;.i

"r

 

LIBRARY

Michigan 5mm

5 University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

4 isms Jl
=

 

 

 

 

 

 

 

:_||
fl—i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

if

 

Q1

 

MSU is An Affirmative Action/Equal Opportunity institution

CMMG-pd

 

THE DEVELOPMENT OF AN INSTRUMENT TO MEASURE
CLIENT STATUS AND PROGRESS IN THERAPY
BASED ON A PSYCHOEDUCATIONAL APPROACH

BY

Joel Lynn Kelley

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

College of Education

Department of Counseling, Educational
Psychology, and Special Education

1991

ABSTRACT

THE DEVELOPMENT OF AN INSTRUMENT TO MEASURE
CLIENT STATUS AND PROGRESS IN THERAPY
BASED ON A PSYCHOEDUCATIONAL APPROACH

BY
Joel Lynn Kelley

Application of quality assurance concepts in the area of
mental health has faced ongoing difficulty due to lack of
instruments to measure client staus and progress in therapy.
Third.party payers as well as therapists and.patients have not
had available to them instruments which are based on sound
theoretical and experimental principles and which have been
shown to have reliability and validity.

The Personal Change Questionnaire was developed to
measure client status and progress in therapy. The purpose
of this study was to evaluate the Personal Change
Questionnaire for internal consistency and to make preliminary
investigation into its construct validity. A 38-item Client
Form was developed as was a 44-item Therapist Form. Each form
was based on an item by item analysis of the Personal Paradigm
Shift. The items on each form were evaluated by a panel of
experts as to content validity and appropriateness. The
Therapist Form included an additional section relating to

motivational dimensions of clients as viewed by therapists.

 

A sample of 115 clients in different stages of therapy
were given the Personal Change Questionnaire. Demographic
data was gathered on each of the participants as well as the
seventeen therapists who participated in the study.

The result of correlational analyses indicate that both
forms of the Personal Change Questionnaire are highly
reliable, demonstrating coeffecient alpha values of .95 for
the Client Form and .97 for the Therapist Form.

Evidence for construct validity was generated through the
observation of a low but significant correlation between
therapist total scores and client total scores (Pearson =
.40). There was a strong correlation between therapist total
scores and therapist rating of client status (Spearman = .79) .
Therapist rating was also positively correlated with client
total scores (Spearman = .32).

No significant correlations were found between total
scores and educational level or length of time in therapy.

A preliminary factor analysis revealed the presence of
two factors on both forms of the Personal Change
Questionnaire.

Comparison of answers between diagnostic groups of
Dysthymia and Cocaine Dependency showed significant

differences in the pattern of answers between the two groups.

ACKNOWLEDGMENTS

To the participants in this study who shared their
thoughts and feelings. To the therapists who took time to
encourage participants, and who gave of themselves and their
expertise to help develop the Personal Change Questionnaire.
To the clients who shared their time and energy as well as
themselves through their answers.

To Mary Ann Brandt, Sandy Canfield, Debbie Wilbur, Mary
Beth Weigand, Bruce Stokely, Lois Werner, Dr. Tom Woodward,
Dr. Lesley Jones, Dr. Martha Whipple, Dr. Peter Fisher,
without whose cooperation and help this study could not have
been completed.

To Dr. Joe Jeney and Dr. William Parker whose earlier
endeavors laid a groundwork for this study and who helped in
the development of the Personal Change Questionnaire. To Dr.
Bill Griz who served as teacher and friend and who served as
evaluator for this instrument.

To Dr. Shaun LaBlance, fellow student and friend,
supporter and participant in this work. Shaun gave me
strength.when things seemed.bleak.and help when it was needed.
He provided friendship in the deepest sense of the word. For

this, and much more, he is well loved.

iv

To the staff of the Rehabilitation Medicine Clinic at
Michigan State University, whose emotional and physical
support made completion of this task possible.

To Jack, Shelley, Jan and Greg for showing me the value
of enlightened self interest.

To Dr. William Hinds, teacher and mentor, editor and
critic, for his expertise and patience in guiding me through
this task. He provided a guiding hand through the many stages
of this work and of my education as a researcher and
therapist. Thank you much more than I can express in words.

To my friends, who both near and far provided a base of
strength from which I could pursue my goals.

To those who made it possible and who have meant so much
to me as I grew into the task. To Ben Ellis, Betty Bechtel,
Joan Babb, and Jim Morse.

To my family who must have had their doubts but who were
always supportive. To :my brother' Tom for his growing
friendship and a place to stay when a break was needed. To
my sister Connie for her ongoing love and friendship.

And finally, to my parents. To my mother, Ann, for her
love and help in times of need, and for her understanding the
importance of this work to me. To my father, Gayle, who
encouraged me to begin a journey which he never saw completed.

To them this paper is dedicated.

TABLE OF CONTENTS

Page
LIST OF TABLESOOOOOOOOOOOOOOOOO0.00.0.0... ......... v
CHAPTER
I. STATEMENT OF THE PROBLEM......... ........ . 1
Introduction............ ..... . ........... 1
Need for the Study....... 0.. 0.. O O. O. O. O 0 O 3
TheorYOOOOIOOOOOOOOCOOOO0.00... ..... .0... 6
Purpose Of the StUdYOOOOOOOOOOOO000...... 11
Research Questions....... ...... . ......... 12
summarYOOOCOOOOOOCOO0...... ......... .0... 13
II. REVIEW OF THE LITERATURE.. ................ 14
Historical development of
Psychoeducational models............ 14
Components of the Psychoeducational
“OdelOOOOOOOO00.000.00.00...00...... 21

Research with Psychoeducational Models... 24
Research with the Personal Paradigm

Shift............................... 34
Historical Development of Quality

Assurance........................... 41
Current Concepts and Problems in Mental

Health Quality Assurance............ 47
Research in Quality Assurance............ 53
Summary......................... ......... 61

III. DESIGN OF THE STUDYOOOOOOOOOOIOOOOOOOO0.0. 62

Description of the Personal Paradigm
Shift and the Personal Change
Questionnaire....................... 62
Personal Awareness....... ...... 64
Decision Making.......... ...... 69
Oppositional Alternatives...... 71
Treatment Planning and Action.. 74
Motivational Dimension......... 79
Summary of the PPS and PCQ..... 80
Selection and Description of the Sample.. 81
Reliability and Validity................. 87
Measure Used in the Study................ 94
Procedures for Data Collection........... 95
Design.............................. ..... 97
Research Hypotheses...................... 98
Procedures Used in Data Analysis......... 101
Supplementary Data Analysis.............. 109

vi

summarYCOOOOOOOOOOOOOOOOOOOO0.0.0.0000...
IV. RESULTS.OOOOOOOCOOCOOOOOOOOO0.0.00....O...

Tests of the Research Hypotheses.........
Hypotheses Related to Reliability
Hypothesis I...................
Hypothesis II..................
Hypothesis III.................
Hypothesis Related to Validity......
Hypothesis IV..................
Hypothesis V...................
Hypothesis VI..................
Hypothesis VII.................
Hypothesis VIII................
Hypothesis IX..................
Hypothesis X...................
Hypothesis XI..................
Hypothesis XII.................
Supplementary Data Analysis.........
Summary..................................

V. SUMMARY, DISCUSSION and IMPLICATIONS.....

Summary..................................
Discussion...............................
Reliability of the Instrument.......
Validity of the Instrument..........
Limitations of the Study.................
Implications of the Research.............
Directions for Future Research...........
Instrument Revision.................
Psychoeducation in Therapy..........
Conclusion...............................

APPENDICESOOOOOCCOOCCOOOCOOOOOO00.0.0... ...........

Introduction for Expert Judges...........
Content Validity Evaluation Form.........
Item Summary for the Personal Change

Questionnaire.......................
Personal Change Questionnaire

Form S..............................

Therapist Form......................
Information Packet.......................
Consent Form.............................
Demographic Page.........................
Directions for Administration of the PCQ.
Therapist Information Form...............

BIBLIOGRAPHY...0....OOOOOOOOOOOOOOOOOOOOOOO0....0..

vii

110
113

113

113
114
116
116
116
119
120
121
122
124
124
125
129
131
135

141

141
146
146
149
153
156
158
158
159
160

161
162

164

165
170
174
175
176
177
178

179

LIST OF TABLES

Description Page

Age Range of Subjects................... 81

Sex of Subjects........... ..... ......... 82
Marital Status of Subjects....... ...... . 82
Educational Level of Subjects..... ...... 83
Employment Status of Subjects.... ..... .. 83
Income Level of Subjects.... ............ 84

Number of Sessions Attended............. 84
Insurance Coverage of Subjects. ......... 85
Reason for Seeking Counseling........... 86
Length of Time Subject Has Had Problem.. 86
Therapist Treatment Orientation.. ....... 87
Sex of Therapists....................... 87
Internal Consistency of the Personal

Change Questionnaire for Total

Scale and Four Subscales..... ..... . 115
Correlation Matrix of the Total Scale

and the Four Subscales of the

Personal Change Questionnaire

Client Form SOOOOOOOOOOOOOOOOOOOOOO 117
Correlation Matrix of the Total Scale

and the Four Subscales of the

Personal Change Questionnaire

Therapist FormOOOOOOOOOOIOOOOO ..... 118

Correlation of Client Total Scores
with Therapist Total Scores........ 120

Correlation of Therapist Total Scores vs
Therapist Rating of Client Status. 121

viii

LIST OF TABLES
continued

Correlation of Therapist Rating of
Client Status in Therapy vs
Client Total Score................. 122

Comparison of Means of Subscale Scores
Client Form SCOOOCOOOOOOOOOOOOOOOOO 123

Comparison of Means of Subscale Scores
Therapist Form..................... 123

Correlation of Client Total Scores
vs Educational Level............... 124

Correlation of Clients Total Scores
vs Number of Sessions.............. 125

Factor Matrix for Form S................ 127
Factor Matrix for Therapist Form........ 128

Comparison of Means of Diagnostic Groups:
Cocaine Dependency vs Dysthymia
Client Form S...................... 130
131

Comparison of Means of Diagnostic Groups:
Cocaine Dependency vs Dysthymia
Therapist Form..................... 132

133

Comparison of Items, Subscales and
Total Scores with Therapist
Rating of Client Status............ 134

Comparison of Means of Subscale Scores
and Total Scores between Client
Form S and Therapist Form of the
Personal Change Questionnaire...... 135

Correlations Between Motivational

Dimensions and Items and Subscale
scoreSOOOOOOOOOOOOOOIOOOOOOOOOOO... 136

ix

CHAPTER I

STATEMENT OF THE PROBLEM

W104;

Quality assurance refers to any activity implemented
within the service delivery system to improve the outcome of
physical or mental health care. Shuman and Troy (1988)
indicate that the most pressing problem in the area of quality
assurance is the lack of specificity evidenced by providers
of mental health services. An essential step in overcoming
this problem would be the delineation of what is being
attempted in treatment, with the goals stated in clear,
observable terms. The most significant advances in quality
assurance need to be made in the procedures which would
document assessment and treatment. This requires an explicit
approach for conceptualizing and expressing what is being
attempted in the therapeutic process.

The quality assurance process can be broken down into two
general areas of consideration. The first area deals with
standards of care. The second is those additional activities
which are designed to eliminate any deficiencies which are
identified during the quality assurance process. Quality
assurance in mental health care is extremely difficult due to
the nature of the services offered and the different

theoretical orientations employed. This complexity requires

2
special attention to the types of assessment and the diversity
of treatment approaches which may be used. While treatment
of a medical condition such as acute appendicitis has a very
limited number of treatments, emotional or psychological
conditions have a diversity of modalities for assessment and
treatment.

While a universally accepted definition of quality
assurance is not available, an optimal definition would be
oriented toward assuring the highest level of care for all
patients. Care would be available, appropriate, and
efficacious for all clients, and it would be provided
regardless of cost (Rodriguez, 1988). Such idealistic
concepts are tempered by factors such as economics, resource
availability, and consumer (patient/client) attitudes.
Quality assurance methodology includes consideration of
providers who want a professional-scientific basis of care and
insurance carriers, i.e., payers, who have an economic view.
Finally, there are patients/clients who seek a legal and
protectionistic position and who ultimately seek high quality,
safe and effective treatment.

The Personal Change Questionnaire used in this study"was
developed, to ‘measure clients' personal change status in
therapy. The questionnaire is based on a programmed model of
change known as the Personal Paradigm Shift which was
developed by Dr. William Hinds at Michigan State University

based on his experience with clients as a health psychologist.

The Personal Paradigm Shift, referred to as the PPS, was
designed according to psychoeducational models of health care
management as an organized and planned method of teaching
individuals about, and how to, undergo change.

The Personal Change Questionnaire, the PCQ, was designed
specifically to evaluate the twenty-four steps of the PPS and
is designed to measure clients' readiness for treatment with
the PPS program. It is expected to be an accurate measure of
progress in the treatment as well.

The purpose of this study was to evaluate the reliability
and the validity of the Personal Change Questionnaire as an
instrument which attempts to measure treatment efficacy under

the general rubric of quality assurance.

H§§Q_£Q£_Lh§_§LEQ!

Quality assurance can be described from three different
perspectives: theoretical, practical and professional.

From a theoretical perspective, according to Donabedian
(1980), quality assurance is concerned with three aspects of
the health. care system: structure, process and outcome.
Structure refers to the resources which are available: that
is, the relatively stable characteristics of the health care
system.which influence the quality of care provided. Process
refers to the activities which are performed in the patient
management process. This includes demographic and diagnostic

information, therapeutic procedures, and follow-up monitoring

4
of the patient. Outcome refers to assessments which examine
the effects of care on patients. This generally includes
measures of health status and patient satisfaction.

The questions regarding structure are generally covered
by governmental regulations and professional associations
which are charged with licensing, certifying, and accrediting
of health care facilities or personnel. Questions about
process and outcome are addressed in treatment decisions and
implementation.

Success of the therapeutic process in mental health has
been a critical issue since before the landmark article of
Eysenck (1953). It.has become increasingly important to third
party providers in the emerging health care network in the
nineteen eighties. Increased demands and costs have raised
the issue to a position of critical importance.

In terms of the practicality of treatment, the
traditional psychoanalytic approach to therapy is clearly
inadequate when applied to large numbers of people. While
behavioral approaches are more cost effective, they still are
expensive and require considerable amounts of therapist time.
Innovative techniques are often not accompanied by valid or
reliable evaluation instruments, and therefore it is difficult
to measure outcome effectively. Questions of effectiveness
and appropriateness of therapies remain unanswerable until
evaluation methods are developed, verified and implemented.

Guidelines set by the American Psychological Association

5

in the Standards for Providers of Psychological Services (APA,
1981) state that "there shall be a written service delivery
plan for every consumer for whom psychological services are
provided ...which will analyze the problems, state priorities
among established goals, and outline systematic procedures for
the implementation of the plan" (p .17). This statement of
standards makes it imperative that treatment programs be
structured to provide a diagnosis, establish treatment
priorities, and delineate procedures for implementation.
Methods for the assessment of progress and the evaluation of
outcomes must be provided.

Because professional standards are expressed as broad
statements of policy, they may be implemented in a variety of
ways by different providers. Any method chosen must employ
established and scientifically verified tools for evaluation
of treatment effectiveness. Evaluation of treatment progress
or outcome in such a manner is essential to the financial,
legal and. ethical justification. of ‘treatment choice .and
implementation.

Gibson (1988) enumerated five essential components of
quality assurance in applied programs: Problem identification,
assessment, correction, resolution, and follow-up monitoring.
The Personal Paradigm Shift incorporates these into its basic
structure. As a measure of the PPS, the Personal Change
Questionnaire constructed for this study assesses these

components. While there appears to be a certain "face

6
validity" to the PCQ, there has been no formal research to
determine the reliability or validity of the instrument.
This study is the first formal attempt to evaluate the
PCQ for reliability and validity in order that the PPS will

meet the three types of standards discussed above.

Inger!

The Personal Paradigm Shift and its associated
measurement form, the Personal Change Questionnaire, are based
on the ideas of psychoeducational models of therapy. Such
models make three essential assumptions:

1) The counselor is viewed as a teacher, rather than

physician or healer.

2) The client's problems are conceptualized as learning
deficits rather than symptoms of illness.

3) The client is seen as an active and self-responsible
agent, capable of directing their own learning,
rather than as a passive recipient of treatment.
(Larson, 1984)

While some models assume a treatment program based on
patient illness, the psychoeducational model begins with
client dissatisfaction or a need for change. As contrasted
by Authier, et.al. , (1975) and noted by others including Habeck
and Fuller (1986), the medical model proceeds from illness to
prescription to therapy, hopefully resulting in a cure. The

psychoeducational model begins with a dissatisfaction and

7
proceeds to goal setting followed by teaching of skills. This
results in satisfaction or goal attainment.

The counselor's role is clearly defined as that of
teacher, one who helps with problem identification, goal
setting, skills acquisition and practice, and self evaluation
procedures for the client. The client, an active participant
as. a student, learns functional skills ‘which. will help
overcome deficiencies which may be causing a problem or
dissatisfaction in their life. The programs are highly
structured with learning based on sound educational principles
such as feedback, mastery and maintenance of specific
behaviors.

Larson (1984) has summarized common themes in psycho-
educational programs. They strive to elicit active
participation from clients in the learning process, hence they
minimize mystification of psychological knowledge. These
designs are based on established learning principles such as
observation, discrimination, reinforcement and generalization.
They consist of both didactic and experiential components.
The programs are highly structured with identified and
specified goals. Progress is evaluated and reported to
clients as part of an ongoing process in the program. Clients
are taught new methods of dealing with problem situations and
learn to internalize these methods.

Alschuler and Ivey (1973) have identified five principles

that serve to maximize long term internalization in

8

psychoeducational programs. They are: 1) gaining clients
interest and attention, 2) providing practical experience of
the skill, 3) developing a clear conceptualization of the
problem.and the skill, 4) relating the learning to other areas
of experience, and 5) providing the client with the
opportunity to meaningfully practice the skill.

The Personal Paradigm Shift (Hinds, 1983) is designed as
a psychoeducational model of personal health care.management.
It is based on the treatment of psychosomatic problems,
particularly those concerning stress related disorders. Its
development was based on observing the dynamics involved when
people learn about the process of making personal change.

This systematic approach to self-health-care is designed
to assist clients in achieving their own goals. In contrast
to many health care intervention programs that are based on
client compliance, this program promotes internalization as
the means of altering and maintaining behavior change.
Internalization as described by Suedfeld (1982) is the change
in attitudes and associated behaviors that result from the
realization that a change to a new pattern will be beneficial,
rational and worthy of acceptance. This usually involves a
change in personal values that then results in behavior
change.

Benfari, Eaken, and Stole (1981) indicate that effective
lifestyle intervention strategies meet six general conditions:

1) The client knows what change is desirable and acquires

9
the knowledge required for the change.

2) The client has the skill or is learning the skill to
implement the knowledge for change.

3) The change is in the client's best interest.

4) The change is in the best interest of primary related
groups such as family, friends, peers, etc.

5) Both. internal and. external environmental systems
require change.

6) Internal and external change agents give their support
to the change and provide feedback through the
process.

The PPS addresses the sixth condition only with regard
to internal supports and support. within. the therapeutic
setting. The fourth condition is addressed only in that
emphasis is on awareness and development of personal controls
rather' than. environmental controls (Jeney, 1985: Parker,
1987). The PPS provides a highly organized and systematic
method of assisting a client to recognize significant
psychological, social and environmental lifestyle variables
that affect their problems. The primary focus of the program
is to engage the client in observing, experiencing and
determining their own behavior. Treatment consists of an
educational process that assists the client in learning to
make a personal change.

Like other problem solving interventions, the PPS assumes

that behavioral change proceeds through a logical sequence of

10
phases. The PPS change process involves experiencing (a)
personal awareness of the problem, (b) decision making
processes, (c) oppositional alternatives to present behaviors,
and (d) planning for and implementing actions. The PPS is
composed of twenty-four steps which focus on various aspects
of the change process such as goals, thoughts, feelings,
images, rewards and punishments.

The Personal Change Questionnaire consists of thirty-
seven items which correspond to components of the PPS. Each
statement on the client form requires a self report of the
client's self understanding of such things as their thoughts,
feelings, goals and plans. The forced choice format requires
the client to determine how they perceive their present
understanding of the problem situation they experience.

The therapist form of the PCQ requires the therapist to
give their own perceptions of the clients' self understanding.
This form also asks the therapist to evaluate motivational
factors of defensiveness, introspection, self disclosure,
commitment, and emotional readiness to learn a program.

Since the PCQ is based on the PPS which incorporates both
process and outcome, the PCQ has the potential to evaluate two
of the three parameters listed earlier by Donabedian (1980)
which are regarded as essential components of quality
assurance, i.e., process and outcome.

Quality assurance programs depend on effective evaluation

of treatment planning, implementation and outcome. Those

11

conversant with quality assurance emphasize the importance of
effective measurement and evaluation tools in.determining how
well these aspects of the programs are realized (Rodriguez,
1988: Wells and Brook, 1988: Gibson, 1988; Cohen, 1988:
Sechrest and Rosenblatt, 1988). Quality assurance requires
strategies for monitoring and assessing treatment process and
outcome. They are essential in the context of justifying the
choice and expense of a program for a specific client.
Decisions about such choices depend on reliable and valid
assessments (Shueman and Troy, 1988).

In selecting outcome measures, program developers must
first enumerate specific outcomes that are relevant to the
target condition. The subsequent outcome measures should be
evaluated for both reliability and validity before they are
matched with specific clinical conditions (Wells and Brook,
1988).

The critical step which is often missing is the
validation of the instruments used in the assessment and

treatment phases of therapy and in the outcome measures.

ose o tu
The purpose of this study was to evaluate the usefulness
and appropriateness of the Personal Change questionnaire as
a measure of status in therapy. In order to do this, it was
necessary to provide a measure of the internal consistency of

the Personal Change Questionnaire and to establish an estimate

12
of both its content and construct validity. The study also
evaluated the PCQ for its validity as a measure of client
status in therapy. The study employed Cronbach's Alpha as a
'measure of the internal consistency; .A factor analysis of the
PCQ was performed, and several measures of construct validity
were employed. Client and therapist views of client self-
understanding were compared. Therapist total scores were
compared to their intuitive evaluations of client status.
Demographic and clinical information were also gathered and

comparisons made between diagnostic groups.

Research Questions

The following research questions were addressed:

1) Is the PCQ a reliable measure of clients' self
understanding?

2) Does the Personal Change Questionnaire show evidence
of content validity?

3) Is there evidence of construct validity for the
Personal Change Questionnaire?

4) Do the items on the Personal Change Questionnaire
fall into four distinct factor areas?

5) Is there evidence of validation of the Personal
Change questionnaire in terms of its relation to
therapists' clinical judgement of status in therapy?

6) What relationships exist between client understanding

as measured by the PCQ and diagnostic categories,

13
and other factors such as length of treatment and

education level.

Summary

In Chapter I, the problem investigated was presented,
along with the need for the study and theory about the
research. Chapter II is a review of the literature in both
quality assurance and psychoeducational designs. Chapter III
presents the design and the methodology for the study. This
includes explanation of and rationale for use of certain
analytical methods. Formal presentation of the Personal
Paradigm Shift and the Personal Change Questionnaire is also
in this chapter. Chapter IV presents the results of the
study. Chapter V discusses these results and conclusions
drawn from them. Implications of the results and directions

for further research are presented.

Chapter II

REVIEW OF THE LITERATURE

The areas to be covered in the literature review are:
theory and.historical development of psychoeducational models
of therapy, definitions of the psychoeducational model,
research in the application of these models, research with the
Personal Paradigm.Shift, theory and historical development of
quality assurance, definitions of quality assurance, and
research in quality assurance. A summary is provided with

conclusions drawn from the research.

0 c D v me catio s

"In order to meet the ever increasing demands of preven-
tative and creative mental health, the field of psychology
must adopt the attitudes of public health and look for
effective ways to administer to a growing population of
clients."(Sanford, 1959, p. 832) . In his presidential address
to the American Psychological Association in 1959, Sanford
allowed that this new direction would include more education
and less absolute mandate, more advice and less prescription,
and more facts with fewer arcane announcements. Psychologists
would become more like teachers and less like doctors. The
psychologist would become the translator for the field of

behavior science, finding ways to apply that knowledge for

14

15
those who could benefit from it. Sanford advocated
psychologists as proponents of the do-it-yourself’movement in
the behavioral sciences.

"Psychology, like consultation, is teaching and
learning. The client must learn his way out of his problem
and into independence. He must control himself, make his own
decisions, be his own boss, run his own life. He must not
learn long term dependency on the counselor. The counselor,
through subtle manipulation of knowledge and skill, helps the
client free himself, forever, from counselors." (p. 832)

Sanford clearly delineates the need for the egalitarian
role as opposed to the authoritarian role and concludes that
an essential step for the long term success of psychology is
the active dissemination of technical knowledge about human
beings. Believing that knowledge is power, Sanford proposes
that the preservation of a healthy democracy demands that such
knowledge be shared and increased.whenever possible. He sets
the tone for much subsequent theory, research and practice in
psychoeducation.

Hobbs (1964) relates three revolutions in the history of
psychology. He attributes one major change to Philippe Pinel
in France, William Tuke in England, and Benjamin Bush and
Dorothea Lynde Dix in the United States. This revolution
involves the recognition that the mentally ill be treated
humanely. They should be treated with kindness and dignity.

Another revolution took place with the advent of Freud's

16
emphasis of the intrapsychic life of humankind.

The other revolution is that which began when concepts
of public health were applied to the field of mental health.
Mental health once was synonymous with its opposite, that is,
mental illness. Now it has genuinely come to mean not just
illness, but in a broader sense, mental well-being. It was
acknowledged that mental illness is not just the private
misery of an individual but that it is a social, ethical and
moral problem, a responsibility of the total community.

Proponents of increased coverage of psychological
services have argued for some time that the traditional
medical model is inadequate to deal with the scope and
immensity of human problems encountered in mental health
services (Carkhuff, 1971, 1972; Guerney, Stollak and Guerney,
1970a, 1971b: Guerney, Guerney and Stollak, 1972: Ivey and
Alschuler, 1973: Authier, 1977). Ivey and Alschuler (1973)
advocated that the counselor be described as a psychological
educator who actively intervenes, teaching healthy skills to
others. This departed from traditional models which assume
a knowledgeable, authoritarian figure who actively cures the
passive patient. This change in the concept of patient-healer
has come about for a variety of reasons.

While psychology had previously been influenced primarily
by the basic sciences, it has more recently come to be
influenced by educational psychology and the social sciences.

