‘” Illilllllllllllllllllllfillllfilllflilil

3 1293 00895 2529

This is to certify that the

thesis entitled

(DNGRUENCE OF PERCEPTION ON
CLIENT mm:
A IDNG—TERM FOLUM-UP

presented by

ALEXIS JAMES VLAHOS

has been accepted towards fulfillment
of the requirements for

M. A. PSYCHOHEY

n- W

Major professor

degree in

 

 

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

f

 

k

LIBRARY
Michigan State
University

WV ‘1

 

I

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

 

 

____J|

 

 

 

 

 

 

 

 

IEI
fl

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1L
I

 

1

 

 

 

 

 

 

 

 

 

 

I___

=71

 

MSU Is An Affirmative Action/Equal Opportunity Institution

czmms-nt

CONGRUENCE OF PERCEPTION ON
CLIENT SYMPTOMATOLOGY:
A LONG TERM FOLLOW-UP
BY

ALEXIS JAMES VLAHOS

A THESIS

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

MASTER OF ARTS

Department of Psychology

1 990

ABSTRACT
CONGRUENCE OF PERCEPTION ON
CLIENT SYMPTOMATOLOGY
A LONG TERM FOLLOW-UP
BY

Alexis James Vlahos

This study examined the relationship between congruence of perception
and long-term client ratings of psychotherapy. Congruence of perception was
defined as the degree of agreement between the therapist and client through
independent ratings based on the Symptom Distress Checklist (SOL-90) ten
dimensions. It was hypothesized that client/therapist congruence of perception,
measured at the time of termination, would be related to client ratings of
therapeutic success six months after therapy had been terminated. This
hypothesis was not supported. Additional analyses examining delayed and
deterioration effects occurring during the post therapy period are reported.
While clients reported improvements in functioning during the posttherapy
period, they did not attribute these improvements to their therapies or
therapists. Implications of these findings as well as suggestions for future

research are discussed.

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to my family and thesis
committee for there never ending patience, support, guidance, encouragement
and good humor.

I would like to thank my parents, John and Georgia Vlahos, for their
guidance, support and encouragement.

I would like to thank Dr. Norman Abeles, Chairman of my thesis
committee, for his contributions, patience, and guidance throughout this project.
I would also like to express my appreciation to my committee members: Dr.
Bertram Karon for his support and encouragement, and Dr. Raymond

Frankmann for his knowledge and suggestions on statistical techniques.

TABLE OF CONTENTS

LIST OF TABLES
INTRODUCTION

REVIEW OF THE LITERATURE
PURPOSE AND IMPORTANCE OF
FOLLOW-UP RESEARCH
CRITICISMS OF FOLLOW-UP DESIGNS
FOLLOW-UP STUDIES AND
CLIENT VARIABLES
CONGRUENCE OF CLIENT SYMPTOMS
AND THERAPEUTIC OUTCOME

HYPOTHESES
HYPOTHESIS l
HYPOTHESIS ll
HYPOTHESIS III

METHOD
SUBJECTS
INSTRUMENTS
PROCEDURE

iv

Page

vi

15

21
21
22

24
24
24
26

TABLE OF CONTENTS

RESULTS
CONGRUENCE OF PERCEPTION
HYPOTHESIS I
HYPOTHESIS ll
HYPOTHESIS III

DISCUSSION
CONGRUENCE OF PERCEPTION AND
RATINGS OF THERAPEUTIC SUCCESS
METHODOLOGICAL EXPLANATIONS
CONCLUSION

REFERENCES

28
28

31
31

34
34
40
43

45

LIST OF TABLES

TABLES Page

1. Correlations between client and
therapist SOL-90 responses 30

2. Correlations between congruence of
perception and follow-up ratings
of therapeutic success by the client 30

3. Results of t-tests on client
ratings of therapeutic success 32

vi

INTRODUCTION

Client/therapist congruence of perception on client symptomatology has
been identified as a factor related to client ratings of psychotherapeutic
outcome (Filak and Abeles, 1984). While this relationship has been established,
the long-term effects of congruence of perception has yet to be explored. This
study examined the relationship between client/therapist congruence of
perception and long-term client ratings of therapeutic success. In addition to
this, delayed and deterioration effects of client ratings of therapeutic success
during the post-therapy period were examined. There were two major purposes
to this study. The first was to extend the findings of Filak and Abeles (1984) to
the post-therapy period. Another objective was to add to the growing research
which elucidates variables which contribute to the long-term maintenance and

persistence of psychotherapeutic changes.

REVIEW OF THE LITERATURE
Eysenck’s (1952) review questioning the efficacy of psychotherapy sparked
a controversy which laid the groundwork for the myriad of psychotherapy
outcome studies that have been conducted to the present time. About the
same time Eysenck came out with his startling conclusions, Victor Raimy (1952)
addressed the issue of follow-up research. He harshly criticized researchers
stating:

"it might be worthwhile to mention the appalling lack of follow-up in the
treatment of mental disturbances certainly contributes to the lack of
knowledge, the insistence upon personal predilections, and the general
impression that no one knows what is really going on." (pp. 321-322).

While Eysenck’s comments proved to have a great impact on the field of

outcome research, Raimy’s criticism failed to have the same effect. Throughout
the years researchers have echoed Raimy’s criticisms hoping to motivate others
to explore this area of psychotherapy outcome (Sargent, 1960; McNair, Lorr,
Young, Roth & Boyd, 1964; Fiske, & Goodman, 1965; Gossett, Lewis, Lewis, &
Phillips, 1973; May, Tuma, & Dixon, 1976; Mash, & Terdal, 1980; Nicholson, &
Berman, 1983; Schramski, Beutler, Lauver, Arizmendi, & Shanfield, 1984).
Raimy’s criticism becomes even more poignant when one examines the lack of
follow-up research in the area of therapeutic factors that contribute to the
maintenance of therapeutic gains.

