without using a
less homogeneous sample.

A recently published study describes the development of the Modified
Erikson Psychosocial Stage Inventory (Darling-Fisher 8 Leidy, 1988).
This instrument was a modification of one developed by Rosenthal et al.
(1981) which measured successful and unsuccessful resolution of Stages
1-6. New scales were created for the final two stages, as well as
modifications of the existing items. An 80-item instrument using a
5-point Likert scale for responses to short sentences, which represented
either successful or unsuccessful resolutions of each stage crisis, was
evaluated for its psychometric properties. A convenience sample of 168
adults whose ages ranged from 19-86 was administered this instrument.
The alpha reliabilities for the eight subscales are good (.75 - .88) and
construct validity was supported by positive relationships between

chronological age and mean scales on the last two stages. This new

79

measure appears promising; however, there have been no studies which have
demonstrated its predictive or criterion validity. In addition, the very
restricted range of scores for each stage (total stage scores range from
1-5) would make this instrument somewhat less useful in the present

primarily correlational study.

W

This study was descriptive in nature. It was concerned with the
relationship between religious orientation and Eriksonian psychosocial
development. The independent variables were religious orientations and
the dependent variables were scores on the Eriksonian stages as obtained
from the AAAP and highest stage mastered on the AAAP. While causality is
not within the scope of this correlational study, the research provides
information on religiousness and development that may enable future

research and theory development.

W

The specific hypotheses to be tested in this study were divided into
two clusters: (A) those hypotheses that concern the use of separate
scores on I and E and their relationship to outcomes on the Eriksonian
measure; and (B) those hypotheses that concern the groups of subjects
determined on the basis of the Religious Orientation typology using each
subject's scores on both scales as a basis of assignment to an Intrinsic,
Extrinsic, Indiscriminately Religious or Non-religious group. Each of
the research hypotheses are restated in testable form, including the null

hypothesis and its alternative.

80

Set A: Hypotheses With Scores on I and E

Hypothesis 1

H01: Scores on the Intrinsic Scale will be uncorrelated with
scores for each of the eight Eriksonian stages on the AAAP.
HA1: Scores on the Intrinsic Scale will be positively correlated

with scores for each of the eight Eriksonian stages on the
AAAP.

Hypothesis 2

H02: Scores on the Intrinsic Scale will be uncorrelated with a
subject's highest Eriksonian stage mastered in order.

Scores on the Intrinsic Scale will be positively correlated
with a subject's highest Eriksonian stage mastered in order.

HA2:

Hypothesis 3

H03: Scores on the Extrinsic Scale will be uncorrelated with
scores for each of the eight Eriksonian stages on the AAAP.

Scores on the Extrinsic Scale will be negatively correlated
with scores for each of the eight Eriksonian stages on the
AAAP.

HA3:

Hypothesis 4

H04: Scores on the Extrinsic Scale will be uncorrelated with a
subject's highest Eriksonian stage mastered in order.

HA4: Scores on the Extrinsic Scale will be negatively correlated
with a subject's highest Eriksonian stage mastered in order.

Set B: Hypotheses Using the Religious Orientation Typology

Hypothesis 5

H05: There will be no difference between the mean scores on each
of the Eriksonian stages for the Intrinsically Religious
group, the Extrinsically Religious group, the
Indiscriminately Religious group, and the Non-religious
group.

HA5: The mean scores of the Intrinsically Religious group for each
of the Eriksonian stages will be equal to or higher than the
mean stage scores of the Non-religious group, which will be
higher than the scores for the Extrinsically Religious
groups, which will be higher than scores for the
Indiscriminately Religious group.

81

Diagrammatically, this hypothesis looks like:

 

II

V

VI

VII

fTWILmF—

VIII

 

Religious
Orientation

Type

Intrinsic

 

Non-religious

 

Extrinsic

 

 

Indiscrim-
inately
Religious

 

 

 

 

 

 

 

 

 

Hypothesis 6

H06 ‘
mastered in order.
HA6:
mastered in order.
We

There will be no difference between the Intrinsic, Extrinsic,
Indiscriminately Religious and Non-religious groups in the
frequency distribution of the number of Eriksonian stages

There will be a difference between the Intrinsic, Extrinsic,
Indiscriminately Religious, and Non-religious groups in the
frequency distribution of the number of Eriksonian stages

Hypotheses 1 through 4 were analyzed by computation of the Pearson

product moment correlation coefficient.

For Hypotheses 1 and 3, eight

separate correlation coefficients were generated, corresponding to each

of the eight Eriksonian stages.
for I involves the assumption that the sample is chosen at random and
that the two measures are distributed normally. Because of the sample

size, these two assumptions were met.

The use of the table of critical values

 

82

Hypothesis 5 was analyzed utilizing multivariate analysis of variance
(MANOVA) which is often used for research designs in which multiple
outcome measures are collected. MANOVA allows the data to be seen in
multivariate perspective and simultaneously analyze the nature of
multiple influences. MANOVA produces one probability statement for the
entire set of variables, allowing a thorough analysis of complex data.
MANOVA offers two distinct advantages over performing multiple ANOVA for
this hypothesis. First, by utilizing a single alpha level, it controls
for an increase in experiment-wise Type I error rate which preserves
statistical power. As the number of dependent variables increases, the
use of multiple ANOVAs increase the Type I error rate, also increasing
the Type II error rate and thereby decreasing power. Second, by analyzing
all of the comparisons simultaneously, MANOVA takes into consideration
possible intercorrelations between the dependent variables. If the
result of the MANOVA is statistically significant, than a series of
one-way analysis of variance (ANOVA) are performed to determine where the
differences are significant between the four groups on the eight stages.

Hypothesis 6 was analyzed through the construction of a frequency
distribution of highest stage mastered within each religious orientation
category. A chi-square test of distribution, with degrees of freedom
equal to one less than the number of groups, was performed to determine
if there were significant differences between the four religious
orientations. The assumption of chi-square is independent sampling for
each group, that is, that assignment of a subject to one group does not

effect assignment of any of the other subjects.

83

For all statistical procedures the probability level for significance
was set at .05. The influence of relevant demographic variables upon
' score on the ROS or on stage scores of the AAAP was examined using

analysis of variance and other statistical procedures.

imam

In Chapter III, the nature of the sample and the procedures for data
collection were described. Each of the instruments to be used in this
study were presented, including a discussion of their psychometric
properties, their validation and the techniques for scoring each of
them. Two sets of hypotheses were presented, the first set having to do
with correlations between the I and E scales and the stage scores and
highest stage mastered on the AAAP; and the second set of hypotheses
relating the religious orientation groups (types) to the stage scores and
highest stage mastered on the AAAP. The methods of analysis of the data
were reviewed. In Chapter IV, the results of the analyses will be

presented.

CHAPTER IV

RESULTS

This chapter will present the results of the data analysis. The
first section of the chapter will restate the original hypotheses and
present the results of the analyses presented in the previous chapter.
The rest of this chapter will present post hoc analyses using demographic
information and other subject information along with the results on the
ROS and AAAP as a way to understand and explain relationships between

religious orientation and healthy psychosocial characteristics.

WW

0 -4

The original hypotheses were organized into two sets. The first set
consisted of four hypotheses considering relationships between (a) a
subjects' score on the Intrinsic and Extrinsic religious orientation
subscales and (b) the stage scores and the highest stage mastered in
order on the AAAP. Table 4.1 presents a summary of the Pearson
correlations that were the statistical procedures for testing the first

four hypotheses.

84

85

 

5.8 0.8.8 EN... 898 8...... E98 RN... ~38 54......

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6.37.86
~n5.- mm..- nqu 9...- «5...- mo..- 0mo.. 50...- 03.-
...z-..-.--:7-i-i-. ,-. iii-:1: I 0.3on
8..
80.8 No.8 «nu-so 53.8 $5.8 35.8 35.8 gin ~nn.l
6.875.:
MB. 3... 0m... woo.- 559. 30.. 05o.- 583 who.-
cooco c. o 5 0 n e n w 5
stun! . . ii. 1;... . 1-..... I -
omen» one»... .2:
~85...

 

 

.88 5 88:! «out 85:. as 8.84. 83... .32 as «88» 3883. 8.. 95.: 8.58:3». .3. seistoa 8:328 Etta
5... 3...:

86

Mails—l
801 Scores on the Intrinsic scale will be uncorrelated with
scores for each of the eight Eriksonian stages on the
AAAP.
HA1 Scores on the Intrinsic scale will be positively

correlated with scores for each of the eight Eriksonian

stages on the AAAP.

Hypothesis 1 is really eight separate hypotheses and eight separate
correlations between the subjects' scores on the Intrinsic religious
subscale and each of the eight stage scales of the AAAP. As the results
in Table 4.1 indicate, it was not possible to reject the null hypothesis
for AAAP Stages 1-7. For Stage 8 the null hypothesis was rejected and
the alternative hypothesis accepted. A subject's score on the Intrinsic
religious subscale is significantly positively correlated (p < .05) with
that subject's score on Stage 8. With the exception of Stage 8, all of
the correlation coefficients are very small and near zero, with some
slightly negative and some slightly positive. The Stage 8 correlation
coefficient is not particularly large, but, given the large sample size
(n - 184), is statistically significant and does indicate that the higher
a subject's score on I, there is a tendency for that subject's score on

Stage 8 to be higher as well.

mm;

302 Scores on the Intrinsic scale will be uncorrelated with a

subject's highest Eriksonian stage mastered in order.

87

HA2 Scores on the Intrinsic scale will be positively
correlated with a subject's highest Eriksonian stage

mastered in order.

The results of the correlation between the subjects' Intrinsic scores
and the highest stages mastered in order (r - .003; p - .485) were not
significant and did not allow the rejection of the null hypothesis.

There is no relationship in this sample between a subject‘s mastery in
order of the Eriksonian stages as measured by the AAAP and his/her degree

of Intrinsic religiousness.

W
H03 Scores on the Extrinsic scale will be uncorrelated with
scores for each of the eight Eriksonian stages on the
AAAP.
HA3 Scores on the Extrinsic scale will be negatively

correlated with scores for each of the eight Eriksonian

stages on the AAAP.

As with Hypothesis 1, Hypothesis 3 is really eight distinct
comparisons between the subjects' degrees of Extrinsic religiousness and
their scores on each of the eight stages of the AAAP. The results of the
correlations are reported in Table 4.1. As predicted in the alternative

hypothesis, each of the correlations was negative; however, four of the

88

eight were not statistically significant. The results do not allow the
rejection of the null hypothesis for Stages 1, 3, 4, and 7.

Though the relative size of the Pearson correlation coefficient is
small in each case, given the sample size (n - 184), there is a
statistically significant relationship between a subject's Extrinsic
score and his/her scores on Stages 2, 5, 6, and 8. The results allow the
rejection of the null hypothesis for Stages 2, S, 6, and 8 and the
acceptance of the alternative hypothesis. For this sample, there is an
inverse relationship between Extrinsic religiousness and the degree of
successful resolution of the tasks of Stage 2 (autonomy), Stage 5
(identity), Stage 6 (intimacy), and Stage 8 (wisdom). As a subject's
score on E increases, there is a corresponding tendency for that

subject's scores on each of those stages to be lower.

W
H04 Scores on the Extrinsic scale will be uncorrelated with a
subject's highest Eriksonian stage mastered in order.
HA4 Scores on the Extrinsic scale will be negatively

correlated with a subject's highest Eriksonian stage

mastered in order.

The Pearson correlation coefficient for the relationship between a
subject's Extrinsic score and the highest stage mastered in order (r -
-.132; p - .037) allowed the rejection of the null hypothesis and the

acceptance of the alternative hypothesis. As a subject's degree of

89

Extrinsic religiousness increases, that is, the score on E increases,
there is a tendency to master fewer stages in Erikson's developmental

order.

W

The second set of hypotheses utilized a fourfold typology of
religious orientations determined by a median split of the sample based
on scores on both the Extrinsic and Intrinsic scales. The four religious
orientation types were Intrinsic (High I, Low E), Extrinsic (Low 1, High

E), Indiscriminately Religious (High I, High E), and Non-religious (Low

I, Low E).
W
H05 There will be no difference among the mean scores on each
of the Eriksonian stages for the Intrinsically religious
group, the Extrinsically religious group, the
Indiscriminately Religious group, and the Non-religious
group.
HA5 The mean scores of the Intrinsically religious group for

each of the Eriksonian stages will be equal to or higher
than the mean stage scores of the Non-religious group,
which will be higher than the scores for the
Extrinsically religious group, which will be higher than

scores for the Indiscriminately Religious group.

90

Three separate multivariate analyses of variance (MANOVAs) were
computed, and their results are reported in Table 4.2. Each of these
MANOVA tests was significant (p < .05), and univariate analyses of
variance (ANOVAs) were run to determine the source of significance.

Table 4.3 summarizes the results of the ANOVAs for each of the mean stage
scores by each of the four religious orientation types. Significant
differences (p < .01) were found among the four religious orientation
groups' mean scores on Stage 2 (autonomy). Results on Stage 4 (industry)
approach statistical significance (p - .06). Results for the other
stages (1, 3, 5, 6, 7, 8) were not statistically significant. Table 4.4
summarizes the mean stage scores on Stages 2 and 4 for each of the

religious orientation groups.

Table 4.2
Multivariate Analyses of Variance Utilizing Four Religious Orientation
Groups and Mean Scores for Eight Stages of the AAAP

 

st e Slabs W Manes
Pillais .217 1. 704 .020
Hotellings .241 1.723 .018

Wilks .796 1.714 .019

 

 

91

Table 4.3
Univariate Analyses of Variance for Mean Stage Scores by Religious
Orientation Groups

 

unlit: Significant;
Stage 1 .983 .402
Stage 2 4.195 .007
Stage 3 1.234 .299
Stage 4 2.501 .061
Stage 5 2.098 .102
Stage 6 1.599 .191
Stage 7 1.661 .177
Stage 8 .811 .489

 

 

Table 4.4
Mean Scores on Stage 2 and Stage 4 by Religious Orientation Group

 

Wigwam mm Magma—tateh
Non-religious 111.14 199.06
Intrinsic 110.29 193.03
Extrinsic 108.73 190.67
Indiscriminately Religious 103.57 184.23

 

 

The mean stage scores for both Stage 2 (autonomy) (p < .01) and Stage
4 (industry) (p - .06) are different across the religious orientation
groups in the direction predicted in the alternative hypothesis. The
Non-religious and Intrinsically religious groups are higher than the
Extrinsically religious group which, in turn, is higher than the

Indiscriminately Religious group. For this sample, Non-religious persons

92

and Intrinsically religious persons report more autonomous functioning
than Extrinsically religious or Indiscriminately Religious subjects.
There is a strong tendency, though not statistically significant, for
these same subjects (Non-religious and Intrinsically religious) to report
a greater sense of industry and competence than those who are Extrinsic

and Indiscriminately Religious.

Wu
H06 There will be no difference among the Intrinsic,
Extrinsic, Indiscriminately Religious, and Non-religious
groups in the frequency distribution of the number of
Eriksonian stages mastered in order.
HA6 There will be a difference among the Intrinsic,

Extrinsic, Indiscriminately Religious, and Non-religious
groups in the frequency distribution of the number of

Eriksonian stages mastered in order.

The chi-square computed for the frequency distribution of the highest
stages mastered in order by the four religious orientations was 24.921 (p
- .410). This non-significant result did not allow the rejection of the
null hypothesis. There was no difference in the frequency distribution
for the highest stage mastered in order by the Intrinsic, Extrinsic,
Indiscriminately Religious, and Non-religious groups. This chi-square
statistical test was actually invalid, given a large number of cells

either with no subjects or with an expected frequency of less than five.

93

W

M

Several analyses were done to assess the influence of the demographic
variables of gender and age on the stage scores of the AAAP. The
differences among mean stage scores for male and female subjects were
compared using t-tests. Significant differences (p < .05) were found
between the average stage scores for men and women on Stage 1 and Stage
4. On both stages male scores were higher than female scores, indicating
a greater degree of maStery of these early developmental tasks of (a)
basic trust in self and confidence about life and (b) sense of
competence, skill, and ability to apply oneself to a task. The other
t-tests for differences between men and women were not significant for
Stages 2, 3, 5, 6, 7,8, and highest stage mastered in order.

Age differences on the stage scores were studied in two ways. First,
the sample was split into two groups at the median age: 24-44 (n - 87)
and 45-86 (n - 97). The mean stage scores for each of the AAAP stages
for these two age groups were compared. There were no age differences
found for Stages 1, 3, 5, 7, and highest stage mastered in order. There
were statistically significant differences (p < .05) for Stages 2, 4, 6,
and 8. In each of these cases, the younger age group scored
significantly higher (better) on the stage. Assuming that a higher stage
score indicates a healthier person or a better resolution of the
task/crisis of that stage, then for this sample, younger persons were
more independent (autonomous), more task oriented and felt more

competent, placed more importance on intimacy and commitment to a person

94

of the opposite sex, and had a deeper sense of meaning and purpose in
life.