Cole and Lacefield (1974) list three factors relevant to this

17

change in emphasis. The first factor is the increased use of
psychometric methods, especially the use of intelligence
testing, beginning in World War I and continuing since that
time. The measurement of human traits and capabilities has
become a major tool in the assessment of individuals for
college placement and career management and planning. A
second factor was the employment of persons in World War II
who were knowledgeable in learning theory and who designed
and implemented programs for the military. New psychologies
of instruction were developed, and these new approaches paved
the way for changes in teaching methodology and.content~ This
type of assessment and training has become institutionalized
in Veterans Administration hospitals.

A third factor was the work in experimental psychology
that was carried on with government funds or at government
facilities. The National Defense Education Act (NDEA)
provided funds for educational and psychological research
carried out in those years following the launch of Sputnik I.
Programs for teaching psychological skills developed from new
concepts of educational design resulting from such research.

Authier, Gustafson, Guerney and Kasdorf (1975) trace
several sources for the development of psychoeducational
models. Freud emphasized the importance of experience in
development, and his concepts, often altered in several ways,
have prevailed over the previously dominant models based on

instinct (Watson and MacDougall, 1929) and genetics (Galton,

18

1869: Goddard, 1913). Pavlov (1927) and Watson (1916) were
other early pioneers whose work was instrumental in models of
human learning.0thers such as Mowrer and Mowrer (1938) and
Wolpe (1958) are credited in part for espousing the idea that
therapy is essentially about learning new behaviors. Dollard
and Miller (1950) were among those who advocated the idea that
clients are capable of genuine learning rather than subjects
who are to be conditioned.

Four factors are cited for the development of present
models of psychoeducational design. First was the public
awareness and acceptance of the need for psychologically
oriented services and the recognition that new models of
delivery are needed to fulfill present demands. A second
factor was the growing dissatisfaction and disillusionment
with the traditional approach's ability to effectively serve
large numbers of people. Studies such as those of Landis
(1937) and Eysenck (1952) cast serious doubts on the efficacy
of traditional treatment. Despite criticism these articles
received, questions about therapeutic effectiveness remained.

A third factor was the increasing recognition that
therapeutic goals would have to include prevention. This
includes the additional recognition that prevention is
impossible using traditional models of intervention.
Psychological education has become an integral part of any
comprehensive program.

Finally, there was the realization that parapsychological

19

activities provide beneficial services. Organizations such
as Alcoholics Anonymous provide essential supportive functions
and as such can be an integral part of a psychological health
care system. This support, as well as skills training by peer
group members, appear to be effective. The development of
sensitivity and encounter groups provided a method by which
large numbers of people could learn interpersonal and
psychological skills. While these groups were often not goal
directed, they provided a format that represented a step
toward delivery of services based on education rather than on
treatment. A generalizable effect was the breaking down of
the concept of "sick" versus "well" , a mainstay of the
traditional medical model of psychological illness.

Many other theorists have helped to lay a foundation for
present psychoeducational models. Carl Rogers' approach
included a willingness to respect and encourage the client to
choose their own goals. The high degree of respect associated
with this attitude is essential in the "therapist as teacher"
movement (Rogers and Dymond, 1954). Bu F. Skinner allows that
the therapist, not the client, must be the first to change
their ways of approaching a program if things are not turning
out right. An attitude commonly found among good teachers is
quite similar in that adjustments must be made by the teacher
rather than blaming the student's resistance. Skinner also
sees the client as a consumer who sets the goals which

environmental manipulations are designed to accomplish

20
(Skinner, 1948, 1971).

Thomas Szasz made an entire generation of therapists
aware of the injustices of the medical classifications of
"sick” versus "well". He greatly contributed to the awareness
that tampering with people's right to behave as they wish must
be done with great caution (Szasz, 1961a, 1961b, 1966).

The cognitive-emotional therapy of Albert Ellis involves
learning which can be used for altering personal behaviors.
Ellis has been a leader in organizational aspects of providing
this type of service on a broad basis to the consuming public
(Ellis, 1963: Ellis and Harper, 1961) . He also advocates
allowing the client to chose the kinds of values and goals
they want to pursue (Ellis, 1977).

In summary, psychoeducational models of treatment are the
result of an evolution in psychological and educational theory
over the past fifty years. Theories of learning, development,
motivation and personality have become combined with
counseling theories and practice. There has been a general
dissatisfaction with the medical model and there has been a
growing recognition that this model is inadequate in the face
of increased demands. New approaches for the dissemination
of psychological knowledge have developed. Therapists using
these approaches treat psychological and emotional deficits
rather than illnesses. Such skills as decision making,
problem solving, impulse control, and interpersonal

sensitivity and communication can be taught, practiced and

21
learned (Guerney, 1977).
Psychoeducational approaches were developed to promote
client involvement and increase therapeutic effectiveness by
making the client a student participant while the therapist

becomes a teacher of learnable skills.

e ts t o a ode

Most advocates of the psychoeducational approach assume
that psychological practitioners see their function, not in
terms of the medical model, but in terms of client
dissatisfactions. The teaching and learning involved in
therapy includes not only the symbolic and cognitive
processes, but it also includes instruction in overt behavior,
especially interpersonal behavior (Guerney, 1977).

The affective-behavioral-interpersonal nature of the
teaching process separates this experience from typical
classroom education. The practitioner is forced to deal with
questions of personal and cultural ethics and moral values,
and the teaching is closely tied to the self concept and
emotions of the client. The nature of the teaching requires
that the therapist be knowledgeable in the overall field of
psychology and education, as well as moral, ethical and
cultural values. In addition, the therapist-as—teacher must
be continually working at improving their own self awareness
and knowledge as part of the overall growing process (Authier,

et. al., 1975).

22

Psychoeducational teaching can be very problem specific
and requires that the instructional materials and methods make
use of the knowledge of psychodynamics and defense mechanisms
acquired during generations of psychological research.
Treatment programs can be developed and shared with the
therapeutic community. Authier (1977) lists thirty-nine
different programs developed in this manner.

The therapist takes on.the role as teacher, one who helps
the client define problems, set goals, acquire skills
necessary to pursue those goals, and evaluate progress. The
client becomes the student who has functional skill
deficiencies which can be corrected with the client as an
active participant in the learning process (Habeck and Fuller,
1986). Learning the skills is facilitated by providing an
organized and often highly structured plan of training. This
plan involves experiential learning as a critical element and
focuses on long term internalization rather than short term
knowledge and satisfaction (Ivey and Alschuler, 1973).

In summary, psychoeducational models exhibit the
following general characteristics:

1) The therapist acts as teacher, helping set up a

program which is followed by the client.

2) The client acts as an active and participating

student, who, after stating a problem, sets a goal
and pursues that goal by following the plan or

program.

23

3) Psychoeducational programs demystify therapeutic
principles and learning processes by eliciting
active participation.

4) The programs are based on established learning
principles and they consist of both didactic and
experiential components.

5) The programs are highly structured with specified
goals and methods of evaluating progress.

6) The programs focus on identification, mastery and

maintenance of specific behaviors.

The Personal Paradigm Shift was developed as a program
of treatment based on the experience of watching the dynamics
of personal change (Hinds, 1983). It is programmed and.highly
structured and includes various aspects of the change process.
It focuses on the identification of problem behaviors and the
specification of outcome goals by the client. The therapist
acts as teacher and guide through the process which includes
determination of the problem situation as a first step. The
behavior of the client in that situation is identified, and
skills involved in decision making are enhanced or taught
while the client. works on identifying oppositional
alternatives and sets up plans of action for changing
behaviors.

The Personal Paradigm Shift satisfies the criteria for

psychoeducational models of change. The therapist acts as

24

teacher while the client is in the role of active student.
The client sets goals and pursues those goals through
participation in the program. The program itself is based on
sound learning principles and includes the opportunity for
both didactic and experiential learning. The program is
structured and provides ongoing methods of evaluating progress
as part of working toward specified goals. The program
focuses on specific behaviors, and there is a continual effort
at identification of problem behaviors and their replacement
with acceptable and desirable behaviors.

The unit on decision making is unique to the PPS, for
often other programs do not focus on the process of deciding.
The client is required to examine both the rewards and
punishments for maintaining or changing the problem behavior.
The effects of these rewards or punishments on the client's
ability and desire to change is examined in detail as part of
the program. Self understanding of personal change is an

integral part of the Personal Paradigm Shift.

es w' u
Early researchers and theorists had a tendency to
theorize without research into the practical application of
their ideas. Psychoeducation as psychotherapy was not
accepted until the development of general skill training
programs of the late sixties and seventies. While Dollard and

Miller (1950), Mowrer (1950), and Murray (1954) proposed the

25
therapist could act as teacher, it was the development and
application of such techniques as counter-conditioning,
aversive conditioning, etc., that paved the way for the use
of educational concepts in therapy.

While these behavioral techniques set the stage, they
dealt more with the manipulation of contingencies with clients
as subjects rather than with clients as students who were
actively involved in understanding the process of changing.
The emphasis was on the "patient receiving a cure", a
reflection of the medical model. Therapists' adherence to
this model limited their ability to see themselves as teachers
in the psychoeducational sense (Authier, 1977).

Maslow (1955) was among those who believed that if
systematic training programs for counselors could be applied
to paraprofessionals, they might be applied to psychiatric
patients as well.

One early application of psychoeducation concepts was
that of Pierce and Drasgow (1969). Their experimental group
of seven psychiatric patients was given training in basic
listening skills after having being rated on their helpfulness
prior to training. Their performance was compared to that of
four control groups: a medication group (on medication) , a
therapy group, an individual therapy group, and a random group
that was receiving a variety of treatments. Analysis using
the Wilcoxin paired-comparison signed rank test revealed that

the training group showed significant changes from pre-

26

training measures on both post taping and standard
interviewing measures. The control groups showed no
significant differences from pre to post measures. 0n
helpfulness, the training group differed from three of the
control groups at the .01 level of significance and from the
fourth group at the .15 level. The results indicated that
psychiatric inpatients could improve their level of
interpersonal functioning with only twenty hours of training
time. They also showed this method to be more effective than
all of the control groups on increasing the level of
interpersonal functioning.

Their study was based on the training model established
by Truax and Carkhuff (1967) which dealt with the enhancement
of the dimensions of empathy, positive regard, genuineness and
concreteness as an integral part of the therapeutic process.

Gutride, Goldstein and Hunter (1973), using a technique
they called Structured Learning Therapy, tested twenty-seven
psychiatric inpatients on their level of social interaction.
The experimental group was then exposed to a four week program
of training in four areas: interacting with someone who
approaches them, interacting when they approach another,
interacting when they approach a group, and resuming old
relationships with family and friends. The control group
received no training of special nature. A post training
measure was made of all subjects after the four week period.

All subjects took both the FIRO-B and the Psychiatric

27

Outpatient Mood Scales. All were subsequently asked to wait
in a room with another patient (an experimental accomplice)
and their interactive behaviors were measured on seven
behaviors including eye contact, forward leaning and seconds
of conversation. Their Structured Learning Therapy yielded
significant differences in the predicted direction on four of
the seven specific behavior variables. Analysis of covariance
yielded a moderate number of main effects. F-scores at the
.001 level of significance were found for social behaviors of
forward leaning, responding to conservation, sitting with
others and eye contact. The results indicate that this
training model is effective in both acute and chronic
patients, and that it works well in the absence of additional
therapy.

Also using the Structured Learning Therapy approach,
Gutride, Goldstein, Hunter, Carrol, Lower, Clark and FMria
(1974) performed a similar experiment with one hundred and
twenty chronic impatients in an attempt to foster social
interactions at mealtimes with more seriously impaired
patients, i.e., longer time in the hospital. An experimental
group was trained with films and group interactions on a
weekly basis. The films depiCted a series of social mealtime
behaviors, while group sessions featured discussions of the
films, role playing activities, video feedback and more
discussions. The group was trained for differing periods of

time from two to seven weeks and was compared on post measures

28

to groups with some or none of the training plus some social
interaction therapy. Post measures included the FIRO scale
of interpersonal behavior and the POMS mood scale. Each
patient was additionally rated by independent raters on their
social interaction skills at a meal during the week following
termination of training. ANOVAs performed indicated fifteen
instances of significance in the social behaviors, ten at the
.01 and five at the .05 level of significance. lThese included
both simple and complex social behaviors.

Goldstein, Gershaw and Sprafkin (1984) present the model
of Structured. Learning* Therapy in.«detail including case
examples. The method has been applied in fifty or more
studies with uniformly positive results. These studies
include patients in the following categories: chronic adult
schizophrenics, behavior disordered adolescents, geriatric
inpatients, and child-abusing parents, among others. The
method has also been applied to the teaching of teachers and
the training of police officers in interpersonal skills. The
authors point out three advantages of their method: 1) the
acquisition of new skills has clearly been demonstrated as
viable and reliable, 2) there is a transfer of learning into
areas not specifically addressed in the original training, and
3) the large number of studies indicates that there are state,
trait, cognitive, demographic and sociometric predictors of
high levels of skills acquisition. and transferu Their

psychoeducational model has now become a research tool in

29
itself.

A different approach was taken by Ely, Guerney and Stover
(1971) who used a psychoeducational model in training couples
in conjugal therapy. Twenty-three couples were randomly
assigned to treatment and control groups after having been
assessed according to the Ely Feeling Questionnaire, the
Primary Communication Inventory, and a series of twelve role
playing situations which were scored by independent judges.
Experimental subjects were trained in each of two roles,
speaker and listener. This training included explanations,
demonstrations, discussions and role plays by each couple.
The control groups received no trainingu Increases in Feeling
Expression statements were significant at the .05 level on the
Ely Feeling Expression Questionnaire. In the role play
situation, these statements increased significantly more for
the experimental group than for the control with a
significance at .05 level. Feeling Clarification and Primary
Communication were both increased significantly at the .05 and
the .001 level respectively. The training was effective in
expanding subjects' capacity to emit desired responses in the
experimental setting. They forecast that longer term
application of the program would provide greater opportunity
for supervision of the application of these skills and their
more integrated use on a daily basis.

A unique application.of psychoeducational principles was

seen in the study of Christensen and Arkowitz (1974) in which

30
the therapist was not in direct training contact at all but
acted only as an information passageway between clients. In
an effort to deal with dating problems, fourteen men and
fourteen women on a college campus were set up with random
practice dates matched only for age, height, race and distance
from campus. Subjects were evaluated prior to the practice
dates with three social anxiety inventories and each reported
their own levels of anxiety in both dating and non-dating
situations. After each date, all of the subjects completed
evaluation forms of themselves which covered such things as
perceived anxiety, self-perceived skill, partner's anxiety and
partner's skill. Each subject also completed a form which
included aspects of their date's appearance and behavior which
they liked or disliked. After the experiment, subjects
completed a satisfaction questionnaire. Forms were forwarded
after each date to all subjects so there was constant feedback
and opportunity for change. Changes in self perceived anxiety
and social distress were significant at the .01 level using
one-tailed t-tests with matching samples. This program did
not teach special skills, but it allowed feedback about
activities and the desire to change to act together as the
teaching medium. The feedback gave information about the
types of changes which would be desirable and it apparently
worked well enough to prompt some positive change. Since
there were no controls, the results should be viewed as

preliminary. The authors see this as a relatively inexpensive

31
way to reach large numbers of people with this specific
program.

A study by Archer and Kagan ( 1973) points out the
usefulness of another psychoeducational approach to skills
training in the realm of interpersonal communication. Based
on the model known as Interpersonal Process Recall (Kagan and
Krathwohl, 1967) and Affect Simulation (Kagan, Schauble,
Resnikoff, Danish, and Krathwohl, 1969) , the study involves
the teaching of interpersonal skills to volunteers from a
student population. Volunteers were placed into Interpersonal
Process Recall, encounter-developmental and control groups.
After the experimental period was over, peer relationship
inventories were used as part of the assessment of
interpersonal skills levels. A two way analysis of variance
with the dependent variables as repeated measures assessed the
general effect of treatments across all dependent variables.
Tukey post hoc comparison tests evaluated the differences
between specific treatments. There were significant treatment
effects at the .05 level. There was significant evidence that
Interpersonal Process Recall is a viable and effective way in
which to provide learning in interpersonal communication
skills.

Kagan (1984) describes the method of Interpersonal Process
Recall in detail. He reports the scope of the research on IPR
and relates the breadth of its applications. He refers to

more than twenty studies confirming the efficacy of the method

32
as a means of influencing human interaction.

Hersen and Bellack (1976), repeating the techniques of
Hersen, Turner, Edelstein and Pinkston (1975), taught social
skills to chronic schizophrenic patients who were very non
social. They targeted several social behaviors, such as
speech duration, number of requests, appropriate affect and
overall assertiveness, among others, and proceeded to train
the subjects in these skills. After measuring baseline rates
for these skills, training was administered by video tape
presentation of skills followed by role playing and practice.
Four weeks of five 20 to 40 minute sessions were offered. Due
to small N's, statistical methods were not extensively
employed beyond graphing the response patterns from the
baseline rates. Subjects showed improvement in social skills
which.were:maintained at follow-up times of two, four, six and
eight weeks. This training of chronic schizophrenic patients
is noted in.Gambrill (1984) in.her discussion of social skills
training. In more than sixty studies, the training of various
populations of subjects has been shown to be effective for
social skills, regardless of the nature of the client.
Results vary across subjects and specific methodology, but
there is consistency of effect. According to Gambrill, this
type of training works, and additionally, it generalizes
across settings into non-experimental areas. Training in
specific skills becomes internalized and generalized, a

requirement for successful learning according to

33
psychoeducational models.

Using a video based format known as Microcounseling
(Ivey, 1969, 1971), Moreland, Ivey and.Phillips (1973) trained
two groups of subjects. One group was trained by use of
interviews with video taped feedback and didactic instruction.
The other was trained with practice interviews and review with
suggestions for improvement of skills. While both groups
showed improvement in some skill areas, the microcounseling
group was significantly higher in attending behaviors and in
reflection of feeling statements (F-scores significant at the
.05 level). More importantly, the microcounseling group
showed a much greater generalization of skills across the six
basic interviewer skills measured at posttest. The level of
significance varied from .05 to .001, the latter being for
attending and reflection of feeling scores. While both groups
improved, the experimental format group produced greater
change. The result is not inconsistent with other studies in
which specific instruction, followed by discussion, role-
playing and practice with video tape and feedback, produce
more profound changes.

In their summary of Microcounseling, Ivey and Garwin
(1984) conclude that there is an abundance of evidence that
teaching skills with this method is successful and can be
applied. with confidence in. a ‘variety of settings. In
addition, Kasdorf and Gustafson (1978) review more than one

hundred and fifty data based studies which are in agreement

34
with this conclusion.

In summary, the number of studies conducted with various
psychoeducational models of treatment is impressive. It is
also conclusive in showing that specific skills, both personal
and interpersonal, can be effectively taught to a wide range
of clients/patients by a wide range of teachers/therapists.
There is ample evidence that therapeutic gains can be made by
the therapist who, acting as a teacher, relays knowledge and
provides experience which enhances client learning. There is
ample evidence that clients can learn to recognize problem
.behaviors, practice and.acquire new'behaviors and.act as their

own monitors after the training process is completed.

ese 't h a a d'

The Personal paradigm Shift (Hinds, 1983) has been used
only sparingly in research settings although it has been used
in. counseling' settings for’ several years *with anecdotal
reports of success. Jeney (1985) investigated the impact of
applying the PPS as an intervention in the treatment of
chronic pain patients. The effectiveness of the PPS was
tested for (a) reducing chronic pain perceptions, (b)
promoting higher incidence of health care behaviors, (3)
modifying certain types of psychological disturbance toward
"normality", and (4) encouraging the maintenance of treatment
gains. Twelve groups of patients (N=51) in a multidisci-

plinary pain clinic were randomly assigned to one of two

35

treatments: training in the structured PPS intervention, or
group discussion about pain and stress management. Treatment
was administered during a one week clinic stay. All groups
received the customary treatment for chronic pain established
at the pain clinic. The design of the study was a pretest-
posttest control group design with follow-up measures.
Subjects completed the McGill Pain Questionnaire, a measure
of pain perception, at pretest, posttest and three month
follow-up. Subjects also completed the Minnesota Multiphasic
Personality Inventory (MMPI) and the Lifestyle Coping
Inventory (Hinds, 1983), a measure of health behavior
performance, at pretest, posttest and three month follow-up.
The follow-up measures were collected by mail with an 80%
(N=41) return rate, with equal mortality in each of the
treatment groups. The groups, not the individual, were the
units of analyses, and the analyses were performed on gain
scores obtained from the pretest-posttest and posttest-follow-
up comparisons.

A.multivariate analysis of variance was performed on the
pain data and found to be non-significant for both the
pretest-posttest and the posttest-follow-up comparisons at the
p=.05 level. A.multivariate analysis of variance was used on
four scales of the MMPI (Hypochondriasis, Depression, Hysteria
and Psychasthenia) and was found to be non-significant at the
.05 level. A one way analysis of variance was performed on

the health behavior data and found to be non-significant at

36
the jp=.05 level. Chi-square ‘tests on. each. of ‘the six
behavioral outcome measures collected at follow-up were found
also to be non-significant at the p=.05 level.

Discussion of the negative results centered around
several features. Small sample size was a methodological
limitation. Analysis was based on only eleven degrees of
freedom since the unit of analysis was the group and not the
individuals. This effectively limited the power of the
statistical tests, and the treatment would have to have had
a very profound effect to ever be detectable with such a small
sample size.

Since the nature of the effect of the PPS on chronic pain
patients is unknown, there was some question as to the use of
the MMPI, the MPQ and the Lifestyle Coping Inventory as being
appropriate for the study.

There was some question as to the group therapy format
since some groups became a therapeutic exploration of feelings
and problems when they had been designed only as control for
group interactions within treatment groups. This confounding
effect may have masked treatment gains or enhanced certain
control groups.

Because both groups in the study demonstrated improvement
on several of the dependent measures, it was difficult to
obtain statistical significance and conclude that one
treatment results in greater gains than the other. Groups

that obtained very good cohesiveness may well have thrown off

37
the experimental results with cohesiveness acting as a
confounding variable.

Jeney felt that the twenty-four steps of the PPS may
require more time to be effectively implemented. The complex
nature of these many steps created some difficulty in those
clients whose early comprehension was lower, and there was
insufficient time to correct for the difficulty.

It was also observed that many of the clients clung to
a more traditional medical model of their treatment, that is
they expected to be cured and did not expect to play an
interactive or proactive part in the treatment process. Since
active involvement is a basic tenet of the program, clients
who viewed themselves as passive recipients of a cure would
not benefit from a program which requires their active
participation.

Pretest differences of individuals in the PPS group
indicated that this group was composed of subjects who were
more psychologically disturbed than the control groups. This
high within-group variance would make it difficult to observe
small differences between groups. In fact, the within-group
variance was almost as high as the between-group difference.

In a similarly designed study Parker (1987) studied the
effect of the PPS program on stress, health behaviors and
academic achievement. The effectiveness of the PPS was
measured for (a) reducing psychological stress, (b) increasing

health care behaviors, and (c) increasing academic achievement

38
as measured by grade point average.

A total of 106 students on academic probation at a state
university were randomly assigned to one of two treatment
groups: training in the PPS and no treatment. Both groups
were allowed to participate in any normal university sponsored
program for those on academic probation. There was a level
of severity’ of’ academic 'underachievement. nested. in. both
groups. The study used a pretest-posttest control group
design with follow-up measures.

Subjects completed the Taylor Manifest Anxiety Scale, the
General Stress Chart (Hinds, 1983) and the Lifestyle Coping
Inventory at pretest and posttest. Quarterly and cumulative
grade point averages were obtained at pretest, posttest and
follow-up after three months. A 99% return rate was obtained
for each of the treatment groups. The group, not the
individual, was the unit of analysis, and analyses were
performed on gain scores obtained from pretest-posttest and
posttest-follow-up comparisons.

A.multivariate analysis of variance was performed on the
psychological stress perceptions and was found not to be
significant for the pretest-posttest comparison (p=.05). A
two way analysis of variance was performed on the health care
behavior performance data and was found to be not significant
at the p=.05 level. Several two way analyses of variance were
performed on the grade point average comparisons and found not

to be significant for either the pretest-posttest comparison

39

or the posttest-follow-up comparison of cumulative grade point
averages (p=.05). A two way analysis of variance performed
on the posttest-follow-up comparison of quarterly grade point
average was found to be significant at the p=.05 level.

Difficulties in the evaluation of stress perceptions of
the two groups may well have been effected by the times chosen
for the posttest and follow-up measures. Both were measured
at times in the class term when students were under maximum
stress due to class requirements. .All students in the program
were also facing severely increased pressures due to the
nature of their' academic difficulty in that they faced
dismissal if improvement was not dramatic. As a result,
pressure was very great throughout the term.of the experiment.

Difficulties in measuring health behaviors was attributed
to learning and maturation effects associated with the
assessment instrument. The Lifestyle Coping Inventory offers
numerous examples in its questions about ways in which health
behaviors may be impacted. Knowledge provided in these
statements makes an indirect, concrete behavioral assessment
that may itself promote change. This may occur regardless of
participating in the PPS program and may well have been a
confounding variable in the experiment.

Use of the grand mean may well have obscured any more
subtle differences that may have occurred between the two
groups. Parker notes that it is also true that the relatively

short elapsed time may not have been long enough for more

4O
dramatic changes to be noted.

The difference between the changes in the two groups from
pretreatment to post treatment in academic performance was
found not to be significant. This was true for each of the
hypotheses concerned with quarterly and cumulative averages.
The long term nature of cumulative averages would not be
expected to change dramatically in such a short period of time
even if treatment was highly successful.

The quarterly differences would be expected to change
more readily, and such differences were noted at the p=.08
level, but not the at p=.05 level. This indicates a trend in
the desired direction, and perhaps this can best be understood
by recognition that the effects of the PPS require time to be
integrated in measurable terms.

It was found that the changes in the two groups from
pretreatment to follow-up were significant in the quarterly
grade point averages. 'This is indicative that the PPS program
does have a positive effect that influences an individual's
future behavior. The complex integration of the many aspects
of the PPS takes time and can be expected to show up after
internalization has had time to take place successfully. The
significant differences noted at the three month follow-up
does lend credence to the notion that additional time did
provide a consolidation of the learning from the PPS program.

In summary, Parker asserts that the dependent measures,

that is the General Stress Inventory, the Taylor MAS and the

41

Lifestyle Coping Inventory may have been inappropriate and
that the time for the program and its effects may have been
too short. He also suggests that a population of engineering
students may have a personal orientation toward concrete
thinking and prescriptive types of interventions. This may
be a very real handicap in learning to apply some of the
abstract and reflective thinking that is required for the PPS.
These personality related factors may well add to the time
required for proper integration of the PPS intervention.

In conclusion, it is suggested that the PPS might best
be applied over longer time periods to allow for better
integration of its many concepts. It was also suggested that
some work be done to determine the optimal size of groups for

training in the PPS.