One area that has shown some promising results in psychotherapy

outcome studies has been congruence of perception research (Filak & Abeles,

1984). Congruence of perception has been shown to correlate positively with

2

3

successful therapy outcomes; yet the persistence and maintenance of symptom
relief after termination has yet to be studied. The present study is designed to
explore the construct of congruence of perception of client symptoms in the
posttherapy period. Before discussing the details of this study, it is useful to
explore the relevant literature. This review begins with an examination of the
purpose and importance of follow-up studies to outcome research. From that
starting point the criticisms and methodological difficulties surrounding follow-up
studies will be explored. Follow-up studies which explore client and therapeutic
factors will then be considered. Finally research on congruence of perception
in the clinical environment will be examined. This literature was chosen as a
means to present the rationale and methodology of the present study.
P r 0 An lm rtan e f Follow- Re r h

Despite the modest number of follow-up studies, researchers recognize the
clinical relevance and scientific merit of such designs. Follow-up studies
provide an added dimension to the results found in psychotherapy research.
These studies allow researchers to determine the natural history of disorders
and to identify factors which may predict or influence itscourse (May, Tuma &
Kraude; 1965). Follow-up research can also provide inferential material by
evaluating a type of treatment or by comparing the effects of two or more
treatments over time (Paul, 1967; Liberman, 1978). In addition to these types,
Mash and Terdal (1980) suggest that follow-up designs can serve diagnostic

and therapeutic purposes. The former represents studies whose primary goal

4

is to maximize gains by clients following their treatment period, the latter
represents follow-ups which may be treatment oriented in nature.

Follow-up designs have also been praised for providing useful information
concerning the long-term effects of psychotherapy. Cowen and Combs (1950),
in reporting their results of a follow-up study of 32 cases of clients treated with
nondirective psychotherapy, contend that the passage of time is a necessary
prerequisite to the validation of the success or failure of psychotherapy.
Kolotkin and Johnson (1983) assert that the length of a follow-up measures
different variables. Long-term follow-ups reflect types of lasting change made
possible by early therapeutic gains; whereas shorter designs reflect changes
that have taken place during the period of contact (Kolotkin & Johnson, 1983).
Other researchers have demanded the use of long-term follow-up evaluations in
the absence of an adequate definition of psychotherapy and its long-term
process (Fiske, Hunt, Luborsky, Orne, Parloff, Reiser, & Tuma, 1970) and the
expectation that long term follow-up will measure therapeutic gains.

Mash and Terdal (1980) assert that the importance of investigating these
long-term effects is emphasized by the “discrepancy often found between status
at termination and status at follow-up" (p.103). This discrepancy can either
represent therapeutic effects which were delayed and surfaced only after
treatment was terminated (Sargent, 1960; McNair et al., 1964), or the
deterioration effects of time (Nicholson & Berman, 1983), or changes due to

intervening events. In either case the information provided by such designs

5

plays a valuable role in aiding researchers and clinicians to ascertain the merit
of a certain type of therapy or therapeutic technique. As such these studies
can play an important role in the planning, initiation, and modification of
treatment programs (Gossett et al., 1973).

Garfield (1983) contends that follow-up designs add to the scientific merit of
an outcome study. He asserts that follow-up designs provide a better
understanding of variables that produce change in psychotherapy as well as
serving as a means to more efficient research methodology. Cohen (1979)
sought to gain empirical evidence between methodological characteristics of
psychotherapy outcome research and clinical psychologists' judgments of its
scientific merit. In his study Cohen mailed a description of one of four
psychotherapy studies to 200 academic and clinical pSychologists. These
professionals were asked to evaluate the study in terms of its scientific merit.
Among the variables studied was the inclusion or exclusion of follow-up data.
Through a factor analysis of the psychologists’ responses, Cohen found a
positive trend, F (1,232) = 3.02, p < .10, for the scientific merit of studies which
included follow-up data. These results lend some indirect support to those
researchers who have stressed the importance of follow-up designs for both
their practical and scientific significance in the area of psychotherapy research.
Criticisms Qf Follow-up Designs

Despite the implied clinical and scientific importance of follow-up research,

this type of analysis has been criticized on both methodological and practical

6

grounds. Sargent (1960) warns that the importance of follow-up research is
"equaled only by the magnitude of the methodological problems it presents"
(p.495). Other critics have weighed the costs and benefits of such
investigations in order to determine their practical relevance and contributions to
the field (Fiske, & Goodman, 1965; Stone, Frank, Nash, & lmber, 1961;
Nicholson, & Berman, 1983).

With regard to these methodological problems, investigators have focused
their criticisms on two intertwined areas: data collection and data interpretation.
Time is the principal factor that effects both these areas. As the length of a
follow-up study increases so too does the attrition of subjects from the original
sample (Sargent, 1960; May et al., 1965; Gottman, & Markman, 1978). This
attrition can be attributed to a variety of causes: subjects may be reluctant to
relive painful experiences, subjects may have relocated, or subjects may fear
that others will find out about their past experiences. Besides attrition, data can
be lost due to selective forgetting by the patient. Sargent (1960) maintains that
repression and denial may operate in the patient making it difficult for the
person to recall aspects of treatment that may be related to present
functioning. In both instances the researcher is faced with a loss of data which
can have a adverse effect.