The second method for analysis of the differences in stage scores
based on age was to split the sample into three groups: younger adults,
24-40 (n - 64); middle age adults, 41-60 (n - 6S); and older adults, 61
and older (n - 55). ANOVAs for each of the stages found non-significant
results for the mean stage scores across these three age groups for
Stages 1, 3, 4, 5, 7, 8, and highest stage mastered in order. On Stage 2
and Stage 6, significant (p < .05) differences were found across the
three age groups. In each of these comparisons, the oldest group had the
lowest mean stage score. Table 4.5 records the rank order of the three
age groups for the mean scores for the eight AAAP stages. As noted
above, only two of these rank orderings (Stages 2 and 6) represent
statistically significant differences in the mean stage scores across the
three age groupings, the pattern of these rank orderings is interesting
to observe. The Stage 8 results are opposite of what would be predicted
on the basis of the theory, since for this sample the youngest adults
scored highest on this stage and the oldest adults scored lowest.
Consistently across seven of the eight stages and the highest stage
mastered in order, the oldest group of adults scored lowest. This
pattern of results is contrary to what one would predict on the basis of
the Eriksonian developmental theory.

Consideration was given to the concept of mastery of the stages using
the procedures developed by Azar (1982) and discussed in Chapter III.
Table 4.6 lists the AAAP stages and (a) the percentage of subjects who

mastered each of the stages regardless of chronological order and (b) the

95

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Iliulual! -34..- u ...... 513;-.. 4?: f , - .-i-.. . . . --i- - - .
n n ~ n n n n n n 2+3 23¢ .330
p N F ~ ~ p _. F p 80...: 23¢ 33:.
I‘ I IQ'-‘ 3: I Jr 1.. I 1 1;! 1|“! Jrlul
~ p n _. — ~ ~ N m .343 :3! .89-8»
canto c. o h o m .4 m ~ — 0.8.5 02
gun-t
«mean 083 a<<<
u323..

 

 

 

 

8.6.5 5 “.2382 83m 323:. 05 use 893 :53 05 we zoom .3. go 02 09.5 05 me .395 xix
m3 3Q»

96

percentage of subjects for whom the particular stage was the highest

stage mastered in order.

Table 4.6

Percentage of Subjects Who Mastered Each Stage, Regardless of
Chronological Order, and Percentage of Subjects Who Mastered up to Each
Stage in Order

 

Percentage Cumulative

Percentage Mastered Percentage

M Jam 42.91.69;
No Stage ‘ 16 16 16
Stage 1 84 51 67
Stage 2 34 3 70
Stage 3 76 4 73
Stage 4 42 13 86
Stage 5 20 .5 87
Stage 6 52 l 88
Stage 7 64 4 92
Stage 8 l7 8 100

 

 

Considering the mastery of any stage regardless of the order, several
observations appear contrary to the theoretically expected outcome. The
sample for this research was drawn from a normal population assumed to be
psychologically healthy, and yet 16% of this sample did not master any of
the AAAP stages at all. It would be expected from Eriksonian theory that
fewer persons would master the higher stages, since successful completion
of later psychosocial tasks is based on successful resolution of earlier

crises and the development of earlier competencies. However, fewer of

97

this sample's subjects mastered Stages 2, 4, and 5 than mastered Stages 6
and 7.

When consideration is given to the mastery of stages in order, the
evidence presented in Table 4.6 again appears contrary to what would be
expected theoretically. Eriksonian theory would predict that a
decreasing number of subjects would master successive stages and that a
population of normal adults would have a large percentage of persons who
have completed the issues related to childhood and adolescence. However,
for this sample 67% only mastered up to Stage 1 (trust), and 86% only
mastered through Stage 4 (industry)--both tasks of childhood, according
to Eriksonian theory. Said somewhat differently, this would mean that a
large portion of the subjects in this sample was dealing unsuccessfully
with psychosocial tasks from childhood. These results from this sample
will be discussed in the next chapter, where consideration will be given
to questions of the construct validity of the AAAP based on these
results.

In part because of the preceding observations and analyses of the
AAAP stage scores and highest stage mastered in order, this researcher
considered each subject's total stages mastered regardless of order in
several post hoc analyses. Looking at each subject's total stages
mastered allowed the stage scores to represent healthy personality
characteristics without considering them as sequentially acquired.
Subjects' total stages mastered ranged from none (16% of sample) to eight
that is, all of the stages (8% of sample). When total stages were
correlated with subjects' scores on I and E as was done for Hypotheses

1-4, the results were similar to the results reported earlier and

98

predictable given those results. No correlation was found between I and
total stages mastered; however, a negative correlation was found for E
and total stages mastered (r - -.16; p - .015). As a subject's Extrinsic
religiousness increases, there is a tendency to master fewer of the

stages.

WWW

Post hoc analyses were conducted on the ROS subscales to consider the
relationship between certain demographic variables and the I and E
subscales. Additional analyses sought to understand the relationship
between I and E and subjects' self report of conversion experience,
disillusionment with religion, and experience profoundly affecting life.

Spearman non-parametric correlations were run between the ROS
subscales and various self-reported religious measures. In terms of
frequency of church attendance, I is positively correlated (r - .43, p
< .001) and E negatively correlated (r - -.19, p < .01) to a subject's
attendance at church worship services. Subjects were asked the
straightforward question, ”How religious are you?” and their responses on
a four-point Likert scale from “not at all” to "very religious" were
highly related to religious orientation. I was significantly positively
related (r - .57, p < .001) and E negatively related (r - -.33, p < .001)
to this self-described degree of religiousness. Subjects were asked to
label their own personal theological orientation and these descriptors on
a continuum from ”fundamentalist" to "liberal" were not significantly
related to E but were negatively related to I (r - -.27, p < .001),

indicating that I is correlated to a more conservative theological

99

orientation. These results are consistent with those summarized in
Chapter II. The higher a score on I the more likely the person will see
him/herself as very religious, be a frequent church attendee, and be more
conservative theologically. The higher the subject's score on E, the
more likely the person will be a less-frequent church attendee and see
him/herself as less religious.

Male and female differences were found on the Intrinsic subscale but
not on the Extrinsic subscale. The t-tests for differences between group
means were not significant (t - -.24, p - .81) for males and females on
E. However, it was found that the women's scores on I were significantly
higher than men's scores (t - - 2.98, p - .003). This finding,
consistent with some previous research with ROS (Donahue, 1985a) would
indicate that women tend to have a religious faith which is more
integrated into their whole life and which forms a more central core for
their identity, experiences, and relationships than men.

When ROS subscale scores were correlated with level of education, the
Spearman coefficients were not significant. For this sample there was no
relationship between either I or E and a subject's educational level.
Some previous research with R08 had found that I may tend to be related
to socially-desirable responding on self-report measures (Batson et al.,
1978; Watson et al., 1984), though Donahue (1985a) has argued that
religious orientation is not related to social desirability. Pearson
correlations between I and E and the social desirability subscale of the
AAAP were both insignificant. The results from this sample support
Donahue, although the relationship of I to social desirability obviously

requires further analysis beyond the scope of this study.

100

Another demographic variable of interest is the subject's age and its
relationship to religious orientation. An initial correlational analysis
was performed which found a positive relationship between both age and E
(r - .16, p - .01) and I (r - .21, p < .01). As a subject's age
increased the subscale scores on I and E both tended to increase.

Further specification of the relationship of age to religious orientation
lead to additional tests. As was done with the AAAP, the subjects were
divided first into two age groups, 24-44 (n - 87) and 45-86 (n - 97), and
then into three groups, 24-40 (n - 64), 41-60 (n - 65), and 61+ (n -

55). T-tests and ANOVAs were then run with the main effect of age being
considered in the analysis.

When two age groups were used, the mean score for the older group was
significantly higher on both I (p - .02) and E (p - .04). When three age
groupings were used, the ANOVA for E was not significant (p - .30),
although the mean score for the oldest age group was highest and the
youngest age group was lowest. Using three age groupings for the ANOVA
and I mean scores was significant (F - 6.45, p < .01) and the difference
was between the oldest age group which was significantly higher than the
other two age groupings. These results indicate that both Extrinsic and
Intrinsic religiousness increase as a function of age, although the

relationship is much stronger for I than for E.

W
W

There were three sets of questions that became the focus for many post
hoc analyses. These questions asked each subject whether or not s/he had

(a) a religious conversion experience, (b) a period of disillusionment

101

with personal religious faith, and (c) an experience which profoundly
affected his/her view of self, others, and/or life. If a person answered
“yes" to these questions, s/he was then requested to state the age of the
experience and to briefly describe what happened and how the particular
experience affected him/her. Some statistical procedures utilized two
groups for each of these questions: those who said “no" and those who
said "yes.” For other post hoc procedures, respondents who said "yes"
were subdivided into two groups, one in which the person checked "yes"
but did not elaborate and one in which the person not only reported the
experience but wrote some detail about the way in which it affected
his/her life. For some analyses, there were three respondent groups:
"no," "yes,“ and "yes with detail.” The distinction between "yes" and
”yes with detail“ was a judgment made by the researcher based upon his
assessment of the quality of the subject's written response. If a
subject only checked ”yes” and gave no written response or if the written
response about the experience was not specific about the nature of the
event or clear about its subsequent effect, that subject's response was
coded as "yes." A ”yes with detail" response required that the subject
provided specific information about the event find about the quality of
the influence it had on the subject. This three-way distinction, it was
hypothesized, would provide potentially meaningful differences. As will
be seen in the following results, this distinction between "yes" and ”yes
with detail" was not particularly meaningful and provided fewer
significant results than the two group analyses.

A set of ANOVAs (with three groups--”no,” ”yes,“ and “yes with

detail") and t-tests (with two groups--'no" and 'yes') were performed

102

with scores on E and I as dependent variables and the independent
variables being conversion experience, disillusionment experience, and
profound experience. Presence or absence of a disillusionment experience
did not significantly affect a person's I or E scores. The same result
(for both t-tests and ANOVAs) was true for report of an experience
profoundly affecting one's life. Profound experience or lack of it had
no affect on a subject's I or E score. Apparently a person's religious
orientation is not influenced either by the experience of having been
disillusioned with one's religious faith or by a life event regarded by
the person as profoundly affecting one's view of life.

However, ROS scores were significantly different when the presence or
absence of a conversion experience was the variable of interest. The
mean E score was significantly lower (t - 3.82, p < .000) for those
persons who reported a conversion experience than for those who did not
report a religious conversion. When the "yes with detail“ group was
created, the results were also significant (F - 8.45, p < .001). Persons
who clearly described the way(s) in which their religious conversion(s)
had affected their lives had significantly lower E than members of the
subject group who only checked ”yes," who in turn were lower on E than
members of the group that reported they had not had religious
conversions. In addition, the mean I score was significantly higher (t -
5.61, p < .001) for those persons who reported religious conversion
experiences than for those who did not report conversion experiences.
Utilizing the three-group comparison also generated significant results
(F - 13.56, p < .001). Subjects who described the way(s) their

conversions influenced their lives ("yes with detail") had higher mean I

103

scores than those who only reported conversions (‘yes'), who in turn were
higher on I than those who reported not having conversion experiences.

An attempt was made to understand the relationship, if any, between
these three types of experiences (religious conversion, disillusionment
with religious, and profound experience affecting life) and the stage
scores on the AAAP. When using the three response groups ("no,' ”yes,"
and 'yes with detail"), no significant differences were found for the
mean scores for any of the Stages 1-7. Conversion experience also did
not significantly affect Stage 8 scores. However, ANOVAs were
statistically significant for disillusionment experience and Stage 8 mean
scores (F - 7.68, p < .001) and for profound experience and Stage 8 mean
scores (F - 5.84, p < .01). In both cases, the mean Stage 8 scores were
ordered as follows: "yes with details" > ”yes" > ”no."

Comparisons were also made utilizing two groups by combining the two
"yes" responses into a single group. When t-tests for the differences
between mean stage scores for each of the eight AAAP stages were
computed, several statistically significant differences were found. The
group that had conversion experiences scored significantly lower on Stage
3 (t - 2.36, p - .02) and on Stage 7 (t - 2.23, p - .03) than the group
that did not have conversion experiences. All other stage scores were
not significantly different; however, for Stages 1, 2, 4, 5, and 6, the
mean conversion group scores were lower than for the no conversion
group. Taken by itself, this result would indicate that a dramatic
religious conversion experience may be associated with less-healthy

psychosocial development, particularly in relationship to the ability to

104

care and guide others (Stage 7) and the presence of guilt or
self-punishing attitudes (Stage 3).

When the variable of interest is simple presence ("yes‘ plus ”yes
with details") or absence of disillusionment with religion,
non-significant t-test results were obtained for mean stage scores for
Stages 1, 3, 5, 6, and 7. Barely statistically-significant results were
obtained for Stage 2 (t - 2.09, p - .04) and Stage 4 (t - 2.01, p - .05)
with members of the group that reported disillusionment experiences
scoring higher. Presence of a reported period of disillusionment with
one's personal religion is apparently related to higher scores and,
therefore, better resolution of the tasks of autonomy (Stage 2) and
industry (Stage 4). Results for Stage 8 (t - 3.97, p < .001) were highly
statistically significant, with the disillusionment group's mean stage
score higher than the no disillusionment group's. Presence of a
disillusionment experience is related to a higher Stage 8 score and more
satisfactory mastery of the task of wisdom (Stage 8).

When the variable of interest is the presence ("yes“ and "yes with
details”) or absence ("no”) of the report of an experience profoundly
affecting one's view of self, life, or others, the t-test is
insignificant for the differences between groups in each stage mean score
for all stages except Stages 5 and 8. Stage 5 mean scores are barely
statistically significant (t - 1.95, p - .05), indicating that
experiences when reflected upon and integrated into one's life can have a
positive effect on one's sense of self and one's identity in relationship
to one's peers. More meaningful perhaps is the result for Stage 8 (t-

3.30, p - .001). The presence of a profound experience affecting one's

105

life is related to a higher score on Stage 8 and a better resolution of
the task of developing wisdom and a coherent philosophy of life.

Based on the above results, it was predicted that persons who
reported both disillusionment experiences and profound life experiences
would score higher on I and the eight stages of the AAAP and lower on E.
Three groups were created based on subjects' report of these two
experiences: a group of subjects who reported they had had both of these
experiences ("yes' on both, n - 54), a group that reported only one of
these experiences ("yes" on one/"no" on one, n - 52), and a group of
subjects who reported they had not had either of these types of
experiences ("no“ on both, n - 56). Those persons who failed to answer
one or the other of these questions were dropped from these analyses.

ANOVA procedures were run using these three groups as independent
variables and the ROS scale scores and AAAP stage scores as dependent
variables. The results were significant for scores on E (F - 3.25, p -
.01) with the "yes on both experiences“ group scoring significantly lower
on the Extrinsic religious scale. Results for I were in the predicted
direction with the “yes on both experiences“ group scoring higher;
however, the F test was not statistically significant. The ANOVA results
for Stages 1-7 were not statistically significant, though it may be noted
that for five of the seven stages, the mean stage score for the "yes on
both experiences" group was higher than the other two groups, with the
“yes on one experience" group's mean stage score the second highest on
five of the seven stages. The results on Stage 8 were highly significant
(F - 5.51, p < .001) and in the direction predicted. Those subjects who

reported both disillusionment experiences with their religious faith and

106

also experiences profoundly affecting their lives scored significantly

higher on Stage 8 than those who reported they had neither experience.

Wm

As presented earlier in the results of univariate analyses of
variance for hypotheses on the mean stage scores for each of the four
religious orientation groups, one was statistically significant (Stage 2)
and one approached significance (Stage 4). Both of these results were in
the predicted direction. An examination of the mean stage scores for
each of the religious orientation groups found an interesting pattern
emerging. Though it must be borne in mind that the data for the other
six stages were not statistically significant, they reflected a very
similar pattern of relationships as those that were significant. Table
4.7 summarizes the rank ordering of the mean stage scores across the four
religious orientations. The table also includes the rank ordering of
mean highest stage mastered in order and mean total stages mastered. In
six of the eight stages, the Non-religious group scored highest, followed
by the Intrinsic group in each case. On the two stages, in which the
Non-religious group's mean score was second highest, the Intrinsic
religious group scored the highest mean score. For highest stage in
order, the Intrinsic religious group is highest, followed by the
Non-religious group; and for total stages mastered the Non-religious
group is highest, with the Intrinsic group second. In no case were the
Extrinsic and Indiscriminately Religious groups rank ordered higher than

third.

107

Table 4.7
Rank Order of the Mean Stage Scores for Each of the Four Religious
Orientation Groups

 

Highest
Stage Total
AAA£_§;§gg Mastered Stages
EELS—Gram lzléifilfi 111.9113 we
Non-religious 2 1 l 1 2 1 2 1
Intrinsic l 2 2 2 2 1 2 2 1 2
Extrinsic 3 3 3 3 4 3 3 4 3 3
Indiscriminately
Religious 4 4 4 4 3 4 4 3 4 4

 

 

A common factor in the Non-religious group and the Intrinsically
religious group was the presence of low E. A common factor in the
Extrinsically religious group and the Indiscriminately Religious group is
high score on E. Consistent with the original hypotheses, the presence
of a high Extrinsic religious orientation appeared to be related to
less-healthy psychosocial characteristics and personality functioning.
From the four religious orientations, two separate groupings were
formed. First, two groups were formed on the basis of scores on I: high
Intrinsic group (n - 94) and low Intrinsic group (n - 90). None of the
t-tests for differences between two group means was significant for any
of the eight stages of the AAAP. Two groups were formed on the basis of
scores on E: high Extrinsic group (n - 90) and low Extrinsic group (n -
94). When t-tests for differences in group means were calculated, five
of the eight comparisons were statistically significant, and two others
approach significance (p < .10). Table 4.8 summarizes the mean stage

scores, standard deviations, and the t-test and its significance level

108

for the eight AAAP stages. The differences between the high Extrinsic
and low Extrinsic groups for Stages 2, 3, 4, 5, and 7 are statistically
significant. In each case the low Extrinsic group scored better (higher)
on the AAAP stage score. For two other stages (6 and 8), the differences
are not statistically significant, but approach significance (.05 < p <
.10) and are in the same direction, that is, the low Extrinsic group mean

scores were higher.