's o ' ve e u ss a

The development of the concepts of quality assurance in
mental health is a logical result of the quest for quality in
medicine. Since as early as 1500 B.C. the quality and
efficacy of medical treatment has been a concern in the
various schools of medicine in different cultures such those
found in as Egypt, China, India and Mexico. The development
of codes of conduct and ethical behavior which are with us
today originated with Greek prescientific medicine and the
Hippocratic Oath of the Fourth Century B.C. (Wells and Brook,

1988).

42

The description, classification and treatment of mentally
ill patients has taken place since the time of Galen, A.D.
138-201. Early treatment generally involved placement in
institutions or support of benefactors who offered care.
There was little concern from the medical community about
quality of such care. Wells and Brook ( 1988) credit a Spanish
merchant named Juan Ciudad Duarte as a forerunner in the
creation of better treatment for mentally disturbed.patients.
Duarte reportedly experienced a transient psychotic episode
and was treated by flogging. He eventually recovered and went
on to found a hospital devoted to the humane treatment of the
mentally ill. He was canonized posthumously as Saint John of
God. Later, the Order of Charity for the Service of the Sick
was established to carry on his methods.

This order influenced the thinking of Jean-Etienne
Esquirol and Phillippe Pinel who are often credited with
establishing the first modern psychiatric hospital. In his
Medico Philosophical Treatise on Insanity (1801), Pinel set
out his development of clear, logical standards of humane
practice for treating the insane. This work had a great deal
of influence since it took place at the time that psychiatry
was emerging as a legitimate part of medicine (Wells and
Brook, 1988).

. Around 1835, Dr. Pierre A. Louis espoused that physicians
should clearly demonstrate the degree of influence of any

therapeutic agent or technique on the duration, progress and

43
termination of disease. This, he claimed, was the only real
way of showing that treatment was effective. This familiar
concept is the basis of modern quality assurance ideas and
devices in both mental and physical medicine of the nineteen
eighties (Tash and Sadler, 1984).

.As universities began to study psychiatry in the mid
eighteen hundreds, there was an attempt to apply medical
expertise to the emerging study of mental disorders. The
etiology of mental illness and the efficacy of treatment for
mental disease became valid academic pursuits. Wilhelm
Griesinger in Germany was an important influence in this time
period. As a variety of medical schools developed in Europe,
efforts were made to establish credentials for medical
education. This was an essential step in delineating
standards for physicians' education and training.

In the late 1850's, studies conducted by Florence
Nightingale on the efficacy of hospital care in the British
Army led to the assertion that sanitation changes could result
in substantial decreases in fatality rates. Even though these
changes were undertaken in purely medical settings, they
established a method by which care could be reviewed on both
individual and group bases. This approach allowed the
investigation of cause and effect between the methods of care
given and results achieved.

Concern over the adequacy of medical schools in the late

eighteen hundreds led to the foundation of the Council on

44

Education of the American Medical Association. .An evaluation
of medical schools co-sponsored by the Carnigie Foundation in
1908 was critical of the organization, facilities and
curricula that were being offered. Suggestionswere made
regarding the creation of more uniform standards based on
scientific methods of research. From then until the 1960's,
the focus of quality assurance was on the improvement of
record.keeping practices, hospital standardization, licensure
requirements which affected staffing, and joint staffing of
cases to assure proper diagnosis and treatment planning.

The American College of Surgeons developed the Hospital
Standardization Program in 1917 which in part implemented the
concept of review as an essential step in reviewing medical
performance.

An essential step in the emerging field of psychoanalysis
was the extensive training and analysis of the trainee, in
association with required ongoing supervision. These two
traditions are still in place and are essential parts of
accredited training programs and governmental licensure
requirements.

The eventual development of the Joint Commission on the
Accreditation of Hospitals (JCAH) in 1951 was the culmination
of an effort by the American College of Physicians, the
American. College of Surgeons, and the American Medical
Association to find ways to control the quality of medical

care by offering voluntary accreditation to health care

45
facilities. This organization has established standards for
long term facilities, facilities for the mentally retarded and
psychiatric inpatients, and child and adolescent facilities.

The greatest influence on mental health quality assurance
in the last fifteen years has been the advent of government
programs for Medicare and Medicaid. Facilities seeking the
reimbursement of money for these clients were required to meet
standards of care as set forth by the government. The
development of professional standard review organizations,
known as PSRO's, as established by the Secretary of Health,
Education and Welfare, led to mandated quality assurance
programs.

Additionally, private third party insurance carriers
instituted similar programs in those agencies which were
providing services for which they expected reimbursement.
While changes in the public law have led the way, financial
pressures in the private sector have had their impact in
leading to these changes in quality assurance concepts and
practices.

Wells and Brook ( 1988) list five social trends which can
be credited with causing or influencing modern concepts of
quality assurance. First, increased medical and psychological
knowledge has made it impossible for a given physician to
master all medical skills. Hence, specialization has
increased. Because specialized technology and procedures

carry with them increased costs, consumers and insurance

46
carriers (payers) have demanded some assurances of proper
care, as well as protection from harm as a result of new
procedures.

Secondly, third party coverage of medical and mental
health services has led to increased demands for information
about both the necessity and effectiveness of treatment. This
information is used to determine eligibility for reimbursement
and has become standard in the medical field.

An increasing sense of social responsibility as witnessed
in the development of public employment programs and federal
entitlement programs such as Medicare and Medicaid is listed
as a. third factor. Private attempts at providing health care
benefits to larger populations were seen in the .Kaiser
Programs of the 1930's, but the influence of more recent
federal programs has been extensive.

A fourth factor is the rising cost of medical care. All
of the previously mentioned factors have had an impact on the
cost of medical care. In general this has had a profound
effect on medicine in that both consumers and payers ‘want
assurance that money is well spent. Rising costs have
stimulated cost-containment efforts such as diagnosis-related-
groups, known as DRG's. These approaches limit the type of
care or the amount of care delivered and thereby limit costs.
Such cost-containment efforts may have negative effects on the
quality of care.

The fifth general trend has been the development of

47
consumer advocacy. The trend toward accountability for
products and services in the market place is reflected by
similar concerns in health care. Litigation and legislation
supporting consumers' rights have been recognized and
considered. in. making' decisions about the application of
procedures and their effects.

.While these trends can be seen as normal consequences of
market development, the health care market is not a normal
market. The reliance on expert opinion and knowledge means
that . the consumer does not have sufficient knowledge and
information to make decisions in many cases. These
difficulties make it incumbent on the medical profession to
regulate itself to a much higher degree than could be done by

consumer pressure alone (Goldschmidt, 1988).

Current Concepts and Problems in Mental

ealt ua ss a ce
Quality assurance is best conceived as any activity in
the health care delivery system.which assures that the system
will achieve the objective of maintaining and improving
health. ‘In the area of mental health, it is generally
regarded as the implementation of any activity that will
improve the outcome of mental health care.
According to Goldschmidt (1988), quality assurance
mechanisms will essentially (a) measure a criterion attribute,

that is, performance of a health care system, service or

48

provider, b) compare it to some normative standard of
performance, and (c) indicate whether corrective action should
be taken. He also suggests that in an ideal system, the
quality assurance mechanisms would indicate the nature of the
corrective actions needed. This would include some method of
obtaining feedback on the corrective action's results. He
points out that the manner in which this is to be accomplished
is the variable that is in serious question.

Tash and Sadler (1984) report four methods of quality
assurance measurement, based on independent but overlapping
dimensions.

The most commonly used approach is that advocated by
Donabedian (1966) which separates quality assurance activities
into three categories: structure, process and outcome. While
structural appraisal assess the adequacy of facilities,
personnel and instrumentation, process evaluation determines
whether the type of service is appropriate to the needs of the
population being served. Outcome review attempts to measure
whether clients improve as a result of a given treatment.

Woy, Luft, Toprac, Tash and Stahler (1983) suggest that
quality assurance programs consist of either concurrent or
retrospective reviews. Concurrent procedures focus on the
treatment activities while the treatment is taking place.
Retrospective procedures take place after treatment is
concluded. One method implements change during treatment

while the other makes plans for future changes. Both include

49
all the dimensions set forth by Donabedian (1966).

Racusin and Krell (1980) discussed quality assurance in
community mental health centers. They concluded that the
easiest separation was into internal and external evaluation
procedures. 'They also indicate that all aspects of structure,
process and outcome must be included in any comprehensive
review.

Donabedian (1966) has also suggested that there are two
ways of looking at any quality assurance data. These two
approaches include reviewing either aggregated data or
individual cases. Individual case review involves looking at
treatment processes and outcome, giving a concurrent review
in most cases. It may be done retrospectively. Aggregate
data review may involve special studies, program review or
profile analysis of care or service.

Given.these'various.perspectives, it.is important.to look
at what activities are included under this general category
of quality assurance.

The most common form of quality assessment is peer review
of cases performed by peer professionals, either internally
or externally. Peer review may involve medical audits that
determine if appropriate care has been provided. It may
involve 'utilization. review' ‘which. requires that. the
organization being evaluated implement a review plan for
determining how facilities were used. This generally includes

such things as examination of admissions, length of stay and

50
professional services provided. Additionally, certification
of admission and desirability of continued stay are often
examined.

While these procedures are most commonly seen in the
inpatient or hospital setting, they have their analogues in
out-patient settings as well. When these aspects of health
care are applied to mental health services, there are several
obstacles to their conversion to psychological or psychiatric
problems.

Generally mental health care systems have iquality
assurance programs that include the following elements:
1) a scope of focus of the QA system, that is, what
should be included:
2) screening procedures for the selection of the cases
to be reviewed:
3) criteria for what constitutes appropriate care:
4) a data collection instrument for obtaining information
regarding compliance with the criteria:
5) a review format for analyzing data:
6) personnel who are trained to do the assessment:
7) a mechanism to provide educational feedback to induce
corrective changes in the system.
(Tash and Stahler, 1984)
While the standards for quality assurance programs are
generally based on these criteria, medical quality assurance

are often not easily applied to outpatient programs. Some

51

standards are inappropriate, while others may be irrelevant
or nearly impossible to implement. For example, many mental
health centers do not generate a diagnosis at intake even
though medical diagnoses are frequently used for hospital
admissions and decisions about length of stay. The questions
about length of stay become moot when the client is
ambulatory. Length of treatment will therefore become more
relevant and decisions about length of treatment may be much
more complex. Admission criteria become less of a factor in
outpatient settings while in hospitals they are often an
essential factor. For mental health centers, assignment to
a particular treatment program or modality is a much more
relevant question.

There are several other problems with the generalization
of medical health standards across mental health settings.
It is more difficult in mental health care to define specific
processes and to define explicit criteria and standards of
treatment. Attempts to formulate treatment standards have met
with controversy and opposition by professionals in the field.
Therapists have less control over clients and the extraneous
factors that may be present in cases of ambulatory care.
Additionally, there is a greater gap between the measure of
process and outcome. The ambulatory nature of treatment
involves a much 'less concentrated or intensive use of
resources than would be available in an inpatient setting.

This means that cost efficiency for the review process in

52
outpatient settings is reduced.

Another problem is the timing of quality assessment. In
mental health centers, clients' treatments may be spread out
over many months, therefore the timing of assessment
activities may well be much more difficult to determine in
ways that will be meaningful.

While these are problems remain, Tash and Stahler (1984)
believe the most crucial problem is the development of
explicit criteria and standards about what constitutes quality
in mental health. .As Brook, Kamburg and Lohr (1982) indicate,
the standard of care for many treatment modalities has not
been agreed upon. Because therapeutic interventions are often
difficult to describe and quantify, it is frequently
problematic to establish explicit standards of adequate or
quality care. The service record of treatment is the normal
tool for making assessments about quality of care. The type
of information determines the kind of study that can be done.
Whether to do process or outcome studies is bound up in the
nature of the information available. An outcome study might
well be focusing only on success or failure of the program
while ignoring all side effects which were an integral part
of the process (Brook, Kamburg and Lohr, 1982, p. 37).

Quality assessment and quality assurance activities that
focus only on the number of therapist hours or the size of
treatment groups often miss an essential point: the

improvement of the individual 's mental health must be observed

53
and measured by specific criteria before overall program
analysis can be effective (Brook, Kamburg and Lohr, 1982: Tash
and Stahler, 1984: Affeldt, Roberts and Walczak, 1983).

As Rodriguez (1984) has pointed out, psychiatric records
are difficult to evaluate, and the problems of measuring
success of treatment remain the critical issue. Improvement
in record keeping does little for the confusion which results
from unmeasurable treatment results. Rodriguez concludes that
data on the efficacy of psychotherapies and related objective
measures is essential in the evaluation of quality assurance

programs (p. 27-28).

Research in Quality Assurance
While the evaluation of quality assurance methods would
ideally include structure, process and outcome as suggested
by Donabedian (1980), most quality assurance research
conducted at hospitals levels have emphasized structural
concepts. This is in direct contrast with mental health
research which generally involves the evaluation of process.
There are very few experimental studies designed as
comparative or controlled studies. Cohen (1988) reviewed the
research in quality assurance, including several that were
implemented to affect quality of services provided, and he

found only two studies that were comparative.
The most extensively studied quality assurance research

is a study set up in 1977 by the American Psychological

54

Association and the Civilian Health Medical Program of the
Uniformed Services. This study, referred to as the
APA/CHAMPUS study, consisted of three levels of case review
within the CHAMPUS plan which consists of approximately eight
million members nationwide. Level 1 review was a fairly
technical review which evaluated the eligibility of
beneficiary and provider in determining if service was a
covered benefit. Level two consisted of case review by
nurses. Using APA process criteria, they completed a
treatment report which included a description of patient
problems, the goals of treatment procedures and the patient's
progress toward these goals. Level 3 review consisted of
random assignment to three psychologists who reviewed each
case. For every claim selected, three reviewers independently
made recommendations concerning reimbursement for previous and
proposed care. This procedure evaluated the clinical process
with explicit screening criteria and professional peer review.

Cohen (1981) conducted. a study' of this. process by
comparing the responses of psychodynamic and non-psychodynamic
therapists in the review process. Two hundred and twenty
APA/CHAMPUS reviewers received a treatment report and 171
returned completed review reports (78%). Six clinical
treatment reports were created, specifying that the client was
a forty year old who had undergone either 40 or 60 treatment
sessions and that the number of additional sessions depended

on progress. Each claim specified dynamic therapy on a once-

{55
a-week basis.

Reviewers were asked to rate on a five point scale the
following: (a) necessity of care, (b) evidence of functional
impairment, (c) adequacy of goals, (d) appropriateness of
treatment, (e) adequacy of reported progress, and (f) overall
adequacy of care. A 6x2x2x2 multivariate analysis of variance
was performed.using reviewers ratings of treatment, number of
previous sessions recommended. for' reimbursement, the
percentage of previous care recommended for reimbursement, and
the recommended number of future sessions.

Significant results were found for the factors of the
reviewer's theoretical orientation (p=.001), treatment length
(p=.001) and the interaction of theoretical orientation and
treatment length. The most important finding was that
therapists of psychodynamic orientation were much.more likely
to approve of longer treatment than were eclectic/non-
psychodynamic therapists. They were much more positive in
their ratings of treatment results and more generous in their
recommendations for past and future reimbursement.

This study raised many questions about how peer review
committees should be composed, whether by dynamic therapists
or others or by combined groups. It clearly indicated that
further research was needed to clarify factors relevant to
peer review and to aid in policy making decisions about the
establishment of mental health quality assurance systems.

Similar studies (Cohen, 1981: Cohen and Oyster-Nelson,

56
1981: Cohen and Pizzirosso, 1982) have begun to build a data
base of information about the effectiveness of peer review
activities as seen in the APA/CHAMPUS methodology. The basic
conclusion from this series of studies was that 1) there is
a need for more scientifically based studies of quality
assurance methods and 2) there is an abundance of issues that
bear a closer scrutiny in the area of peer review of services.

A study based on a similar three level review was
conducted by the Psychiatric Utilization Review and Evaluation
Project at Yale University. Heinz, Goldblatt, Flynn and
Garrison (1974) compared three methods of peer review using
the same charts. An open ended evaluation was compared to a
three level evaluation and a structured questionnaire that
contained forty-seven specific questions regarding the case
file. Twenty-seven charts were divided into three categories
with.three charts.per'categoryu They consisted of adolescents
who had been (a) treated as out patients only, (b) talked of
suicidal ideation or made suicidal attempts or (c)
schizophrenics who were being treated as inpatients. The
responses ‘were graded and. weighted so as ‘to ‘provide a
numerical rating scale.

An analysis of variance was employed to determine whether
or not the results obtained using the three methods differed
concerning assessment of general adequacy, specific adequacy
of patient care and recording of patient information. The

results indicated that there were no significant differences

57
due to method of review. A.measure of the time for completing
evaluations did indicate that the questionnaire method took
only 67% of the time of the individual review and 75% of the
time required for the three level review.

Heinz and his associates suggest that the three level
review'process works as well as the other'methods and suggests
that it is worthy of further review. They make little of the
25% difference in time allowed for the review process.

Another study of the relative merits of different review
styles was done by Hays (1977). This study compared three
methods of case review for admission criteria into a state
mental hospital, The reviewvmethods consisted.of 1) reviewers
who were in house hospital staff, 2) local mental health
professionals as consultant reviewers and 3) researchers from
the Texas Institute of Mental Services as reviewers. Each
team consisted of a psychiatrist, a clinical psychologist, a
psychiatric nurse and a clinical social worker.

Results using a binomial test comparison found that
global judgements of admission were in agreement at the p=.01
level of significance at one facility and at p=.05 level at
two other facilities. The research team concluded that review
of admissions. procedure and. process could. be adequately
managed by a combination of in house and traveling review
teams.

Luft, Sampson and Newman (1976) conducted a therapist and

client follow-up survey to assess the impact of quality

58

assurance peer review on the therapy process. Eighty
therapists and fifty patients were surveyed regarding therapy
and results. They were also asked about.how they felt therapy
and its results were affected by the peer review of their
cases. While the results of many specific issues were
discussed, there was remarkable agreement between clients and
therapists on the progress made. Therapist and clients were
in agreement 77% of the time on the amount of therapeutic gain
which was made (Chi-squared tests with p=.002). Clients and
therapists agreed to a significant level about who needed more
therapy and who did not (Chi-squared tests at .04). There was
no indication that peer review was seen as a deterrent in the
formation of therapeutic relationship by either therapists or
clients. The therapist choice of treatment modality was
reported to have been negatively affected by the peer review
process in only ten percent of the cases. One fourth of the
therapists who had presented.ten.or more cases for review said
they thought more concretely and systematically about their
treatment plans and goals as a 'result of the peer review
process.

Luft, Sampson and Newman conclude that the peer review
process was not as disruptive as some might fear and that it
appears to help therapists focus on treatment goals more
concretely and clearly. They concluded that the presence of
third party payment requires fluidity about treatment goals.

One third of the clients were concerned about confidentiality

59
while another third felt they would benefit from review of
their cases by a committee composed of other professionals.
Therapists agreed three fourths of the time that the review
of their cases was a benefit in helping them confirm diagnoses
and resolve difficulties involving transference and
countertransference.

It was also discovered that clients used less than three
fourth of their allotted visits, and therapists and clients
were in remarkable agreement about the length and adequacy of
treatment. The study concluded that peer review for this
system effectively monitored the appropriateness of treatment
and allocated limited treatment funds without unduly
interfering with treatment patterns, processes or outcomes
(Luft, Sampson and Newman, 1976, p. 895).

Dall and Claiborn (1982) evaluated the Aetna peer review
pilot study in which Aetna Insurance Company had requested
cases be submitted for peer review by APA reviewers. This
involved sending to a selected board of APA reviewers
approximately 4% of the cases covered in a given time.
Benefits for clients who had reached a level of 24, 40 or 60
treatment sessions were used. Forty-five cases were reviewed
in the six month time frame selected. Results of the study
indicated that 90% of past care cases reviewed were
recommended for payment, a tacit agreement of diagnosis and
treatment plan. Reviewers were less reluctant to "intervene"

in planned treatment, and termination of treatment was

60
recommended in only 27% of the cases. Partial Approval (24%)
and Approval (18%) were also noted. Reviewers were unanimous
in 27% of the cases and there was a 2/3 majority in 77% of the
cases.

It was also evident that there were severe problems with
information available in the charts. Often such things as
marital status were not present in the charts, and 76% of them
had either no progress report or had an extremely vague
reference to treatment goals. These were regarded as major
deficiencies in the charts.

Dall and Claiborn concluded that the reliability of the
reviewers was less than might have been expected or preferred.
The inadequate treatment forms did not elicit adequate or
relevant descriptive information, but reviewers appeared to
give the providers the benefit of the doubt. This was in
direct contradiction to the reviewer instructions which
indicated that inadequate information should be an indicator
of less than adequate care. The findings suggest some
interesting and troubling insights into practice and peer
review. They could not explain why female reviewers seemed
to be "tougher" than their male counterparts or why
dynamically oriented care was reviewed.with greater leniency.
Differences in the adjudged adequacy of treatment reports as
a function of the reviewers gender raises a question about
possible gender determined inappropriate treatment. Further

study may determine some answers to these questions.

61
Education as to the nature and conduct of the peer review
process may enlighten future reviewers as to some of their
responsibilities and potential problems in reviewing cases.
Finally, they conclude that the peer review process is an
integral part of the commitment by professional psychologists

to quality assurance (Dall and Claiborn, 1982, p. 8).

Summerx

As Cohen (1989) has indicated, the basis of treatment
review in.quality assurance requires the presence of valid.and
reliable information about screening criteria, process
judgements and treatment documentation. The critical
ingredient is valid and reliable measures of the assessment
and treatment plans as they are implemented. It is for this
reason that the present study was conducted.

The Personal Paradigm Shift provides a structured, goal-
oriented, education-based treatment program that has shown
itself to be viable in the treatment of stress, psychosomatic
problems and in the positive changing of student grades.
Proper implementation of the PPS to other problem areas
requires the establishment of a measurement instrument for
determining starting points and for measuring progress
throughout the program. The Personal Change Questionnaire has

been designed to accomplish this purpose.

CHAPTER III
DESIGN OF THE STUDY

The methodology of the study is described in chapter III.
The chapter includes a description of the Personal Paradigm
Shift and the Personal Change Questionnaire. The documents
are compared, section by section, and the derivation of the
items on the PCQ is explained. The chapter also includes a
description of the sample used in the study and.the procedures
followed in obtaining the sample. Finally there is a
discussion of the research hypotheses and an explanation of
the logic for the choice of the statistical and procedural

methods.

sc ° ' 0 so
e e so a 'o 'r

The Personal Paradigm Shift was developed by Hinds (1983)
as a psychoeducational approach to personal behavior change
and health care management. It is a planned and organized
method of teaching individuals about the process of personal
change. It was developed from the observations about the
process that people go through when making personal changes.
It follows established psychoeducational principles and it
requires the psychologist to become a teacher while the client

becomes an actively involved student. It assumes that health

62

63
care Ibehaviors are functional relationships that can Ibe
learned or changed (Jeney, 1985; Parker, 1987).

The Personal Paradigm Shift moves through a logical
sequence of stages. The change process requires that clients
experience (a) personal awareness, (b) decision making
processes, (c) oppositional alternatives to present behaviors,
and (d) planning and action that leads to change. Each of
these stages is composed.of detailed steps.designed.toieducate
the client about the relationships between external behaviors
and internal phenomena such as thoughts, feelings and bodily
reactions. Internal processes often maintain or exacerbate
personal problems or distress. Understanding how this takes
place is a critical step in the client's education about
change.

The Personal Change Questionnaire consists of two forms
which are designed to evaluate both client and therapist
knowledge about the client's understanding of personal change.
Since the test instrument is based on the Personal Paradigm
Shift, the PCQ draws its items directly from the individual
steps of the PPS. The following sections discuss each of the
four stages and their individual steps, and each question on
the PCQ is directly linked to the appropriate step from which
it was derived.

There is an additional section on the therapist form
which is based on observed motivational dynamics and it will

be discussed at the end of the last section which is common

64
to both the client and therapist forms.

The Personal Paradigm Shift begins with a stage referred
to as Personal Awareness.
Sta : rs w ss

The Awareness stage is designed to guide clients through
a self analysis of their present behaviors and to enable them
to confront the personal dynamics of their present problem or
difficulty. This self analysis of their problem and its
antecedent. or concurrent 'thoughts and feelings is often
difficult because clients are not familiar with this type of
self observation. It is essential in Stage 1 that clients
become aware of and examine relationships between internal and
external variables that lead to the creation and maintenance
of their personal problems. Step 1 requires that clients
describe the problem on which they want to womk. At this
point, problems are usually described in general terms such
as being "over-weight" or having "test anxiety".

The Personal Change Questionnaire is a series of
statements about personal self knowledge to which the patient
is asked to respond. Each part of the PCQ relates directly
to one stage of the PPS or to a specific aspect of one stage.
So for this stage of the PPS, the PCQ states directly: "I know
what my main problem is." The clients are asked to rate
themselves on a scale of self knowledge from 0 to 7, with 0
being "Don't Know" to 7 meaning "Know Very Well".

Additionally, the PCQ provides an example of the type of

65
response that might be given. The PCQ states: "I get depressed
and can't work."

This section of the PCQ appears in this form:

DON'T KNOW KNOW
1. I know what my main problem is.
For example: I get depressed and can't work. 0 1 2 3 4 5 6 7

The client is requested to circle the number which corre-
sponds to their own personal level of self knowledge. They
are directed.to respond.according to their own.actions, goals,
thoughts, feelings, etc. They are told that goals refers to
plans for changing their old actions.

The therapist form of the PCQ is divided into similar
sections but each section is labeled. The section labels do
not appear on the client form. The statements are similar to
the client form, and.the therapist.has been instructed to rate
the client on how sure the therapist believes the client is
about each of the statements. Therapist ratings are based
solely on the therapist's knowledge of the client. The form
appears in this format:

A. Personel Awareness:
How sure are you that the client: doesn't know knows
1. what their main problem is? 0 1 2 3 4 5 6 7

The format remains the same until question #27. In the
interest of space, therapist form answers will not be given
for each part of the PCQ. The change in format starting with

statement #27 will be presented at the point it appears.

66

Step two of Stage 1 asks the client to set some initial
goal or goals related to the presenting problem. Stating
specific behavioral goals forces the client to focus on
possible favorable outcomes. Once again they are usually
stated in general terms, such as "I want to return to work."

The PCQ states in item #2: ”I know specific goals that
will help me solve my main problem". For example:"I will lose
twenty pounds and have more energy for work."

Step three of Stage 1 requires the client to identify
specific events that precede the occurrence of their problem.
This can be a«difficult task since many people feel that their
problems just "seem to happen." A record may be generated in
the PPS of the places, people, times and events that appear
to go with the problem situation.

The PCQ states in item #3:"I know the situation in which
my problem takes place." For example:"I get depressed when
I am alone too much."