The loss of data presents the researcher with major obstacles when it
comes to interpreting data. Primarily, the issue of sample bias is of great

concern in studies with attrition. Sargent (1960) questions the multiple

7

motivations of patients who return follow-up assessments. Studies that have
examined sample bias with regard to those subjects who return follow-up
assessments and those who do not have yielded conflicting and inconclusive
results. Stone et al. (1961) found no differences on important dimensions of the
30 out of 54 subjects who participated in a five year follow-up. Other
researchers have found nonsignificant trends indicating the following: Returnees
show greater number of symptoms after therapy (Tobias, 1987), are less
improved on their target complaint (Liberman, Frank, Hoehn-Saric, Stone,
lmber, & Panda, 1972), have a more favorable status at termination (Fiske, &
Goodman, 1965), and have a higher level of satisfaction at termination (Tobias,
1987). These results make it difficult to draw any conclusions. It seems
necessary to examine returnee bias on an individual basis.

From a practical viewpoint, criticism has been leveled at follow-up studies for
the cost, time, and for the limited results obtained (Nicholson, & Berman, 1983).
Follow-up designs take up a lot of time and use many resources; as such,
researchers have looked into the efficacy of follow-up designs to determine their
usefulness as a routine part of an outcome study.

Fiske and Goodman (1965), in a study examining the posttherapy period,
reassessed 69 of an original 93 patients 18 months after psychotherapy was
terminated. They found that the follow-up group, as a whole, showed neither
significant gains nor losses during this period. They view this as a positive sign

stating that "one need not be concerned about the possibility that measures at

8

termination of therapy are temporarily elevated by factors associated with the
termination itself' (p.176). These results, they argue, should be viewed as
demonstrating the efficacy of psychotherapy.

In a more comprehensive view, Nicholson and Berman (1983) utilized meta-
analysis to review 67 psychotherapy outcome studies which contained follow-up
designs. This prominent investigation evaluated whether follow-up designs
should be a necessary part of outcome studies. The results of this extensive
investigation illustrated that patient outcome at the end of therapy was highly
correlated with outcome at follow-up (r(68)=.66, p<.001; and the difference in
the mean effect size for group comparisons was -.01, SD=.26). The authors
contend that these results show the durability of therapeutic gains; as such they
assert that, due to limited monetary resources and time constraints, follow-up
studies need not be adopted routinely as part of an outcome design.

The various types of criticism bring some valid arguments to the forefront.
Yes, follow-up designs are complex, filled with methodological difficulties, and,
as Nicholson and Berman (1983) contend, add little to the evaluation of
psychotherapeutic outcome studies. Yet, as will be seen in the next section,
follow-up studies provide researchers and clinicians with the opportunity to learn
about variables which contribute to the maintenance of therapeutic gains.

F LL W- P T DIES AND CLIENT VARIABLE
The Fiske and Goodman (1965) and Nicholson and Berman (1983) studies

reflect a general trend where follow-up studies have concentrated on the

9
evaluation of the efficacy of different types of therapy. This type of study takes

a global approach in which the value of follow-up designs is judged and
decisions about future use of follow-up studies may be based on these
judgments.

Interestingly, despite their comprehensive conclusions, Nicholson and
Berman recommend that follow-up designs be adopted “where the specific
focus is on the maintenance of treatment gains" (p.275). This suggestion is
targeted at answering elucidatory questions. As opposed to evaluative studies,
elucidatory inquiries attempt to obtain generalizable knowledge and
explanations about phenomena by investigating relationships among variables
(Smith, Glass, and Miller, 1980). Studies which focus on these types of
questions have been largely overlooked in the follow-up literature (McNair et al.,
1964; Fiske, & Goodman, 1965; Schramski et al., 1984). Schramski et al.
(1984) contend that due to the limited research in this specific domain the
"identity of those variables that contribute to the maintenance of therapeutic
change is still largely unknown" (p.78).

This section examines those studies which elucidate variables that
contribute to the persistence and maintenance of therapeutic change. While
some of these studies were covered in Nicholson and Berman’s (1983) meta-
analytic investigation (Stone et al., 1961; lmber et al., 1968; Liberman et al.,
1972), the global approach that this meta-analysis took overlooked some of the

more specific findings of these studies. In addition to evaluating patient status

 

10

at various follow-up periods, these studies also took a microscopic perspective
elucidating specific factors which contribute to the persistence of therapeutic
change. It is due to these specific factors that a more in depth examination of
these studies is justified.

Early studies focused on the clients’ ability to maintain change or even
improve in the posttherapy period; to a lesser extent, they began to identify
variables which are related to posttherapy change (Stone et al., 1961; McNair et
al., 1964; lmber et al., 1968; Bergin, & Lambert, 1978). More recent research
has begun to focus specifically on variables which contribute to the
maintenance of therapeutic change (Gossett et al., 1973; Gossett et al., 1977;
Garfield, 1978; Schramski et al. 1984).

A line of research conducted at of John Hopkins Hospital explored the
variables of long-term improvement, patient experiences in the posttherapy
period and the relationship between the two. The groundwork for this line of
research began with an intensive five-year follow-up of psychiatric outpatients
conducted by Stone et al. (1961). Individuals were assigned to one of three
treatment conditions (individual therapy, group therapy; or minimal contact
therapy) for six months of treatment. Patients were tested 6 months after
discharge, then once a year until they reached 5 years. Thirty, of an original 54
patients, returned for the 5 year follow-up which consisted of examining

changes in personal discomfort and social ineffectiveness.