Samara:

This chapter has presented the results of the analyses for the
hypotheses originally proposed for this study in Chapter 3. Additional
analyses were performed post hoc which considered questions of the
validity of the AAAP as a developmental measure, the ROS scales in
relation to subjects' demographic information, the relationships of
conversion, disillusionment, and profound experiences to ROS scores and
AAAP stage scores, the four-fold religious orientation groups' rank
orderings and the generation of analyses of high Extrinsic and low

Extrinsic groups.

109

 

 

0.... 3.0.. 0.2.5.6 30..

3. v a m... 8.0 ~04... 0.8.38 i... 0 83m
~o.~. 3.8 0.2.33 30..

8. v a 3.. 8.... .30 0.8.3.... i... .. 83m
3.». 3.8 0.2.3.... :3

2. v a 3.. 3.... 8.8 82.3.... 5... o 83...
a... 2.8. 0.2.3.6 :3

n... v a 5.... an... 0...... 0.8.3... i... m 88....
2.5 8.8. 0.2.3.6 :3

.8. v a ..: 3...... 2.8. 0.8.5.3 5... e 88...
k.» 8.3. 0.2.00.6 30..

3. v a 8.... mm.» 9.... 0.8.3.... i... n 83...
5.... .06.. 0.8.3.0 8..

8... v a 3.... .5. ~33. 22...... 6... ~ 83...
no.» 33¢. 0.8.00.6 30..

2. A a 8.. 3.0 8...... 0.8.3.3 8... . 83...

«98:8... 8.2.-. 8.8.2.... IS»:
.0 .03.. 2.8.8

 

090.5 0.8.086
so. use 0.8.00.6 an... 0..» E 00030 .2: 22m 0... ..o. 0.26.. 00:33.83 .80 .0032... 65.00.30 Eons-um .050:
a... 030..

CHAPTER V
SUMMARY AND CONCLUSIONS

In this chapter a summary of the study is presented. The limitations
to the study are discussed. The major conclusions based on the results
obtained in this study are listed and discussed. Finally, implications
for future research and improvements in the design of this research

project are discussed.

0 he tud

This study was a descriptive study attempting to understand the
relationships between adult intrinsic and extrinsic religious
orientations and the psychosocial development tasks of Eriksonian theory.

A body of theory and research over the past 25 years has delineated
differences between an intrinsic and an extrinsic religious orientation.
These orientations are different ways of being religious or attitudes
toward one's personal religious faith. An intrinsic religious
orientation views religious faith as an integrative, master motive to all
of life, giving meaning and purpose to life and unifying the self.
Intrinsic religiousness is autonomous, inner directed, tolerant, self
transcending, personal, and committed. In contrast, the extrinsic

religious orientation is more self serving and utilitarian. Extrinsic

110

111

religious orientation leads to a more compartmentalized religious life
from the rest of the person's life. The extrinsic orientation tends to
be dependent, defensive, security seeking, and prejudicial. Religion for
the extrinsically oriented is useful as a means to other ends.

Previous research by others with the ROS had indicated that the
extrinsic scale was positively correlated to a variety of less desirable
or healthy psychological traits (i.e., anxiety, prejudice, external
motivation) and that the intrinsic scale was positively related to some
psychologically desirable traits (i.e., self-control, sense of well
being, responsibility) and uncorrelated or negatively related to
undesirable or unhealthy traits (i.e., dogmatism, prejudice, fear of
death). There was some research evidence that the intrinsic and
extrinsic religious orientations were related to other aspects of
personality development and considerable speculation about the
developmental nature of these orientations.

Building on this previous theorizing and research, the present study
considered the relationship of the two religious orientations as measured
by subjects' scores on the Religious Orientation Scales (ROS) to the
Eriksonian psychosocial developmental tasks as measured by the subjects'
scores on the eight stages of the Assessment of Adult Adjustment Patterns
(AAAP). A demographic sheet was developed for this study which asked
information from the subjects such as conversion experience; education
level; profound experience affecting one's view of self, others, and
life, which it was thought might be variables influencing the outcome on

either the ROS or AAAP.

112

A random sample of 373 religiously affiliated adults was obtained
from several Lansing area churches. These persons were asked to complete
and anonymously return the demographic sheet, the ROS and the AAAP. The
subjects were told that the study was concerned with normal adult
development and ways of being religious. Of those who received these
materials by mail, 184 completed and returned all three of the
instruments.

After the data were collected, the AAAP scores for each of the eight
Eriksonian stages were computed. The highest stage mastered in order on
the AAAP was determined for each subject. These were the dependent
variables in this study. Each subject's score on the I and E subscales
of the ROS were computed, and these scores were the independent variables
utilized in the analytic procedures. Subjects were divided on the basis
of their scores on both I and E into a four-fold religious orientation
classification. The four groups thus formed were Intrinsic (High 1, Low
E), Non-religious (Low 1, Low E), Extrinsic (Low 1, High E), and
Indiscriminately Religious (High 1, High E). The original hypotheses
were clustered according to whether they utilized the I and E scores for
each of the subjects or the four-fold classification of the subjects into
religious orientation groups.

Hypothesis 1 concerned the correlation between the subjects' scores
on the intrinsic religious scale and on each of the eight stages of the
AAAP. No significant relationship was found for Stages 1-7. For Stage 8
(wisdom), however, a statistically significant positive relationship was
found with I. A higher score on I tends to be related to a higher score

on the Stage 8 task of wisdom.

113

Hypothesis 2 concerned the relationship between a subject's intrinsic
religious score and the highest stage mastered in order on the AAAP.
This correlation was not statistically significant.

Hypothesis 3 considered the correlation between subjects' scores on
the extrinsic religious scale and on each of the eight stages of the
AAAP. Non-significant results were obtained for Stages 1, 3, 4, and 7,
although each of these correlations was in the predicted negative
direction. There was a statistically significant negative relationship
between a subject's score on E and score on Stage 2 (autonomy), Stage 5
(identity), Stage 6 (intimacy), and Stage 8 (wisdom). As a subject's on
E increased, there is a tendency for that subject to score lower on
Stages 2, 5, 6, and 8.

Hypothesis 4 concerned the relationship between a subject's extrinsic
religious score and the highest stage mastered in order on the AAAP.

This correlation was significant, indicating that as a subject's score on
E increased, there was a tendency for that subject to master in order
fewer of the AAAP stages.

Hypothesis 5 considered the four religious orientations groups and
the mean stage score for each of the eight AAAP stages. Statistically
significant differences across the four religious orientation groups were
found on Stage 2 (autonomy) and the differences approached significance
(p - .06) on Stage 4 (industry). On both of these stages, the results
were in the predicted direction with the Non-religious and Intrinsic
groups scoring higher than the Extrinsic group which, in turn, scored

higher than the Indiscriminately Religious group.

114

Hypothesis 6 concerned the differences among the four religious
orientation groups and highest stage mastered in order. The results were
not significant.

Several post-hoc analyses were performed to provide additional
understanding of the results obtained for the original hypotheses and to
explore additional questions which emerged as the data were analyzed.

A cluster of post hoc analyses were performed utilizing the AAAP.
Several of the analyses using subjects' ages and AAAP stage scores found
that younger age groups in this sample had higher mean stages scores than
the oldest age group. In addition, it was observed that a large
percentage of this sample had only mastered through the earliest stages
associated with infancy and childhood.

The demographic variables were considered in relation to subjects'
scores on I and E. I correlates with the other measures of religiousness
and with a tendency toward more conservative theology. E is unrelated or
negatively related to these other religious measures. Neither E nor I is
related to social desirability or educational level. Women score higher
than men on I, while there is no gender difference on E. Both I and E
were found to increase with subject's age.

Presence of a religious conversion experience appeared to lower the
stage scores on the AAAP for Stages 1-7, although only the differences on
Stages 3 and 7 were statistically significant. For subjects who reported
they had had religious conversions, E scores were lower and I scores were
higher.

The reported presence or absence of an experience of disillusionment

with one's personal religious faith did not affect the subject's score on

115

E and I or on Stages 1, 3, 5, 6, and 7 of the AAAP. However, subjects
who had gone through a period of disillusionment with their religious
faith scored higher on Stage 2 (autonomy), Stage 4 (industry), and Stage
8 (wisdom).

Subjects' report of an experience profoundly affecting their view of
life did not affect scores on E or I. On the AAAP stage scores, subjects
who reported profound experiences scored higher on Stage 5 (identity) and
Stage 8 (wisdom) than those who did not report such experiences. No
significant differences were found for the other AAAP stages.

For subjects who reported both disillusionment experiences and
profound experiences affecting their life, scores on E were significantly
lower, and scores on Stage 8 were higher than those who reported only one
of these experiences or neither of them.

When two groups were formed based on subjects' scores on E, the group
that was low on E scored better (higher) on all of the eight stages of
the AAAP than the High E group, with five of the eight comparisons
statistically significant. No significant differences were observed when

High I and Low I groups were formed.

W
W

As the data for this study were being analyzed, several results taken
together seemed to call into question the use of the AAAP as a
chronologically sequenced psychosocial developmental measure. These
results will be discussed relative to their implications for the

construct validity of the AAAP.

116

When two age groups were formed, the younger age group was
significantly higher than the older group on Stages 2, 4, 6, and 8. When
three age groups were formed, the mean stage scores for Stages 2 and 6
were significantly different. In each of these comparisons, the oldest
group had the lowest mean stage score. What may be being observed in a
lower Stage 2 score is a loss of independent and autonomous functioning
by older adults, as some of them become more dependent. The differences
in Stage 6 (intimacy) scores may also reflect for the oldest adults the
loss of their life partners to death and a lower stage score due to
non-endorsement of these stage items on the AAAP. The construct validity
of the AAAP's Stage 6 (intimacy) was already called into question by
Azar's (1982) results finding no difference between a normal and a
psychotic population. The difference on Stage 4 (industry) may reflect
changes in the productivity level of an older person or a lessened
importance of work -related task accomplishments or a lowered sense of
competence due to aging effects. The difference on Stage 8 contradicts
the Eriksonian notion of wisdom and the acceptance of one's life cycle as
a developmental task of older adults.

The results of this study add further evidence that the AAAP Stage 6
is of questionable validity in the context of a measure of psychosocially
related developmental tasks. Beyond this, however, the age-related
differences on stage scores raise questions about the usefulness of the
AAAP as a developmental measure.

Table 4.5 records the rank order of the three age groups for the
means scores for the eight AAAP stages. Though only two of these rank

orderings are statistically significant, the pattern of these rank

117

orderings is interesting and suggestive. It is difficult to explain
these results based on Eriksonian theory. As previously noted, the Stage
8 results are opposite of what would be predicted by the theory. Stage 8
is the task of older adults, and it would be expected that they would
score higher on that stage and that younger adults would be lowest. For
this sample the youngest adults score highest and the oldest adults score
lowest. It would also be predicted by the theory that a sample of
healthy older adults would manifest mastery of a greater number of the
developmental stages in order than younger adults, but, again, this is
not the case for this sample.

It is possible that what is being observed is a cohort effect of this
cross-sectional research design, in which older adults at present will
score lower on the AAAP and that if this study were repeated in 30 years
the older adult group at that time would score much higher across the
stage scores. It may be that what occurs developmentally is a peaking of
stage scores (and their related developmental task competencies) more
toward middle life with a gradual diminution of those stage-related
competencies over time. Eriksonian theory does not consider this
possibility, particularly for psychologically healthy development across
the life span. Rather, developmental theory would posit that basic life
tasks of trust, industry, identity, and the like once accomplished are
typically maintained and become the building blocks for later
developmental tasks.

In Table 3.14, a summary was presented of the mean stage scores for
this research sample and several other studies. Examination of these

averages across the studies demonstrates a high degree of similarity.

118

The samples include adults from age 19 to 69 (Azar, 1982) and a sample of
college-aged students, aged 18-22 (Valdez, 1984). The stage score means
for the college-aged sample are not particularly different from those of
the other, broader age-span samples. The college-aged sample, according
to the theory, has not yet struggled with the crises of Stages 7 and 8
and has only begun to work through the issues of Stage 6. How is it that
their Stages 7 and 8 scores, then, are so similar to those of a.
broader-based and older sample? This is a similar question to that which
was raised above in discussing this research sample and the observation
that young adults scored higher than older adults on several stages.

When we consider the concept of mastery of the stages using the
procedure developed by Azar (1982) and discussed in Chapter III, there
are additional challenges to the use of the AAAP as a psychosocial
developmental measure. Table 4.6 lists the AAAP stages and (a) the
percentage of subjects who mastered each stage regardless of the
chronological order and (b) the percentage of subjects for whom the
particular stage was the highest stage mastered in order.

Considering the mastery of any stage regardless of the order, several
observations appear contrary to the theoretically expected outcome, if
the AAAP is indeed a developmental measure. The sample for this research
was drawn from a normal population assumed to be psychologically healthy,
and yet 16% of this sample did not master any of the AAAP stages at all.
It would be expected from Eriksonian theory that fewer persons would
master the higher stages, since successful completion of later
psychosocial tasks is based on successful resolution of earlier crises

and the development of earlier competencies. However, fewer of this

119

sample's subjects mastered Stages 2, 4, and 5 than mastered Stages 6 and
7.

When consideration is given to the mastery of stages in order, the
evidence from this sample does not support the utility of the AAAP as a
measure of developmentally sequenced tasks. Eriksonian theory would
predict that a decreasing number of subjects would master successive
stages and that a population of normal adults would have a large
percentage of persons who have completed the issues related to childhood
and adolescence. However, for this sample 67% only mastered up to Stage
1 (trust), and 86% only mastered through Stage 4 (industry)--both tasks
of childhood, according to Eriksonian theory. Said somewhat differently,
this would mean that a large portion of the subjects in this sample was
dealing unsuccessfully with psychosocial tasks from childhood.

The preceding analysis leads this researcher to question the use of
the AAAP as a chronological developmental measure. The results with this
sample do not support this use of the AAAP; and, based on the observed
similarities with other samples using the AAAP, it would be expected to
find the same phenomena with those samples as well. The AAAP clearly
needs refinement and modification before its continued use as a research
tool and developmental measure. The directions that this additional
research might take would include the establishment of age-related norms
for each AAAP stage; modification of the mastery levels; longitudinal
research to assess any cohort effects; reexamination of the items
utilized for each stage for their content, relevance to Eriksonian

theory, and relationship to each other.

120

Having discussed this major limitation in the AAAP, it is necessary
to ask what is being measured by the stage scores of the AAAP. The scale
reliabilities for each of the stages are very high (see Table 3.14), and
the item-to-total scale reliabilities are relatively high (Azar, 1982),
These scale properties would suggest that the stage scores are measuring
some characteristic personality trait or psychosocial task. Rather than
measuring the sequential acquisition of these characteristics, this
researcher concludes that these Stage scores can be understood better as
measuring aspects of a normal or healthy adult personality. It will be
in this way that the subsequent discussion of results will be treating

the stage scores.

Wham

There are several other important limitations to the present study
which must be kept in mind as one considers the results of the data
analysis and attempts to draw conclusions about the meaning of the
results.

1. The sample of this study was composed of adults who were
religiously affiliated. The sample was somewhat narrow. It
was an essentially all white, Protestant, middle-class sample,
in general better educated than the general population of the
country, state, or local area. Conclusions reached from this
study's results apply only to this type of population: white,
middle class, and Protestant church affiliated. Having said
this, it is nonetheless true that the results reported for

AAAP scores and ROS scores of this sample compare favorably

121

with results from other research studies summarized in
Chapters II and III on diverse populations.

A major limitation to these results and their interpretation
is the cross-sectional nature of the design of the study.
With a cross-sectional study, the differences across the age
groupings may be cohort effects, reflecting the cultural and
historical situation of that age group. For example, the
generally lower AAAP stage scores or the increase in E scores
observed in the older age grouping of this sample, may be a
cohort effect. Our culture may be promoting better
psychological health and a more healthy religiousness so that
when the younger age group becomes the older age group (a
generation from now), they will score higher than the older
age group of this study. There have been no longitudinal
studies with either of the major instruments of this design
and without longitudinal data one can never be certain that
any observed differences are not a cohort effect. Age-related
results must be considered with this caution in mind.

The study is primarily descriptive in nature. In this type of
survey research, specific statements of causation are not
possible, although descriptions of apparent effects of
variables on other variables are possible.