Step four of Stage 1 is designed to develop understanding
of the process by which the antecedent events cause an effect
in the client's internal reactions. The client is instructed
to identify and list internal reactions such as beliefs,
feelings, mental images and physiological reactions that they
commonly experience in the problem situation. The long term
goal of this step is to create and nurture an understanding
of the relationship between external events and the internal

reactions.

67
The PCQ contains two statements related to this step:
#4: I know my own feelings in the problem situation. For
example: I feel lost and very nervous.
#5: I know my own thoughts in the problem situation. For
example: I wonder what is wrong with me.

Step five asks the client to identify specific behavioral
actions taken as a result of antecedent events and internal
reactions. The process herein is designed to assist the
client in understanding how behaviors result from a sequence
of events, both external and internal.

The PCQ states in #6:"1 know what actions I take when I
experience my thoughts and feelings in the problem situation. "
For example: I spend a lot of time reading to escape these
thoughts and feelings.

Step six addresses the consequences of the actions taken
in the problem situation as discussed in step five. These
consequences are discussed in both internal and external
terms. Links are established between the consequences of
behavior and personal self judgements. This link is seen as
essential since people often 'make self judgements based on the
consequences of their actions. This step of the PPS also
addresses the needs that the client is meeting by their
present problem behavior. The acknowledgement of these needs
becomes a foundation of client self understanding.

The PCQ addresses the consequences of behavior with two

statements:

68

#7: I know the consequences of my own actions in the
problem situation. For example: The more I sit and read, the
more I feel isolated and alone.

#8: I know whether I am praising or criticizing myself
in the problem situation. For example: I usually put myself
down for feeling so low.

Step seven, the final step in Stage 1, is termed "taking
a loss". The client is asked to identify how needs are met
or which belief must be "given up" in order that change take
place. A central concern of the PPS is that personal change
involves giving something up as well as gaining something.
This focus is unique to the PPS and is not elaborated upon in
other models of behavior change.

Step seven also addresses in detail the personal beliefs
about what a client feels is needed for survival in life. The
ways in which a person holds to and acts upon their beliefs
is important to their outlook on the world. Adjusting these
beliefs is essential to the process of change.

The PCthakes.two statements related.to losses and needs:

#9: I know what I will lose if I give up my present
actions in the problem situation. For example: I will lose
some of the attention I will get from some of my friends when
I am depressed.

#10: I know what needs of mine are being met by my
present actions in the problem situation. For example: Some

of my needs for attention are being met.

69
W

This stage requires that the client use the information
they have gained from the awareness stage. They determine how
this information will influence their decisions about making
changes. The process consists of three steps: constructing
a balance sheet, making a commitment to change and confronting
the barrier of fear; These are steps 8, 9 and 10. This stage
of the PPS is similar to other behavior change programs.

In an attempt to understand specific inhibiting
conflicts, the client is required to construct a balance sheet
of thoughts and feelings surrounding personal changes. The
writing down of these factors makes it difficult to avoid
confronting them. In this way, unconscious forces are raised
into conscious awareness and their influence on change can be
mitigated. Specifically, the perceived punishments and
rewards for changing or for remaining the same are brought
forth. Clients can then determine more readily some of the
forces which may have been hidden but still acting in previous
attempts at change.

The PCQ addresses these issues with four statements:

#11: I know what punishments I will get when I change my
actions in the problem situation. For example: It would feel
like a punishment to lose some of that special attention.

#12: I know the rewards I will get when I change my
actions in the problem situation. For example: If I change

my actions I won't be alone and depressed as much.

7O

#13: I know the punishments I will get if I don't change
my actions in the problem situation. For example: If I don't
change I can expect to feel isolated and alone a lot.

#14: I know what rewards I will get if I don't change my
actions in the problem situation. For example: I can remain
depressed and not have to go to work.

Step nine requires a commitment from the client, both
cognitive and emotional, to making a change. This change is
based.on information.from previous steps“ The client is asked
for a decision to continue working on making personal change.
It is possible that the current situation is satisfactory to
the client. In such a case, no such commitment is made.

The PCQ addresses the question of commitment directly.

#15: I know how much I am committed to making a change
in my actions. For example: I am willing to miss some of my
favorite activities to attend therapy sessions.

The inhibiting nature of fear on change is a central
concept to the PPS. At this point in the PPS program, step
10, the client confronts the resistance to change by examining
fears about change. The experience of these unpleasant
emotions is discussed and the client identifies the defenses
normally used to avoid these feelings.

The PCQ addresses fears in two ways.

#16: I know the fears I will have to face if I don't
change my actions. For example: I will have to face the fear

that I am a failure.

71

#17: I know how to avoid my fears. For example: I can
avoid my fears by staying very busy, reading or playing
pinball.

ta e : O 05' A te 'ves

This stage of the PPS is designed to help the client
replace old internal reactions with new ones. Replacement is
essential since failure to learn new, replacement behaviors
will result in the re-emergence of the old behaviors. In this
stage, the client will create new internal reactions that are
different and often counter to the old internal reactions
identified in the awareness stage. The steps of stage three
also fosters a sense of self control over internal reactions
and subsequent behaviors.

Step 11 is essentially a repeat of step 3 in which the
client is asked to record information about the stressful
conditions which normally surround the problem. This step
is repeated to emphasize that the client cannot necessarily
control external events, but that they can learn to control
internal processes which are a result of external people,
places or events.

Since this step is a repeat, it is not directly addressed
as a separate statement on the PCQ.

Step 12 is the most essential and difficult step in the
PPS. Changing the internal reactions of the client requires
that they experience a shift in the way they regard events.

This is the "paradigm shift" phase and requires a concentrated

72
effort to develop new feelings, thoughts, mental images,
physiological reactions and self judgements. Each of these
internal reactions is addressed directly by the PCQ.

The PCQ states:

#18: I know'I can bring about new feelings in the problem
situation. For example: I know'I can develop feelings of self
confidence even if I do fail.

#19: I know I can bring about new thoughts in the problem
situation. For example: I can think about making new friends
when I do things rather than being alone.

#20: I know I can bring about new mental images in the
problem situation. I can see myself being congratulated for
doing a good job and being successful.

#21: I know I can recognize new bodily reactions in the
problem situation. For example: I'll be able to feel lighter
and more energetic when I lose twenty pounds.

#22: I know I can bring about new self judgements in the
problem situation. For example: I can think of my depression
as a stage which I can successfully get through.

Step 13 has the client develop alternative behaviors to
the problem situation. At this point, any alternatives are
acceptable. No attempt is made to determine whether these
alternatives are appropriate. The results of these
alternative behaviors are dealt with in step 14. In the
realistic application of the PPS, these two steps may well be

taking place almost simultaneously as alternatives and the

73
results of alternatives are discussed.

The PCQ states:

#23: I know I can bring about new personal actions to
solve my own problems. For example: I can do something active
rather than bury my head in a book for two weeks.

Determining the results of these alternative behaviors
is the essence of step 14. While some possible alternative
behaviors may be negative, the emphasis is on developing new
behaviors which will result in maximizing personal gains.
While new or different losses may be expected, gains made in
new behaviors will outweigh any new losses. These new gains
may include such internal reactions as more self-control,
self-esteem.and self-confidence. IMeasurable external factors
may emerge such as weight loss or reductions in test anxiety.
The PCQ addresses consequences, self evaluations and needs
with two statements.

(#24: I know I can recognize the consequences of my new
actions in the problem situation. For example: I'll be able
to see that staying active helps me avoid unnecessary
depression.

#25: I know I can bring about new self evaluations as a
result of my new actions in the problem situation. For
example: I can feel more in control and feel good about taking
positive actions.

#26: I know I can recognize needs that will be met by my

new actions in the problem situation. For example: Being

74
active will help me meet new people and satisfy my needs for
attention.
W

This final stage of the Personal Paradigm Shift is
designed to provide a foundation for change. Realistic steps
are provided for making and carrying out plans to bring about
desired changes.

It incorporates reasonable steps such as goal setting,
finding role models and support groups, and devising specific
methods of record keeping and self evaluation. This approach
is set up to reduce fears about change and.to enhance feelings
of control about the ongoing process. This foundation for
change is reality based and provides for feedback about
change.

Step 15 requires that a specific, realistic short-term
goal to be defined by the client. The step provides for the
establishment of behavioral goals which are attainable. This
provides ongoing motivation for change.

The PCQ addresses this step as follows:

#27: I know I can choose a realistic starting goal for
myself. For example: I can choose to lose two pounds in the
first month of dieting.

At this point there is a slight change in the therapist
form of the PCQ. The choices of "doesn't know" and "knows"
are changed to "can" and “can't". The therapist choices now

appear as: How sure are you that the client "can" or "can't"

75
(#27) ”choose a realistic first goal for themselves?". This
new format remains unchanged through item #37.

Step 16 addresses the identification and development of
support for the client's desired change. Clients learn that
dependency on others is an essential part of life and that
they can use others as a source of strength and support in
enacting change in their lives. This includes the gathering
of appropriate information about what actually changes.

The PCQ addresses this step with two statements:

#28: I know I can find someone to support me in making
a change. For example: I could attend Weight Watchers or
arrange a conference with my doctor.

#29: I know I can seek information for making some
changes. For example: My doctor can give me some good
information about losing weight.

The learning concept of modeling is the focus of step 17.
Clients are encouraged to identify and seek information from
or about others who have made similar changes. This step
requires the client to identify a model whom they can use to
enhance their own personal changes.

The PCQ covers this step as follows:

#30: I know I can find a role model. For example: A.good
friend at work lost forty pounds and he looks great.

Step 18 deals with learning to provide self rewards for
the accomplishment of the initial change goal. Positive

reinforcement which is immediate will act as motivation for

76
the continuance of the change process. This is a critical
component of operant learning and it provides a method by
which the client can learn self motivation and self control.

The PCQ states:

#31: I know I can give myself rewards for making these
changes. For example: When I lose weight, I can get that new
leather jacket I've been wanting.

Organizing a specific plan of attack for working on their
problem is the goal of step 19. Clients determine the
sequence of behaviors they expect to perform in order to
achieve the desired change goal. The essence of this step is
that a predictable plan or blueprint increases the client's
control and helps them maintain motivation.

The PCQ addresses this directly:

#32: I know I can make an organized plan for achieving
these changes. For example: I can make a daily schedule for
my exercise and food intake.

Step 20 deals with developing a feedback system for the
client to measure their progress toward their change goal.
The development of effective feedback mechanisms is important
for making changes so that the client not be discouraged by
slow or seemingly small changes. A lack of consistent
positive feedback can be a negative influence on the change
process. Teaching ways of self evaluation which are effective
will provide positive feedback and thereby enhance motivation.

The PCQ states:

77

#33: I know I can give myself feedback on my progress.
For example: I can keep a record of my weight each Monday to
see how I am doing.

Teaching the client about establishing a time line for
change is the purpose of step 21. This encourages the client
to maintain their efforts and maximize the chance of success
by staying on their time schedule. This scheduling sets time
oriented goals that establish a sense of continuity and
security for the attainment of change. A set schedule also
diminishes the opportunity for procrastination about the goal.

This is directly addressed in the PCQ as follows:

#34: I know I can set a realistic period of time for
change to take place. For example: If I work on losing a
pound a week, I'll reach my goal by the new year.

Step 22 involves learning making backup plans. Since all
plans include the possibility of failure, clients are taught
to develop alternates to their first plan. Knowing how to
deal with failure enhances motivation and feelings of self
control. I

The PCQ states:

#35: I know I can make a backup plan if I fail to reach
my goal. For example: If my weight loss is too slow, I can
increase my daily exercises a bit at a time.

Fear of success is often an unconscious motivator that
can undermine the most earnest attempts to change. This is

addressed in step 23 as clients are taught about the fears

78
associated with the risk of making change. Expressing and
learning ways of dealing with these fears enhances self
control and maintains motivation.

The PCQ states:

#36: I know I can make a plan to avoid sabotaging or
undercutting my success. For example: Each week I will review
what I have eaten to see if I have followed my plan.

The last step instructs the client in how to keep a
record of their results. This record in the form of a diary
or chart will provide specific and concrete information about
gains or losses as the plans are put in action. This tends
to provide ongoing support and motivation for the client.

The PCQ states:

#37: I can keep a diary or chart to follow my change.
For example: I can make a chart that includes my daily calorie
intake and exercise schedule.

This is the end of the client form.

It was noted in application of the Personal Paradigm
Shift that certain clients did much better in completing the
program successfully. It was proposed that motivational
factors appeared to be present in some individuals who
appeared to be more successful. These motivational factors
are included on the therapist form and discussed in the

following section.

79
Meriyerienel_pineneienr Therapist Form Section C)

1. Non-defensiveness: How non-defensive is the client?
On. the basis of‘ past interaction ‘with. the client, the
therapist is asked to evaluate how defensive the client is in
the therapeutic situation. There is some belief that less
defensive clients work.through.the program more successfully.
.It is expected that a higher non-defensiveness rating
corresponds to better success. Therapists may be able to
determine some aspects of projected outcome by observation of
the client's defensive or non-defensive attitude in the first
few sessions.

2. Introspection: The therapist is asked to rate the
client on how‘well they look inside themselves. It is thought
that those who can be more introspective may be more
successful in the program.

3. Engagement: How well does the client engage the
therapist in the sessions. Basically, are they attentive and
communicative about their problem or do they remain distant
or distracted in their interactions in therapy?

4. Self-disclosure: The therapist is asked to rate the
client on how well the client can open up to the therapist
about themselves and their problem. How well can they discuss
the various, highly personal aspects of their problem?

5. Commitment: How committed is the client to making a
change? Based on their interaction with the client, the

therapist is asked to rate the client on the client's

8O
commitment to making a change.

6. Emotionally’ ready to ‘make a change: Does the
therapist feel that the client is ready emotionally to become
involved in the program?

Each of these dimensions is based on both deductive and
intuitive processes within the therapist. While it is true
that therapists operate in different ways, it is also true
that therapists' intuitions may be valid reflections of client
thinking. In addition, they always affect the interactions
of client and therapist. An attempt is being made on the PCQ
to measure these intuitive factors of the therapist and

compare them to the other client factors.

umma o t P d the P

The Personal Paradigm Shift was designed to teach clients
how to change. It employs a psychoeducational approach and
it is highly structured and detailed in addressing a number
of variables about change. While psychoeducational models
have been applied with success in many settings, and the PPS
has shown success in limited settings, there has been no way
to measure client status for this program. Additionally, no
method for determining specific progress through the program
has been developed. The Personal Change Questionnaire is
based on the tenets of the PPS and has been designed to
measure the status of clients in therapy. In addition, it is

hoped that the PCQ will prove to be an effective measure of

81
client progress through the PPS program. Each part of the PPS
is directly addressed by a corresponding part of the PCQ. The

PCQ is designed to be administered at any time in therapy.

MW

Subjects for the study were selected from the clientele
of several mental health facilities. These sources included
at least one drug inpatient treatment facility for criminal
offenders with drug addictions and one community mental health
facility as well as several private practice settings.

The client pool consisted of 115 subjects. Of these
subjects, 86% were between the ages of 21 and 45, with 51%
female and 49% male. See Table 3.1 below and Table 3.2
on page 82.

Table 3.1: Age of Subjects

 

Age Range Frequency Percent
16-20 7 6.1
21-25 18 15.7
26-30 21 18.3
31-35 19 16.5
36-40 19 16.5
41-45 22 19.1
46-50 6 5.2

50+ 3 2.7

 

82

Table 3.2: Sex of Subjects

 

Sex Frequency Percent
Female 59 51
Male 55 49

 

Of this sample, 44.3% were single, 37.4% married, 12.2%
divorced and 1.7% widowed and 4.4% in other categories.

Table 3.3: Marital Status of Subjects

 

Status Frequency Percent
Single 51 44.3
'Married 43 37.4
Divorced 14 12.2
Widowed 2 1.7
Other 5 4.4

 

Thirty-nine percent had a high school diploma or GED,
while 11.3% had not attained a high school diploma. Those
with two years of college made up 15.7% while those with a
four year degree made up 19.1%. Those with graduate work
comprised the final 14.8%. This is summarized in Table 3.4
on page 83.

In the area of employment, 32.2% were employed on an
hourly basis. Salaried persons made up 29.6% while the self

employed accounted for 7% of the sample. Housewives comprised

83
7% and 20.9% were unemployed. See Table 3.5 below.

Table 3.4: Educational Level of Subjects

 

Category Frequency Percent
No HS Diploma 13 11.3
HS or GED 45 39.1
2 yr College 18 15.7
4 yr College 22 19.1
Grad School 17 14.8

 

Table 3.5: Employment Status of Subjects

 

Category Frequency Percent
Self Employed 8 7.0
Hourly Wage 37 32.2
Salaried 34 29.6
Housewife 8 7.0
Unemployed 24 20.9
Other 4 3.5

 

Income levels are reported in Table 3.6 on page 84.
Twenty-four percent of the subjects were in the 0 - $10,000
range while 9.6% were in the $10,000 - $20,000 range. .Another
15.7% were between $20,000 and $30,000 with 13% between
$30,000 and $40,000. A group of 9.6% were in the $40,000 to

$50,000 range and 26.1% were above $50,000. See Table 3.6.

84

Table 3.6: Income Level of Subjects

 

Range
0 - $10,000
$10,000-$20,000
$2o,ooo-$3o,ooo
$30,000-$40,000
$40,000-$50,000

$50,000 or more

Frequency
28
11
18
15
11

30

Percent

24.3

9.6

15.7

13.0

9.6

26.1

 

With regard to therapy sessions attended, 13.2% had from
0-9 sessions and 8.3% had between 10-19 sessions.
28.3% had from 20-29 sessions.
had between 30-39 sessions.
comprised 24.5% of the sample.

Table 3.7:

Eighteen point nine percent
Those with forty sessions or more
See Table 3.7 below.

Number of Sessions

 

No. of Sessions

0 - 9
10 - 19
20 - 29
3O - 39
4O - 49
50+

Frequency

14

30

16

20

11

15

Percent
13.2
28.3
15.1
18.9
10.3

18.9

 

Another

85
Insurance status of the subjects is reported in Table 3.8
below. As indicated, 50.4% had insurance and 36.5% did not.
.Another 13.1% did not know or did not respond to the question.

Table 3.8: Covered by Insurance

 

Category Frequency Percent
Yes 58 50.4
No 42 36.5

Don't Know/

No answer 15 13.1

 

Reasons for counseling were reported as follows: 22.6%
reported substance abuse, 20.9% reported a family problem,
15.7% reported anxiety and 14.8% indicated.that.depressionnwas
the problem. A 2.6% sample reported a physical problem, 9.6%
reported they were required to come, and 13.9% reported
another, non specified problem as their reason for therapy.
See Table 3.9 on page 86.

Subjects also reported on the length of time they had
been experiencing their problem. Of the respondents, 78.3%
reported having had the problem more than two years. Only
9.6% reported a time period of 1 - 2 years and 7.0% for three
months to a year. Less than one per cent reported a time
period of less than 3 months. A summary of this information
is found in Table 3.10 on page 86.

Therapists were asked to describe their treatment

86

Table 3.9: Reason for Seeking Counseling

 

Category Frequency Percent
Anxiety 18 15.7
Depression 17 14.8
Fam. Prob. 24 20.9
Sub. Abuse 26 22.6
Phys. Prob. 3 2.6
Had to Come 11 9.6
Other 16 13.9

 

Table 3.10: Time Subject had the Problem

 

Time Period Frequency Percent
0 - 3 months 1 .9
3mo-1yr a 7.0
1 yr - 2 yrs 11 9.6
2 yrs or more 90 78.3

 

orientation and did so in the following manner. Only 10.4%

were self described as Psychodynamic while 24.3% were
Cognitive Behavioral. Gestalt and Psychoeducational both
reported.only 1.7% and.the remaining 60.9% were self described
as eclectic. Each subject was coded on the client form and
had a treatment orientation though there were in fact less

than 115 therapists. The information was coded in this manner

87
when it was noted that therapists did not necessarily describe
themselves the same for each client. This information is
summarized in table 3.11.
Table 3.11: Therapist Treatment Orientation

(per Subject

 

Category Frequency Percent
Psychodynamic 12 10.4
Cog-Behavioral 28 24.3
Psychoeducational 2 1.7
Gestalt 2 1.7
Eclectic 70 60.9
other 1 .9

 

The female/male distribution of the therapists is
reported in Table 3.12 below. Of the participating
therapists, 53% were female and 47% were male.

Table 3.12: Sex of Therapists

 

Female 53%

Male 47%

 

1'ab' ' V 'd't
In establishing instruments such as the Personal Change

Questionnaire, attention.must be paid to both the reliability

88

and validity of the instrument. These characteristics are
related, but they are significantly different. Reliability
is concerned with the precision of measurement and the ability
of the instrument to repeatedly make measurements of the same
precision. Validity deals with the question about whether
quantities which are being measured have any usefulness. It
is clear that a precise, repeatable measure, ie., one with
high reliability, may not be useful for a particular purpose
and in that sense have limited validity. On the other hand,
if a ‘measure is not reliable, it can never be valid.
Reliability is a necessary but not a sufficient requirement
of validity (Nunnally, 1970).

' Reliability can best be conceptualized by recognizing
that every measurement has some error in it. This error may
be small and may be due to a variety of causes in either the
instrument or in the measuring process or in both" The larger
this measurement error becomes, the less precise is the
measurement. The greater the error, the less likely is an
accurate repetition of any given measurement. Reliability can
be increased by decreasing the measurement error.

When trying to understand the relationship between events
and behavior, the social scientist is often looking at very
small correlations. If the reliability of an instrument is
low, the measurement error may well obscure evidence of
correlation. It is essential to minimize measurement error

in order to maximize the detection of correlational evidence.

89

Reliability also refers to the generalizability of
measurements. A reliable instrument is expected to give
consistent results across settings such that similar groups
can expect to attain similar results in a variety of
situations. Any reliable test must meet the criteria of
applicability across testing situations, times and subjects
(Nunnally, 1970, p .118).

In summary, all tests have some "zone of uncertainty"
that surrounds it's measured scores. Highly reliable tests
minimize this zone of uncertainty by decreasing measurement
error and increasing precision of measurement.

Validity is a multidimensional term which in general has
come to refer to the test's ability to measure what it
purports to measure. While this may appear straightforward,
validity is not easily determined nor does it have only one
meaning. The physical or hard sciences may provide reliable
and valid measures such as height, weight or age. The social
sciences often. have less ‘well defined concepts such as
intelligence or motivation. Validation of measurements for
these abstract concepts is more complex.

"Validity is a matter of degree rather than an all-or-

none property, and validation is a never ending pro-

cess....New evidence may suggest modification of an
existing measure or the development of a new and better
approach to measuring the attribute in question"

(Nunnally, 1970, p. 133).

90

Validity of a measurement device involves the use to
which it is being put rather than just the nature of the
instrument itself. Nunnally (1970) lists three major purposes
of test instruments and three corresponding types of validity
that need to be considered. These are predictive validity,
content validity and construct validity.

When the purpose of an instrument is to establish an
estimate of behavior, predictive validity is important. This
involves the establishment of some criterion against which
other factors are compared in order to determine if these
factors can predict how well one can expect to do on the
criterion measure. This has been referred to alternatively
as "postdiction" or "concurrent" validity. In all cases it
involves the use of a predictor variable and its relation to
the criterion measure. The predictive nature relates only to
the correlation between the two variables. It does not
include the notion of causality.

Content validity is the term applied to the adequacy with
which a certain test measures the universe of information it
is supposed to measure. Content validity is most applicable
to measures of achievement. It is best assured by obtaining
a representative sample of test items from the larger universe
of possible items and by employing reasonable methods of test
construction. Test construction must control for both
sampling and reasonableness. If not, critical elements within

the universe of items will be ignored, over emphasized or

91
improperly addressed. The degree of content validity "rests
mainly on appeals to reason regarding the adequacy with which
important content has been sampled and on the adequacy with
which the content has been cast in test items."(1970, p. 137)

.According to Nunnally: "If it is agreed by most potential
users of the test, or at least persons in positions of respon-
sibility, that the plan was sound and was well carried out,
the test has a high degree of content validity."(1978, p. 92)
Expert inspection of the items and the logic behind their
selection is an acceptable indication of content validity and
is often used as a first step in test construction and
validation. Additional steps might well include successive
pilot studies to test the usefulness of items.

Content validity is also positively impacted by the
presence of high internal consistency as measured by such
statistical methods as Cronbach's Alpha. A moderate level of
internal consistency is needed as further evidence of content
validity.

Nunnally (1978) also suggests that performance on a test
both before and after training increases the likelihood of
content reliability, as does similar or equivalent performance
on tests that purport to measure the same things.

Construct validity is essential to most psychological
test validation, and it is central to this study. Abstract
concepts approached in social sciences are often known as

constructs since they include a variety of different and

92

possibly divergent thoughts, feelings and behaviors. Such
constructs represent.hypotheses about how these things relate
together or how they might be affected by, particular
treatments. Most constructs are relatively diffuse and have
indistinct boundaries. They may have several central ideas
and other peripheral ideas which are directly or marginally
related. Because of their nebulous nature, construct
validation requires evidence from a variety of directions.
This leads to the notion that no one test or questionnaire can
be sufficient proof of the validity of the construct.

A collection of evidence may lead to the acceptance of
the construct, without the presence of absolute proof.
Nunnally (1970) delineates three requirements necessary for
the development and validation of constructs.

First, the domain of the observable behaviors must be
specified. This requires defining the terms and behaviors to
be observed, and specifying what relationships one would
expect to find among the behaviors. This amounts to outlining
the domain of the observables so that one can examine what
relationships exist among the various observables.

Secondly, construct validity requires some delineation
of what relationships one would expect to find among the
observables if the construct exists. They might all correlate
together, they might cluster into different groups of
correlations or they might fail to correlate at all. Attempts

at validation should include predicted relationships as

93
dictated by the construct as a whole.

Finally, sufficient evidence for construct validity is
the fact that supposed measures of the construct (either a
single 'measure or' a combination. of measures) behave as
expected. While there might be alternate explanations of
their behavior, the collective action along predicted lines
is considered sufficient evidence of existence of the
construct.(Nunnally, 1978, p. 103)

Constructs in the social sciences are often nebulous and
require a winnowing process that may expand or contract the
construct as evidence is collected. In all cases, construct
validity is never achieved as much as it is pursued. In this
light, it is more likely that one observes evidence for
construct validity rather than attaining certainty or absolute
proof that a construct is real. The pursuit of construct
validity involves a statistical approach to correlations and
predicted interrelations of the various components of the
construct.

One additional aspect of validity is that of the general-
izability of results. In order for results to be generalized,
it must be shown that a particular instrument gives results
which are comparable to other tests purporting to measure the
same thing. If this is not the case, one has to recognize
that.there may only'be limited applicability of the instrument

to the specific criteria of the original study.