11

In the area of discomfort, the authors found early improvement (6 months to
2 years) for anxiety and depressive symptoms, and substantial relief in all
symptoms by the end of the five-year period. Improvement in social
ineffectiveness showed slower gains. At six months 67% of the 30 patients
reported improvement, 30% were worse, and only one remained unchanged;
but at the five-year mark 97% (29 out of 30) had improved whereas only 1
remained unchanged.

In addition to these variables, the authors explored factors which
contributed to changes within the individuals at the five-year point. In terms of
discomfort, the authors found that the most improved at the end of the five-
years were younger (28.5 vs. 33.7 yrs), more likely to be sicker initially, and
showed a tendency (p<.20) to be Catholic or Lutheran as opposed to other
religious groups. The least improved were those who reported many somatic
symptoms. Patients most improved on social ineffectiveness showed greater
personal comfort, and seemed to gain greater spontaneity in their intimate
relations.

As a way of extending and adding to this information, additional follow-up
studies were conducted. Two of these focused on the same population: one at
ten-years (lmber et al., 1968), and the other at twenty-years (Liberman, 1978); a
third study examined a five-year follow-up of neurotic outpatients at the same
1 clinic (Liberman et al., 1972). These studies confirmed the original finding of the

five-year follow-up. In addition to this, lmber and his colleagues (1968) explored

 

12
to what patients attributed their change. They found that the most improved

patients associated their condition to either change in socioeconomic conditions
or acceptance of and adjustment to life circumstances, including their
symptoms Ge. learning to live with their pain). In contrast, the unimproved
individuals "were unable to specify the reasons for their condition.” (p. 80)

It would appear that this research made some important contributions to the
understanding of therapeutic changes and the posttherapy period. It
demonstrated the ability of patients to maintain and even improve in the
posttherapy period, and it began to identify factors that contribute to maintained
patient improvement. It marked early steps to finding factors that contribute to
the maintenance and growth of therapeutic gains.

Despite the wealth of information provided by this study, there are some
limitations to the interpretations that can be drawn from the results. In
conducting their study the investigators were interested in exploring the patients
life events subsequent to treatment. A client reentering treatment would be
noted in the study; but such action has a confounding effect on the variables
which originally contributed to client change. Researchers have shown the
influence of intervening life events during the posttherapy period to the
maintenance of therapeutic change (Fiske & Goodman, 1965; Paul, 1967). In
addition to this, individual client differences were not explored and the original

period of treatment was relatively short. Thus care must be taken when

13

interpreting these results in relation to changes obtained from the original
therapy.

McNair et al. (1964) also investigated the therapeutic effects subsequent to
termination. In a three-year follow-up of 81 male psychiatric outpatients these
researchers examined: 1. whether improvement over pre-therapy status is
maintained over the three years, 2. if patient status at follow-up is related to
months of treatment and 3. if patient status at follow-up is related to number of
therapy sessions. Similar to the findings of Stone et al. (1961), patients showed
significant changes from their pre-therapy status at the three-year point.
Subjects were less anxious, less hostile, less dependent and more self-
accepting at this point. The investigators also found a continued trend for
improvement from the point of the one-year follow-up to the three-year point.

In exploring factors that contributed to this outcome, McNair and his
colleagues (1964) report a significant relationship between the number of
treatments and client reports of favorable changes. On the other hand, number
of months of therapy did not relate to client change at follow-up. The authors
also found follow-up status significantly associated with initial status, measures
of social attainment, and measures of intelligence. In discussing the findings of
this report the authors state: " one bit of evidence supporting the proposition
that the gains were due to therapy is that most of the therapy and most of the
gains were achieved during the first year... it does suggest that therapy has an

impact" (p. 263).

14

Another approach to examining variables which correlate with long-term
outcome was taken by Gossett et al. (1977). Following the results of a literature
review on adolescents treated in psychiatric hospitals (Gossett et al., 1973),
these researchers investigated the patients’ pretreatment status (the severity of
psychopathology, onset of symptoms, the patients' energy level and physically
destructive behavior before hospitalization), type of treatment, and aftercare
(continuation of individual therapy). Their population consisted of 55 teenagers
who were treated at Timberlawn Psychiatric Hospital between July 1966 and
November 1968. Follow-ups were conducted twenty months to four years after
treatment was completed.

The results of this study show that ”better outcome was experienced by
patients who had more reactive onset, higher energy level, little physically
threatening behavior, less severe symptoms, completed treatment, and
psychotherapy following discharge" (p.1039). These results complement those
by McNair et al. (1965) in regards to the patients pretherapy adjustment; yet
contrast the findings of Stone et al. (1961) which showed that subjects who
were sicker initially were more improved.

Schramski et al. (1984) set out to advance the knowledge about factors
which are relative to and predictive of therapeutic change. They studied thirty
outpatient, psychotherapy clients for their change of status at termination and at
a six month follow-up. Their results demonstrated persistence of therapeutic

change for 77% of their subjects. In evaluating which factors contributed to this

15

change they found that socioeconomic status, age, initial neuroticism, and initial
extraversion were all significant. In their concluding comments these
researchers stressed how studies on these variables are important to helping
therapists provide better services.

Other client variables which have been linked to the maintenance of
therapeutic change are: age and change in marital status (Garfield, 1978), and
locus of control (as cited in Schramski et al. 1984).