The correlational statistics for the original hypotheses,
though considered statistically significant, are not
especially large. The relationships between variables in

those cases may not be particularly meaningful or

122

interpretable. When this is the situation, collaborative
evidence must be found to support specific conclusions and/or
any conclusions drawn must be qualified.

A relatively large number of different types of analyses were
conducted after the original hypotheses were tested, and few
of these post hoc results were statistically significant.
There is an increased probability of a Type II error as the
number of analyses increases. That is, it is possible that
some of the significant results are the result of chance and
do not represent true significant results. Alternatively, the
small number of significant results may be related to the
measurement problems and validity considerations which were
discussed concerning the AAAP, or they may be related to the
sampling procedure utilized and/or some unknown measurement
difficulty with the ROS scales or four-fold typology.

Whatever the reason, the fact that few of the analyses reached
a statistically significant level requires caution about the
interpretation of the results and the conclusions drawn from
the study.

Having stated that few significant results are a
limitation to this study and its results, it is nevertheless
the case that none of the results obtained either for the
original hypotheses or for the additional analyses when
utilizing ROS and AAAP scores were significant in the opposite
direction from that which was predicted on the basis of theory

or previous research with the exception of the age-related

123

considerations with the AAAP. While one must be very cautious
about drawing conclusions on the basis of “trends" in the
pattern of results or results that ”approach significance,"
the pattern of results from this study do conform to the
general pattern of expected results.

A final limitation of the present study was in the general
formulation of the underlying theory and original hypotheses.
The theoretical and research underpinnings to the present
study did not allow for a detailed specification of the
relationship between the independent and dependent variables,
particularly when consideration is given to the Eriksonian
stages. The broad sweep of the hypotheses was not able to
take into account either individual differences or nomothetic
changes across the life span on the manifestation of either
religious orientation or successful resolution of the
psychosocial developmental tasks. The present stage of
research in the psychology of religion in general and the
developmental sub-specialty of the psychology of religion in
particular does not have a sufficient body of knowledge upon
which to be more specific than the early chapters of this
dissertation; however, the general nature of the hypotheses
did require some additional attempts to analyze the results,

to explain the findings, and to search for relationships.

124

W

The results of this study indicate that intrinsic religious
orientation is unrelated to the specific normal or healthy psychosocial
characteristics which are measured by the AAAP. This result could be an
artifact of the measurement difficulties discussed regarding the AAAP.
It may be that scores on the stages became distributed in such a way that
differences between subjects are flattened out at the healthy end of the
scales. The scope of this study does not allow for the exploration of
measurement issues. 1

However, a reexamination of the original hypotheses regarding a
positive correlation between I and each of the stages of the AAAP would
lead to a different prediction of the relationship between the two sets
of variables. As stated, the original hypotheses l and 2 imply that
higher I is positively related to better than average, that is, more
psychologically healthy personality characteristics of trust, autonomy,
identity, and so forth. Presence of High I would be expected to be
related to a higher than average functioning person. Intrinsically
religious persons according to the hypotheses were expected to do better
than the general population of adults. When the previous research
utilizing ROS is re-considered it shows that most studies have drawn
comparisons between different subgroups of religiously affiliated or
religiously identified populations. So the comparisons were between
religious persons' scores on I and E and some other variable(s) of
interest and not between I and a normal population not identified or
classified as to degree of religiousness or church affiliation (Meadow &

Kahoe, 1984; Donahue, 1985a; Bergin et al., 1988). Across many of the

125

studies summarized in Chapter II, E was positively correlated to less
desirable or healthy characteristics, and I was often uncorrelated with
either less desirable traits 2; more healthy traits. In this regard,
then, non-significant results with I are consistent with previous
research utilizing other independent variables of interest.

In addition, the nature of the correlational statistical procedure
meant that the I score was indirectly confounded with the subjects' score
on E. Persons could score high on both I and E. These subjects,
identified as Indiscriminately Religious, would be expected to be the
least healthy and to be lowest on the stage scores, which was generally
the case. These subjects' lower scores would tend to cancel out any
positive correlation between high I and high stage score, if this
relationship existed for subjects with high I and lower E.

On one stage of the AAAP, there was a significant positive
relationship with I. The higher a subject's score on I, the higher that
subject's score on Stage 8 (wisdom). These two scales and the constructs
they are measuring appear to be related. An item analysis of both Stage
8 items and I items did not reveal similar content, which would have then
explained this relationship.

It seems intuitively that an intrinsic religious faith and Stage 8
(wisdom) would be similar or related. Intrinsic religious orientation
represents an attempt to integrate one's faith into one's life. For this
person religious or personal faith is a way of living, of viewing life,
of making sense of the world. Faith, God, religion are regarded as
personally relevant. It can be said that with an intrinsic orientation

to one's religion, one has a philosophy of life. I may be associated

126

with the development of the wisdom of Stage 8. Though an intrinsic
orientation may be associated with a more rigid or inflexible person,
conservative theologically and even indiscriminate in religious
preferences, it still issues in a relatively consistent orientation to
life and the events of life. So high I may help a person to make sense
of life and its varied events. Alternatively, it may be that persons who
have a general sense of the meaning of their lives and an acceptance of
their life processes may be drawn to a more intrinsic expression of
religious faith.

When consideration is given to an extrinsic religious orientation by
itself, a different picture emerges. For both the original hypotheses
and the post hoc analyses, more statistically significant results were
obtained. The results from this study were consistent with previous
research in which E was related to less healthy or desirable traits
(Headow & Kahoe, 1984; Donahue, 1985a). In this study, E was negatively
correlated with scores on Stages 2, 5, 6, and 8. While it may be that
these results and/or the non-significant results are related to some
unknown measurement problem of the AAAP, such as the variability of
subject responses and the range of scores on a particular stage, there
are theory-based explanations of these significant results that make
sense.

E is described as a more dependent and immature religious
orientation, and this may contribute to a less independent personality,
one in which the individual struggles with the will to be oneself.
Persons who are less autonomous, less individuated may not be able to

develop a more committed faith which requires them to stand alone. It

127

may be that E is related to the less healthy (and frequently
caricaturized) aspects of some very religious persons and denominations:
self-punishing, self-critical, experiencing shame or guilt and doubts
about oneself and one's worth. I, by contrast, would represent a
movement beyond these primitive intrapunitive measures to
self-acceptance, forgiveness, and empowerment.

Relative to Stage 5 (identity), extrinsic religiousness appears
related to a lowered self confidence in both the internal experience of
the person and in the interpersonal realm. Extrinsic religiousness has
been described as defensive, security-seeking, prejudicial, intolerant,
utilitarian, uncommitted. This sounds much like a description of a
person of low self concept and with little comfort about his/her
identity. A lack of clarity about oneself may lead to either hesitation
and uncertainty in one's commitments and involvements (religious or
otherwise) or else chargability about one's participations and
involvements. Both of these are characteristic of higher E.
Additionally, it may be that lacking a clear sense of self leads to the
adoption of a looser, less-coherent religious philosophy.

Significant results for Stage 6 must be very cautiously discussed
because of the discriminate validity problem of stage 6 on the AAAP found
in the original research. Since Azar (1982) found that Stage 6 scores
were not different between a normal and a psychiatric population, any
conclusions regarding Stage 6 must be qualified. Having said this, the
negative relationship between high E and lower score on Stage 6 was
consistent with that predicted on the basis of theory. Persons with high

E may tend to begin less satisfactory relationships with others or create

128

them out of their own neediness. Extrinsic religious orientation tends
to be more egocentric, self-serving, self-absorbed, and manipulative.
Clearly these traits are unlikely to facilitate a warm interpersonal
bond. Building on the results discussed above for Stages 2 and S,
developmental theory would posit the need for a sense of independence and
identity before one can enter into an intimate, long-term, faithful
relationship with another person. One must be in relationship to oneself
before one can be in relationship to others. Stages 2 and 5 are
concerned with intrapersonal and interpersonal competencies and
unsuccessful resolutions. These would be predictive of unsuccessful
resolution on Stage 6. Interestingly, the non-significant results of
Stages 3 (initiative) and 4 (identity) are consistent with and supportive
of this interpretation. Those two ”stages" have to do with tasks related
to a sense of competence in work and productivity not so much in
interpersonal relationships.

Stage 8 (wisdom) was found to have an inverse relationship to E. As
E becomes higher, score on Stage 8 tends to become lower. E is a more
fragmented or compartmentalized approach to religion, such that personal
faith is not a motivator in the individual's life. Extrinsic religious
orientation is also less reflective or introspective and related to a
less examined life. So that a lack of a more committed, individualized,
personalized religious faith may lead to a reduced ability to develop a
meaningful philosophy of life. Alternatively, it may be that the lack of
a sense of coherence about life in general, a lack of acceptance of one's
life course may lead to a more detached, compartmentalized, extrinsic

religion. As with the discussion of the previous significant results,

129

these two aspects of personality appear to be intertwined and mutually
reinforcing.

Using the four-fold typology, only one comparison was statistically
significant. On Stage 2 (autonomy), the Non-religious and Intrinsic
groups scored higher than the Extrinsic and Indiscriminately Religious.
These differences were in the direction predicted on the basis of the
religious orientation theory. Several of the different analyses using
Stage 2 were significant, and this may indicate that as a measurement
issue, ”autonomy" is easier to operationalize than some of the other AAAP
characteristics. I requires a certain level of independence, a capacity
to choose for oneself and the ability to act on one's choices. Intrinsic
religiousness could develop out of a personality in which Stage 2 issues
were successfully resolved. Persons who are intrinsic in their religious
orientation would have ways to deal with failures that would not lead to
a sense of shame, guilt, or self-doubt, but could be viewed as
self-renewing, since mistakes and failures are forgiven and acceptance is
unconditional in an intrinsic orientation. It could be argued then that
I would tend to promote personal responsibility and autonomous

functioning.

921321211211;

The major conclusions to be drawn from this study are the following.
1. Intrinsic religiousness is relatively unrelated to
normal/healthy psychosocial development as described by

Eriksonian theory and measured by the AAAP.

130

Intrinsic religiousness does appear to be positively related
to the task of Stage 8 (wisdom). Intrinsic religiousness
tends to be related to higher scores on Stage 8 of the AAAP.
Intrinsic religiousness is an attempt to integrate the
personal faith and beliefs into the whole of a person's life
and experiencing and is, therefore, consistent with the tasks
of developing a philosophy of life, finding meaning and
purpose in life, and accepting one's life cycle.

. Extrinsic religiousness by itself is not related to the
psychosocial tasks or traits of basic trust (Stage 1),
initiative and purpose (Stage 3), industry and competence
(Stage 4), and generativity (Stage 7).

. Extrinsic religiousness considered by itself appears to be
negatively related to the psychosocial tasks of autonomy
(Stage 2), identity (Stage 5), intimacy (Stage 6), and wisdom
(Stage 8). The greater the presence of an extrinsic
religious orientation, the more likely that a person will
experience guilt, self-doubt, and self-punishing attitudes;
lack of clarity about the self; difficulty in or less
satisfying relationships with others; and a sense of lack of
clarity about life's meaning and about a guiding philosophy
for one's life.

. Persons who have high extrinsic religious orientation are
overall less psychologically healthy than persons who are low
extrinsic. The evidences for this conclusion include not

only the results for Hypotheses 3 and 4, but also the post

131

hoc analyses with High E and Low E groups. This negative
relationship between a highly extrinsic religious orientation
and nearly all of the healthy characteristics of the AAAP
allows the conclusion that this is a generalized phenomena.
If the stage tasks are somewhat related to chronological age
of ascendancy of the crisis and mastery or resolution of that
crisis, than one can conclude that extrinsic religiousness is
related to less healthy overall psychological or personality
development. ‘At the very least, high extrinsic religious
orientation is related to less healthy psychological
functioning than low extrinsic religiousness.

. There is no evidence that intrinsically religious people (or
any other kind of religious people) are more psychologically
healthy than non-religious people.

. Intrinsic and non-religious persons appear to do better on
the psychological task of autonomous functioning (Stage 2).
These two types of persons are, therefore, probably less
concerned with self doubts, feelings of interpersonal
inferiority, and lack of a sense of power, will, and
capability.

. There is some evidence which would indicate that in general
intrinsic and non-religious persons enjoy better
psychological functioning than do extrinsic and
indiscriminately religious persons across all of the

psychosocial characteristics studied.

10.

132

Both self-reported intrinsic and extrinsic religious
orientations show a tendency to increase with age. This may
be a cohort effect of this cross sectional study, which might
disappear if a longitudinal study were done tracking the same
cohorts' changes in I and E over time. The findings of this
study contradict the proposal of Gorsuch and Venable (1983)
that E might decrease with age and I increase with
chronological age. This study does not provide support for
the idea that I and E are chronological developmental
phenomenon. There is evidence that I and E are
differentially related to other personality characteristics
and measures of psychological health. It may be suggested
that I and E are related to pervasive personality bases
emerging in experiences and intra-psychic processes during
the course of psychosocial development. So that for adults,
E is positively related to a less healthy or perhaps more
dysfunctional personality. In contrast, I by itself is not
related to psychological functioning or health except for the
influence it has on the development of a coherent guiding
philosophy of life.

Conversion experience correlates positively with higher I and
lower E, perhaps because of the personal nature of the
experience of the believer with God that characterizes most
conversion experiences. I and E are not affected by either
presence or absence of (a) a reported experience of

disillusionment with one's personal religious faith 2; (b) an

11.

133

experience profoundly affecting one's view of self, others,
and life.

Both a report of a period of disillusionment with one's
religious faith and report of an experience profoundly
affecting one's life are related to a higher Stage 8
(wisdom), to higher I and lower E, and to a generally
healthier functioning personality.

Persons who report one or the other or both of these
experiences tend to have a better resolution of the principle
tasks of later adulthood: developing a guiding philosophy of
life and accepting one's life cycle. A period of
questioning, scepticism, and doubts about one's religion can
lead to a deeper commitment and a more meaningful, integrated
personal faith, which then forms a set of values and
principles for living. This intrinsic religious orientation
can contribute to a sense of life having ultimate meaning and
personal satisfaction and to a more clearly articulated sense
of acceptance of one's life, that is, to wisdom.
Alternatively, a carefully considered philosophy of life
(task of Stage 8) can facilitate a person's movement through
a period of disillusionment or through a profound experience
endowing those experiences with significance and enhancing
overall subsequent development. Clearly, a life that is
reflected upon and which derives value and meaning from its

experiences will likely be more integrated, committed,

134

healthy, and in a traditional sense ”wise." A self examined
life is probably a healthier life.

12. In a broad sense, the question raised in Chapter I can be
answered affirmatively: there is a healthier way of being
religious (Low Extrinsic) and a less healthy way of being
religious (High Extrinsic). The evidence of this study both
supports and extends previous research in explicating healthy

and unhealthy religiousness.

W

The results of this study could be a source of encouragement to
religious educators, ministers, and other religious professionals, for
the results are consistent with much of what is preached and taught by
religious professionals regarding the value of a committed religious
faith which is integrated into a person's life and forms a motivational
core of the individual's identity and lifestyle. Host religious
traditions would support the notion that an intrinsic religious
orientation is more desirable than an extrinsic orientation. This
generalization is supported by the research in this dissertation.

Though specific causation was beyond the scope of the design of this
study, it can be said that a greater extrinsic religiousness is related
to a generally less-psychologically healthy personality. No conclusion
can be reached to decide whether the unhealthy personality
causes/predisposes the person to be more extrinsically religious or
whether the extrinsic religious orientation causes/predisposes toward a

less healthy personality. Despite this inability to settle this age-old

135

dispute, religious professionals can utilize the obvious relationship
between high E and lower mental health to encourage the development of an
intrinsic religiousness both for adults and children. Teaching,
preaching, and various learning opportunities which challenge E and
facilitate movement toward greater I are supported by this research.
Additionally, for religious professions to have background in human
development and the psychological sciences and to actively seek to
integrate the insights of psychology into their teaching, preaching, and
pastoral care may promote better mental health, better relationships,
better parenting, which in turn may foster the development of a more
committed intrinsic religious faith orientation. One further cautionary
note, this research does not address the question of whether by helping
an adult to score higher on I (that is, become more intrinsic) that
person will become more psychologically healthy; nor that if one helps a
person become more psychologically healthy, then that individual will
become more intrinsic. Those changes would mean that these two phenomena
were not only related (a conclusion reachable on the basis of this study)
but were also causally and directionally related (a conclusion not
possible on the basis of this study).

As will be suggested below, the ROS has the potential to be utilized
by counseling professionals with religious clients as a rather crude,
differential diagnostic tool, if norms based on large, diverse samples
were available. At this point it is not possible to use the ROS in this
way, other than perhaps by a sensitive clinician using the pattern of

scores on I and E intuitively.

136

Clinicians do need to develop a sense of these two religious
orientations so that with religious clients a valuable diagnostic
distinction can be made between those clients who are High E and those
who are Low E. This differentiation can influence the therapist's way of
relating to the client, the degree of self-disclosure of the therapist's
own religious orientation and theological perspectives, and by suggestion 4
of the manner of handling the religious content discussed by the client.