94
asu i t e St d

The Personal Change Questionnaire was developed as a
measure of client status and change in therapy based on
psychoeducational principles. The PCQ was developed and
administered to clients and a corresponding therapist form was
completed by their therapists. The PCQ, Form S, is a 38 item
self report instrument that measures psychoeducational aspects
of therapy. Items for the PCQ are responded to on an eight
point scale with a low end of zero and a maximum of seven.
The scales range from "Don't Know" to "Know Very Well" for
some items and from "Can't" to "Can" for other items. Each
response is based.on the individual's self knowledge about.the
content of the item.

The Therapist Form of the PCQ has forty-four (44) items.
The first 38 correspond to the Form S items and require the
therapist to evaluate the client based on the therapist's
perceptions of the client's ability to deal with the content
of the items. The Therapist Form has six additional questions
that deal with motivational aspects of the client.

Higher scores are thought to delineate greater client
self understanding and ability to deal with their problems.

The PCQ is comprised of four sub-scales developed from
a model of personal change known as the Personal Paradigm
Shift (Hinds, 1983). These scales are Personal Awareness,
Oppositional Alternatives, Decision Making and Planning and

Action. The development of the items is discussed at length

95
at.the beginning of Chapter III (p. 62-81). Empirical support
for this instrument's reliability and validity are a major
component of this study are presented in Chapter IV. The
design used to test questions of reliability and.validity are

presented later in this chapter.

3 t o e 0

Clients were approached by their therapist after no fewer
than four sessions and asked to participate. They were told
that the survey was part of a study on quality assurance and
that it was designed to evaluate the effectiveness of care
provided by that facility. It was explained that their
participation in the survey was essential in order to evaluate
the survey itself. It was pointed out that it was the survey
which was being evaluated, not the clients. The test itself
was being checked to determine if it will effectively measure
quality in the services which clients were receiving.

They were informed that they were not to put their name
or any other identifying information on the surveyu EThey'were
asked to complete some demographic information as part of the
evaluation, and they were told that this information was to
be used only as part of the evaluation. Such information
would not be sufficient for identification of any individual.
They were informed that all information gathered would be
confidential. Information sheets and test forms were coded

so that statistical analyses could be run later, but anonymity

96
of response was maintained.

All participants in the study signed a release form
stating that they understood the purpose to which the
information was to be put. All persons asked to be in the
study had the option to refuse to participate, a condition
explained to them before they began the test. They were
allowed to change their'minds and.drop out of the study at any
time prior to handing in the evaluation form. The consent
forms were collected prior to the collection of the evaluation
forms. They were placed in different envelopes than the ones
containing the survey forms and the demographic forms.
Participants were reminded that for the survey to be
effectively evaluated, their answers should be as honest and
accurate as possible. They were encouraged to answer to the
best of their ability, and they were told to work as fast as
they could without feeling hurried.

The first attempts at gathering information in the form
of completed tests did not control for race, sex, age or any
other demographic variables. Participants were asked to
report the number of sessions they had attended as part of
their therapy.

Therapist selection was achieved by presenting the nature
of the study to a variety of therapists and asking for their
participation. No attempt was made to select or control for
sex or treatment orientation in this initial study.

Therapists volunteered to participate and select clients as

97

subjects according to guidelines set forth in the research.
These included: 1) clients were to be of legal age, 2) clients
were to be in therapy for reasons that would be judged
primarily psychological, and 3) clients were to be able to
think clearly enough to participate in the study: ie, no
thought disorders serious enough to impair the ability to
complete the questionnaire.

Each therapist was given instructions for administration
of the PCQ which they agreed to follow. These included how
to distribute and collect the PCQ, the demographic forms and
the consent forms. The nature of the research was never
secret and therapists were encouraged to ask questions of the
primary researcher about the nature of the research and plans
for the completed PCQ and possible outcome reports that might

be generated.

£81311

The overall design of the study is descriptive, with
major emphasis on the development of reliability and validity
estimates for the Personal Change Questionnaire. An addition,
the study was designed to determine if there is evidence for
a psychoeducational component of psychotherapy and.to develop
an instrument to measure that component. Development of such
an instrument provides one method of determining quality of
therapeutic services by being able to determine clients'

status in therapy based on the psychoeducational model.

98

The PCQ was developed specifically to be the instrument
to make such a measurement. It was empirically tested for
reliability and validity in order to fulfill this role.
Statistical procedures employed include central tendency,
correlation, t-tests, Cronbach's Alpha and factor analysis.

The descriptive nature of this research does not lend
itself to discussion of cause. Results from the present study
will hopefully provide a basis for further research into use
of psychoeducational principles in therapy and quality

assurance .

Research Hypotheses

Research hypotheses were developed around both aspects
of the study. These included the need for validity and
reliability estimates for the PCQ and the search.for'a psycho-
educational component to psychotherapy. Several research
hypotheses were generated.

The first three dealt with questions of reliability of
the Personal Change Questionnaire.

Reliabilipy of the PCQ:

Hypothesis I: The internal consistency of the total items

of the Personal Change Questionnaire will be sufficiently

high to infer homogeneity of the construct of a

psychoeducational basis for personal change in therapy.

Hyporhesis II: The internal consistency of each of the

99
four subsections of the Personal Change Questionnaire
will be sufficiently high to infer that each subscale

demonstrates internal consistency reliability.

Hypothesis III: The correlations between the total scale
and the four subscales of the Personal Change
Questionnaire will be sufficiently high to infer that the
Personal Change Questionnaire is measuring a uni-

dimensional concept.

Initiel Validity of the Personal ghange Questionnaire
Hypothesis IV: The content of the items on the Personal
Change Questionnaire will be indicative of a
psychoeducational basis of personal change in

psychotherapy.

Hypothesis V: A linear correlation will be observed
between therapist total scores on the Therapist Form and
client total scores on the Client Form S of the Personal

Change Questionnaire.

Hypothesis VI: A.linear correlation will be observed
between the Therapist Form total scores and the therapist

rating of client status in therapy.

Hypothesis VII: A linear correlation will be observed

100
between the therapist rating of client's overall status

in therapy and the client's total scores.

Hypothesis VIII: The mean scores for the items in the
subscales will be ordered such that the scores will be
highest in section I with decreasing scores in each
following section, ‘with 'the lowest scores ‘being in

section IV.

Hypothesis IX: There will be a linear correlation
between the total scores of items on the Personal Change

Questionnaire and the educational level of subjects.

Hypothesis X: There will be a linear correlation between
the total scores of items on the Personal Change
Questionnaire and the number of sessions of therapy

reported by clients.

Hypothesis HI: Four distinct factors will be detected
corresponding to the four subdivisions of the Personal

Change Questionnaire.

Hyporhesis XII: There will be differences in the mean
score for items, subscales and total scores on the

Personal Change Questionnaire between diagnostic groups.

101
This section presented the research hypotheses. Next the

procedures for analyzing the results are presented.

Prpeegures Heeg in Dete Analysis

Statistical procedures were used to test eleven of the
original research hypotheses. Research statistics used to
test the research hypotheses included: Cronbach's coefficient
Alpha, the Spearman correlation coefficient, the Pearson
product-moment correlation and the t-test statistic. A factor
analysis was also performed. Procedures used for analysis of
each hypothesis are discussed below.

Hypotheses I ang II refer directly to the reliability of
the Personal Change Questionnaire. According to Anastasi
(1961), in a broad sense, test reliability indicates the
extent to which individual differences in test scores are
attributable to chance errors of measurement and the extent
to which they are attributable to true differences in the
characteristic under consideration. Mehrens and Lehmen define
reliability as the degree of consistency between two measures
of the same thing (1978). Measurements are reliable to the
extent that they are repeatable (Nunnally, 1978). To attain
an idea of how repeatable a measure is, a common method is to
administer'a given.test.more than once to the same individual.
That being unwieldy or inappropriate, an alternative method
is to administer two comparable halves of the test at the same

time. This is known as the split halves method.

102

Split thalves is considered a measure of internal
consistency since the two administrations of equivalent forms
of the test are contained in one single test. How to split
a test can become a problem since splitting it different ways
might exclude some types of items from one half or the other.
To avoid this difficulty, Cronbach (1951) developed a
technique for finding the average correlation between all
possible. split half combinations. It essentially measures the
extent tO‘WhiCh each itemicorrelateS‘with.the'total test score
and the extent to which each item is tapping a similar
concept. It employs methods by which all item variances are
added together. This provides a way to calculate the internal
consistency without having to devise multiple tests.

"Coefficient Alpha provides a good estimate of reliabil-
ity in most situations, since the major source of measurement
error is because of the sampling content "(Nunnally, 1978, p.
230).

Mehrens and Lehman (1978) suggest that Alpha coefficients
of tests used in making decisions about people should be in
the range of .85, but that for group<decisions, a range around
.65 may be acceptable (p. 122). Anastasi puts the acceptable
figure in the .80's to the .90's (p. 118). Nunnally proposes
a somewhat higher figure for decisions about people, depending
on the particular use to which the test is being put. He
suggests a level of .90 to .95. He suggests a level of .70

for the Alpha value in basic research.

103

For this pilot study, it was determined that values of
.6 would be acceptable for correlations involved in the
research setting.

Hyperneere_III tests the relationship between the PCQ
subscales and the total scale. A Pearson product moment
correlation was computed. This is the single most often used
measure of a relationship between two variables. When a
psychological variable is measurable on eleven.or more levels
they can be said to be continuous and little information is
lost in analysis if this is not the case. For comparison of
the subscales to the total scores, such a correlation is
appropriate.

The Pearson product-moment correlation takes into account
not only the item's position in the group but also the amount
of the deviation from the group mean. It generates a measure
of direction and magnitude of the relationship between the
variables being tested. Values range from +1 for a perfect
one-to-one positive correlation to a -1 for a perfect negative
correlation. This coefficient is the most stable measure of
correlation and has the smallest standard of error of all
correlation techniques (Sesan, 1983).

In order to infer a unidimensional construct, there must
be observed a significant level of correlation between the
total scale score and the sub-scale scores of the test
(Anastasi, 1961, p. 146). The Personal Change Questionnaire

must also meet these criteria.

104

Hypgtheeifly deals with the content validity of the PCQ
and was not tested statistically. Content validity is related
to how adequately the content of the test items samples the
domain about which inferences are to be made (Mehrens and
Lehman, 1978, p. 124). There is no statistical method for
determining .content validity since it is determined by a
thorough inspection of the items. The content area must be
systematically analyzed to insure that all major aspects are
adequately covered by the test items. In addition, the items
must be sampled in the correct proportions (Anastasi, 1961,
p. 136).

As discussed previously (p. 86), content validity rests
mainly upon appeals to reason as to the appropriateness of
content and the adequacy to which content has been addressed
in the items. A scale can be assumed to have content validity
if (1) the content domain was clearly defined, (2) specific
objectives were developed and items generated around these
objectives, and (3) there is high inter-rater reliability
among expert judges who rate the content of the test as
reflective of the defined content domain (Sesan, 1983, p.
110) . Procedure for this hypothesis verification is discussed
in Chapter 4.

Hypptnesie V states that there will be observed a
relationship between therapists scores and client scores.
This hypothesis is partially a reflection of construct

validity and partly a measure of the applicability of the

105
Personal Change Questionnaire to therapy settings as a measure
of Quality Assurance. Construct validity has been addressed
elsewhere and will be discussed with regard to Hypotheses VI.

A major goal in the development of the PCQ was to provide
an instrument which would reliably predict client preparedness
and status in therapy. This hypothesis is one attempt to
establish links that apply to therapeutic settings.

This relationship was examined using both the Pearson
product-moment correlation and the Spearmen correlation. ‘The
Spearman correlation is used with information which is ordinal
in nature, that is, ordered from low to high, but not
necessarily in even intervals. The format of the PCQ provides
for scores from 0 to 7, but there is no way to adequately
determine if the intervals are measuring the same amount of
change. For example, does a change from 3 to 4 on a scale of
knowing represent the same change as the change from 5 to 6?
This question is presently unanswerable. Since there are less
than eleven levels in the measurement scale of the PCQ, it was
decided to look for correlations between individual items with
the Spearman correlation first.

Since the information on the scales of the PCQ is
ordinal, but not necessarily interval, the Spearman.was used.
It provides a measure of the direction and magnitude of the
relationship between variables. In general, Spearman values
tend to be higher that Pearson values for the same set of data

(Gravetter and Wallnau, 1985).

 

106

In a separate analysis, the Pearson product moment was
also computed for this relationship. In this case, the
additive nature of the total scales being compared pushed the
number of levels above eleven and rules regarding use of the
Pearson apply.

Hyperheeieryl relates to construct validity and again to
the applicability of the measure for Quality Assurance. One
aspect of construct validity is how well certain expected
relationships are predicted and explained by the construct.
One measure of status in therapy is therapist opinion. This
is being directly compared by searching for the relationship
between therapist opinion and the total scale scores on the
PCQ. In this study, it was predicted that a positive
correlation would exist between therapists total scores and
therapist rating of clients status in therapy.

The Spearman correlation was again used since therapist
rating had only eight levels.

' Hypornesie VII relates to the applicability of the PCQ
client form, FORM S, to decisions about status in therapy.
A positive relationship between therapist rating of how well
the client is dealing with their problem and client self
rating total scores would enable the instrument to be employed
as a meaningful self report instrument. The Spearman
correlation was used in this analysis also.

W relates to an expected relationship

between the subscales of the PCQ. Since the Personal Paradigm

107

Shift was designed as a progressive series of steps leading
to self change, it was postulated that higher scores would be
expected on earlier subscales as a person moved through the
progression of stages. Initial inspection of score means of
the subsections suggested that the expected relationship would
not be present. It was decided that a two-way t-test would
more appropriately determine any significant differences
between the means of the subscale scores.

Hypothesis 13 was descriptive in nature and it was
believed that there would exist a relationship between
educational levels and.the total scale scores of the subjects.
Again the Spearman correlation was used.

Hypothesis H sought to determine if there was a
relationship between length of time in therapy and total
scores. Therapy based on the Personal Paradigm Shift
progresses sequentially through the steps as therapy
continues. Therefore, longer therapy, ie., more sessions,
was expected to relate to higher overall scores. This
hypothesis was designed to investigate the possibility that
clients in therapy based on other assumptions would follow a
similar progressive pattern. The Spearman Correlation was
used with this hypothesis also.

Hypothesis XI relates to the performance of a factor
analysis of the PCQ and is concerned with questions of
construct validity. Essentially, factor analysis is a refined

technique for analyzing the interrelationships of behavior

108

data (Anastasi, 1961, p. 147) or for analyzing the inter-
relationships between items on a test (Mehrens and Lehman,
1978, p. 565). The process involves reducing the number of
variables or categories which are first used to describe
traits or behavior. Geometrically, this method involves
finding reference axes in terms of which each item can be
plotted. Mathematically, this is accomplished by determining
the linear combinations of all the variables in a data matrix.
Each such linear combination is said to be a factor.

Analytically, this provides a method of determining the
correlation between any factor and any item on a test such as
the PCQ. The analysis provides a measure of the degree to
which each item is "loaded on" or reflective of that factor.
A factor analysis was done on the items of the PCQ in Order
to \determine if the four subscales correspond to separate
factors. This determination is accomplished by interpretation
of any factors which appear as a result of the analysis.

W was established to investigate differences
between diagnostic groups which were observed in the sample.
It is essentially descriptive in nature, and it has
implications in the establishment of construct validity of
both the instrument and the psychoeducational component of
therapy. Mean scores were compared using the t-test statistic

as discussed previously under Hypothesis VIII.

109
MW

Original data collection was supplemented by data from
two sources. The first was a direct result of correlations
which were computed as part of the Spearman correlation
matrices. One matrix was generated to determine relationships
both within and between' test forms on individual items,
subscale scores and total scores, as well as client self
ratings and' therapist ratings of clients. In .addition, a
matrix was generated to investigate possible relations between
the therapist "Motivational Dimensions" and items, subscales
and total scores on the therapist form of the PCQ.

This information was reported in the ferm of Spearman
correlation coefficients for each of the combinations. For
example, item #2, "How sure are you that the client knows
specific goals that will help solve the problem?", is
correlated with Therapist Rating of Client Status in Therapy
with, a value of .6171. Such information was reported as
correlation values and is contained in tabular form in Chapter
IV. Information relating the Motivational Dimensions and
items, subscales and total scores was also reported and
appears in Chapter IV in tabular form. An example would be
the relationship between Motivational Dimension 6, Emotional
readiness to learn a program of change, and item 15, "How much
are they committed to change?".

Only correlations above the value of .6 were reported.

These individual correlations add to understanding and

110
integrating the overall picture generated by earlier data as
it relates to construct validity. In addition, it provides
evidence as to which items or concepts appear to weigh most
heavily with therapists as they make evaluations of clients
in this study.

A second source of data is taken from.a series of t-tests
for paired samples similar to that reported in Table 4.7 and
4.8. These tables compare the means of subscale scores on
Client and Therapist Forms of the PCQ. The same analysis
generated comparisons between the means of the subscale scores
for the Client and the Therapists Forms. This information is
useful is understanding how client reports and therapist
reports might differ. Integration of such information may
lead to further understanding of similarities and.differences
in client's self rating and therapist ratings of clients.
Such information relates directly to the applicability of the
instrument in Quality Assurance.

This information is reported in tabular form in Chapter
IV and discussed further at that point.

Summer!

The purpose of this study was to evaluate the usefulness
and appropriateness of the Personal Change Questionnaire as
a measure of achievement in therapy. In order to accomplish
this, it ‘was necessary to 'provide 'measures of internal

consistency reliability, content validity and construct

111
validity. Investigation was made into the applicability of
the instrument to questions of Quality Assurance. In
addition, an attempt was made to find evidence of a
psychoeducational component of therapy.

ReSponses to the Personal Change Questionnaire from 115
clients in psychotherapy and their therapists were gathered
from various therapeutic settings including a community mental
health setting, private practice clinics and a halfway house
for criminal offenders with drug addiction problems. The PCQ
is a ‘rationally' constructed instrument composed. of four
subscales developed to measure status in therapy based on a
psychoeducational approach. It consists of forms for both
therapists and clients.

Hypotheses related to the instrument, the applicability
of the instrument and the psychoeducational component of
therapy were developed and procedures to test these hypotheses
were outlined. Cronbach's coefficient alpha was used to
measure both client and therapist forms of the PCQ for
internal consistency in both the overall test and the
subscales. The Pearson product-moment correlation was used
to determine if the PCQ was measuring a unidimensional
construct.

Initial tests of validity for the PCQ focused on both
content and construct validity. Content validity was
established through expert appraisal and acceptance of the

item content and construction. Several tests of construct

112
validity were made, including comparison of therapists overall
scores and therapist opinion about client status.

Several hypothesis about the applicability of the
instrument to therapeutic settings were also set forth.
Demographic variables were compared to test scores in an
effort to locate and identify relationships. .

Additional descriptive and correlational information was
discussed as was its relevance and importance to construct
validity and instrument applicability to Quality Assurance.

Factor analysis was employed to investigate the presence
of underlying factors in the PCQ. This technique provides
further evidence of construct validity.

Statistical procedures included computation of Cronbach's
alpha, the Spearman and Pearson product-moment correlations,
and the t-test statistic. A factor analysis also was done.

Results of the present study provide groundwork for
future adaptations of both the Personal Change Questionnaire
and the Personal Paradigm Shift from which the PCQ was

developed.

 

CHAPTER IV

ANALYSIS OF RESULTS

In this chapter the results of the data analyses are
presented. Each hypothesis is restated in testable form when

appropriate, and the results of the subsequent analysis are

given.

Tests of t e ese h H at eses
ot eses Rel ted Re ' ' ' 0 he e so a

Questionnaire

Hypothesis I:‘The internal consistency of the total items

of the Personal Change Questionnaire will be sufficiently

high to infer homogeneity of the construct of a

psychoeducational basis for personal change in therapy.

The internal consistency of the PCQ was determined by
computation of Cronbach's coefficient alpha. An analysis
employing Cronbach's alpha was conducted initially with n=59.
A value of .95 was determined for the client form and a value
of .96 was determined for the therapist form.

At that time there were no items of zero variance and it
was decided to retain all items for the final analysis.

A final analysis was made at n=115 with the following
results: Form S (client form) had a value of .95 and the

therapist form had a value of .97.

113

114

These values are quite high and an inference can be made
that the PCQ is measuring a homogeneous construct. The PCQ
is an internally consistent measure of a psychoeducational
component of therapy.

HypothesIs II: The internal consistency of each of the

four subsections of the Personal Change Questionnaire

will be sufficiently high to infer that each subscale
demonstrates internal consistency reliability.

An alpha coefficient was computed for each of the four
sub-scales of the PCQ. This was done at n=59 and n=115. For
n=59, the client form had alpha values for the subscales as
follows: .83 for Personal Awareness, .81 for Decision Making,
.92 for Oppositional Alternatives, and .92 for Planning and
Action. The Therapist Form at initial assessment had value
as follows: .92 for Personal Awareness, .86 for Decision
Making, .92 for Oppositional Alternatives, and .92 for
Planning and Action.

At final analysis of n=115, the following values were
obtained: Form 8: .78 for Personal Awareness, .82 for Decision
Making, .92 for Oppositional Alternatives and .93 for Planning
and Action: Therapist Form: .90 for Personal Awareness, .87
for Decision Making, .94 for Oppositional Alternatives and
.93 for Planning and Action.

All alpha coefficients are high enough to infer
homogeneity of the subscale items.

The results are presented in Table 4.1 on page 115.

115
Table 4.1
Internal consistency of the Personal Change Questionnaire
for the Total Scale and Four Subscales

Cronbach's Coefficient Alpha

 

CLIENT FORM S

 

N = 59 . N = 115
Total Scale .95 .94
Personal Awareness .83 .78
Decision Making .81 .82
Oppositional Alternatives .92 .92
Planning and Action .92 - .93

 

THERAPIST FORM

 

Total Scale .96 ' .97
Personal Awareness .92 .90
Decision Making .86 .87
Oppositional Alternatives .92 .94

Planning and Action .92 .93

 

116

WI: The correlations between the total scale

and the four subscales of the Personal Change

Questionnaire will be sufficiently high to infer that the

Personal Change Questionnaire is measuring a

unidimensional concept.

The correlations between the PCQ Total Scale and the four
sub-scales were determined by computing a Pearson product
moment correlation matrix. On the Client Form, the inter-
scale correlations ranged from .33 between Patient Awareness
and Planning and Action to .89 between Total Scale and
Oppositional Alternatives. The Patient Awareness scale is the
scale with the lowest inter-scale correlations. On the
Therapist Form, the inter-scale correlations ranged from .69
between Personal Awareness and Oppositional Alternatives to
.94 between. Total Scale and Planning and .Action. .All
interactions were significant at the .01 alpha level. See

Table 4.2 on page 117 and Table 4.3 on page 118.

tl°°tM ;‘.,’-. e. ' 1‘ o. '. ° !’ " 9°!1 !- "
Questiennsine
W: The content of items on the Personal

Change Questionnaire will be indicative of a

psychoeducational basis of personal change in

psychotherapy.

This hypothesis related to the content validity of the
Personal Change Questionnaire and is not testable by the use

of statistics. The three methods noted previously (p. 104)

 

117
Table 4.2
Correlation Matrix of the Total Scale and the Four Sub-Scales

of the Personal Change Questionnaire - Client Form S

 

 

Scale. Total Personal Decision. Opposit. Planning
Awareness Making Altern. & Action

Total 1.00 .69 .80 .89 .84

Pers.

Awar. .69 1.00 .61 .51 .33

Dec.

Mak. .80 .61 1.00 .69 .47

Opp.

Alt. .89 .51 .69 1.00 .67

Plan/

Act. .84 .33 .47 .67 1.00

 

Note: All correlations significant at the p < .01

as described by Mehrens and Lehman (1978) and.Anastasi (1978)
were used to gain content validity. These criteria are
listed, followed by a discussion of how they were satisfied.

Criteria for content validity include (1) the content
area is clearly identified, (2) specific objectives were
developed and items generated around these objectives and (3)

there is a high degree of inter-rater reliability among expert

 

118
Table 4.3
Correlation Matrix of the Total Scale and the Four Sub-Scales

of the Personal Change Questionnaire - Therapist Form

 

Scale Total Personal Decision Oppos it . Planning

 

Awareness Making Altern. & Action

Total 1.00 .86 .89 .92 .94
Pers.

Awar. .86 1.00 .75 .69 .71
Dec.

Mak. .89 .75 1.00 .74 .77
Opp .

Alt. .92 .69 .74 1.00 .85
Plan/

Act. .94 .71 .77 .85 1.00

 

Note: All correlations significant at the p < .01

judges who rate the content of the test as reflective of the
defined content domain.

The content area was clearly defined as those segments
of the Personal Paradigm Shift which were used for treatment
with the PPS. Each step of the PPS is clearly defined and
structured so as to reflect one part of the overall process

of change. Each item.of the PCQ was derived directly from one

 

119
item of the PPS. This insured that each item was a direct
reflection of a clearly defined content domain.

Three judges were selected who were deemed to be experts
in the use of psychoeducational models. Two of these were
familiar with the PPS. They were asked to read the PPS and
the PCQ and decide if they agreed or disagreed that the item
was a genuine reflection of the content of the corresponding
segment of the PPS. They’ were offered the option of
requesting that the item be rewritten if they felt that in
their judgement that it was unclear or not adequate. The
judges were unanimous in.their agreement that the items on the
Personal Change Questionnaire were true reflections of the
content of the Personal Paradigm Shift.

W: A linear correlation will be observed

between therapist total scores on the Therapist Form

and client total scores on the Client Form S of the

Personal Change Questionnaire.

This can be stated in the following form:

He: P = 0
He: P f 0

The correlation matrix generated by comparing all items
and subsections of the client form, Form 8, and all items and
sub-sections of the therapist form.provides this information.
This matrix generated a Spearman correlation for each matched
set of itemsm The Spearman value generated.was .44 for n=115,

a figure that was significant at the .001 level.

 

120

In addition, a Pearson product moment correlation
coefficient was generated. Since the total scores included
all possible ranges of scores fromieach item, the total is 37,
the number of items, times 7, the number of levels for each
score (minus 1 since a score of zero would not add to the
total). The total 259 is greater than 11 and the Pearson can
be employed. The Pearson value was .40 for n=115, with a
significance level of p > .01.

In this case, the null hypothesis would be rejected since
there is a minimum value of .40 (Pearson) and a maximum value
of .44 (Spearman). There is a moderate linear correlation
between the total scores on the Client Form S of the PCQ and
the total scores on the Therapist Form of the PCQ.

Table 4.4 Correlation of Client Total Scores

with Therapist Total Scores

 

Client Total Scores

 

Therapist .40 (Pearson, n=115, p=.001)

Total Scores .44 (Spearman, n=115, p=.01)

 

Hypornesis VI: A linear correlation will be observed
between the Therapist Form total scores and the therapist
rating of client status in therapy.