On the whole it would appear as if studies in this area have identified some
important factors; but the reality of the situation shows conflicting results for
these factors. Among the factors that have yielded conflicting results are: the
number of therapy sessions in relation to the persistence of therapeutic gains
(McNair et al., 1964 vs. Fiske & Goodman, 1965; Bergin & Lambert, 1978), and
clients pretherapy status (Stone et al., 1961 vs. Gossett et al. 1977). These
inconsistent and inconclusive results show the need for more and better
research elucidating factors that contribute to the maintenance of therapeutic
change. A starting place for such studies can be found in the outcome
literature where various factors that contribute to therapeutic change are
discussed.

n r n f Client m t ms and There e ti tcom

The literature on psychotherapy outcome research reveals one area which

has received little attention yet has shown some promising results. This area

has been labeled congruence of perception, and it explores the relationship

16

between therapists’ accurately perceiving clients symptoms and therapeutic
outcome (Filak, & Abeles, 1984). This section will examine the literature on
congruence of perception in the clinical environment as a means of providing a
rationale for the present study.

While research on congruence in psychotherapy is relatively new and
unexplored, it roots in the "accurate empathy" studies of Truax & Carkhuff
(1967). Accurate empathy was described as the extent to which the therapist is
sensitive to the thoughts and feelings of the client, and has the ability to
communicate his understanding to the client in a language attuned to the client
(Mitchell, Bozarth, & Krauft, 1977). This research showed that empathy is not
related as much to client improvement as accurate empathy (T ruax, & Carkhuff,
1967). As such, it appears that it is important for the client to feel understood
and accepted for improvement to occur.

An earlier study, Cartwright and Lerner (1965), indirectly touched on the
importance of accurate empathy. As part of their study of 28 patients receiving
client-centered counseling, the investigators explored empathic understanding
of the client by the therapist. They found that at the end of therapy that the
therapists understood the self-image of improved subjects better than those
who were unimproved. Therapists also showed significant gains in empathy
scores for the beginning of therapy to termination. The investigators conclude
that these results "showed that there was a significant relation between

improvement and increased understanding." (p. 140) Taking these conclusions

17

under consideration, one can see that this study also demonstrates the
relationship between accurate empathy and psychotherapeutic outcome.

Cooley and Lajoy (1980), in an attempt to examine the therapeutic
relationship as perceived by clients and therapists, investigated the discrepancy
between client and therapist perceptions and correlated this with outcome. The
results of their study show that differences in the areas of understanding and
acceptance correlate more highly with outcome (r=.27, p<.05). These
complement earlier findings in which patient rated improvement was related to
therapist understanding and acceptance (Lorr, 1965). From these results the
Cooley and Lajoy (1980) conclude "when client and therapist perceptions of the
therapeutic relationship agree, this relationship is associated with greater client
improvement" (p.569). However they qualify these conclusions stating that the
nature of the relationship is difficult to identify.

Despite the successes of research in this area, reviews of research have
leveled criticism at these findings suggesting that results in this area are
inconclusive and complex. Luborsky, Chandler, Auerbach, & Cohen (1971), in
a review of 166 outcome studies, found that empathy is a significant factor
related to improvement. Nevertheless they conclude that the causal relationship
is more complex and is, as of yet, undetermined. Garfield and Bergin (1971)
found no relationship between accurate empathy and therapeutic outcome.
Finally, two recent reviews (Mitchell, Bozarth, & Krauft, 1977; Gurman, 1977)

have also questioned the efficacy of accurate empathy research stating the

18

construct may be more complex than originally conceptualized. They claim that
accurate empathy does play a role in improvement; yet assert that as of yet
research has not determined its role.

The studies that were conducted in the area of accurate empathy did provide
some interesting and promising results. The results show that the
understanding and acceptance were important to outcome. The one thing
they did not do is specify how these factors contributed to improvement. In
addition, these investigations laid the groundwork for research in the area of
congruence of perception which attempts to establish the relationship between
agreement on specific clinicians’ judgement and clients’ self-rating to positive
therapy outcome.

A study by Filak and Abeles (1984) was the first to examine the relationship
between congruence of perception on client symptoms and psychotherapy
outcome. They compared client self-ratings on the Symptom Checklist (SCL-
90R) to therapist ratings of clients on the therapist version of the Symptom
Checklist (SOL-90A) of 50 clients treated at the Michigan State University
Psychological Clinic. The results showed a significant relationship between
posttherapy congruence of symptom perception between therapist and patient
and positive outcome (r= .45, p<.001, r=.57, p<.001). Other results reported
demonstrated that pretherapy congruence was not a significant factor and that
“the congruence or accuracy factor was associated with the therapist-client

dyads that had better outcomes" (p. 853). The authors contend that these

19

findings suggest that patient-therapist understanding of client symptoms is
indeed an important factor related to outcome and that increased
correspondence takes place during the course of the treatment.

The study of therapist and patient congruence begins to suggest factors that
may play a role in the success of psychotherapy. Yet, as Filak and Abeles
(1984) noted in their concluding comments, “future research on accuracy or
congruence is needed” (p.854). One area of focus for research is the long-term
effects of this factor on the persistence and maintenance of client ratings of
therapeutic success. This is the area to be investigated in the present study.
The purpose of the present study is to explore the long-term client ratings of
therapeutic success as related to therapist/client congruence of perception. As
such, the present study will add to the growing research elucidating variables

which contribute to the maintenance and persistence of therapeutic gains.

HYPOTHESES

The research concerning the relationship of congruence of perception to
psychotherapy outcome suggests potential for providing information which
should be useful to fuller understanding of the therapeutic process. Questions
remain about the extent to which this factor helps the client. Does this factor
enable the client to function outside the therapy situation or is it important only
as long as therapy goes on? Does therapist-client congruence of perception
during therapy provide a foundation which has a lasting impact on the client?