As with other parts of a theoretical model of mental health and the goals

I

of counseling which intuitively guide any clinician, this research
supports a conceptualization of a more and less healthy religiousness.
The process of psychotherapy encourages self-reflection and
exploration of one's life experiences. Clinicians when they facilitate
personal reflection promote greater overall health and life
satisfaction. Disillusionment with one's religious faith or any life
event that has a profound impact on the client and which can be examined
in the therapeutic process can become integrated into one's outlook on
life and influence one's philosophy of life. Therapy can facilitate the
development of a greater sense of coherence, meaning, and purpose in life

and lead to a more mature religious faith.

1W

There are a variety of implications from this study for future
research. Several of these implications would represent modification to
the research design and methodology of this study, while others would

involve refinement of the two major instruments utilized in this study.

 

 

137

The sample for this study limited the generalizability of the
conclusions. A broader sample from a more heterogeneous population would
allow for applicability of the results to the general population. A
broader sample would be racially diverse, more socioeconomically diverse,
and involve a greater number of denominations. Use of a population not
identified as religiously affiliated would likely produce a broader
spread on the ROS subscales. The present sample of Protestant,
religiously affiliated adults does not allow for generalizations about
persons affiliated with other religions, nor does it allow for
conclusions about non-affiliated religious persons or non-religious
persons.

Though a cross-sectional design has its advantages, it also has
limitations. There was a small positive relationship between both I and
E and age of subject. It is unclear from this cross-sectional study
whether this represents a cohort effect or a general trend toward
increased religiousness (of whatever orientation) as a person grows
older. There is a clear need to specify the nature of the relationship
between I and E and chronological age. A longitudinal study would allow
for the exploration of changes in subjects' manifestations of I and E, as
well as changes on the AAAP stage scores.

This study raised questions about the usefulness and validity of the
AAAP as a chronological developmental measure. Further research using
this instrument is necessary to further examine this tentative
conclusion. Larger samples from broader populations would be useful to
establish age-related and gender-related norms for each of the stage

scores. A reevaluation of the mastery levels for the stages and the use

138

and meaning of those mastery levels would be valuable as an aid to future
research. The AAAP is not without value as a measure of psychologically
healthy personality variables; however, additional normative data is
needed, and further examination and refinement of the psychometric
properties of the stages would make the AAAP a more valuable and useful
developmental measure.

Another study similar in design to this one could choose to utilize a
different measure of psychosocial development. A recently published
(1988) instrument based on Eriksonian theory of personality development
called the ugagg;gg_gf_ngghggggi§1_nggglgpmgnt by Gwen A Hawley, Ph.D.,
appears to have relatively good scale properties, norms for a large
sample (n - 2,480), and construct validity established through a series
of multitrait-multimethod analyses. An instrument with more clearly
established construct validity and with age-related norms would provide a
basis for more specific conclusions about the relationship between
psychosocial development and religious orientation.

The ROS has continuing value as a tool in research regarding the
psychology of religion. The ROS could potentially be used as a brief
screening device by counselors, chaplains, ministers, and other
professionals concerned with mental healthy and religious or spiritual
issues. The ROS has the potential to differentiate healthy and unhealthy
religiousness and be used as an indicator of overall relative
psychological functioning. This potential is only realizable if norms
are established for the ROS subscales with diverse papulations and
denominational affiliations. Age- and gender-related norms are necessary

to determine cut-off scores for both I and E. Continued standardization

139

of research methods and scoring procedures with the ROS will allow for
valid comparisons across studies. Researchers must report various
statistics for the E and I scales so that normative data can be developed
and generalizations across studies can be made.

The four-fold typology of religious orientation groups does not
appear to have explanatory value, at least with these independent
variables. This may be the case for other studies. The use of a median
split of the sample on the basis of E appears to generate more meaningful
comparison groups (High E and Low E). Using a four-fold median split may
require elimination of subjects falling around the median of both I and E
to create what might be considered "purer" religious orientation groups.

Additional research using other measures of psychologically healthy,
personality variables, or psychopathology could build on the
generalizations from this study concerning the religious orientations and

psychological health.

APPENDIX

Dear Friend,

We are writing to encourage you to participate in a scientific research project
being conducted by Jay M. Terbush, a doctoral candidate at Michigan State
University. The purpose of the research is to study normal adult development
and attitudes toward religion. At a recent meeting of our Board of Deacons,
approval was given to Jay to approach persons who are connected with our church
to ask for your help. Your name was selected at random from our mailing
list--no one from our church knows to whom this letter is going.

This is an important study because, unlike much other research that looks at
religious faith and mental pggblggg, this study is looking at normal (mentally
healthy) adults who are church affiliated. Your responses are very important
as part of the study, whether you are very active in church or not, whether you
consider yourself a religious person or not very religious.

In order to participate, you are being asked to fill out the following three
items in the enclosed booklet: (a) a Data Sheet, (b) the Assessment of Adult
Adjustment Patterns-~which asks you questions about many aspects of adult life,
and (c) the Religious Life Inventory--which asks your opinion about a variety
of questions about religion.

It will probably take you about one to one and one-half hours to complete these
forms, which ask you for your own opinions about many aspects of your life. We
encourage you to complete them within the next week. When you have finished
all three sets of questions, please mail the booklet back in the enclosed
stamped, addressed envelope.

All responses to these questions will be entirely anonymous. We ask you no; to
put your name on any of these forms. Of course, your participation is entirely
voluntary. You may choose not to participate at all or to refuse to answer
certain of the questions. Though your participation would be extremely helpful
to this research, there will be no way for anyone to know whether you
participate or not.

You indicate your voluntary agreement to participate by completing and
returning these questionnaires. If you choose not to complete these forms, we
ask you to please mail them back anyway so that they can be reused.

Although no direct benefit to you can be expected from filling out these
questionnaires, we think that the questions and thinking about your answers
will be interesting to you. The researcher will be offering an adult seminar
to all the members of our church to discuss the research results and
implications for adult development and faith. If you have any questions or
concerns about participating in this research or would be interested in the
results of this study, please contact Jay Terbush at (517) 224-3440.

Thank you very much for participating in this important research.

Sincerely,

Church Official Jay H. Terbush, H.A., M.Div.

140

 

141

.0... 50> 00.00.: 0.

so... ..:0 pecans. 00.... 3.300.. 3.0.... .000 00.3 00 60> .—

2. 00>

2.000 300.000.... 30.020. ..o 0.020.. ..:..> c. 00:0...0 00.0.- 0 30:07.09;
..:..000 ER... :03 ..0... 005.00on 5.0.0280 30.2.0.— 0 00.. 30> 2.0..

..00: 0 00:0 :05 0.8.-
..00... 0 00:0

5.8.- 0 00s.. 00.20 .3 030
5:8. 0 00....

23> 0 00.0.. 3.3 3.05.0003
23> 0 00....» n2... >.0¢030...c.
00.6: ..o 30.0..

 

20.5...0 30> .0 000.200 020.3: 9.020 30> 0.. 30:02.0... so:

..< 00

030.030.. 2.0.2.0..
>..0> 000.03 00..
n o n 0 n ~ .

.333 0c: 05 ca .005: 05 0.0...0 0000... 30> 0.0 2.0.3.0.. so:

v0.3.3... >..0>
3.00.0.2. 30.002.9-
00.033... 30530
3.00000... .9.

.5330. c. 000.00... 59> .o 00.30.. 05 0. 00.5

.2.

.2.

.395 <_—<o

 

"3.003. 0050 III.
.28.. l.
082...... II.
03003005“. I
70:00:03 I
3:33... I.

"2.80.0002. .80 0.030.. 30.020.

30> .3. 009.0 ao> 3:0... .000. 00.1.. 3.00.5.8 10.00.00... .0800: .

 

 

..0a00.

".3508 03> .o 5.00:...- .0.8.00c.a0c00 .

 

 

"3.00:...0 15.00585... £0.55 .50» .

 

2.28.... 850 l

.I..00...00\00.£.
20.0.... 50.00! 0300..
.0226
2:82..

so... "00... .

00.50.. .93.:0303130000
00.30.. 02.0000:

00000.0 0.3.2.00-

.oofi. 08...}...8 82.. .II
.86. i... ll

.85. 82.. II

28.3080 .23. .85... .02.... .26. .8288. .

"00¢ .

0.060.

0.08 .

142

000.... .0 .0909. >...l. 0
0.... .0...000\00.000. 0.0000230. .000...00.0
000000 .0.0000.. .20...
.0800... >..:... 0.0.000...
.0 .003000 ...00 0. 02.003. .00.0>..0
.0 000. .0 0000... 3.5.00.0. 0...\.000...00.0
00.00.00. 03030.0 0.0.. 00.30....
am 000. 0.0... 3.030000 :00 0 0. 00.3...
003000 .0 ..00 >0 2.203.900 .0 000.
000... 0.8.2.0030 .0 0.000
3.00 .0 0.000
.00.00 .0 0.000
003000 .0 0.000
00.03000330000 ll.

.30> 0. >.000 .0... >00 ..0000 0000.0 «00.3020. 00. .0 >00 00000....on
30> 2.0.. .0509. ~. .000 00. 0....30 .000. .0 .0.... .0 000020000 .0000.

.30> 00.00..0 ..

:9. 0c. 0883.. .23 3.38.. 3.0.... IIIIIII .0... .2... .0 .0.... ..
0.. 00»

..0.0000 .050 0.... 00.00.3330. .30.. .0 .000000...00 .30> .0... .0 30...

.39. 3.00:0 30.50.80 >00 0.303 30> .0... .0.0000..0000 00 000 30> 2.0..

 

 

60.0.0... 30.2.0. 0.... .0 0.3.00 00. 0....0000 >..0..o

..: 00. 085.8. .30. ..II
80-00. 085.8. :00... I
30.0.. 080...... 052. II
.0...... 8.008.8- I

a. 8. 0.3.8. 086....» II

2.0.0 ..00000 >0050 .00..0I...000 ..00..00 00.300 .0920 >030 .00000.0

.9 ..0... 00.0.8. 030.....0. .0l.0. .30> .0 .008 03000. 39> 0.0 000 .0.) .0 .n.

.0... .30.. 00.00..0 ..

:00 0.0 00000000 .00: 0....0000 >..0..0 III «000 .05 .0 .00> ..
0.. 00»

0.0.0 .000..00.0 00.300 50> 6.030.. 0.30.3.0. .30> 0000:0030 >.030..00

 

 

..: .0 0.... 0000.03.30... 00.0000 30> 0.0.... 0... .30> 0. .500 0 000 39> 0.6.. .~.

143

.u..o..o

.co.uuouo.n >mu>uo>o >I oucoa..c. oca.uo.oo.ncou nao.o..o.
F nca .o..o. c.ou o» a. .o>o.n .o omoa.:a >Lae..a ocF .F~ n ¢ F uo. on oa:.o. . .cou.on «no.9..o. o no . gmaogu.< .oF
.co.unu.uos uco 35930;» aao.u..o. ouo>.oa .ooo.>.ou u:..:t
F c. 8.: .o 83.8.. Hon. 3 o... .3. 2.3.39. a. a. .8 o... >5 a... 32: no 5.8.5 .9822. ..:- 8.58
m e F :3... 3 {.3 05.- lo . :23 >3 . 953.3. of .o
.uc.on...oa u.aacouo
F at. .o.u¢a >3 uuouo.a ou .ov.o c. n.o..on nao.u..o. .0... o» guoocnno o.og:
>5 ou........§.ou 3 P3339. 3. .3... . 3.9.0.330 .2 n e F >3 ..:...on 0.. >23. 3!! o..- o.o..3 30...... >t .n
..gu.:gu .0. cu... >I anon. o.:uo.ou.. too. . .oF .uc.o. oc.>.a us» .0 ace .0 ouconoon
F m e F oz» .0 ego). >.cooa coon o>og . cou.o ou.ao .>
.>u.>.uuo .o.uou .o.coucou o o. go.:cu >I
F “on. .. co.a..o. c. uoo.ouc. >I .o. con-o. >L-l..n < .>F .0... .-.as a too. .
n c F no mac. 0: o>o..oa . «on: sun! on stag-I u.:ooou u. .o
.o... .o oc.coos oz“ anon. aca.uuoac >=os u.ozuco
F .. o-aouon on on acou.onc. >..o.uonuo .. co.o..o¢ .oF .0... .:.ouoon
n e F ac. >naoc - o.:uou on a. .o>o.n mo coongan on» .n
.>o.a ou ugnzou coon o>oc _ onaouon >..o.go >o.a . .nF
F .>u.::§iou
.2332! of c. 59.8 o 5.338 3 3.2. £5.33. ..3.
F .23. o :05 .22... .595 >3: o3; - Eo— n e F :5 a. of £0.55. a 0...... >1 so. coo-0.. 08 .v
0. .o.o.a v.30: . .n:o.o gu.::u o c.0— on o.o3 . F. .qF
.8 8.5.89.9- ..050 v...- .8235...
.go.:cu acouuo .a.caco~.u.u >3 no >o: anon oz» >.uooxo c. >uoouu
. . 58:33.... 3.3.2.5 B 3.83... .8 : .n. a .. . a... 38:8 ..: i 92. 8 I. 3:2. 83:... .n
.o... >.. c. 5...: 353.39... 9.9- >cas .o... c. .238... .23
F o.o o.o;u .oo. . .ca.u..o. >l c. o>o..on . gmaogu.< .~F m e F >l ..o 03:. .o>a co.u..o. >I >L.oo 0» too: >LF . .~
.8293233 .323 .83 ..:—..:. 253...}? u:- «39......
F ouo.:e.o. o» ouo.a a mo acou.oas. woos a. gu.:cu ocF .FF m e F cog: 3.02200 a. cool on «so..o co.o..o. was: .F
«mm mm... Ema «flu...
uc.um ac.um nagum uc.um

60.590 ‘8

..:o> 5.... .235 30> on... 2.3.09... a. a. 6.2.9.. 9.0.3 ..o 35... o... o... 9.2: .83.»: :30 50> 3 “mono... .35: 2: 38.9 69.03.: 39.9.3. 3. 3 8.:
>583... FF. 5.. 33¢ 9.2.... 2.2. 2... 2: ca .5633» 53 5.: «$68.93... .3 32.8....- ..:o> 3... coco... .. .33. ooofi Fo coco 5.: 8.53... 333 39!.
8.3 uco 5.: ooh... 3:3 392. 93... .5... u:- co.u..o.. v.33 83.3- .2}... «:3- 3! Zn..- coCoa o 5...... 35.-33o o..- 33. 8.33.0. of 3.5.5:...

wa<um zo_F<sz_¢O «DO—c.4wu

144

F...

F...

F3

F...

F3

F...

F3

F...

F...

F...

F0.

F...

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fn.

Fm.

Fm.

Fm.

Fn.

F~F

FNF

3.

F3

FNF

F3

2....

F~F

F~F

F~F

F~F

FF.

FFF

FF.

FF.

FF.

FFF

FF.

FF.

FF.

FF.

FF.

FF.

605.0500 .500. .0 :03 :0...- 0FF...0.. F

....0 F C0300
0 05.020 30.. :0 00:000.. 030.. F .30... >:0... 30F.

.>..§....00 >3 .0. 005... 0F. F

.:0.:.00 :0. 6.0. 0. .0....0 5 308...... 0.00.. .005.
>... .30... 52.5.35 .30 .530. 3.03.00 0)... F

630.3... >202... 050.. .673 F

3:00.000 :0 .0 05.0 0 .0
0:00. 0... .0 30...... 0 .0 0...: >8 3.03.5.0) 030.. F

683.... .3... 838.3... .6...» .

.0.-
o._FF >300. 0.0000 .0505 30.... 0. 0.. .0. >.00 ..:

68....
F...0...._ :0... 0.- 3 53.000. 0.0.. 0.0 0.0000

...0...0 >5
3.3.0.8 .000. ..:. 3:03.... 83.03. 30390 F

.05...>:0 .30... .03. 0550080 :0.....:00:00
0000 3:. 00 :00 F .03.-.00 8303... 0... :05

£03.80 >I ..0003. 0.. 0.00. 0.: F .030... F :25

.3.

.-

.20.

F...

F...

F3

F3

F.»

F...

F...

F3

F...

F...

F...

F3

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

 

 

 

.. >05... 03.. 330... 0.0000 .0... .0209 0... 05.00.05
.0300 0... 05.00.05 .05 .045: 0000. 3...... 50.00 ..5

F... Fm.

F~v FF.

>03... 03.. 330... 0.0000 .0... 03020.. F .F "a

0. >..

..:...3... 0... 0. .xo: .. .0... .095: o... 3030.... 3.... :0 5.3 9.03.... 30> 3....
6:03.02. 0.0... 3 3030:. 0:33 .0 .5... 0: 0.0 0.0.:
..0.0..00 5 0... 30... 30> 30.. a... 20.33 30... 30> 30.. 0.5.00... 3 09.0.30 0.03 3:03.... 0..—

..0~< .<... 60...: ..F... 60.06... .3... >0
mzawbp<a hawthgfiF—C p.33 00 hawtwmwmm<

F~F

F~F

Fwy

3.

F3

F3

F3

FNF

24..

F3

FNF

20..