The hypothesis can be stated in the following form:

 

121
Hg: P 8 0
He: P f 0
The Spearman coefficient was calculated for this

hypothesis. In addition, a separate correlation was run aside
from the overall matrix. The values were both at .79 for
n=115 and are significant at the .001 level. The null
hypothesis would be rejected. There is a strong linear
correlation between.Therapist Form total scores and therapist
ratings of client status in therapy.

Table 4.5 Therapist Total Scores vs

Therapist Rating of Client Status

 

Therapist Total Scores

 

Therapist Rating

of Client Status .79 (n=115, p=.001)

 

Hypotneeis VII: A linear correlation will be observed
between the therapist rating of client status in therapy
and the client's total scores.
The hypothesis can be restated:

Hg: P = 0

Hg: P f 0
The Spearman correlation was calculated since the

therapist rating had only a scale of eight. The value was

 

122
determined to be .32 for n=115 and it was significant at the
.001 level. The null hypothesis is rejected. There is a
moderate correlation between therapist rating of client's
status in therapy and the client's total scores.
Table 4.6 Therapist Rating of Client Status

vs Client Total Score

 

Therapist Rating

 

Client Total

Score .32 (n=115, p=.001)

 

Hyperneere_yIII: The mean scores for the items in the
subscales will be ordered such that the scores will be
highest in section I with decreasing scores being in each
following section, with. the lowest scores being in
section IV.

The hypothesis can be stated:

a... s a... s E. s a...

i... > a... > a... > i...

A t-test comparison of the mean scores of the subsections

5'?

of the PCQ indicates that differences between means of the
subsection scores are not significant in the direction of the
hypothesis. The null hypothesis is retained. Tables 4.7 and
4.8 summarize the comparison of mean scores for the subscales.

See page 123.

123

Table 4.7 Comparison of Means of Subscale Scores
Client Form 8

 

 

 

 

 

 

 

 

SUBSCALES MEAN STANDARD t-value 2-tailed
DEVIATION prob.
Per.Aware. 5.1383 .900
-1.72 .087
DEEIMQKiDQ_____512212 1-190
Per.Aware. 5.1383 .900
- .90 .372
Opp.AIt, 5.2345 1.317
Per.Aware. 5.1383 .900
.29 .770
ELanlAct. 5.0979 1.513
Dec.Making 5.2932 1.190
.64 .526
Opn.Alt. 5.2345 1.317
Dec.Making 5.2932 1.190
1.48 .143
Plan/Act 5.0979 1.513
Opp.Alt 5.2345 1.317
1.26 .211
Plan/Act 5.0979 1.513
Note: None of the comparisons are significant.

Table 4.8 Comparison of Means of Subscale Scores
Therapist Form

 

 

 

 

 

 

 

 

SUBSCALES MEAN STANDARD t-value 2-tailed
DEVIATION, DrOb.
Per.Aware. 5.1952 1.006
- .20 .842
Dee.HakIng §,292§ ;.;gz
Per.Aware. 5.1952 1.006
' 2.58 .011**
ODD.A1t. 4.9816 1.199
Per.Aware. 5.1952 1.006
2.44 .016**
Plan/Act. 5.0038 1.165
Dec.Making 5.2096 1.142
2.91 .004**
ODD.A1t. 4.9816 1.199
Dec.Making 5.2096 1.142
2.83 .005**
Plan/Act, 5.0038 1.165
Opp.A1t 4.9816 1.199
- .36 .719
Plan/Act. 5.0038 1.165
Note: ** Indicates the t-values are significant. The null

hypothesis is still retained since the overall order
is PA < DM > CA < PAA not PA > DM > GA > PAA>

 

124
MESH—Di: There will be a linear correlation
between the total scores of items on the Personal Change
Questionnaire on Form S and the educational level of the
subjects.
Restatement of the hypothesis provides:
Hp: P = 0
He: P f 0
The Spearman correlation was calculated and determined
to be -.09 for n=115 with significance at .174. The null
hypothesis is retained. There is no observable linear
relationship between educational level and scores on the
Personal Change Questionnaire.
Table 4.9 Correlation of Client Total Scores

vs Educational Level

 

Client Total Scores

 

Educational Level -.09 (n=115, p=.174)

 

W: There will be a linear correlation between
the total scores of items on the Personal Change
Questionnaire and the number of sessions of therapy
reported by clients.
The hypothesis can be stated:

Hg: P = 0

He: P f 0

 

125
The Spearman correlation was computed to be .16 for
n=115. This was at a significance level of .01. The null
hypothesis is rejected. There is a small linear relationship
between the number of sessions and the total scores on Form
S of the Personal Change Questionnaire.
Table 4.10 Correlation of Clients Total Scores

vs Number of Sessions

 

Clients Total Scores

 

Number of Sessions .16 (n=115, p=.01)

 

Hypernee1e_xl: Four distinct factors will be detected
corresponding to the four subdivisions of the Personal
Change Questionnaire.

An exploratory factor analysis was performed on the data
provided by the respondents to the PCQ. An analysis was
performed on both the therapist and client forms of the PCQ.
All items of the Client Formwwere used.except #38, client self
rating. For the Therapist Form, all items were used with the
exception of #38, therapist rating of client status. In
addition, the Motivational Dimensions were not included in the
analysis of the Therapist Form.

A Varimax rotation provided four factors for each form
of the PCQ. The criteria for determining the importance of

an item to a factor are subjective. It is believed that items

126
with higher loadings are more important than having many items
loaded at lower levels (Kim and Mueller, 1978). A level of
.40 was chosen as a liberal cut-off point since it was felt
that the purpose of an exploratory analysis was to identify
clusters of items rather“ than. eliminate items from. the
instrument.

In each case the results are indicative of the presence
of two factors. The first factor loads heavily in all
questions of the therapist form and also loads heavily in all
questions of the client form.except #1, #3, #4, #5, #6, #7 and
#9. The second factor loads somewhat.heavily on items #4, #5,
#6, and #8. It loads slightly less on items #3 and #7. A
second factor was observed on the client form. This factor
loads heavily on item #6. It also loads slightly less heavily
on items #1, #4, #5, #7 and #16. It was decided to retain
only those factors which loaded on more than three items.

The hypothesis as stated would be rejected since there
are not clearly four factors. corresponding' to ‘the four
subsections of the PCQ. Discussion of the meaning cf the
observed factors will be found in Chapter V.

Table 4.11 on page 127 demonstrates the factor matrix for
the client form factor analysis. ‘Table 4.12 on page 128 shows

the factor matrix for the therapist form.

 

—>—“

127
Table 4.11

Factor Matrix for Form S

 

 

ITEM, FACTOR 1 FACTOR 2
CL 1 .25 .39
CL 2 .57 .05
CL 3 .39 .15
CL 4 .25 .41
CL 5 .18 .38
CL 6 .17 .58
CL 7 .31 .47
CL 8 .53 .25
CL 9 .39 .07
CL 10 .44 .37
C1 11 .57 .24
CL 12 .62 .24
CL 13 .54 .36
CL 14 .56 .17
CL 15 .55 .29
CL 16 .48 .44
CL 18 .71 .24
CL 19 .76 .14
CL 20 .78 .12
CL 21 .51 -.18
CL 22 .79 .13
CL 23 .73 .13
CL 24 .76 .22
CL 25 .75 .00
CL 26 .74 .02
CL 27 .65 -.32
CL 28 .46 -.29
CL 29 .60 -.29
CL 30 .56 -.30
CL 31 .69 -.41
CL 32 .77 -.46
CL 33 .78 -.40
CL 34 .57 -.45
CL 35 .66 -.40
CL 36 .68 -.32
CL 37 .68 -.21

 

128
Table 4.12

Factor Matrix for Therapist Form

 

 

ITEM FACTOR 1 FACTOR 24,
IN 1 .68 .33
IN 2 .76 .05
IN 3 .51 .40
IN 4 .51 .50
IN 5 .60 .57
IN 6 .44 .59
IN 7 .51 .41
IN 8 .43 .57
IN 9 .69 .02
IN 10 .64 -.14
IN 11 .63 -.14
IN 12 .71 .11
IN 13 .53 -.04
IN 14 .68 -.13
IN 15 .70 .78
IN 16 .68 .16
IN 17 .63 -.21
IN 18 .78 -.30
IN 19 .75 -.26
IN 20 .73 -.24
IN 21 .71 -.05
IN 22 .78 -.20
IN 23 .69 -.01
IN 24 .77 -.06
IN 25 .81 -.30
IN 26 .80 .08
IN 27 .76 -.19
IN 28 .69 -.19
IN 29 .75 -.18
IN 30 .62 -.16
IN 31 .64 -.11
IN 32 .81 -.01
IN 33 .78 -.08
IN 34 .68 .16
IN 35 .76 -.11
IN 36 .80 -.08
IN 37 .64 .03

 

Note: The designation "IN 1" refers to question #1 on the
Therapist Form.

129
es's : There will be differences in the
mean score for items, subscales and total scores
between diagnostic groups using the Personal Change
Questionnaire.
The hypothesis can be stated:

'E: aCocaine Dependency = 50”“in

13E: aCocaine Dependency # iiDysthymia

This hypothesis was generated from the data set when it
was observed that, despite an overall n of 115, there were two
clear diagnostic groups which emerged. These were Cocaine
Dependence and Dysthymia. A series of t-tests were computed
to compare the means of the individual items, subsections and
overall scores for the two groups. The was done on the client
and therapist forms.

Of the 43 possible comparisons of means, on the client
form there were 9 comparisons significant at p=.05, 5 were
significant at p=.01 and 1 was significant at p=.001. This
is a total of 15 comparisons that were significant.

On the therapist form, there were 6 that were significant
at p=.05, 6 that were significant at p=.01 and two that were
significant.at.p=.001~ This.is a total of 14 comparisons that
were significant- Observable differences exist between these
diagnostic groups in individual item scores, subsection scores
and total scores. The null hypothesis is rejected.

Significant differences between means are summarized on

pages 130-134 in Table 4.13 and Table 4.14.

 

130

Table 4.13

Comparison of Means of Diagnostic Groups:

Cocaine Dependency Vs Dysthymia

Client Form S

 

 

 

 

 

 

 

 

 

 

 

 

ITEM GROUP MEAN STANDARD t-value 2-ta11ed
DEVIATION prob.
11 Punishment when I change my actions
1 4.1667 2.057
-2.41 .021
2, 5.4091 1.403
12 Rewards when I change my actions
1 5.3200 1.651
-3.42 .002
6.5909 .796
13 Punishment when I don't change
1 5.6400 1.150
-3.25 .002
2, 6.5455 .739
17 How I avoid my fears
1 4.5600 1.873
-3.05 .004
2 6.0000 1.345
20 New mental images in response to problem
1 4.9200 1.891
-2.12 .040
' 6.0000 1.604
23 New, healthier reactions to counter old
1 5.5600 1.294
. -2.10 .042
2 6.223 .869
27 I can choose a realistic first goal
1 4.6800 1.796
- .256 .014
2 5.9545 1.618.
30 I can seek out a role model
1 4.8400 2.055
-2.02 .050
2 6.0455 2.035
32 I can make organized plan of attack
1 4.5600 1.895
-2.61 .012
In: 5.8636 1.521
34 I can set a realistic time period for change
1 4.0000 2.062
-2.38 .022
2 5.4091 1.992
Note: Group 1 is the Cocaine Dependency Group, Group 2 is

the Dysthymia group.

 

131

Table 4.13
continued

 

ITEM GROUP MEAN STANDARD t-value 2-tailed

 

 

 

 

 

 

 

 

DEVIATION. prob.
36 I can make plan to avoid sabotaging success
1 4.3600 2.018
-2.47 .018
2. 5.5455 1:224.
37 I can keep a daily chart to follow change
1 4.9200 1.382
-2.60 .013
24, 5.9091 1.231
Total Scale scores for all items
1 5.2017 .955
-2.31 .026
2. 5.7211 .556
Subscale for Decision Making
1 5.2057 1.031
-3.88 .001
6.1104 .510
Subscale for Planning and Action
1 4.9782 .262
-2.43 .019
2- .r5.8099 .221
Note: Group 1 is the Cocaine Dependency Group: Group 2 is

the Dysthymia Group.

u e t '8

Additional data was generated from the correlation
matrices developed for analyses of previously discussed
hypotheses. This information falls into three categories.
First is information regarding relationships between items,
subscale scores and total scores, both within and.between the
Client and Therapist Forms.

This first type of information reports correlations which
are significant statistically between items, subscale scores
and total scores on the Therapist Form and the Therapist
Rating of Client Status. Each such significant correlation

is reported in Table 4.15 on page 134.

132
Table 4.14

Comparison of Means of Diagnostic Groups:
Cocaine Dependency vs Dysthymia

Therapist Form

 

 

ITEM GROUP MEAN STANDARD t—value 2-tailed
DEVIATION prob
7 Consequences of actions in problem situation
1 5.1600 1.405
-2.69 .010

6.1364 1.082
8 Whether praising or criticizing in problem situation

 

 

 

 

 

 

 

1 5.7200 1.308
2.53 .015
.2 4.9091 ..868
11 Punishment when they change their action
1 4.5200 1.661
-2.57 .014
5.5909 1.182
12 Rewards when they change their actions
1 4.7200 1.137
-6.50 .001
2TH. 6.4091 .590
13 Punishment when they don't change their actions
1 5.5600 1.356
-3.45 .002
6.5909 .590
14 Rewards when they don't change their actions
1 4.9600 1.274
-2.19 .034
2 5:7721____11270
18 New feelings in the problem situation
1 4.1200 1.453
-2.49 .017
4.9545 .785
20 New mental images in response to problem
1 4.4000 1.225
-2.53 .015

5.2273 1.020
23 New, healthier reactions to counter their old

 

 

1 4.9200 1.256
-2.86 .007
2 5.7721. -752
28 Create a support system for self
1 4.8400 1.463
-2.76 .008
1 5.9091 1.192
Note: Group 1 is the Cocaine Dependency Group: Group

2 is the Dysthymia Group.

133

 

 

 

 

 

Table 4.14
continued
ITEM GROUP MEAN STANDARD t-value 2-tailed
DEVIATION prob
30 I can seek out a role model
1 5.4400 1.227
-2.15 .037
205 6.2273 1.270
Total Scale Scores
1 5.0853 .763
-2.03 .050
:2 5.4449 .424
Scale for Decision Making
1 5.9057 .974
-3.89 .001
2. 5.9285 .413
Note: Group 1 is the Cocaine Dependency Group: Group 2 is

the Dysthymia Group.

A second source of additional information was developed
from comparison of the mean scores of subscales from the
Client Form with the mean scores for subscales from the
Therapist Form.

This information relates the average score for each
subscale pair between the two forms and demonstrates by the
use of the t-test stastistic if there are differences in the
mean scores for the two forms. See Table 4.16, page 135.

A third source of supplemental descriptive data results
from the correlation matrix generated by comparing the
Therapist Form "Motivational Dimension" items to all items,
subscale scores and total scores. It was believed that
certain client characteristics best described.as motivational
might be related to therapy as described in the Personal

Paradigm Shift and also to therapy in general.

134

Table 4.15

Comparison of Items, Subscales and Total Scores
with Therapist Rating of Client Status

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ITEM SPEARMAN SIG.LEVEL
2 Goals that will help solve .61 .001

their proplen

15 How much are they committed .73 .001
no change

18 New feelings in the problem .68 .001
geituation

21 New bodily reactions in response .60 .001
to problem

22 New self judgements in the .68 .001
problem situation

24 New consequences of their .62 .001
behavior

25 New self evaluations as a result .59 .001
of 11W

32 Can make an organized plan of .63 .001
attack

33 Can give themselves feedback .62 .001
on their progress

36 Can make a plan to avoid .70 .001
saboraging suceese

Subscale 1: Personal Awareness .66 .001

Subscale 2: Decision Making .62 .001

Subscale 3: Oppositional Alternatives .75 .001

Subscale 4: Planning and Action .75 .001

Total Scores Therapist Form .79 .001

 

Correlations were computed between each item on the

Therapist Form. Those which are significant 'are found in

Table 4.17 on page 136 (only significant correlations noted).

These provide information as to what variables might be

accounting for therapists' ratings as they look at and rate

clients.

135

Table 4.16
Comparison of Means of Subscale and Total Scores
between Client Form 8 and Therapist Form
of the Personnal Change Questionnaire

 

 

 

 

 

 

 

SUBSCALE/ MEAN STANDARD t-value 2-tailed
SCALE DEVIATION PROB.
TH:Pers.Awar. 5.1952 1.006
.54 .590
CL:Per§.Awer. 5.1383 .900
TH:Dec.Mak. 5.2096 1.142
-.69 .494
CL:Pee,HeH, 5,2232 1.190
TH:Opp.Alt. 4.9816 1.199
-1.72 .088
CL:Opp,AIr, 5.2345 1.317
TH:Plan/Act 5.0038 1.165
-.67 .505
CL:Plan/Act 5.0979 1.513
TH:Tot.Score 5.0894 1.018
-.86 .391
CL:Ipr,§core 5.1788 1.011
Summer!

The Personal Change Questionnaire is based on concepts
of psychoeducation as contained in the Personal Paradigm
Shift. This programmed model of self change contains four
subscales. These are (1) Personal Awareness, (2) Decision
Making, (3) Oppositional Alternatives, and (4) Planning and
Action. Procedures for generating items for the PCQ involved
the generation of items around specific objectives and
determining high inter-rater reliability between experts who
judged the items prior to test administration.

Several hypotheses related to the internal consistency
reliability of the PCQ were tested. Hypothesis I was

concerned with total scale reliability. A Cronbach's alpha

 

136

Table 4.17
Correlations Between Motivational Dimensions
and Items and Subscale scores

 

ITEMS epreenen SIG.LEVEL
TH.E5(Commitment)
VS .75 .001

15:How nucn eommirred re enenge
TH.E6(Emot.Read.to lrn.program...)

 

 

vs .71 .001
15:How much commirred re enenge
TH:E6(Emotional Readiness...)

vs .64 .001

22:Bring new self judgements

TH:E6(Emotional Readiness...)
Vs .62 .001

27:Choose real. first goal
TH:E6(Emotional Readiness...)

vs
33:Give themselves feedback
TH:E6(Emotional Readiness...)

vs .60 .001

36:Plan to avoid sabotaging suecess
TH:E6(Emotional Readiness...)

vs .63 .001
375Keep diary to plot progrese
TH:E6(Emotional Readiness...)

vs .68 .001
TH:OpposIrional AIternetives

TH:E6(Emotional Readiness...)

vs .67 .001
TH:Planning and Action
TH:E6(Emotional Readiness...)

vs .67 .001

TH:RatIng of Client Status

.60 .001

coefficient value was determined.to be .95 for the Client form
and .97 for the Therapist form.

Hypothesis II related to the internal consistency of the
subscales of the PCQ. Alpha coefficients for the Client form
were determined to be as follows: Personal Awareness: .78:

Decision Making: .82: Oppositional Alternatives: .92:

137
Planning and Action: .93. For the therapist form, the values
of alpha were: Personal Awareness: .90: Decision Making:
.87: Oppositional Alternatives: .94: Planning and Action:
.93.

Both the total scale and the subscale on both forms have
shown high internal consistency reliability.

Hypothesis III tested whether or not the PCQ was
measuring a unidimensional construct. A Pearson product
moment correlation coefficient computed between the PCQ and
the four subscales demonstrated significant levels of
correlation among the scales. This finding suggests that the
PCQ is measuring a unidimensional construct.

Hypothesis IV’related.to the content‘Validity of the PCQ.
The PCQ is assumed to have content validity because (1) the
content domain was clearly defined, (2) specific objectives
were developed for the areas of content to be samples, and (3)
very high inter-rater reliability among a group of expert
judges rating items was obtained.

Hypothesis V began the testing of the construct validity
of the PCQ. A small linear correlation was found between the
total scores on the Client Form and total scores on the
Therapist Form. A Spearman value was determined to be .44
and was significant at the p=.01 level. .AiPearson correlation
was found to be .40 and was significant at the .001 level.
There is a moderate relationship between therapist ratings of

clients and the clients' rating of themselves.

 

138

Hypothesis VI investigated the relationship between
therapist total scale scores and therapist overall rating of
clients status. This was an additional attempt at determining
construct validity as well as an effort to establish plausible
reasons to use the PCQ as a measure of quality assurance. The
Spearman correlation was determined to be .78 with p=.001.
This is a relatively high value and is further evidence of
construct validity for the PCQ. It appears that the Therapist
Form might provide a useful tool in the determination of
client status in therapy.

Discovering the relationship of therapist rating to
client overall scores was the object of Hypothesis VII. A
Spearman correlation of .32 was found to be significant at
p=.001. This relationship was moderate and positive.

Hypothesis VIII was originally established upon the
belief that clients in therapy would experience higher scores
in early subsections and subsequent decreasing scores in the
following subsections. Preliminary analysis at n=59 indicated
that this was clearly not the case and the hypothesis was
amended to discover what relationships would exist between
the subsections. It was discovered that there are significant
differences between the subscale scores for different
sections.

Hypothesis IX evaluated the relationship between total
score and educational level. This hypothesis was rejected

when the Spearman coefficient was determined to be -.09 with

 

139
a significance level of .174. There is no discernible
difference in overall total scores as a result of difference
in educational level.

The relationship between total score and the number of
sessions in therapy was the subject of Hypothesis X. A
Spearman coefficient of .16 with significance level of p=.01
is indicative of a very small correlation between length of
time in therapy and PCQ total score.

Hypothesis XI projected that there would exist four
distinct factor areas corresponding to the four subsection
areas. The results of a factor analysis indicate that there
are clearly two factors present in both the Client and
Therapist Forms. 'These factor areas do not appear to coincide
with the four subsections of the PCQ. The four factor
hypothesis is rejected.

Hypothesis XII tested the existence of significant
relationships between item means, subscale means and total
score means for two different diagnostic groups who used the
PCQ. Analysis of the client form revealed 15 significant
differences including total score, Decision Making and
Planning and Action. Therapist Form scores differed in 14
categories including total score and Decision Making. There
is clear evidence that for this study, two clearly identified
and separate diagnostic groups, Cocaine Dependency and
Dysthymia, differ in both the way they report their own self

perceptions and the way in which they are reported to be

 

140
perceived by their therapists.

Supplemental data analyses revealed the existence of
several correlations that lend additional evidence to the
notion of construct validity. In addition, supplemental
analysis indicates that the mean scores for the Therapist Form
and the Client Form are not significantly different. There
are as well several strong correlations evident between items
and subscale scores on the Therapist Form and the motivational
dimension known as "Emotional Readiness to learn a Program".

Chapter V will report the summary of the study as well

as conclusions and implications for future research.

 

CHAPTER V

SUMMARY, DISCUSSION AND IMPLICATIONS

This chapter summarizes the research study. The findings
of the study are presented along with a discussion of the
results related to instrument development and the psychoedu-
cational component of therapy. Implications of the findings
are discussed, along with suggested directions for future

research.

§EEEQEY

The purpose of the study was to evaluate the usefulness
and appropriateness of the Personal Change Questionnaire as
a measure of status in therapy. The instrument was developed
around and measures those components of therapy which can be
said to be psychoeducational in nature. Additionally, the
instrument was used to gather descriptive information about
one group identified as Cocaine Dependent and another group
identified as Dysthymic.

The Personal Change Questionnaire is a 38-item self
report measure developed around psychoeducational concepts
embodied in the Personal Paradigm Shift (Hinds, 1983). This
model of personal change incorporates four areas of client

experience in therapy.

141

142

These areas are Personal Awareness, Decision Making,
Oppositional Alternatives and Planning and Action. The
Personal Paradigm Shift is based on sound psychoeducational
principles as set forth in the research of Sanford (1959),
Carkhuff (1971, 1972), Guerney, Guerney and Stollak (1972),
Ivey and Alschuler (1973 and Authier (1977). Each researcher
or research team has set forth a series of guidelines as to
the roles therapist and client play in psychoeducational
counseling. These guidelines include the therapist acting as
a teacher with the client becoming an active and involved
student. The therapist helps the client define problems and
set goals, acquire and practice skills, and learn how to
evaluate progress.

Clients become students who face their problems as
functional deficiencies which can be overcome. The learning
of new skills is facilitated by developing and implementing
an organized and often highly structured plan of training.
Principles of learning and psychology are demystified and
learning is enhanced by the active participation of the
client. Psychoeducational programs focus on identification
of problems, mastery of specific behaviors and development of
feedback mechanisms by which learning can be measured and
reinforced by the client without the long term help of the
therapist. The Personal Paradigm Shift reflects these
principles as does the Personal Change Questionnaire.

Procedures used to develop the PCQ included defining the

143
content from which items were to be taken, that is, the
Personal Paradigm Shift, and having the items evaluated by
expert judges as to their content and structure prior to the
administration of the test. The test battery included a form
for clients and for their therapists who would evaluate them
without reference to the client responses.

A sample of 115 clients was gathered from the clientele
of several private practice settings, one community mental
health facility and one half way house for drug offenders
leaving prison. Seventeen therapists were involved in the
study as the therapist of record for these subjects.

Subjects were surveyed with regard to demographic
information including age, sex, marital status, income level,
duration of their problem and time in therapy. Therapists
were surveyed about their treatment orientation and were asked
to give a diagnosis for each client.

Eleven hypotheses related to instrument development and
the psychoeducational component of therapy were tested. In
addition, a descriptive analysis was done on the differences
between two diagnostic groups which appeared in the sample.
.A quantity of supplementary descriptive data was collected
which relates to several aspects of the study, including
construct validity, applicability of the instrument and
differences in the way clients are seen by therapists and how
they see themselves. This supplementary information will be

presented when applicable. Results regarding reliability and

144

validity of the PCQ are as follows:

1.

The PCQ was determined to be an internally
consistent and reliable instrument for both client
and therapist forms. A Cronbach's alpha coefficient
of .95 was obtained for the Client Form 8 total
scale of the PCQ. An alpha coefficient of .97 was
obtained for the Therapist Form for the PCQ total
scale.

The four sub-scales of the PCQ demonstrated Alpha
values of .78, .82, .92, and .93 for the Client Form
8 on sub-scales of Personal Awareness, Decision
Making, Oppositional Alternatives and Planning and
Action. For the Therapist Form, the corresponding
values were .90, .87, .94, and .93 for the same sub-
scales. All values are sufficiently high for
research purposes, and values may be sufficient for
other uses. Each value is sufficiently high to
infer that each sub-scale demonstrates internal
consistency reliability.

Intercorrelations between.the Total Scale Score and
individual Scale Scores were significantly high
enough to infer that the PCQ is measuring a
unidimensional concept. This was found to be true
for the Client Form 8 and the Therapist Form.

Content validity was established through the use of
a highly defined content area from which specific
items were derived. There was high inter-rater
reliability among expert judges as to the content
being both appropriate and well structured.