In the physical disease domain it is said that patients who remain cancer
free for at least five years are considered cured. Psychological disorders are
often more difficult to evaluate than damages in tissues. Schramski et al. (1984)
assert that studies which look at the relationship between therapist behavior
and the persistence of change would be useful to the growing understanding of
the posttherapy period. The present investigation is designed to evaluate such
variables. In this study the concept of congruence of perception on client
symptomatology is studied as it relates to client ratings of therapeutic success
six months following termination. The purpose of this study is to extend the
findings of Filak and Abeles to the post therapy period and to add to the
growing data pool on therapeutic factors that contribute to the maintenance and
persistence of therapeutic change in the client. Specifically the hypotheses
examine congruence as it relates to persistence or deterioration effects six

months following termination of therapy.

20

21

Before stating the hypotheses of this study it is useful to define the specific
variables which are examined throughout this investigation:

1. Congruence of perception is defined as the degree of agreement
between the therapist and client through independent ratings based on the
Symptom Distress Checklist (SCL-90) ten dimensions (Filak and Abeles, 1984).

2. Successful therapeutic outcome is defined as high ratings on the client
post-therapy questionnaire (discussed in next section).

3. Deterioration effect is defined as a decline in the clients’ ratings in the
post-therapy questions from the post-therapy period to the time of the follow-up
assessment.

HYPOTHESES

Hypothesis 1: Client/therapist congruence of perception, measured at
termination of therapy, will be related to client ratings of therapeutic success six
months after therapy has been terminated.

This hypothesis tests whether congruence of perception on
symptomatology is enough to maintain client ratings of therapeutic success six-
months following therapy.

Hypothesis 2: For cases where therapy was rated as successful but there
was a lack of client-therapist congruence at termination there will be a
deterioration effect of client ratings of therapeutic success at the six-month

follow-up.

22

This hypothesis tests the premise that if both members of the therapeutic
process are not in agreement on client symptomatology at termination the
successful ratings of therapy seen at termination will not persist.

Hypothesis 3: In cases where there is congruence at termination but
nonpositive ratings of therapy there will be higher positive ratings at the follow-
up point.

This hypothesis tests the delayed effects of the psychotherapy. It is testing
whether congruence of perception will have a positive impact on client behavior
during the posttherapy period.

The procedures used to investigate the preceding hypotheses follow those
set forth by Filak and Abeles (1984). A comparison of the congruence of
perception measure is made separately with each of the posttherapy and follow-
up client questions. The responses to these questions range from "extremely
dissatisfied" to "extremely satisfied". The procedure used for testing hypotheses
is to obtain a correlation between the congruence of perception measure and
the responses to each of the client questions for all clients who have had at
least one therapy session and completed both the posttherapy forms and the
follow-up questionnaire.

The congruence of perception measure will be determined using the
following procedures. First, client responses on the SCL-90R will be converted
to the degree of symptomatology present for the nine dimensions and the

global severity index following instructions of Derogatis (1977). These scores

23

will then be converted to T-scores based on norms (Derogatis, 1977).
Clinicians’ scores will also be converted to T-scores, from their direct ratings,
based on norms from Filak and Abeles (1984).

Congruence of perception measures are derived from the client and
therapist ratings. It is a score that reflects the average of the sum of the
absolute value T-score points discrepancies between client and therapist across
the ten dimensions (Filak and Abeles, 1984). For instance, a score of 9.20
represents an average of 9.20 T-score points discrepancy between the clinician
and client ratings of the clients’ degree of symptomology across the ten
dimensions (Filak and Abeles, 1984). As Filak and Abeles (1984) stated, this
number represents the sum of the clinicians overratings and underratings of the
client level of symptom distress in relation to client self-report. These measures
will be acquired at the time of termination thus yielding congruence scores at

that time.

METHOD
Megs

The subjects of this study were all clients of the Michigan State University
Psychological Clinic. The clinic serves non-student members of the Michigan
State University community and the surrounding areas. The clinic is a research
and training facility and provides low cost services to adults, children and
families.

Therapists at the Psychological Clinic are graduate students in clinical
psychology. Their experience ranges from beginning practicum students to
clinicians with a great deal of clinical experience.

The client group in this study was composed of all clients who returned the
follow-up assessment packet sent to them six months following the termination
of therapy. A total of 35 adult clients returned this packet and had given their
consent to partake in this study.

Instruments
A. Symptom Checklist (SOL-90, Derogatis, 1977).

The present study used 2 forms of the SOL-90: the client form, and the
clinician form. Ratings taken at the time of termination were used for the
present study.

1. Client form--SCL-90R. The SCL-90R is a multidimensional self-report
measure in which clients indicate thoughts, feelings, and symptoms that they

have recently experienced. This measure requires the client to rate 90

24

25

statements of problems on a 5 point scale ranging from “not at all“ to
"extremely". The problems on this scale compose and load on nine symptom
dimensions: obsessive-compulsive, somatization, interpersonal sensitivity,
depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychotism.

2. Clinician Form-SCL-90A. This form incorporates the 9 symptom
dimensions. The clinicians were asked to directly rate the client’s level of
pathology on each dimension on a 6 point scale ranging from “none" to
"extreme".

The SCL-90 has been tested for reliability and validity. Reliability studies
have focused on test-retest and internal consistency measures (Derogatis,
1977). The SOL-90 was found to be reliable on both measures. Validity studies
were conducted on 1002 outpatients (Derogatis, & Cleary, 1977). A factor
analytic procedure known as "Procrustes procedure" was used to test the
construct validity of this measure. This measure was tested in terms of its
internal characteristics and its correlation with other measures of the same
construct. Results indicate that the SOL-90 ’s symptom constructs have validity.