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

.30.? .0. .30 .0...... x030 F .~F
..:0..0o.o Z. 5 30> 3 .500 . .F 3.... F .FF

£0. 3.00000
0... F0 .08....- 0 ..:...0 :00 F .3 .:03 F .F .9

....03 3 .080 05300 5 530...... 0.6.. F .0
60.0.20 0.... F .o

.993 0 F0 .5... :F a... F :0..3
09.3.0200 >0 5.3 v3.0.0 .00. >F.0..0:00 F .0.

..F.03 0000 on FF... :00
F .0......0000 .03393033 32.5.3 .0

.00.: 03... F 30.0... .0 ..:... 0... .. 3.. >2 .0

.6833...
3.8.3.. .8 .. 33.. .8 .8... 2... . ..

.:0..F.00 . :0
3.3.8.. 05.03.00 .00F..0.-.00F. 0:00 030.. F .n

69.200!
.00.... .0 3.55.00 0...... 3.0.8.00 F .~

.0000 F30. F .F

 

 

 

 

a... .0 03... .0: 33...... F...
0.. .0 0:... .0: 3.8.. ..0 .0.- .0 03.. .03 Fm.
ol .0 03.. >38: .0 .0.- .0 033 F3

08 F0 0?; 33.5.00 FF.

2.30.. 0.. 0. 0.0 .00.. :<

2.0.. .0 03.. 33.5.00:
.30.... 00.50.... 0... 053 I3. 5.0 .03....

£03.03 0:0 >:0 :0 0.... 203.. 00. 05.3.00. .0: £30.30 {03 .:00 35.000 30> 00 .38.. .0 00
.:00 30> .. >...0:0.. .. 9.5.31. 000... .0330 0.00...

umFBFFuwFFFo

145

F...

F...

F...

F...

F...

F...

F...

Fe.

F...

F...

F..

F...

F...

F...

F...

F...

F...

F...

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fn.

Fm.

Fm.

Fm.

Fn.

Fm.

Fm.

F~.

FF...

3.

3.

FN.

F~.

F~.

F~.

2...

F...

FN.

F~.

F~.

F~.

F~.

F~.

F~.

F~.

FF.

FF.

FF.

FF.

F:

FF.

FF.

FF.

FF.

FF.

FF.

FF.

F..

FF.

FF.

FF.

FF.

FF.

.0... ...:00
030.0 0 .0 .:0.. :. c0000. 0 9...... .0 .3030... 0.:

.3000. .3... 9.0.0 .00 .

60.5.0. 0......)

0 5... 0.. 00.00 F .35 5... 0:0 500. .. 0.
:5... F £0330 03.039... 0 .0 0.0.. .0. F :0...
6.0.00.0 030. 0. 0.... F

..F... .I :. 9 .5300 .0. F

0.00.00... .. >0. 0. 2....
F ..5 ..5 200...... .00. F .90.. . :. 0.. :0...

.90..
0 0.0.0.. .0000. o. 2.0.. F :0... .5... 0o... .0. F

..:0 50.. .:..0 >2

.5... .I .0 03.0 I. F

.>...0:.0..0 0:0 >5.2.0.30... :0... F ..0: .0. :F
3.0.. E. 0. 00....000 .00 F

.20....30 x0...
5.: 0... .00. 3.0.30. 30:000. .00.000.0 ..: F :05

0.0.. .0. ..00 00:0...
>0. 0.0000 3... 05 .0 0:0 g... ....... .0 .00... :F

.0... 5... 50.. 0. 0.... 0.0000
.00. 0.0000 .050 30.. 05:30... >..:0....0 0)... F

.30....- a..300 05 .00 3.0.. 0... 30.
0:0 $030.0 0... 0: .0.... 0:00.00. :05 FF... E00. F

.0.... .0 .30.. :0 :05 0.0.0
.0. 0.0.00.0 .5305 . :0 ..03 0. 0... .0. 0.0.. a. .F

.0. :5. F 00:0... 0)... F .39... .0 .00... :.

.3

.mm

...m

.3

Fe.

F...

Fv.

F...

F...

3.

F...

F...

F...

F...

F...

F...

F...

F...

F...

F...

F...

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fm.

Fn.

Fm.

Fn.

Fm.

Fm.

Fm.

Fn.

Fm.

Fn.

Fm.

F~.

3.

2..

FN.

F~.

F...

F...

F...

F~.

FN.

F...

F...

FN.

Fm.

F...

2...

FN.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

FF.

F..

FF.

FF.

FF.

FF.

FF.

FF.

FF.

.0...
:. 03.33:. 32.2.00 ..0 0.0000 .05 .00. .

0.0.8.0 .005 3.0.. :.0 F

.8. a
c. 03. :2 0... .55. :80 . .. .335 .8. .

..:0........0 0.. 0.05 :05 :0... ..03 :.0 F
..:...00... .0.... :. 203.59.... .00. .
..0....0 >0F:0 F

.060:
.00.. >0... 5. 0. 03:... F ..0959> .0: F :0...

5.0.5.0.
038.38. 2.... 6.... 3...... s... .

.._.0.. ..00.
no.9... .0: .0 .0505 .30 8.0.... >0F:. F

.005 2.... F 0.0.0.. 0....
9.0. 0 .0. ..:.5 0! 0.0. ..5 3.00.0 0.... F

...0.3:.
0.0. F $30.1. 3.90.0 0 0)... ..:.0 F .F

0".“
03.3.0.0. 0 05:..0. :05. 2.00.2.0. I. F

.5... 0. o:.>..
.00 F .23 :0 0.... ..I 000.. 0. 0. .0. 0.... .. .F

.3320 05 530.... >2. .0 8.00.
. 3...... ..- :. 5.0.. 9.... .0: 0.. .055 :0...

2.21.9. a... ..:. .. o... ..:8. z.

...050 5...
9.0....0. >0 :. ..0:.. 0... :0 ..oil .30.. 2.0.. F

.33.. .0. .000. :0 3.0.. :.0 F

.F..

.F...

.8

...n

.3

.Nn

.Fn

.on

146

$3

33

A3

33

A3

33

a3

3v

CV

3.

3.

3.

3.

33

3.

33

any

any

any

any

any

any

any

any

any

any

any

any

.3

as

any

any

any

3v

awv

.3

33

A3

“NV

2.3

3v

“NV

amv

.3

3»

3.

.3

AN.

33

.3

A:

A:

A:

A:

2;

AS

a:

A:

A:

A:

A:

a:

A:

A:

A:

A:

.303 on :00 _ 9:50: a. 0.05 .00. _
.30.. c. 0.300 0530; 3 0|: 32,03 _
003.330 .033 0 :0 v8.33 0.5.. _

.3. 03:00..
no: «08 x3: .6. c2... :03 J03 300.5 0 30.. _

.06 ..3
.0030 300.3 0 02:. 3260...... 3:038 ac. 030.959

3.33
30 03... 02:. _ 353029530 05 .0 .59... a0 _

.2... .0 000:0 8.5.0... ..RI 505.: 3.3: 3 30 on _
do. .3 x03 0 co .8.- .f A.08. 3 30.. 3. ac: _

480.393
u:- voEnE 3.30.30 0. “to: >- 305 a. 3 000 _

.>u_g0gu
..3 >08... 8:90:00 33.3.33 0:00 0)... .

69.3.5 ..3 50309 >230 xi 8.!- 3 to: _
.52: too: >05 :9... 0.39.30: x5 3 00:...» aco— _

£32 52033 05 «0 :30 3302393
0337.05 3 I3. 0050 30 9:530 030 _

.3333. 0.5.5.20 50:30 3 0x: _

.88 u. 300
8 08.53% a. _ .8238. 8 3 8.80 _ :25

.3393
0.3.6... 05 033 3 :0: 033:3 3 0x: .

.3: .o .88. oz 2: 58. i2. _

8. .0 0:3 .8 .300 u 3.
0.0 so 0:3 32. 0 AC

. no

.mo

.3

.8

.3

.3

33

33

a:

a:

a:

a:

3.

33

3.

A:

33

33

3.

a:

a3

33

any

33

any

an.

an.

any

any

any

any

.3

an.

an.

Ac

any

any

an.

AC

as

3v

.3

S»

2.3

a3

a3

3»

2.3

‘3

as

A3

A3

3.

23

a3

2.5

a:

A:

a:

A:

as

A:

a:

a:

a:

A:

A:

a:

A:

A:

2.3

A:

03:85.2
.6 c. 2,: 2... c832“. 2: .85. _

.0... 30030:. :00 0550: 305 9 33:00
on you _ .8233 500. 3 Son :0 . :95

6:23:35
00:35 5 303300 32:03.33 3 0a: _

.30: 0055 El- 8.

6573 .03. 0 $030 I. _

.282... x! ..3 08>: 9:8 >23 _
.3305 5 330000»... I.—

.000.03 .03 _

.0.- .o fin! 03 3009.0 0.30

.058 0. «as... t.

.3003...
30: :0 9.33: 09 $030 339.... I... _

635.20.558- xl ‘0 39:. .03 _

.0.- ..3 u. on 3 05063
300 3 50.0.5 _ £03 05 30.3 0033 0..

.080 no. a 2.38
0....» 30.. 0 26.. _ ..qu 00 ‘50... >30.

.20... :0 on 3.83 _ 305 3:00:80
30. _ :85. c. 00.53 :0: 0 303 _ :05

.2300 c. 03 3 09:5
50.9.0 3 080.50 000... _ .30.. £00. 3 0i .8.

.0381 38.- 3 .823:— _00. _

2. .o 3: . 5
0.! .0 as» ..00 I Apv

.E

.K

.3

147

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

.n.

.n.

.n.

.m.

an.

.m.

an.

.m.

an.

.n.

.n.

an.

.n.

.n.

3.

an.

.n.

3.

.N.

S...

2...

3.

.~.

3..

.N.

3.

3.

.~.

.~.

2....

.~.

.~.

3.

.N.

.w.

2...

3.

3.

3.

3.

3.

2.

3.

3.

2.

2.

3.

2..

3.

3.

2.

3.

2.

3.

3.

60.0.2.0 ..:: 05.00.02... >200 .
.0... .0 ..0: 3...... ..00 3.00000 0... .0 0.000..

.I. . :0...
:0 ..03 30. 0.0 0.0. 0.0000 .0. 0.00000 0000 .00. .

.0......000. 0:0 02...... :30 E 09:20 .
.3. 0.30000 0.... .0 0.00.8... 050.0 0.05.30 .00. .
$8.038... 0000.... c. 0.00.0...00 0. 0.... .

.........
>0... .0.... 3.03.0 >00 00.0....0 «50> >0: 0.... 0.... .

.030... .l ..:: 3:0. 32.00 2.0.. .0>0.. .

.0.... 0... ..0 20;... 5.. . .0... 00.300.
.0... 30.. 00.58.00 0......00. :06. 203.50 .00. .

.0002... .0 00.50
:0... 2.... .0 00.5.. 9.30... .08....00 0.00 .00. .

...
0.0.9.00 . :30. 0 0. :00..- vouuifiu 2.0.. . 09.0

..:...c 3.5.5. ..:: 9 08.. .

.8
. to: o... .8... 88.88... ..:. 2.58.8... .8. .

..0....00 :30 ..- .0n..: :0:- >..0.0 0. 000530.. >2
00.23.. 00.300 .0 0000 .00 .

00.3.... 5... 0000 >.0> .9. .3. 0..

..:. . 000.00 .0 ..:... 0... 0.. 0. 00580.00 :0 .

.0... .0. .00 00.3.3.0 0... 0. 9 3.33 0. 3.0.. >..

.10
_ €00.00 .0 ..:... 0... 00.50.00 0.000... .050 :33

..n.

.on.

.02

.55

.3—

.-.

.20.—

.9.—

.0:

.2—

.0:

.m:

..:

.n:

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

an.

3.

.n.

.m.

an.

an.

an.

.n.

.m.

an.

.n.

.m.

.n.

an.

.n.

3.

an.

3.

3.

.~.

.~.

3.

2....

.w.

3.

.N.

3.

.~.

2...

.N.

3.

20..

.~.

3.

20..

.N.

3.

3.

3.

S.

2..

3.

2..

3.

2.

C.

3.

3.

3.

2.

3.

3.

3.

3.

2.

.00....01. .800 :0. f0.) 0.6.. .
5:30.200 0.. 0. 0.0000 «009.0 00.. on .

60303.0 0.
C0203; 00.. 000.00 0000 0... ..30 >30... 00 .

.00 0. .00.. . .0... 000.. 00...... 0.0... .00 .
.000... x0380 8.50.9.0 >078 .

is... I 3.... .2. 8...... 3.... .
.0.§..B..800 .0 .0 .50.... I0 .

.85... .28.». .8... 0.. 3 ..: 8.0:...
..00. 0.

0... .00. 0... .0... .00. . 60:03.0 0. 3.35.0
10.. 2.0.0:... 00 0.. 0. 0>0.. . 00:000-

.:8... 23:85 .2 8 .

6.0.. xi 000...

.3... .3000 0.8.0.00 5.0 t V020.
0.6.. 0..- >..§l000 ..I .0 0.0000. 0... .00 0.6.. .

.0... 0000.000. 000.3500... 0.00.3.5
6.000.. 050» 5... 00000 . 000.. 0... >200 .

0%“
00...... 0.... .0 .0. 0 00.00 0.0 0.000.. 90.0»

60.3.5 30.30 0.... .
.0.... 9.0. 0 no. 0 5... >03 :00 .
.00 0.0.. 00.3....

6.333.530
.0050 .0 5.5.5.00 c. 03.00 I0 .

.NZ

.5—

.c:

.3.

.3—

33's

148

3.

3.

A...

3.

3.

3.

3.

3.

3.

3.

3.

3.

A...

3.

3.

3.

an.

an.

an.

an.

an.

an.

an.

an.

3.

an.

an.

an.

an.

an.

an.

A~.

AN.

“N.

.w.

3.

3.

3.

«N.

3.

.~.

3.

a.

3.

3...

.~.

AN.

3.

2..

3.

S.

2.

3.

3.

C.

2..

3.

3.

3.

3.

3.

2.

2.

.000... xi 000.. .

.2050 .0 050.0. .0000 0...
.0 030.00. 00.3 003 . :05 00.0.. :02. 0.5.. 0.0...

..000 0.00.» c. 0.0 . :0...
3000.0 0.5... ..03... 00...... .3090 05.2.... 300.... 0.... .

00:03.80 .0 005.00.
.0:... E 0.0.... _ .8 00000.0 00 . 80.00 0... .2...

.0...
.0 0.9.0. ..00 0.... 0.0000 .5 0003...... 000300.00 0.0 .

60.000 .00....- 0..0 0.300: 50..- 00:..00. >2

.0.... £030....» 0... .0 8.50.. 0... ..0. 0:..00. 0 000 .
..:. . :0... 0:23.... ..02. 0.. 00 0.00 00: 0.00.. .
.2050 .0 00:23. 0... 0.0:... .

.00..0>.:: 0...
0:0 0... 03000 0.....- ..0>000.0 . 00 00:20 0030.. >:

.26. 8 0808. 0.... .03 z; .3... :80 _

.09....00.
0... 00.00000 0. 0.0.0.. 0003—0000 0.... 0. 0.0.0000
.0: 0. u. 00... 30008.0 00 000.0 00 :03 2.0... .

08:8.
.0. 00000 4.0. . .00.... ..:: 0:00.80 3 000.0 .00 .

.0... ..0. 00.: 05500.00 00 0.0.20 00.... 0000 .00. .

.0020...- .0...-.0 0.0.. >0... .0...
0.83:. 00 0.0: 50.5 0... :. 0000020000 00.. 0.5.. .

.0.- 5...0.... .0: 0000 5000
..- ... .0 050:0... 0... 0.... .0: 0.. . 5:0... :0).

0... .0 03.. .0: ..00 0 3.
0.0 .0 0:... .0: 0 an.

.no.

. .0—

.3.

.~m—

.03

3.

3.

3.

3.

3.

A...

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

3.

3.

2....

AN.

3.

3.

.N.

3.

AN.

20..

3.

AN.

3.

a.

3.

aw.

3.

3.

3.

2..

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

.05 ..0. 0.0.0890. 00: 0. 0.0000
30.00.50 6058.500. .3... 0..0.0 0:.200

.50 i 00... 0:20....»
>02. 0000. 000.000 0.0000 :2... .00.... 00: I0 .

$003.50 ..00 0.0 .
.0: 000.0300 3:92.050
.3005: 0.... .

00:200. 0.0000500 0......
00... 00.50000 0.00 30.0.3.3. 00: 0.6.. .

60.00.000.000
530:... 0... .0 05:00- 0... ..0. 9.300. 0 .00 .

.0... 0.. 3 000: 300.0 0...

080.36....
2. c. 2.: of 8.0:... o... 0...: ..:: .

6.8.0.000 .30 .I 0... .

:09 300.000.. 0. 0005000.. .0 000.0 0.00.. b.
.8 8 a. 0.8... . .0.... 3 0.: ..:

.0.- 003. 0.!- ..I 00... 0:00.090. >00) 0. 0.
£00.00 0.23500: 0 l0 .

55.50.00,. 0 :. 0:0 .00
.:0....000000000..00.00.. 0000250300»:

0.0000 0030...... 0. >000 u. .0.... .