Some evidence of construct validity was generated
through the observation of a low but significant
correlation. between therapist total scores and
client total scores (Pearson = .40).

Additional evidence for construct validity was found
in the strong correlation between therapist total
scores and therapists rating of client status in
therapy. A Spearman correlation coefficient value
of .78 was computed. This value is significant at
the p=.001 level.

A Spearman value of .32 (p=.001) was computed for
the correlation between therapist rating of client
and client total scores. The statistical test of
the hypothesis as stated was low but significant.

8.

10.

11.

145

A hypothesis relating the the sequencing of mean
score values for the sub-scale scores was disproved.
Additional information about the relationship
between the sub-scale scores was gathered. This
information did not strengthen the evidence for
construct validity of the PCQ as presently set forth
in the model of the Personal Paradigm Shift.
However, there are implications in the results from
the Therapist Form as to how therapists and clients
differ in their judgements which do relate to the
process of therapy.

There was no evidence of a correlation between total
scores and educational level as seen on the Client
Form S of the PCQ.

There was no evidence of a correlation between the
client total score and the number of sessions in
therapy.

An exploratory factor analysis detected a factor on
the PCQ which was highly loaded on.almost every item
of the test. A second factor was detected which was
moderately loaded on several items of sub-scale
Personal Awareness of both forms. No other factors
were deemed significant since they appeared in less
than three items. The factors did not conform to the
four sub-scales of the PCQ. While not confirming
the hypothesis, however, the detection of the two
factors on each test is strong evidence again that
the test measures at least two separate parts of a
unidimensional concept.

Supplemental data analysis revealed the following:

1.

Analysis of the means of items, sub-scales and total
scores showed the existence of fifteen differences
between diagnostic groups on the Client Form and
fourteen on the Therapist Form of the PCQ. The
strong evidence would indicate that Dysthymics and
Cocaine Dependents differ in the way they report
their self perceptions and the manner in which they
are perceived by their therapists. Supplementary
data analyses indicate the presence of differences
between the way each of the two diagnostic groups
understands rewards and.punishments, and the way in
which they handle decision making.

Therapists in this study had high correlations
between certain items on the Therapist Form and
their rating of client status in therapy. These
centered around commitment, ability to make plans,

146

and the ability of clients to perceive themselves
in a new conditions surrounding their problem.

3 . Correlations between subscales and therapist ratings
were different one from another and were ordered in
the following manner: Planning and Action had the
highest correlation with therapist rating.
Oppositional Alternatives had the second highest
correlation with therapist rating. Personal
Awareness had the next highest correlation and
Decision Making had the lowest.

3. There were no significant differences between the
mean scores of the subscales between the Client Form
S and the Therapist Form on the Personal Change
Questionnaire.

4. Analysis of the Motivational Dimensions of the
Therapist Form revealed the following: the dimension
entitled "Readiness to learn a program of change"
was significantly correlated to items relating to
clients' abilities perceive themselves in new
situations surrounding their problem. It was also
highly correlated. with the subscale scores of
Oppositional Alternatives and Planning and Action
and to total scores. This is additional evidence
of which values therapists weighed most heavily when
evaluating clients with this instrument.

Discussion
Reliability of tne instrnnenr. The internal

consistency reliability of the Personal Change Questionnaire,
both Client Form 8 and Therapist Form was determined in the
study. The high alpha coefficient figure of .95 for the
Client Form and .97 for the Therapist Form suggests that both
forms exhibit sufficiently' high internal consistency
reliability for a self report instrument.

Several factors may have contributed to those high

figures. Mehrens and lehman (1978) list four such factors

147
which are considered. Generally, the longer a test, the more
reliable it is. With thirty-eight items, the PCQ is not
overly long. It is not felt that this length would contribute
to the internal consistency in any great measure.

Heterogeneity of the responses tends to increase the
reliability of the test. The sample was fairly heterogeneous
with.respect.to‘client'variables such as age, economic status,
education and length of time in therapy; Both sexes were more
or less evenly represented. This mixture provides for a
greater variability of responses. An additional factor which
may have contributed to the heterogeneity of responses was
that different items had different internal degrees of
difficulty. Knowing the nature of the problem is net as
difficult as understanding the various rewards, punishments
and losses one might encounter when undergoing change. This
variability in difficulty may have added to the heterogeneity
of response.

The scales used on the PCQ are commonly referred to
Likert Scales in that they require a judgement of degree for
the respondent. The use of such scales as opposed to a
dichotomous scale also adds heterogeneity to the responses and
hence adds to the overall reliability of the test.

The scales as chosen provide the respondent a choice of
scores of from 0 to 7. This set of responses does not require
that a difference from 3 to 4 is the same as from 4 to 5.

While it seems logical that such intervals reflect similar

 

148

divisions, this has not been.proven. In addition, a scale of
only eight is not sufficient to assume a continuous variable.
Therefore, the scale is ordinal, not interval. Ordinal scales
require use of the Spearman rank order correlation coefficient
for developing the correlation matrix on which alpha
coefficients are based. Spearman values tend to be higher
than Pearson product moment values. This may result in higher
alpha values since it is true that the Spearman coefficient
provides the highest correlational values.

The development of the items on the PCQ directly from
corresponding segments of the Personal Change Questionnaire
may have added to the reliability estimate. This is due to
the direct correspondence to items on the PPS which are well
founded in psychoeducational theory and empirical experience
in providing therapy. Such close correspondence adds to the
development of a homogeneous construct and hence increases
internal consistency.

It is likely that the PCQ is measuring different aspects
of a single construct. A consistent factor appears to be
present throughout the test, perhaps appearing in different
ways. High inter-scale correlations between all scales is
suggestive that the PCQ is measuring a unidimensional
construct. This is further supported by the results of the
factor analysis which shows but two factors and one of them
highly loaded on all items of the instrument.

The scale with the lowest inter-scale correlations may

 

149
well be measuring a different aspect of therapy which is not
related to the other sub-scales. This is also supported by
evidence which shows most items of the sub-scale known as
Personal Awareness to be loaded on the second factor which
appears in the factor analyses for both forms of the Personal
Change Questionnaire.

In summary, the PCQ, both Client Form S and Therapist
Form, demonstrates sufficiently high internal consistency
reliability. The instrument appears to be measuring one
construct. with. two different. dimensions, one related. to
internal, perhaps dynamic, processes and one related to other
learning processes.

Validity of the instrument. Content validity for the PCQ
was established by the development of items on a one to one
correspondence with segments of the Personal Paradigm Shift.
Additional support for content validity is derived from high
inter—rater reliability of expert judges who rated the PCQ on
the basis of a psychoeducational component as found on the
Personal Paradigm Shift. Content validity is also positively
impacted by the high level of internal consistency reliability
discussed in the preceding section.

Initial steps at the establishment of construct validity
were also undertaken in this study. Determination of
construct validity requires an ongoing evaluation of
information about the construct from a variety of sources.

In this study there were several attempts at gathering such

 

150
evidence.

One aspect of construct validity is that expected
relationships should appear. It was assumed that therapist
rating of client status in therapy, a somewhat intuitive and
personalized.measure, would be positively correlated with the
therapist total scores from the PCQ. This relationship was
found to be highly positive, indicating that therapist opinion
correlated with the psychoeducational component as measured
by the PCQ.

There was also observed to be a positive correlation
between therapist total scores and client total scores. While
the coefficient of correlation was not high, .40 (Pearson),
it was positive. For the purpose of this study, this does not
add significantly to the evidence for construct validity.
One explanation for this moderate correlation is that clients
are not trained to be as astute in their observations as are
therapists. In addition, they are intimately involved in
their situations and may not be as objective about their
perceptions as are therapists.

An additional factor may well have been the low variance
of scores per item. Few clients in the study had been in
therapy for a short period. In fact most had been in therapy
for some time. This may well have resulted in the absence of
lower end scores and thus reduced the variability of those
scores. Low variance generally results in lower correlations.

A hypothesized correlation between.therapist ratings and

151
client total scores was determined to have a value of .32.
While this was in the correct direction, the value was lower
than that acceptable to this study. Again the factors
discussed above may be important: that is, differences exist
between therapist and client judgement and low variability
reduces to ability to achieve higher correlations.

Of interest to construct validity is the factor analysis
which was performed. The results clearly indicated the
presence of two major factors. The first might well be
labeled the "deciding-committing-acting" dimension of change.
The other major factor might be labeled "internal responses"
and it presents itself in items located in the Personal
Awareness sub-scale. These could also be related as the
"what" and the "why" components of therapy.

Two other factors which were detected in the items did
not appear in three or more items and were not recognized as
significant in this study. They may be only experimental
artifacts or accidental by-products of the data analysis.

Construct validity can be advanced by the observation
that expected relationships behave as predicted. This has
been partially substantiated by some aspects of the study.
It is further advanced by noting the differences in responses
between Dysthymics and Cocaine Dependents. Denial, whether
conscious or unconscious, is a major aspect of substance
abuse. Failure to see the problem for what it is places

severe limits on a client's ability to understand many aspects

 

152

of their situation. Comparison of the two treatment groups
indicates that the Cocaine Dependency group had significant
differences from the Dysthymia group in their self
perceptions. On the Client Form S, Cocaine Dependents were
significantly lower on mean scores on fifteen of forty-five
possible comparisons with the Dysthymia group. On the
Therapist Form they were significantly lower on thirteen of
the scores. The groups were different on the PCQ in ways
suggestive of differences in self perception and self
understanding.

The two diagnostic groups differed significantly on their
self perceptions of rewards and punishments. This was true
on both the Client Form S and the Therapist Form. It is
consistent, and therefore adds to construct validity, that
they also differed significantly in. the Decision. Making
subscale scores on both forms. There was no .significant
difference on Personal Awareness between the diagnostic
groups.

Evidence for construct validity of a psychoeducational
component of therapy is incomplete. This study does begin to
establish that such a component is present and is measurable.

In summary, the study has shown the PCQ to have a high
internal consistency reliability, with high reliability for
the subscales as well. The PCQ appears to 'measure a
unidimensional concept with several aspects which are slightly

different. While there is some evidence of construct

 

153
validity, the issue is far from settled and requires ongoing

investigation for additional confirmation.

Wm

One limitation of the study was the lack of variation in
the treatment orientation of therapists who were willing to
participate. The psychoeducational nature of the study may
have been partially responsible for the low participation rate
among therapists who are Psychodynamic in orientation.
Anecdotal reports of interaction with therapists of this
orientation often centered around concern about "transference
issues" if clients participated. In addition, there may have
been a resistance on the part of some psychodynamic therapists
to engage in a study which is psychoeducational in nature.

It was noted that of the therapists who declined to
participate in the study, many expressed great enthusiasm over
the notion of an educational component of therapy and were
curious abOut results. Of the psychodynamic therapists known
to this researcher, there was an almost complete disdain and
lack of interest either in participating or knowing how the
study progressed. This lack of participation limits the
generalizability of the results to the population of clients
in therapy and to the population of therapists doing therapy.

Similar to this limitation is that, of the participating
therapists, 60.9% were self described as eclectic. This

appears to be a disproportionately high number for one

 

154
theoretical orientation. 'Therapist characteristics which are
unknown may have influenced the results of the study. This
large number from one orientation further reduces the
generalizability of the results.

A similar concern about therapist selection was that
there may be unknown therapist variables which determine who
will be inclined to participate in any study, much less one
such as this. Therapists self-selected themselves into the
study and then also selected subjects to participate. There
may be unknown or even unconscious forces that govern how this
selection takes place.

It was also observed that therapists did not follow
directions for administration. of the test. Many reported
allowing subjects to take the test home and mail it back.
While this may not have influenced the results, there may
exist other unidentified and uncontrolled factors in the
administration of the PCQ.

Additionally, there was no system for evaluating or even
knowing which clients may have declined to participate in the
study after being asked to do so by their therapists.

The client sample was fairly evenly distributed across
the demographic variables. While it was not a specified
demographic variable, it is also true that one facility
provided approximately 30% of the subjects. This was a
halfway house for men returning to the general population

after criminal proceedings against them. These men made up

155
80% of the Cocaine Dependency group. Once again, the ability
to generalize to the population of all clients is limited if
the results of the study are too heavily influenced by this
one large group of subjects.

There is anecdotal evidence of problems with the format
of the PCQ. Several therapists reported that it looked rather
formidable and that clients and therapists alike had "balked
at filling out a booklet." Several clients commented on the
test itself that certain examples did not apply to them.
There were several comments that.the.examples for items on the
PCQ were not necessary and even confusing.

One additional problem with therapists was that several
expressed. concern about what might happen if insurance
companies decided. that. they' could. accurately limit
authorization for visits with such an instrument. Several
therapists who voiced this opinion were less than responsive
to the study.

At the analysis stage of the study, a larger group of
subjects would have provided more power in, the factor
analysis. It is common to expect a ratio of 5:1 for Subjects
to items in the factor analysis. The ratio for this study was
closer to 3:1. While there is no evidence that this ratio
invalidated the results, extrapolation from such an analysis

should be done with caution.

 

156
1W

The Personal Change Questionnaire was established to
measure client status and progress in therapy. This study has
established that the PCQ has internal consistency and content
validity. It shows evidence of construct validity and the
therapist form appears to have some applicability in the
judgement of client status in therapy.

The study has shown that there is a psychoeducational
component to psychotherapy, even when therapy is not based on
principles of psychoeducation as such. It was also shown that
diagnostic groups differ in self evaluation and therapist
evaluation using the PCQ.

Several implications can be seen from this study.

First, this instrument does provide a way to measure
differences between diagnostic groups as to how they perceive
themselves and are perceived by therapists. This appears to
provide clues as to how problems might be approached. The
instrument measures client self perceptions and to some degree
compares them with therapist perceptions. Perhaps there are
ways to note the differences between these perceptions and
incorporate learning about these differences as a part of
therapy.

Second, this instrument points out that the assumption
of placing Personal AwareneSs first in the process of change
in the Personal Paradigm Shift is not necessarily true as a

general rule in therapy. The study appears to indicate that

157
developing a clear plan of action, understanding the choices
and making a commitment for change may be the critical step
for change to begin, regardless of how well one understands
their "internal responses."

Next, it has been shown that in this study, therapists
appeared to place a high _ value on several aspects of the
clients they evaluated. Therapists perceived emotional
readiness of the client as a critical factor in clients'
ability to participate in and learn a new program to change
their behavior. Therapists also perceived clients' ability
to make a commitment to such a program as critical to their
rating of the client. Additionally, therapists perceived as
important the clients' ability to perceive new reactions in
the problem situation as they implemented new behaviors and
responses.

Therapists perceived Personal Awareness to be more
important as compared to Decision.Making, but both therapists
and clients rated Planning and Action and Oppositional
Alternatives as more important than Personal Awareness.
Client and therapist ratings of clients resulted in similar
scores for subscales and total scores for the Personal Change
Questionnaire.

It was shown that both forms of the Personal Change
Questionnaire were meaningful measures of client and therapist
understanding of clients. Therapist rating of client status

in therapy was shown to be highly correlated with total

158
scores. Therapist rating was also positively correlated with
client rating and client total scores.

Finally, the study has shown that there is a factor‘which
was labeled the "deciding-commiting-acting" or "what"
component of therapy. Tests of the research hypotheses have
indicated that this factor explains 50% of the variance
between a therapist's intuitive opinion of client status and
the therapist's total scores on the Personal Change
Questionnaire. Another 25% of the variance was explained by
a factor labeled the "internal responses" or "why" component

of therapy.

pirectipne for Pnpnre Heseeren

st m ev s

The Personal Change Questionnaire in its present form
appears long and tedious. Time for administration is reported
to be less than fifteen minutes for subjects and less than
that for therapists. However, it does look like a booklet and
that often discourages potential users.

Revised versions might be produced which are designed
differently to appear less formidable. A trial run comparing
various forms might determine a better form of the test. One
form could be given without examples for each question.
Others might include a change of format with True/False
answers. Another might include the revision in a different

size type so that it would appear shorter.

159

With the format unchanged, this study might be replicated
with specific diagnostic groups or therapists of specific
orientations so that their response patterns could be
compared.1

Since the PCQ was designed to measure status and progress
in therapy, a logical next study would be longitudinal in
nature and would measure changes in responses for the same
clients or groups of clients over time. This approach might
provide a different estimate of test-retest reliability as
well as a way to measure progress of specific groups or
clients. Success of such a study would provide further
evidence of construct validity.

Future research should focus on involving clients with
less theapy time in order to provide some variability in
responses. A study comparing responses of groups with five
or less sessions to a group with fifty sessions might well
show differences in response patterns which could be
attributable to therapy.

Comparison of the Personal Change Questionnaire to other
tests designed to measure client status or progress in therapy
might ’well demonstrate the possibilities or limitations of

both instruments.

W
The Personal Paradigm Shift and the Personal Change

Questionnaire both purport to involve a psychoeducational

160
basis or component. Other instruments for the evaluation of
therapy which.purport to involve.psychoeducational components
should be brought together. A critical step in construct
validity is the comparison of measures which claim to measure
the same construct. Any ongoing research into the validity

of this instrument must include such a comparison.

Qenclueien

The results of this study arejpositive with regard.to the
development of an instrument to measure client status in
therapy. The Personal Change Questionnaire has been shown to
possess high internal consistency and content validity. The
search for construct validity has been moderately successful
if limited. The instrument has been shown to be capable of
detecting significant differences between diagnostic groups.
The Therapist Form has shown a high correlation with therapist
intuitive evaluation of client status. Each of these positive
results lends itself ‘to the further’ development of the
Personal Change Questionnaire as a reliable and valid
instrument to help insure quality in the delivery of

psychological services to the therapeutic community.

161
Introduction

The Personal Change Questionnaire was designed to measure
the status and progress of clients in psychotherapy. This
instrument was developed from the concepts of psychoeduca-
tional models of treatment. More specifically, it was derived
from the programmed model of change known as the Personal
Paradigm Shift developed by Dr. William C. Hinds of Michigan
State University. One of the critical steps in the develop-
ment of such a measurement instrument is the establishment of
content validity of the items. Content validity may be based
on an analysis of the items as compared to the content of the
model on which the instrument is based. It is essential to
determine that each item on the Personal Change Questionnaire
(PCQ) is representative.of one aspect.of the Personal Paradigm
Shift (PPS). .

Enclosed you will find a copy of the test instrument as
well as a copy of the Personal Paradigm Shift. In addition
will be a form which asks you to decide if each item on the
PCQ reflects the content of one part of the PPS. There is a
form.provided which asks you.to make a choice about each item.
Your choices are:

A) AGREE D) DISAGREE R) REVISE

Choice (A) Agree means that you agree that the content
of the item reflects the content of part of the PPS. Choice
(D) Disagree means that you disagree, that you feel that the
item does not reflect the content of that part of the PPS.
Choice (R) Revise means that you feel the item should be
revised to reflect the content in a more effective manner.

You are also provided with a form which shows the part
of the PPS from which each item of the PCQ was derived. This
will enable you to quick reference each item of the PCQ
without having to read the entire PPS to find a particular
item. The items of the PCQ follow the same order as do the
corresponding sections of the PPS.

Please read each item and the appropriate section of the
PPS and decide if you.agree, disagree or think.the item should
be revised. If you have any questions, please contact me at
517-355-7648 during office hours (8:00-5:00) Monday through
Thursday or at home at 517-336-0786 at other times. ‘The speed
of your response will be appreciated.

Thank you for your help in this matter.

Sincerely,

Joel L. Kelley, M.A.

162

on V u

The following is a list of each item of the Personal
Change Questionnaire and a brief indicator of the content of

that item.

In the space to the right of the item and its

content, indicate whether you agree or disagree that the item
content is reflective of the content of the Personal Paradigm
Shift. You may also choose to indicate that the item needs
to be revised to better reflect the content of the PPS.
Indicate your preference by placing a circle around the
appropriate response.

ITEM
NUMBER

1

2

10
11
12

13'

14
15
16
17

18

ITEM
CONTENT

main problem
specific goals
situation
feelings

thoughts

actions
consequences
praise/criticize
will lose

needs

punishments if change
rewards if change

punishments if don't
change

rewards if don't change
committed

fears

how to avoid fears

new feelings

> 3’ iv V >' 3’ :5 V >' 3’ iV »

V >' 3’ :9 > >'

UUUUUOUUUUUU

UUUUUU

AGREE DI SAGREE REVISE

w 21 :o w 21 50 w 2! :0 w 21 :0

w I! I! :0 w w

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

Please return all materials in the enclosed envelope.

Thank you for you help with this project.

163
new beliefs
new mental images
new bodily reactions
new self judgements
new, healthy reactions

recognize new
consequences

bring new self
evaluations

recognize needs being
met

realistic first goal
create support system
seek new information
seek role model

reward self for change
make plan

give self feedback

set realistic time for
making a change

make a backup plan
plan to avoid sabotage

keep diary

V F >' 3' 3’

V 3’ iv 5 3' 3’ {P >

> 3’ :5 >

UUUDD

UUUOUUUU

U C! (3 U

m w w 21 50 w

w

w I! in m 21 50 w 2!

w 21 in w

164

 

 

 

SUMMARY
Personal Change Questionnaire Personal Paradigm Shift
Section Question Question Section Step
Title Number Content
Personal 1 main problem A 1
Awareness 2 specific goals A 2
3 situation A 3
4 feelings A 4
5 thoughts A 4
6 actions A 5
7 consequences A 6
8 praise/criticize A 6
9 lose A 7
10 needs A 7
Decision 11 punishments if change B 8
Making 12 rewards if change B 8
13 punishments if don't change B 8
14 rewards if don't change B 8
15 committed B 9
16 fears B 10
17 avoid fears B 10
Oppositional 18 new feelings C 12 **
Alternatives 19 new beliefs C 12
20 new mental images C 12
21 new bodily reactions C 12
22 new self judgements C 12
23 new, healthy reactions C 13
24 recognize new consequences C 14
25 bring new self evaluations C 14
26 recognize needs being met C 14
Planning 27 realistic first goal D 15
and 28 create support system D 16
Action 29 seek new information D 17
30 seek role model D 17
31 reward self for change D 18
32 make plan D 19
33 give self feedback D 20
34 realistic time for change D 21
35 backup plan D 22
36 plan to avoid sabotage D 23
37 keep diary D 24

165

PERSONAL CHANGE QUESTIONNAIRE
FORM S

Quality Assurance Project
William C. Hinds, Ed.D.
Joel L. Kelley, M.A.

Michigan State University
1989

166

Directions: For each statement, rate how well you know yourself on a scale of knowing from
0 to 7, with '0' meaning that you “don’t know' to '7' meaning that you know “very well“.
Statements refer to your own actions, thoughts feelings, goals, etc. The word “Goals” refers to
plans for changing your old actions.

1.

10.

I know what my main problem is.
For example: I get depressed and can’t work.

. I know specific goals that will help me solve

my problem.
For example: I will lose twenty pounds and
have more energy for work.

. I know the situation in which my problem

takes place.
For example: I get depressed when I am alone
too much.

. I know my own feelings in the problem

situation.
For example: I feel lost and very nervous.

. I know my own thoughts in the problem

situation.
For example: I wonder what is wrong with me.

. I know what actions I take when I experience
my thoughts and feelings in the problem situation.

For example: I spend a lot of time reading to
escape these thoughts and feelings.

. I know the consequences of my own actions in

the problem situation.
For example: The more I sit and read, the
more I feel isolated and alone.

. I know whether I am praising or criticizing

myself in the problem situation.
For example: I usually put myself down for
feeling so low.

.Iknowwhatlwillloseiflgivcupmy

present actions in the problem situation.
For example: I will lose some of the
attention I get from some of my friends
when I am depressed.

I know what needs of mine are being met by my
present actions in the problem situation.

For example: Some of my needs for attention
are being met.

DON’T KNOW
0 1 2 3

KNOW
67

167

11. I know what punishments I will get when

I change my actions in the problem situaction.

For example: It would feel a punishment to
lose some of that special attention.

12. I know the rewards I will get when I change
my actions in the problem situation.
For example: If I change my actions I won’t
be alone and depressed as much.

13. I know the punishments I will get if I don’t
change my actions in the problem situation.
For example: Ifl don’t change I can expect
to feel isolated and alone a lot.

14. I know what rewards I will get if I don’t
change my actions in the problem situation.
For example: I can stay depressed and not
have to go to work.

15. I know how much I am committed to making
a change in my actions.
For example: I am willing to miss some of my
favorite activities to attend therapy sessions.

16. I know the fears I will have to face if I
don’t change my actions.
For example: I will have to face the fear
that I am a failure.

17. I know how to avoid my fears.
For example: I can avoid my fears by staying
very busy, reading or playing pinball.

18. I know I can bring about new feelings in the
problem situation.
For example: I know I can develop feelings
of self confidence even if I do fail.

19. I know I can bring about new thoughts in the
problem situation.
For example: I can think about making some
new friends when I do things rather than
being alone.

20. I know I can bring about new mental images
in the problem situation.
For example: I can see myself being
congratulated for doing a good job and
being successful.

DON’T KNOW

0123

KNOW
67

168

21. I know I can recognize new bodily reactions
in the problem situation.
For example: I’ll be able to feel lighter and
more energetic when I lose twenty pounds.

22. I know I can bring about new self judgements
in the problem situation.
For example: I can think of my depression as
a stage I can successfully get through.

23. I know I can bring about new personal actions
to solve my own problems.
For example: I can do something active rather
than bury my head in a book for two weeks.

24. I know I can recognize the consequences of my
new actions in the problem situation.
For example: I’ll be able to see that staying
active helps me avoid unnecessary depression.

25. I know I can bring about new self evaluations
as a result of my new actions in the problem
situation.

For example: I can feel more in control and
can feel positive for taking positive actions.

26.1knowlcanrecognizeneedsthatwillbemet
by my new actions in the problem situation.
For example: Being active will help me meet
new people and satisfy my needs for attention.

27. I know I can choose a realistic starting
goal for myself.
For example: I can choose to lose two pounds
in the first month of dieting.

28. I know I can find someone to support me in
making a change.
For example: I could attend weight watchers
or talk to my doctor.

29. I know I can seek out information for making
some changes.

For example: My doctor can give me some good
information about losing weight.

30. I know I can find a role model.
For example: A friend at work lost forty
pounds and he looks great.

DON’T KNOW

0123

KNOW

67

 

169

31. I know that I can give myself rewards for
making these changes.
For example: When I lose weight, I can get
that new leather jacket I’ve been wanting.

32. I know I can make an organized plan for
achieving these changes.
For example: I can make a daily schedule
for my exercise and food intake.

33. I know I can give myself feedback on my
progress.
For example: I can keep a record of my weight
each Monday to see how I am doing.

34. I know I can set a realistic period of time
for change to take place.
For example: If I work on losing a pound a
week, I’ll reach my goal by the new year.

35. I know that I can make a backup plan ifI
fail to reach my goal.
For example: If my weight loss is too slow,
I can increase my daily exercise a bit at a
time.