8. Client Post-Therapy Questionnaire

The post therapy questionnaire is a 56 item self-report measure given to
clients at the time of termination. This questionnaire is designed to get at the
clients’ beliefs about the effectiveness of therapy. Of the fifty-six questions in

this instrument, the six questions which were deemed as the best

26

representatives of relevant process and outcome variables in the Filak and
Abeles (1984) study were selected for use in this study. These questions are:

1) How much have you benefitted from your therapy?

2) Everything considered, how satisfied are you with the results of your

psychotherapy experience?

3) What impression did you have of your therapist’s level of experience?

4) How much do you feel you have changed as a result of therapy?

5) On the whole, how well do you feel you are getting along now?

6) To what extent have your complaints or symptoms that brought you

to therapy changed as a result of treatment?

C. Client Follow-up Questionnaire

Six months following termination, clients were sent a follow-up assessment
packet. Included in this packet was the post-therapy questionnaire. The same
six questions used in the post-therapy packet were once again used to assess
the clients status at follow-up time.
Pr ur
Clients seeking psychotherapy at the Michigan State. University

Psychological Clinic were given the opportunity to participate in the evaluation
of its services. Clients who agreed to participate in the research were given
questionnaire packets to complete. The SOL-90 is among these questionnaires.

Clients then entered therapy with clinicians based on their availability.

27

At termination, clients and therapists were asked to complete post-
therapy packets. Included in these packets are the instruments to be used in
this study (previously discussed).

Six months following the termination of therapy clients were sent follow-
up questionnaire packet and asked to send it back once they complete it. The

data from these questionnaires was analyzed as discussed.

RESULTS
n r n P r ti n

Congruence of perception was defined as the degree of agreement
between the therapist and client through independent ratings based on the
Symptom Distress Checklist’s (SCL-90) ten dimensions. Operationally
congruence of perception involved taking the average of the sum of the
absolute value T-score points discrepancies between client and therapist
responses across the ten dimensions of the SCL-90. As such, this measure
represents the algebraic sum of the clinicians overratings and underratings of
the client level of symptom distress in relation to client self-report. For the
thirty-two cases used in the present study, the congruence of perception
measure yielded a mean of 8.52 (S._. = 2.77). In essence, this means that for
this group of clients there was a difference of 8.52 T-score points between
his/her ratings of symptom distress and clinician ratings of symptom distress
across the nine dimensions and global pathology index of the SCL-90.

In addition to the above calculation, a Pearson Product Moment
Correlation (PPMC) was performed in order to analyze the relationship between
clinician and client responses for each individual dimension of the SCL-90.
Results of correlational analyses indicated statistically significant relationships
between client and clinician ratings on nine of the ten dimensions, with

psychoticism yielding the only nonsignificant results. Table 1 presents the

28

29

correlation co-efficient and level of significance for each of the dimensions of the
SOL-90.
YP E l

The first hypothesis proposed that:

Client/therapist congruence of perception, measured at the
termination of therapy, will be related to client ratings of therapeutic
success six months six months after therapy has been terminated.
Congruence of perception was correlated with each of the six measures

of client ratings of therapeutic success using a Pearson Product Moment
Correlation Coefficient (PPMC). The results of this analysis yielded no
statistically significant correlations between congruence of perception and the
six measures of therapeutic success. Table 2 presents the correlation
coefficient and level of significance for the correlation of congruence of
perception to each of the measures of therapeutic success.

In order to gain some insight into the characteristics of the sample and
the potential for sample bias, a second analysis was conducted on the group of
subjects who were determined to have high congruence of perception and who
rated therapy as successful at the time of termination. This analysis used a t-
test to examine the differences in the means of client from the time of
termination to the six month follow-up for each of the six measures chosen in
this study. Twenty cases met the qualifications to be analyzed
in this category. The results of this analysis indicate statistically significant t-

scores in client ratings of: 1) the amount of benefit from therapy, 2) level of

TABLE 1
CORRELATIONS BETWEEN CLIENT AND THERAPIST SOL-90 RESPONSES.

 

SOL-90 DIMENSION N r. LEV OF SIG.
SOMATIZATION 32 .433 .01
OBSESSIVE-COMPULSIVE 32 .417 .01
INTERPERSONAL SENSITIVITY 32 .598 .001
DEPRESSION 32 .478 .01
ANXIETY 32 .669 .001
HOSTILITY 32 .413 .01
PHOBIC ANXIETY 32 .301 .05
PARANOID IDEATION 32 .430 .01
PSYCHOTICISM 32 .087 (NS)
GLOBAL PATHOLOGY 32 .448 .01
TABLE 2

CORRELATIONS BETWEEN CONGRUENCE OF PERCEPTION AND
FOLLOW-UP RATINGS OF THERAPEUTIC SUCCESS BY THE CLIENT.

FOLLOW-UP QUESTIONS N r. LEV OF SIG.

HOW MUCH HAVE YOU BENEFITTED 25 .192 .18 (NS)
FROM YOUR THERAPY?

EVERYTHING CONSIDERED, HOW 25 -.175 .20 (NS)
SATISFIED ARE YOU WITH THE

RESULTS OF YOUR PSYCHOTHERAPY

EXPERIENCE?

WHAT IMPRESSION DID YOU HAVE 25 -.215 .15 (NS)
OF YOUR THERAPIST’S EXPERIENCE
LEVEL?

HOW MUCH Do YOU FEEL YOU 25 .181 .19 (NS)
HAVE CHANGED AS A RESULT OF .
PSYCHOTHERAPY?

ON THE WHOLE, HOW WELL DO 25 -.013 .48 (NS)
YOU FEEL YOU ARE GETTING ALONG NOW?