0... .0 0:... 0 S.
0.- .0 0...... ..00 u 2.

.5...

.3.

.~:

. 3.

.on.

.3.

.00
Km.

149

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3..

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

RN.

.3

a.

3.

3.

2...

3.

.3

A8

3.

3.

2....

3...

20..

a.m.

3.

a.

a.

3.

3.

3.

3.

3.

2..

2..

2.

2..

2..

3.

3.

3.

3.

2.

3.

0:30 90.09 0.9303600 0.0 .

63.00030 0.30:2:
:0 .30.... :30 >00 3 00.0... 00 530:0 9.0000 0.0.

.0.- 0:03.950
00 0:03 3:03 0:0>:0 >5 :00000 0.00.. 000 :00 .

.02303030
030 9.000: 30000 0. 8:00:00 .0 >093 000 . :05

.0...—5:0 00 00.... 000: :0 005 2000009... >00) 0. 0.
3.00: . :05 0300000000 3.00.00: 0.0 .

.300: 05 00 c0230 0 :0 .

.202... 3 053 3.: .

.0820 . 0900.00
05 8:. ..:. 00.0 0:00.00 00505 >000: 5:00 .

60030.00 0:00: 0:0 230 0.90.0 00,0:
00 0.000 000.00. a: 00.0.0000 >5..- 30: 00 00000.0 .00 .

.200? k. 0:20.000 0... 00.0... :00 .
0.00 >30 .

6:20:00 0:0 0020.500
00E. 0..- 000:0 . .8 00000.0 :0 . 000000 05 5.:

.00.!- . 0:30.000 0300:3300 05 00:00 .

.0903000 0000 3 .00....0..100>0
:00... 0>0.. . :05 0:. :. 03.00.0093 00.. 0):. .

$0: >00>0 .0 0000 0 0.0020 >2 0: 00 00:. .

£03000
£03005 0:. .0 05:00... 05 000 90:00.. 0 000 .

$00 be 026
0:0. 00 530:0 000.0 .00. . .55 5.: 0:0 0: 0):. _

.50.

{0.9

. .6—

.3—

.8.

3.

3.

3.

3.

3v

3.

3.

3.

3.

3.

3v

3.

3.

3.

3.

3.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

3.

a.

a.

20..

a.

a.

20..

3.

2....

2...

“N.

3.

«N.

3.

2....

20..

A:

2.

A:

3.

C.

3.

A:

A:

3.

3.

3.

A:

3.

3.

2..

3.

.0...:
0.0050 0.000 0000 3:00 0: I 00 00.0000 0.
00.0.0 00 005 0:00:00 00 000.0 00 :03 006.. .

.03.- 00 00 .0000 0.0.

.0055 05:02.0
0.... 0.00 . :00: 80.00000 :00: 006.. 000..»

.00.0.0.00 030300200 0.00» :000. .

.0...:0:
00 0.00:»... 0. 00:0... .0:0.0 0' . :25

..000 0.0000 0050 30.. 00 030.200 :0 .

£00000 0050 05 00» 08:03
00,0; 00 00,0; . .0000 .0300 0 :0 000 00 00 000

.0:. 00» 00 0.0 . .000-
0 0002.0 . 00:0 .2000! 0:000 :05. . 00 000 0<

.305. :05 0.0.000. 505.:
030:. ..B . 6:00:00 00 >005 000 . :05

.0031! 08.0.0 000000 020.00.00.00
0:0 005 0:00:00 3 000.0 00 :00: 006.. .

$00300 >078 .

.0: 90.0.. 0300002000
:0 . 65.09000 .003. 00 00. 0 000:

0030):. 3.0000 000 0006: :00 .
{6.5 3.000 . 0550.30 8.0.... :0 . :05 :05

.0000 0...» c. 0003 0:. r 00 000.2 0000 05

. 603009000 )- 008. .
.0: 5000 10.0.00 0:200:00 00.00 0:00.000 :05

60:0..0 0:000 00::00 .

.02.

.nk—

.K—

Ko—

.3—

.9:

150

A...

3.

3.

3.

A...

3.

3.

A...

3.

3.

3.

3.

3.

3.

a...

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

3.

AN.

2...

2....

3.

3.

.N.

2....

20..

20..

20..

3.

.N.

3.

Am.

2...

A:

2..

3.

2.

3.

2..

3.

C.

3.

2.

3.

S.

A:

A:

2..

.0:0>:0 03.0.0 30300:. 00: 00 .

£00000
0.00 03003 :05 00500 :03 000 00 >00 00230.30 .

000.5003
0:0 30. .00. . .000. ..: 0. 0... 0020.300 0:00.80 0.

.00....» 00 0.0000 00 0.... .
0.8.... 0. 8 8 00.000 . :03 0... .0.... 0020.. 0..
.003. 0 00.00 05.0.. 0 :05 0.0: 03.0 .

6550.000
.0 0:0 .00 0. 10.0 0.0.3.0.. 00450.00 :00 .

.5... .0030:
0 00 30.000 .0 0:... 0 00.05.0005 003 . 0....

0.0. . :05 .0005002. 00:00:. .0 00:00.0... 00.. 03.. .

$200035 05 c.
08.5 00 3.00.980 05 5000 030 .0 00:00 0 03.. .

0:000 0 00.00 :00 .

.000000 300...! 3.00.02.
0 0.0 00.2. 005 000 .0580 3 0:03.. 000503.003

68.00050 :0 .

$00.00..
.0:. 000.000 0... 0005 00:0.000x0 :0 00.. 03... .

65.0000
000.00 0... 03000 9:30 000 >05 00.! 00 3000.0
c000: . .000. 3. 00 0.- 00202.00 0:00.80 2

.0330: 0000 0 I; .5... 9.3.00 1.. 0.3 0000!. 0..

0.0 .0 0:00 00: ..00 0 3.
0: .0 0:00 00.. 0 2”.

.o-

. -~

. —-

.03

.03

.s—~

.05

.mE

.05

.np~

.~P~

3.

3.

3.

3.

3.

3.

3.

3.

2..

A...

3.

3.

3.

A...

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

20..

as

3.

AN.

3.

2...

a.

3.

20..

as

“N.

a.

8.

3.

3.

3.

3.

2..

a:

C.

3v

3.

3.

2.

3.

A:

2.

3.

.220 0003 >05 305. . 5:05
:03 3.00500 :. 0.0000 005000 9.2.0000
0.... 0.00 . :05 00.0.0 .0000 03.. 0005

60:200. 005..
3. 0005 . .00 00000.0 ...0 . :00000 05 5.:

.0000 05 00 00 00.000 033.0 3 . 000 003000..)

.0:.530 0:30 0:00 :00 . 0.2.3
5.3 0... 3 c. 8200 0:0 0000. 00 0. 0005

.003 0... 0:0 .0.
03000 0.05 ..00 . 6:00:00 00 30:0 000 . :05

.0030 000 . 00 00000.. 0000 0:.

.0030:
5:005 00.. 00 05:00: 05 00» 9.200. 0 000 .

.0: .0
5.05305 00 00' 3 005 0500090. >003 0. 0.

.0055 00.00 0:30 30:00.0 .00. .

.0: 0000.0
00 000000 003: 0.0000 00 0000.3 005.. 05

.3003
.0 0000000 30: 8.0030030 033.0 0.0 . .I0 .
2.3 5.3 5:00 :. 5:0 30000 0.0 . 5:05 :03

6:30 00 00 0.003
2... c. .0...: S. .0 .2 0 0.... .00... 0... _

.020 0553003 5.3 0... ..:—5303
30.. 000 . 000... 05 .0055 00 0.00. . 0000 05

.000 03000 00:200. 0000000
A 3... 0:30 3.00 .005 5.3 :00000 0 03.. .

a. .o S... . a.
8. .o 0.5 ..8 .. 2.

.p—N

.o—N

. SN

.02

151

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

.n.

.n.

3.

An.

2.

3.

2.

3.

an.

an.

3.

an.

2.

3.

.n.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

2.

2.

2.

2..

2.

2.

2.

2.

2.

2.

2.

2.

2..

2.

2.

2.

000500.00. .0 00:..00.
00:... ..0 002.0 . .00 3000.0 00 . :00000 05 5.3

.00....00. 0.0.8250 .03.. >.000000..00 . 00.0.0200

.>.0.0.5 005.50 :000 030.. .

.0:..0300. 0.0 . 00.
30.. 0030.0 0: .000 0.0 .0 3.0.00 05 ..0000 >.0000 .

0000000300 0:.00 0.030 3 >03 .0 .0 0:0 00 .

.00 000.0
.0: 003 . 0:02.00 5.... ..00 90.30.. >078 0.5.303 .

.00050 .0 3200:0000... 000 .

0.0000 0050 .0 0.8.0.000 0:0
if?! .353... o... .5 02.008... 2. 0.0.0.. 3..

0g.“
05 00 0.00. 0.300 . 1.0.0.000 .000 05 0003 0005
00:00.. . .>.u:000...0 00.0.5 0:00 030.. 000.0. .
..0... . 5.85.0 6:20.000 0000 .I 00 ..000 ..00. . 00

.0030 00 :00 . 0.0.3 5.3 :00000 0 050. .0: 030.. .

0.0.00
.08.... ..: .o 8.5.... o... 0o. 8:8. . .2. .

.0.- 0000.:0.00 ..00 .3000 05...... 00 8.000..

.0000. 0:0 0.00.500
3... 0005 3 0. 30.00 0059.0 3 .00 . .0000... 0..»

.005003 00...... 3.20.. 000 030. 0:0 ..00 .2. 00.
0.0000 3 0.... 0030: .
0.00.3.5 3... .0 2.00 .0 02.0000 0.0 .

.20 . :00000
0 .:0000.0 30... :0 00.00000 030.. . 00:0... >:0.0 30..

. m3

. 3.3

. n3

33

. 2mm

.93

.03

. n3

. N3...

.33

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

an.

an.

3.

an.

3.

.n.

3.

.n.

.m.

3.

2...

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

2.

2.

2..

2.

2..

2.

2.

2.

2.

2.

2.

2.

2.

2..

.2. 3 200000.... >23. 0. ..00 00.00000
05 .0 080.80 5.3 00203:. >.0000 0:.00

.00:0.0. 3... 3 00:..00.
0.200.. 0:0 .0003 500 000.000 3 000. .00. .

0‘09““
.0.... .85.. So... 2. 0.. 00.08. B 0..- .... ..88
3 00.... 0.... .0 .05 5.0.009... >003 0. a.

.303
>0. .00 3:00 . :0... 33:000. .00. 000.0200 .

.000 3.00000 05 5.3
02:... 2.... . 00.58....8 080 o... 5.2. .

.090300. 00.00.
.0 0:...05 0:00: 80.00 0059.0 00 000.0 8.00

.05.. 0300. 00 08:03
. :05 02... 0003 0005 .050> 003 . :00...

00>er I . .:30 >0
0.00. 3000.20 >003 .0000. 0000000 0.0000 :05

6.000000020
000 003 0.0000 0. 030000300 00 ..:-0 0|... .

.09...00.
000 0.0000 0050 30.. 0:30.005 00 I000 .

0005.003 .0 00:..00. 00.0..
..- 0005 . .00 00000.0 0.0 . :00000 05 5.3

.0... .3 0. 05:00: 03.0 00 I05 30003:.
. 0:0 .0000300 >5.- 5... 00100. 000 003.03 >3

000.5000 0:0
00.3.0.0 3.5.0000 0. ..003 ..- 005 0. 00 000 _

$0002.00 5.3 00000 00 0.0.300
0305.3 05.0 3... 00.0000 00 0.00 0.0.00 0.... .

.oqw

. 03

.53

. m3

. n3

33

. SN

. u-

.03

JMSJZ

3.

.3

A:

A3

3.

.3

A:

a:

A:

3.

a:

3.

3.

3v

3.

3.

3y

3.

any

any

an

AC

.3

any

any

any

any

any

any

any

any

3U

any

any

any

any

“NV

«3

AN.

23

3y

3v

«3

2.2

22

3v

«3

3...

.N.

.8

9.2

3V

3.

2.:

A:

.3

2L

2.

A:

A:

A:

A:

a:

a:

A:

A:

3.

a:

2;

A:

A:

A:

£53.28“. 0... 39!. ac: _

.33: v3.3.3 you _

$223 :35...» _

.02:- E 5 330965 05 8.3 _

.: 2.22.. 52:... 85.8. E :2 38:

.8238“.
9.9- . 5.5 5.: noon. .3 32. 2: 5.: >32. to _

.89. u. a: an: 3 8.: _ >33:

.9 @5395 no: _ «:5 o.- nooggous >33... >2
.838 Rani! :- >38.qu I. _

.8050 “3.3 .

.955. 2.» 53- 3.80 .o 3:293 02:. _
.593: so)»: an...“ 355 530 >....oz _

.u. v03
>2: :2! 23:38 E. So: 3 8.3 :33 «>2. _

.2303 _ no on aca‘gu 3 03. 9:3 8:. _
.839. 08¢ a a. 805 .033 .93 _ :25

.395 3.. 9.2;:- ou 3:3 _ «95:00. 50
£599.23 you. .ggal-uuotggcof

.98 33 05 a. 5 cl _ 0:. 3 003a 52,325
it... >33... 0... 3:39.; >1

8. .o 3: .2 ..oo . ..:
a... .o 3.: .2. u .C

.~o~

. —o~

.oo~

.03

3V

3y

3y

3y

3v

3v

.3

3y

3y

3v

3v

3v

3v

.3

3v

7:

A:

23

any

any

any

any

any

”3

any

any

any

any

any

AC

any

3U

any

any

3.

3w

9.3

A3

:3

.3

a.

22

a»

3.

as

as

3y

3v

3.

3v

3.

a:

A:

A5

CL

A:

A:

a:

A:

a:

:v

A:

2..

a:

A:

A:

2L

A:

2. .o 8: .. a.
8 no 02v .VOO I A—v

6000 3.333 a... 0.8K 30:3 _ JR
5&2 I _ .E~
6.- 5.: :33 ac:
0.98 .85.. cos: 28>.- Ccouso ..o >....oa _ .NR
68.82: >- :5. _ .§~
6:3 .6 .. v.3: >2 .E~
.ucl: _
an...) 9:8 3... no... I. 32. 3.3 .0 «8:0: .ooN
$.33...- ucunu .. on: 8.: >2 63.
..:. SB. «2..-:8 .8. _ £3
5:3
«:33. ~35... :9... >3 3 a: so. as... ..: .3...
.18.qu :2. >- 233 “I! _ .3...
.833: :- 5 v9.23:- 3 no:
:3 no.5 ..:. 383:. no... .8383 o..- 22: .3...
.0.- c357: 3 we. a «9.3 a. .no~
60.46.: >l .833 8:33 >2 .3...
.l. 8:39.85 «5 2. .5...
.0.- S... 390.. 63
.5 8.!- u.:o: _ Cc: .3378: o
59: 3005.. «£233 :3 23 ‘0 «Co 9.2. _ Z .03
.9:- >.33980 :3 9!. _ =5 uooo o.
:3 26... _ :23 3: 5 382.396 v2. 26.. _ .omw.

1.513

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

.N.

.3

a.

2...

3.

a.

3...

3.

A3

3.

A3

3.

.8

3.

A:

C.

3.

3.

3.

2.

3.

2.

3.

3.

3.

3.

£9.33 0.39.2.3 .0...... a so .

.>....-» u.- vouoo... 3.8.2. >2

.>..moo n03... ..: . ..8... 3 oo 3 0.... a... :25
.>..I.. >93... c 5 v3.9. 3.. .

.8. no... 82.... 3 80.52.90 u..- 833 >2
$3.326 9.!- . :2... 33>! 5.5.. .

...o....o
o... 39.3”. . .85.... 53 3. o... «2.... ....o.o - 2

63...... 3.095.-..05 :- so .

.39...
feet . 0893a >9 Vow-.22.; 3.3....» 9:09 >07... .

$558.3
39... 55.. . 5.3 a. 3.5. 3 2...... n...

.o... ..o. a. on
3 9.02.3 you 3 .30... . 3.23 o... .2... .038. oz

.33 so .

46.9.0 .o .0. a 23.. .

.o~n

.05

.0;

.5;

.0;

.m—n

Jpn

.npn

.mwn

.Sn

.05

.03

ion

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

3.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

an.

3.

3.

3.

a.

3.

a.

a.

a.

3v

3.

Au.

3.

3.

3...

A3

A:

2..

3.

2.

3.

3.

3.

2..

A:

A:

3.

A...

3.

A:

3.

Jon . >2. 2.. .3 3.8.2.- 3 $8.. feat . .son
.98. st 52.- >..u.a In . .03

do. .83 a on . 0333 3...... xi .899. «.90.. .Sn

$2....- ..I
96.. 2... 0.88 .8: 8 an: 3 3...... Eco. . Jan
.2523 >32... >2... 3.... >3 2.: 0.80.. S... . .non

.3038. .03 . £23.3- .~cn

.880. vooa
>39... - >23»: a. 9.0... €2.22. .00. . :2... .Sn

_ .8 . .5.
of 2.2.2. . >4. .3. 8398 not. >.u.lu!.ou .

don
.39. o..- no...“ :2. 05 28:2. . .024.
£39.33 26.. 3 loo. . 3:03 a a. 2.3 .02..