36. I know I can make a plan to avoid sabotaging
or undercutting myself.
For example: Each week I will review what I
have eaten and how I exercised to see if I
have followed my plan.

37. I know I can keep a diary or chart to follow

my change.
For example: I can make a chart that includes
my daily calorie intake and exercise schedule.

DON’T KNOW

01

2345

2345

2345

2345

2345

2345

2345

KNOW
67

67

67

67

67

67

67

This next question asks you to rate the present level at which you are dealing with your
personal situation or problem, the one for which you are seeking counseling. Please rate it
from 0 to 7, with 0 meaning that you are not dealing with it at all, and 7 meaning that you are

dealing with it very well.

38. At present, I am dealing with my problem.
END

Thank you for your help.

very
well

01234567

 

170

PERSONAL CHANGE QUESTIONNAIRE
INTERVIEWER FORM
FORM 8

QUALITY ASSURANCE PROJECT
William C. Hinds, Ed.D.

Joel L. Kelley, M.A.

Michigan State University
1989

171

PERSONAL CHANGE QUESTIONNAIRE - INTERVIEWER FORM

Rate the client on how sure you are that they have an understanding about themselves in their
problem situation.

AW:

How sure are you that the client: DOESN’T KNOW KNOWS
1. what their main problem is? 0 1 2 3 4 5 6 7
2. specific goals that will help solve

their problem? 0 1 2 3 4 5 6 7
3. the situation in which their problem

behavior occurs? 0 1 2 3 4 5 6 7
4. their feelings in the problem situation? 0 1 2 3 4 5 6 7
5. their thoughts when they are in the

problem situation? 0 1 2 3 4 5 6 7
6. what actions they take when experiencing

their thoughts and feelings in the

problem situation? 0 1 2 3 4 5 6 7
7. the consequences of their actions in

the problem situation? 0 l 2 3 4 5 6 7
8. whether they are praising or criticizing

themselves in the problem situation? 0 1 2 3 4 5 6 7
9. what they would lose if they gave up

their present problem solving behavior? 0 1 2 3 4 5 6 7
10. the needs which are being met by their

present problem solving behaviors? 0 1 2 3 4 5 6 7
B. Wins:
11. the punishments they will receive when

they change their action? 0 1 2 3 4 5 6 7

12. the rewards they will get when they
change their actions? 0 1 2 3 4 5 6 7

13. the punishments they will get if they
don’t change their actions? 0 1 2 3 4 5 6 7

14. the rewards they will get if they don’t
change their actions? 0 l 2 3 4 5 6 7

172

How sure are you that the client: DOESN’T KNOW
15. how much they are committed to change? 0 1 2 3 4
16. the fears they will have to face if

they don’t change? 0 1 2 3 4
17. how they avoid their fears? 0 1 2 3 4
C. W: DOESN’T KNOW

THEY CAN

18. bring about new feelings in the problem
situation? 0 1 2 3 4

19. bring about new beliefs in the problem
situation? 0 1 2 3 4

20. bring about new mental images in response
to the problem situation? 0 1 2 3 4

21. recognize new bodily reactions in response
to the problem situation? 0 1 2 3 4

22. bring about new self judgements in the
problem situation? 0 1 2 3 4

23. bring about new, healthy reactions which
will counter their old reactions in the
problem situation? 0 1 2 3 4

24. recognize new consequences of their
behavior? 0 1 2 3 4

25. bring about new self evaluations as a
result of their new actions? 0 1 2 3 4

26. recognize the needs that will be met by

their new actions? 0 1 2 3 4
D. W
How sure are you that the client: CAN’T

27. can choose a realistic first goal for
themselves? 0 1 2 3 4

28. can create a support system for
themselves? 0 1 2 3 4

29. can seek out information about new
behaviors? 0 l 2 3 4

KNOWS
6 7

6 7

6 7
KNOWS
THEYCAN

67

67

67

67

173

How sure are you that the client:
30. can seek out a role model?

31. can give rewards to themselves for
making changes?

32. can make an organized plan of attack?

33. can give themselves feedback on their
progress?

34. can set a realistic period of time for
making a change?

35. can make a backup plan if they fail to
reach their goal with the first plan?

36. can make a plan to avoid sabotaging
their success?

37. can keep a diary or chart to follow
their change?

5 WW:
Rate the client on the following:
1) Defensiveness

2) Introspection

3) Engagement

4) Self-disclosure

5) Commitment

6) Emotional readiness to learn a program
of change

@6069

H
NNNNN
wwwuw

##bb
UIMUtMUI

1234

HIGH
67
67
67
67
67

567

F. On a scale of 0 to 7 below, rate your client on how well they are dealing with their problem

at this time.

POORLY
01234567

WELL

 

174

Quality Assurance Research Project
Michigan State University
Information Package and Consent Form

Dear Participant,

The following form allows us to use the results of this
survey as part of a research project at Michigan State
University. We are attempting to devise a survey that will
help evaluate the quality of services at mental health
facilities and counseling centers. A major part of this
evaluation is the establishment of accurate and reliable
instruments. The survey you are being asked to complete is
being evaluated to see if it‘will be acceptableu In order for
us to do this, we need your help.

We are asking you to complete the survey as accurately
and honestly as possible. Your responses will be kept
confidential. In fact, your name will never appear on the
answer sheet itself, nor will it appear on the information
sheet. We need your answers so that we can make an evaluation
about validity of the survey. Information on the other sheet
allows us to make comparisons about how groups of people will
respond. This extra information is vital to us, but it will
not be detailed enough to identify any one individual person.
In addition, since your name is not on the survey or the
information sheet, your answers will remain strickly
confidential.

You are being asked to take this survey, but you are
under no obligation to do»so. ‘You may change your mind at any
time and decide not to complete the survey. If you choose not
to complete it, just turn it in to the person who has given
it to you. There is no obligation on your part.

If you agree to complete the survey, we hope you will be
as honest and accurate as possible so that the survey can be
evaluated effectively. It should take you about 30 minutes
to complete the three forms included, that is the information
form, the consent form and the survey.

If you are willing to complete the survey, please read
and sign the consent form on the next page. When you have
completed the survey and the information form, return all the
forms to the person who has asked you to participate.

Thank you for your help and participation.

William C. Hinds, Ed.D. Joel L. Kelley, M.A.

Project Director Research Assistant

175
Quality Assurance Research Project
Michigan State University
Consent Form
To be filled out by each participant:

I, ,doherebygive
my consent to Michigan State University, William C. Hinds,
Ed.D., and Joel L. Kelley, M.A., to use my answers on the
Personal Change Questionnaire as part of their study on
Quality Assurance.

I understand that I am under no obligation to participate
in the study by completing this survey. I further understand
that at any time during the taking of the survey, I can change
my mind and choose not to complete it.

I understand that the results will be used solely for the
study of the instrument itself and that my individual answers
will remain confidential.

I understand that the results of the completed project
will be public knowledge and that they will be used to
evaluate this survey as a measurement device for quality
assurance programs. 3

I am completing this survey of my own free will.

Signature Date

 

176

Quality Assurance Research Project
Michigan State University

Please answer by checking the choice that applies to you:

AGE: 16-20 SEX: FEMALE __ FAMILY INCOME LEVEL
21-25 MALE __ (Check one)
26-30 0-$10,000___
31-35 MARITAL STATUS: $10,000-$20,000
36-40 SINGLE $20,000-$30,000
41-45 MARRIED $30,000-$40,000

46-50 DIVORCED __ $40,000-$50,000
51-55 WIDOWED _ $50,000 or more
56-60 OTHER _
60-
EDUCATION:
(Check highest level you attained) EMPLOYMENT: (check one)
DID NOT FINISH HIGH SCHOOL __ SELF EMPLOYED __
FINISHED HIGH SCHOOL OR GED __ HOURLY WORKER __
COMPLETED TWO YEARS COLLEGE __ SALARIED WORKER __
COMPLETED FOUR YEARS COLLEGE __ HOUSEWIFE __
COMPLETED GRADUATE DEGREE __ RETIRED __
‘ UNEMPLOYED _
OTHER (£111 in) _
NUMER OF COUNSELING SESSIONS YOU HAVE HAD HERE: (If you do

not know the number, write in the approximate number.)

ARE YOU COVERED BY INSURANCE FOR YOUR SESSIONS HERE: YES __

no _
DON "r KNOW _
REASON FOR SEEKING COUNSELING: HOW LONG HAVE YOU HAD
(Check the most important one) THIS PROBLEM?
ANXIETY O-3months
DEPRESSION 3 mo. - 1 year

MARITAL OR FAMILY PROBLEM

ALCOHOL/SUBSTANCE ABUSE

PHYSICAL PROBLEM

I WAS TOLD I HAD TO COME
By Whom?

OTHER REASON FOR BEING HERE FOR COUNSELING __
Explain:

1 year - 2 years
2 years or more

 

 

 

177

Directions for administration of the Personal Change
Questionnaire:

1)

2)

3)

4)

5)

6)

Give the client/subject the packet containing the
information form, consent form , the demographic
information sheet and the Personal Change Questionnaire.

Have the client read the information form. If they agree
to participate. have them sign the consent form and fill
out demographic sheet. Then have them complete the
Personal Change Questionnaire.

When they have completed.all forms, they should place the
demographic information sheet and the Personal Change
Questionnaire in the envelope provided. They should
attach the consent form to the outside of the envelope
and return it to you. This allows you to identify the
client/subject so that you can complete your portion of
the Personal Change Questionnaire without seeing their
answers. It is important for statistical and
measurement purposes that their self evaluation be
correctly matched with your evaluation of them.

After they are finished, you identify the client/subject
in question and then fill out the Therapist information
form and the Therapist Form of the Personal Change
Questionnaire. Place them in the same envelope with the
client forms.

Remove their consent form and place it in a separate
envelope with all the other consent forms. It was on
their envelope originally to help you identify them so
you could effectively fill out your part of the survey
and get it in the correct envelope.

After completing all forms from all clients/subjects,
place all the client envelopes and the envelope
containing all the consent forms in the mailing envelope
and return to:

Joel L. Kelley
P.O.Box 26292
Lansing, Michigan 48909

or call me at 517-355-7648 (W) or 517-336-0786 (H) and
I'll pick them up.

Thank you for your help.

178

Quality Assurance Research Project
Michigan State University
Therapist Form

This form is to be completed by the therapist of someone
who has agreed to participate in this research project. These
four pages should come to you in a packet of material which
has your client's consent form attached. Please identify the
client to your own satisfaction without looking at their
answers. Then complete your questionnaire, ie. , these sheets.
When you have done this, please place your answer form (this
sheet and the attached sheets) in the same envelope as your
client has placed their answer form. Seal the envelope with

the consent form outside and save both for the researcher to
collect. '

Thank you for your help.

 

1) Please check your theoretical/treatment orientation:
a) Psychodynamic
b) Cognitive/Behavioral
c) Psychoeducational
d) Gestaldt
e) Eclectic
f) other, please describe:

 

2) Please give your best DSM-III diagnostic code for the
client, Axis I and II only, in the space provided.
If you do not have such a diagnosis, please briefly
describe your diagnosis of their problem.
DSM-III Axis I:
DSM-III Axis II:
Your own Diagnosis:

 

 

 

 

 

 

179

BIBLIOGRAPHY

Affeldt, J.E., Roberts, J.S., and Walzak, R.M. (1983) . Quality
Assurance: Its origins, status and future direction: A

J CAH perspective. MW.
g, 245- 255.

Alschuler, A.S., and Ivey, A.E. (1973a) . Getting into psycho-

logical education. WW. £14.19 .
682-691.

Alschuler, A.S., and Ivey, A.E. (1973b). Internalization: The

outcome of psychological education, Pgrsoggel and
Guidance Jourggl, 51(2), 607-610. -

Anastasi, A., (1976). Psychological testing, New York:
MacMillan Publishing Coompany.

Archer, J. , and Kagan, N (1973) . Teaching interpersonal

skills on campus: A pyramid approach, ,2. Cgugseling
PS (31)., 2Q, 335-341.

Authier, J. (1977). The psychoeducational model: definition,

contemporary roots and content. Qanadian.§22nselor. 12.
15-22.

Authier, J., Gustafson, Guerney, B., and Kasdorf, J.A. (1975) .
The psychological practitioner as a teacher: A
theoretical, historical and practical review, 991mm
Psychologist, 5(2), 31-50.

Benfari, R. , Eaken, E. , and Stole, J. (1981) . Behavioral
Intervention and compliance to treatment regimes, Annual
Review of Egbllc Health, 2, 431-471.

Bent, R.J. (1988). Education for quality assurance, in G.

Stricker and A.R.Rodriguez, (Eds.) , MW
Assuran2e_in_nental_fleelth. New York: Plenum Press.

Brook, R.M., Kamberg, C.J., and Lohr, K.N. (1983). Quality

assessment in mental health..Erofessional_2sxchelogx..5.
415-439.

Carkhuff, R.R. ( 1971) . Training as a preferred mode of treat-
ment, Jgurgal of ngnsgllng Esygnglggy, 1§(2), 123-131.

Carkhuff, R.R. (1972). The development of systematic human
resource development models, Cgunseligg EsxgnglggLIS .

3(2), 4-11.

 

 

 

 

180

Christensen, A., and Arkowitz, H. (1974). Preliminary report
on practice dating as a treatment for college dating

problenS. I2urna1_of_gounseling_£sxchologx. 21121. 92-
95.

Cohen, L.H. ( 1974) Peer review of psychoanalytic psycho-
therapy: An experimental study of the APA/CHAMPUS
program. Professional_2sxchologx.12. 776-784-

Cohen, L.H. (1988). Research on quality assurance, in G.

Stricker and A.R.Rodriguez, (Eds.), Hgggbggg_gfi_gggll§y
Assurence.1n_uental_nea1tn. New York: Plenum Press.

Cohen, L.H., and Oyster-Nelson, C. (1981). Clinician's
evaluations of psychodynamic psychotherapy: Experimental

data on psychological peer review; figg;ngl_gf_§gg§gltigg
an__§11n19a1_£s22h21291.$2. 583- 589.

Cohen, L.H., and Pizzirusso, D. (1982). Peer review of
psychoanalytic psychotherapy: Experimental studies of the

APA/CHAMPUS program, Eyglggglgg and thg figgltn ngfes-
close. 2111. 415- 436.

Cole, H.P., and Lacefeild, W.E. (1974). Theories of learning,
development and psychoeducational design: Origins and

applications in non-school settings, Jggzggl_g£_Agpllgd
flgélfih. fig: 5'16-

Cronbach, L., (1951). Coefficient Alpha and the internal
structure of tests, Egyghggggzlkg, lg, 297-334.

Dall, O., and Claiborn, W. (1982). .An evaluation of the
Aetna pilot peer review project. Psychoth_rapxi_1heorxi
Ee_earch_and_£ractics. 12. 6-16-

Dollard, J., and Miller, N.,
New York: McGraw-Hill, 1950.

Donabedian, A. (1966). Evaluating the quality of medical
care. E111pank_nem2ria1_£und_ouarterlx. 4. 166-204-

Donabedian. A. (1980)- WWW

M1t__g_§g§sm_nt, Ann Arbor: Health Administration
Press.

Ellis. A- (1963)- Beas2n_and_emotion_1n_p§29h2tberap¥. New
York: Lyle Stewart.

Ellis, A. (1977). Skill training in counseling and psycho-
therapy. Qanadian_£22nselor. 12. 30-35-

 

181

Ellis, A., and Harper, R.A., (1961). A gnlge £9 nnnionnl
living, Englewood Cliffs, New Jersey: Prentice Hall.

Ely, A.L., Guerney, B.G., and Stover, L. (1973). Efficacy of

the training phase of conjugal therapy, Egygngnngnnny;
The2r21_Besearch_and_2ract1ce. 19111. 201-207-

Eysenck, H.J., (1952) . The effects of psychotherapy: An
evaluation, on o s t’ Ps lo , lg, 319-
324.

Galton. R-. (1952). nered11ar1_Qen12s1_An_1nguirx_1nto_1ts
W, New York: Horizon Press (Reprint).

Gambrill, E. (1988). Social skills training, in D. Larson,
(Ed-). Ieach1ng___p§xchologica1___skills. Monterey.

California: Brooks/Cole.

Gibson, R.W. (1988). The influence of external forces on the
quality assessment process, in G.Stricker and.A. R. Rod-
riguez, (Eds.), ' ss ra c ' en
nealth, New York: Plenum Press.

Goddard, H.H, (1913). Ing_§nlllknn_£nnily, New York: Mac-
Millan Company.

Goldschmidt, P. G. (1976). A cost-effectiveness model for
evaluating health care programs: Applications to drug

abuse treatment, lngniny, lglnnnl, 29-47.

Goldstein, AIP., Gershaw, N.J., and Sprafkin, R.P. (1984).
Structured Learning Therapy: Background, procedures and
evaluation, in D.Larson, (Ed.),
gkillg, Monterey, California: Brooks/Cole, 1984.

Gravetter, F. J. and Wallnau, L.B., (1985). finn§31§§12§_£gr
the_behaxioral_§ciences. New York: West Publishing Co.

Guerney, B.G. (1977). Should teachers treat illiteracy, hypo-
calliography, and dysmathematica?, angglnn_§gnn§glgn,

1;, 9-14.

Guerney, B.G., Guerney, L.F., and Stollak, G. (1972). The
potential advantages of changing from a medical to an

educational model in practicing psychology, W1
Dexelopment. 2. 238-245-

Guerney, B.G., Stollak, G., and Guerney, L.F. (1970). A
formal model for a new'mode of psycholgical practice: or

How to escape a zombie. Qounseling_£sxsh21291st. 2121.
97-104.

 

 

182

Guerney, B. G., Stollak, G., and Guerney, L. F. (1971). The
practicing psychologist as educator - an alternative to

the medical model. Professional_2sxchologx. 2. 276- -282-

Gutride, M.E., Goldstein, A.P.,.Hunter, G., Carrol, 8., Lower,
W., Clark, L., and Furia, R. (1974). Structured Learning
Therapy with transfer training for chronic inpatients,

l_nrna1__f_§11n1911_2sxch2123L 12. 277-279.

Gutride, M.E., Goldstein, A.P., and Hunter, P.E. (1973). The
use of modeling and role playing to increase social in-
teraction among asocial psychiatric patients, Jgurnal of

Q2nsu111ng_and_911nical_£s2chologL 12111. 43-47-

Habeck, R.V., and Fuller, T.C. (1985). Rehabilitation Coun-
seling: A psychoeducational perspective, W
Beh12111111122_§22nse1129. 11121. 43-47-

Hays, J. (1977). Three methods of peer review in a state men-

tal hOSpital system..Eszchelogica1_8eper_s...11. 519- 525-

Henisz, J., Goldblatt, P.B., Flynn, H.R., and Garrison, V.
(1974) . A comparison of three approaches to patient care

appraisal based on chart review, W

Psychaitrx. 111. 1142-1152-
Hersen, M., Turner, S.M., Edelstein, A.B., and Pinkston, 8.6.
(1975). Effects of phenothiazines and social skills

training in a withdrawn schizophrenic, ignnnnl__gf
911n1211_2_xchol_gL 11. 588- 594.

Hersen, M., and Bellack, A. S. (1976). A multiple baseline
analysis of social skills training in chronic schizo-

phrenics. 2_2rna1_of_Applied_Beha_12r_Analxs1s. 2. 239-

246.
Hinds. N. c., (1983). Ihe_2ers2nal_Earad19m_§hift1_a_life:
s l- i - - t°01 tog oaCr . ge._tg ,ar- no1-°’n‘n ,

East Lansing, Michigan: Michigan State University, 1983.

Hobbs, N. (1964). Mental health's third revolution, Anenlgnn
12urn11_21_9rtnopsxchiatrx. 11. 822-833-

Ivey, A.E. (1969). The intentional individual: a process-
outcome view of behavioral psychology, ggnnsgllng

Esxchologist. 1. 59-60-

Ivey. A- E-. (1971) n1cr9c9unsel1ng1.1nnoxation§_1n_cgunse1:
lng__n_ln1ng, Springfield, Illinois: C. C. Thomas, 1971.

 

183

Ivey, A.E. and Alschuler, A.S. (1973). Psychological educa-

tion is.---.. W. 5.11.21. 588-
597.

Ivey, A.E., and Galvin, M, (1984). Microcounseling: A meta-
model for counseling, therapy, business and medical
interviews, in D.Larson, (Ed.), '

Skillg, Monterey, California: Brooks/Cole, 1984.

Jeney. 3.. (1985)- WW

0 a er s l b av'ors d s h -
logitsuiflsmbanmmmiepamnatianm unpublished
doctoral dissertation, Michigan State University.

Kagan, N., and Krathwohl, D.R., (1967). Stnglg§_1n_nnmgn_in;

e act on: nte e so cess Recall stimulated
ylgggnnng, final report of grant OE 7-32-0410-270, East
Lansing, Michigan: Educational Publications Services.

Kagan, N., Schaube, P.G., Resnikoff, A., Danish, S. J., and
Krathwahl, D.R. (1969). Interpersonal Process Recall,
www.1fl. 365-374-

Kagan, N. (1984). Interpersonal Process Recall: Basic methods
and recent research, in, D. Larson, (Ed. ), gagging

Esyghologlga al §k ills, Monterey, California: Brook/Cole.

Kasdorf, Guustafson, K. (1978). Research related to micro-

counseling, in A. E. Ivey and A. S. eAlschuler, (Eds. ), 1112119:
ounsel -° , ov °., :. . s: r- :0
psxshstnarapx_snd_adusat12n. Springfield. Illinois: C-

C. Thomas.

Kim, J. O. and Mueller, C. W. (1978). Factor analysis:
Statistical methods and practical issues. In J. L.
Sullivan & R. G. Memi (Eds.), W

' ' ' c . Beverly Hills: Sage
Publications.

Landis, G. (1937). A statistical analysis of psychotherapeu-
tic methods, in L.Hensie, (Ed.), 0
1n nsycnotngrnny, New York: Columbia University Press.

Larson. 0.. (Edd. (1984)- WWW.
Monterey, California: Brooks/Cole, 1984.

Luft, L., Sampson, L., and.Newman, D. (9176). Effects of peer
review on outpatient psychotherapy: Therapist and patient
review. WM. 121. 891- -895-

184

Maslow, A. (1955). Defeciency motivation and growth motiva-
tion. in R-Jones. (Edd. MW:
tlgn, Lincoln, Nebraska: University of Nebraska Press.

MehrenS. W- and Lehman. I-. (1978). Measurement_and_ayalna:

W. New York: Holt. Rhine-
hart and Winston.

Moreland,, J. R., Ivey, A. E., and Phillips, J. (9173). An.eval-
uation of microcounseling as an interview training tool,

MW. 11. 294-
300.

Mowrer. 0.. (1950)- W.
New York: Roland, 1950.

Mowrer, O., and Mowrer, W. (1938). Enuresis: A method for its

study and treatment. MW.
g, 436-459.

Murray, E.J. (1954). A case study in a behavioral analysis

of psychotherapy. WWW:
m, 52, 305-310.

Nunnally, J., (1970). o s o -
t, New York: MaGraw-Hill.

Nunnally, J. (1978). Egygngngtnlg_tnggny, New York: MaGraw-
Hill.

Parker. 91-1... (1987)- W

n st ss he t e v'o a ac em' c i veme
WW. unpub-
lished doctoral dissertation, Michigan State University,
East Lansing, Michigan.

PavloV. 1.. (1927) WW

4‘ 'I F ° '1. ' '9 Q- - ‘9 '. '1‘.,

London: Oxford University Press.

Pierce, R.M., and Drasgow, J. (9169). Teaching facilitative
interpersonal functioning' to jpsychiatric inpatients,

WW. 111.41. 295-298-

Rodriguez, A.R. (1988). An introduction to quality assurance
in mental health, in G. Stricker and A. R. Rodriguez,
(Eds.), . . - - 1 - -
New York: Plenum Press.

 

RogerS. C-. and Dymond. Ro. (Eds.). W
nllny_gnnnge, Chicago: University of Chicago Press.

185

Rucusin, R., and Krell, H. (1980). Quality assurance in

community mental health centers, WW
Hgnlnn, 2(snn), 292-303.

Sanford, F. (1959). Creative mental health and the principle
of Habeus Mentum. MW. 112. 829-835-

Sechrest, L., and Rosenblatt, A. (1988). Evaluating peer
review, in. G. Stricker, (Ed.),

W: New York: Plenum Press-

Shueman, S. A., and Troy, W. G. (1988). Quality assurance in
outpatient.psychotherapy, inLG. Stricker, (Ed. ), finngnggk

_f_gna11ty_ssstransa_1n_mantal_nasltn. New York: Plenum
Press .
Skinner, B.F., (1948). Wnlggn Two, New York: MacMillan.
Skinner. B-F-. (1971)- Baysnd_fraadsn_and_dign1ty. New York:

Alfred A. Knopf.

Stricker, G., and Rodriguez, A.R., (Eds.), (1988). finngbggk
of_gna11ty_assnranss_1n_nantal_haalth. New York: Plenum
Press.

Suedfeld, P. (1982). Environmental factors influencing main-
tenance of lifestyle change, in R. Stuart, (Ed.), Agngz;
enc c e ti e v'o
ed e, New York: Brunner/Mazel Publishers.

Szasz, T., (1961). s ° 0 i

fon a tnegzy g: nensonnl gongucn, New York: Harper and
Harper.

Szasz, T. (1961). The use of naming and the origin of the myth
of mental illness. American_£sysnslsgist. 18. 59-65.

Szasz, T. (1966). The mental health ethic, in A.J.George,
(Ed-). Etniss_snd_§221atx. New York: DoubledaY-

Tash, W. R., and Strahler, G. J. (1984). Current status of

quality assurance in mental health, Angnlgnn_fignnyl_nnl
Scientist. 21121. 608- 629-

Truax, C., and Carkhuff, R.R., (1967). W
05.15 - :10. S O 2‘ :0 _-.t 0 11°. .- -. ,
Chicago: Aldine.

Watson, J., and MacDougall, W., (1929). Ing_nntnlg_gfi_bgnny;
' 'sm, New York: Norton.

186

Wells, K.B., and Brook, R.H. (1988). Historical trends in
quality assurance for mental health services, in G.

stricker. (Edd . WW
ngnltn,New York: Plenum Press.

Wolpe. Jo. (1969)- W. New

York: Pergamon Press.

Woy, J. R., Lnft, L., Toprac, M., Tash, W., and Stahler, G.
(1983). Evaluating the quality of clinical processes,
in. P.Glazer’ and. K.Kirkhart, (Eds.), Qgrnenstones. 9f

e o co: v' ' me a
systgns, Austin: Human Interaction Research Institute.y

HICHIGRN STRTE UN V.

HIV Will H

I LIBRARIES
"m H"WWIIWIIIHWI
264

31293008857