.06 :35...» «9. on 5.. .8 >9...- 8...:.: >2 .3...

.0: .39
3 09.3.25: .3. .o ..:—533 no...» u. .8...

.95»... >I 53a >....o.. . .no~

..:... 8 .5: >2.» .2... .05 ..3
8 a. 382. 3.82. 8 >8 :3 o... .5... . Jo...

603.33. >5 .0 >9.
0... c. you .9. on 2.0.333... >3-3>au >x .no~

154

FOLLOW-UP POSTCARD

You are very important! A few days ago you received questionnaires concerning
normal adult development and attitudes toward religion. Your participation in
this research (by completing those questionnaires) will help educators,
ministers, and counselors to better understand healthy adults of all ages and
the issues they (and we!) face.

If you have already completed and returned those questionnaires--thank you! If
not, could you please do so today? Because these questionnaires are being sent
to only a small, randomly chosen sample of people from several churches, your
response is extremely important to the study and the accuracy of the results.

If by some chance you did not receive the questionnaires or if they have gotten
misplaced, please call Barbara Reeves (517/355-8447) or me (517/224-3440) so we
can put another set in the mail today.

Jay M. Terbush, M.A., M.Div.
Project Director

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Acklin, M.W., Brown, E.C., & Mauger, P.A. (1983). The role of religious

values in coping with cancer. MW. 2.2.
322-333.

Aden, L. (1976). Faith and the developmental cycle. Pastoral
25152119132. 23. 215-230.

Alker, H.A., & Gawin, F. (1978). On the intrapsychic specificity of
happiness. Journal ofi Eersoggligy, 3;, 311-322.

Allen, R.O., & Spilka, B. (1967). Committed and consensual religion: A
specification of religion-prejudice relationships. lgg;ng1_£gr_§hg

figientifiig Study 9: Religion, g, 191-206.

Allport. GM. (1950). WM. London:
Macmillan.

Allport, G.W. (1954). t ce. Reading, MA:

Addison-Wesley.

Allport, G.W. (1959). Religion and prejudice. Ih§_§1§ng_figxigg, Z,

1-10.

Allport, G.W. (1963). Behavioral science, religion and mental health.
l22;na1_2£_Bclizien_and_fleal£h. 2. 187-197.

Allport, G.W. (1966). The religious context of prejudice. Jo rn or
W. i. 447-457.

Allport, G.W., & Ross, J.M. (1967). Personal religious orientation and
prejudice. l2urna1_2i_2cra2nali£x_and_fiecial_21xch2128x. 2. 432-443.

Argyle. M-. & Bait-Hallahmi. B. (1975). Ih2_§2£iel_2£!£h21281_2£
religion. Boston: Routledge and Kegan Paul.

Arulpragasam, D.S. (1986). ° v to
son' us 0 c e e s
e ta 0 - . Unpublished Ph.D.
dissertation, Michigan State University.

155

156

Athy, J. E. (1986). v 1coholics

erns.

 

Unpublished Ph. D. dissertation, Michigan State University.

Azar, J.A. (1982). s n ive
!‘ 1_!' Q " ‘ '41" ° " 1‘ °! 1 ‘ ° ° . ‘ e916 etic
ghoozy. Unpublished Ph. D. dissertation, Michigan State University.

Baker, M., & Gorsuch, R. (1982). Trait anxiety and intrinsic-extrinsic

religiousness. MW. 2.1.
119-122.

Batson, C.D. (1976). Religion as prosocial: Agent or double agent?
MW. 15. 29-45.

Batson, C.D. (1980). And the truth shall make you free: Religion as
personally liberating or enslaving? Unpublished manuscript,
University of Kansas.

Batson, C.D., Naifeh, S.J., & Pate, S. (1978). Social desirability,
religious orientation, and racial prejudice. Jouroo; go; the

MW. 11. 31-41.

Batson, C.D., & Raynor-Prince, L. (1983). Religious orientation and
complexity of thought about existential concerns. log;nol_fo;_§ho

52W. 22. 38-50-

Batson. C.D.. 6: Ventis. W.L. (1982). WWW—A
W- New York: Oxford University

Press.

Becker, R.J. (l971). Religion and psychological health. In M. Strommer
(Ed.), '-_ - - - , ; .
New York: Rawthorn.

 

Bergin, A. (1983). Religiosity and mental health: A critical

reeveleution end mete-analysis. W
W. 19.. 170-184.

Bergin, A.E., Masters, K.S., a Richards, P.S. (1987). Religiousness and
mental health reconsidered: A study of an intrinsically religious

semple- W. 3.4. 197-204.

Bergin, A.E., Stinchfield, R.D., Gaskin, T.A., Masters, K.S., & Sullivan,
C.E. (1988). Religious life-styles and mental health: An

exploratory study- W. 3.5. 91-98.

Bolt, M. Purpose in life and religious orientation. loorool_o£
WW. 2. 116-118.

157

Boyd, R.D., & Koskela, R.N. (1970). A test of Erikson's theory of
ego-stage development by means of a self-report instrument. Inn

i22rna1_2£_£xperimental_fiducation. 2&. 1-14.

Brown, L.B. (1964). Classifications of religious orientation. Jougnal
i2r_the_§cicntific_§£sdx_2£_8elizien. a. 91-99.

Chirban, J.T. (1980). v ith
ooyoloonono. Unpublished Th.D. dissertation, Harvard University.

Conn, W.E. (1986). Adult conversions. 2§§£21§l_2§1£h21281. 2&. 225-235.

Constantinople, A. (1969). An Eriksonian measure of personality

development in college students. Doyoloonon;ol_£§y§hglggx. 1.
357-372. -

Crandall, J.E., and Rasmussen, R.D. Purpose in life as related to
specific values. Jougnal of Clinical Poycnology,,fil, 483-485.

Crowne, D., & Marlowe, D. (1960). A new scale of social desirability

independent of psychopathology- i9urnal_2f.§2nsultina_£§xcholesx. a.
349-354.

Darling-Fisher, C.S., & Leidy, N.K. (1988). Measuring Eriksonian
development in the adult: The modified Eriksonian psychosocial stage

inventorY- £§XEh21281£§l_82221£§. £2. 747-754-

Davis. J.A. (1965). Educati2n_f2t_pesitizc_mental_heelth. Chicago:
Aldine.

Dodrill, 0., Bean, P., & Bostrom, S. (1973). The assessment of
religiosity in evangelical college students and its relationship to

prior family religious involvement. 1on;nol_o£_£§yohology_ono
Theology, 1, 52-57.

Donahue, M.J. (1985a). Intrinsic and extrinsic religiousness: Review

and mete-enelysie- l2u:naI_2f;Pers2nali£1.2nd.§2&ial_2§¥sh2lesx. 38.
400-419.

Donahue, M.J. (1985b). Intrinsic and extrinsic religiousness: The
empirical research. c t e n,
25, 418-423.

Donaldson. W.J. (Ed.). (1976). Bcsearsh_in_ncntal_heelth_and_telisieu§

e av ° u o o
be v . Atlanta: Psychological

Studies Institute.

158

Droege, T.A. (1978). - a t ° e m n
in_rolo;1on_§o_§oo. Chicago: Lutheran Education Association.

Ellis, A. (1980). Psychotherapy and atheistic values: A response to A.
E. Bergin's “Psychotherapy and religious values.” Jongnol_of
t , 5o, 635-639.

Erikson, E. (1963). 9hildh22d_and_§2£ietx. New York: Norton.

Erikson, E. (1968). loonoloyy_yon§h_ono_oz1o1§. New York: Norton.

Ernsberger, D.J., & Manaster, G.J. (1981). Moral development,
intrinsic/extrinsic religious orientation and denominational

teethinse WW. 1%. 23-41.

Farquhar, W.W., Wilson, F.R., & Azar, J.A. (1983). Assessment of adult
adjustment patterns (AAAP) research instrument: First report.
Unpublished document.

Fleck, J.R. (1976). Dimensions of personal religion: A dichotomy or
trichotomy? In W.J. Donaldson (Ed.), Roseagon 1n nontol noalth and

ous av . Atlanta: Psychological Studies Institute.

Fowler. J.W. (1981). WWW

ev 0 me a the e . San Francisco: Harper and
Row.

Fowler, J.W. (1984). e o co 1 . San
Francisco: Harper and Row.

Frankl, V. (1962). Monlo_§on;on_£o;_noon1ng. Boston: Beacon Press.

Freud, S. (1927). The future of an illusion. In J. Strachey (Ed. &

Trans.). WWW
, 21. London: Hogarth, 1961. Cited from W.D.

mums!
Robson-Scott (Trans.). Garden City, NY: doubleday Anchor, 1961.

Gibbs, H.W., 6 Achterberg-Lewlis, J. (1978). Spiritual values and death
anxiety: Implications for counseling with terminal cancer patients.

WW. 2:. 563-569.

Clock. C.Y.. 6: Stark. R. (1970). WWW
Iollglono_oonmlonon§. Berkeley: University of California Press.
Goldman, R. (1964). s c o d s e e.

New York: Seabury Press.

159

Gorsuch, R.L., & Venable, G.D. (1983). Development of an “age universal"

I-E scale. MW. 22.
181-187.

Griffin, R., & Thompson, J. (1983). Does denominational affiliation make
a difference? Intrinsic and extrinsic religiousness revisited.

MW. 19. 92-116.
Hewley. G.A. (1988). NW. Odessa. FL:

Psychological Assessment Resources.

Hiltner, S. (1966). Toward a theology of conversion in light of
psychology. W. 11. 35-42.

Hoge, D.R. (1972). A validated intrinsic religious motivation scale.
Won. 11. 369-376.

Hoge, D.R., & Carroll, J.U. (1973). Religiosity and prejudice in
northern and southern churches. WW

821181211. .12. 181-197.

Hood, R.W. (1985). The conceptualization of religious parity in
Allport' s typology. MW.

25, 413-417.

Hunt, R.A., & King, M. (1971). The intrisic-extrinsic concept: A review
and evaluation. WM. .19.
339-356.

Jackson, 8.3. (1981). Measurement of commitment to role identities.
MW. 9.0. 138-146.

James. W. (1961). MW. London:

Collier Books.

Joe, V.C., McGee, S.J., 5 Dazey, D. (1977). Religiousness and
devaluation of a rape Victim. MW. 13..

64.
Jung, 6.6. (1933). MoooInnoon_1n_ooozoh_o£_o_oonl. New York: Harcourt.
Kahoe, R.D. (1974). Personality and achievement correlates of intrinsic
and extrinsic religious orientations. Jon:nol_of_£o;oonolloy_ono

W. 22. 812-818.

Kahoe, R.D. (1975). Intrinsic religion and authoritarianism: A
differentiated relationship. St 0

ReLisLon. 1.6.. 179-133-

160

Kahoe, R.D. (1985). The development of intrinsic and extrinsic religious

orientations. WM. 23.
408-412.

Kahoe, R.D., & Dunn, R.F. (1975). The fear of death and religious

attitudes and behavior. l2uIna1_f2r_the_§sientifis_fitudx_2f
821181239 l5. 379-382-

Kahoe, R.D., & Meadow, M.J. (1981). A developmental perspective on

religious orientation dimensions. Jonznol_o£_3ollglon_ono_flool£h,
29, 8-17.

King, M.B., & Hunt, R.A. (1972). Measuring the religious variable:

Replieation. l2uIna1_f2r_ths_Scientifis_§£ud1_2f_zslisien. 11.
240-251. »

Kraft, W.A., Litwin, W.J., & Barber, S.E. (1986). Relationship of
intrinsic-extrinsic religiousness and assertiveness. Poyohologlool

Boooggo, 52, 1115-1118.

LeBlanc, S. (1988). e ve n ce busers
' ev o

Unpublished Ph.D. dissertation, Michigan State University.

Martin, C., & Nichols, R.C. (1962). Personality and religious belief.
l2urnal_2£;fiesial_£§1chslezx. is. 3-8.

McClain, E.W. (1978). Personality differences between intrinsically
religious and non-religious students: A factor analysis study.

l2u;nal_2f_£ers2nalitx_8s§essment. 92. 159-166.

McConahay, J.B., & Hough, J.C. (1973). Love and guilt-oriented
dimensions of Christian belief. a e ifi Stu of
Bfili‘ignv 12! 53-64-

Meadow, M.J., & Kahoe, R.D. (1984). ' e o in

lnolyloool_llnoo. New York: Harper and Row.

Morris, R.J., & Hood, R.U. (1981). The generalizability and specificity

of intrinsic-extrinsic orientation. Boylow of fiellglouo Reoeazcn,
22, 245-254.

Paragament, K.I., Steele, R.E., & Tyler, F.B. (1978, August). Religious
practice, religious motivation, religious identification and
individual psychosocial competence. Paper presented at the annual
convention of the American Psychological Association, Toronto.

Patrick, J.W. (1979). Personal faith and the fear of death among
diversent relisioue pepuletions- l2urna1_f2r_ths_§sisntifis_§tudx_2£
Belizien. 18. 298-305.

161

Picciotto. H. (1987). MW.
Unpublished Ph.D. dissertation, Michigan State University.

Pruyser, P.W. (1977). The seamy side of current religious beliefs.
Winn. £11. 329-348.

Rice, C.A. (1971). a o o 0 en t ons
MW. Unpublished Ph.D.

dissertation, Boston University.

Robinson. J.P.. & Shaver. P.R. (1973). W
noolonooo (Rev. Ed.). Ann Arbor, MI: Institute for Social Research.

Rokeach, M. (1960). e en d ed d. New York: Basic Books.

Rosenthal, D.A., Gurney, R.M., & Moore, S.M. (1981). From trust to
intimacy: A new inventory for examining Erikson's stages of

psychosocial develoment. W. 1s.

525-537.

Sanua, V.D. (1969). Religion, mental health, and personality: A review
of empirical studies. www.125.
1203-1213.

Scabbo, R.E. (1984). v te f e e
1:9. :9-' s .1; nea~ '0 9 1‘ -:-;::!9‘1 0 :e 3.‘ -- 11‘1,

-- s v . Unpublished Ph.D. dissertation,

Michigan State University.

Shoemaker, A., & Bolt, M. (1977). The Rokeach value survey and perceived

Christian values. Walnut. 5. 139-142.
Sparkman, G.T. (1986). Proposals on religious development: A brief
review. W. .82. 93-109.
Spilka, B. (1977). Utilitarianism and personal faith. ou of

W. i. 226-233.

Spilka, B., Hood, R.W., & Gorsuch, R.L. (1985). Ibo_ooyohology_of

Iol1g1onl__An_omol:1ool_ooo;oooh. Englewood Cliffs, NJ: Prentice
Hall.

Spilka, B., Kojetin, B., & McIntosh, D. (1985). Forms and measures of
personal faith: Questions, correlates and distinctions. longnol_£o;

W. 29.. 437-442.

162

Spilka, B., Pelligrini, R.J., 5 Dailey, K. (1968). Religion, American
values, and death perspectives. fioolologlool_§ymoo§lnn, 1, 57-66.

Spilka, B., Stout, L., Minton, B., 5 Sizemore, D. (1977). Death and
personal faith: A psychometric investigation. lonrnol_fo;_§no

SEW. 1.6. 169-178.

Stark, R. (1971). Psychopathology and religious commitment. Boylon_of
KW. 12. 165-176.

Strickland, B.R., 5 Shaffer, S. (1971). I-E, I-E, and F. Jonnnol_fo;
Win. 19. 366-369.

Strommen.M.P. (1971). SW
W. New York: Hawthorn.

Tate, E.D., 5 Miller, G.R. (1971). Differences in value systems of
persons with varying religious orientations. Jougnal to; one

MW. 19. 357-365.
Valdez, R. (1983). de d

RMWWW

identitie_. ‘Unpublished Ph.D. dissertation, Michigan State
University.

Venable. G.D. (1984). WW
v e ve. Unpublished Ph.D. dissertation, Fuller

Theological Seminary.

Ventis, W.L., Batson, C.D., 5 Burke, D. (1980). Religious orientation
and self-actualization: A psychometric study. Unpublished
manuscript, University of Kansas.

Watson, P.J., Hood, R.W., 5 Morris, R.J. Religious orientation,

humanistic values, and narcissism. Boy1oy_o£_3oligions_flo§ooroh, 25.
257-264.

Whitehead, R.E., 5 Whitehead, J.D. (1982). QhI1oolon_llfoyooooo1n§;__1no

o- ,0 0' ‘ g; ;g';: :90 ; 0 Q : 0" o :0

Garden City, NY: Doubleday Image Books.

Wiebe, K.F., 5 Fleck, J.R. (1980). Personality correlates of intrinsic,
extrinsic, and non-religious orientations. u o 5 ch ,
19:, 181-187.

Wilcox, M. (1980). Doyoloononool_1onznoy. Nashville: Abingdon Press.

Wilson, W.C. (1960). Extrinsic religious values and prejudice. ionnnnl
W. 6.9. 286-288.

163

Wright. D. (1972). WW1
fignoloo. Highcroft House, England: Association for Religious
Education.

"lilifllfll’ll’lillll“