TO WHAT EXTENT HAVE YOUR 24 .144 .25 (NS)
COMPLAINTS OR SYMPTOMS THAT

BROUGHT YOU TO THERAPY CHANGED

AS A RESULT OF TREATMENT?

30

31
satisfaction with therapy, 3) impression of therapist’s level of experience, and 4)
on how well the client is getting along. Table 3 presents the t-scores and levels
of significance for this group.
H P THE I II

The second hypothesis proposed that:

For cases where therapy was rated as successful and where there

was a lack of client/therapist congruence at termination there will be a

deterioration effect of client ratings of therapeutic success at the six-

month follow-up.

Of the thirty-two cases chosen for this study, seven cases met the
qualifications for inclusion in this analysis. A t-test was used to analyze the
mean differences between client ratings of therapeutic success at the time of
termination and follow-up period. The results of the t-test indicate that there is
no statistically significant difference in client ratings for any of the six questions

chosen as measures of therapeutic success.

HYPOTHESIS Ill
The third hypothesis proposed that:

In cases where there is client/therapist congruence of perception
and nonpositive ratings of therapeutic success at termination, there will
be higher positive ratings of therapeutic success at the follow-up point.
Four cases met the qualifications for this analysis. Statistical t-tests were

used to analyze the mean differences for each of the six questions chosen to

measure level of therapeutic success. This analysis produced statistically

significant results for client ratings of level of satisfaction with their therapy

CLIENT
QUESIIQN§

HOW MUCH HAVE
YOU BENEFITTED
FROM YOUR THERAPY?

EVERYTHING
CONSIDERED,HOW
SATISFIED ARE

YOU WITH THE RESULTS
OF YOUR PSYCHO-
THERAPY EXPERIENCE?

WHAT IMPRESSIONS
DID YOU HAVE OF

YOUR THERAPIST'S
EXPERIENCE LEVEL?

HOW MUCH DO YOU
FEEL YOU HAVE
CHANGED AS A
RESULT OF PSYCHO-
THERAPY?

ON THE WHOLE,

HOW WELL DO YOU
FEEL YOU ARE
GETTING ALONG NOW?

TO WHAT EXTENT
HAVE YOUR SYMP-
TOMS OR COMPLAINTS
THAT BROUGHT YOU
TO THERAPY CHANGED
AS A RESULT OF
TREATMENT?

TABLE 3
RESULTS OF T-TESTS ON CLIENT RATINGS OF
THERAPEUTIC SUCCESS.

POST-THERAPY FOLLOW-UP
MEAN S.D.

2.25

2.25

1.62

1.05

1.35

.639

.696

.937

32

2.90

MEAN s.D.

1.62

1.60

.550

.671

O 737

DF

19

19

19

19

19

18

-2.16

2.94

4.00

.00

-.75

.008

.001

(NS)

.025

.465
(NS)

33

experience (t= 5.20, p>.05, df=3). This result indicates a decline in the
clients’level of satisfaction over the post treatment period. No statistically

significant results were found for the other measures of therapeutic satisfaction.

 

DISCUSSION
The purpose of this study was to explore and elucidate a variable which
might contribute to the persistence and maintenance of client ratings of
therapeutic success during the post-therapy period. Specifically, this study
examined the relationship between congruence of perception of client symptom
distress levels, rated by clients and therapists, and long-term client ratings of

therapeutic success. Congruence of perception was defined by the degree of

-i-

agreement between client and therapist ratings of the level of client symptom
distress. Client and therapist ratings of the nine dimensions and the global
distress index of the Symptom Checklist (SCL-90) served as a measure of client
symptom distress levels. Overall, analyses of the data failed to Show a
Statistically significant relationship between client/therapist congruence of
perception and ratings of therapeutic success at a six month follow-up period.

In this section the writer discusses the results of data analyses from both
theoretical and methodological perspectives. First, this section examines the
results in light of the congruence of perception theory; then, the methodological
limits and their implications are discussed. Finally, this section attempts to tie
the two together and take a general look at future research in this area.
Cpngruenpe pf perception and Ratings of Therapeutic Success

The first hypothesis examined the relationship between client/therapist
congruence of perception and client ratings of therapeutic success six months

after termination. This hypothesis was designed to investigate whether

34

35

client/therapist congruence of perception on client symptomology might be a
variable which contributes to the persistence and maintenance of client ratings
of therapeutic success during the post therapy period.

The predicted relationship between congruence of perception at the time
of termination and client ratings of therapeutic success and a six month follow-
up point was not supported by the data. The statistical analysis did not
establish a significant correlation between client/therapist congruence and client
ratings on the amount of benefit from therapy, satisfaction with therapy,
impression of therapist’s experience level, change resulting from therapy,
current functioning level, and symptom relief.

The lack of a Significant correlation suggests that congruence of
perception at the time of termination is not a factor which could be related to
client ratings of therapeutic success at a follow-up point in time. While it has
been determined that congruence of perception is a variable associated with
positive therapeutic outcomes (Filak and Abeles, 1984), these results indicate
that the effects of this variable do not persist in time beyond the course of
therapy. It may be that once symptom relief has been attained, client
assessments of assessments of satisfaction with therapy, change resulting from
therapy, benefits from therapy, and change in symptom levels begin to fade and
are no longer attributed to therapy in a statistically significant manner.

Another explanation might be found in the numerous and undefined

variables that occur, and can not be controlled, in the post-therapy period. The

I

36

client does not live in a vacuum after therapy. Events occur that could have
great impact on the client and his/her ability to cope (e.g. marriage or divorce,
changes in occupation, death of family member, etc.). The impact that these
events could have on an individual varies; as such, one would expect to see the
impact of these events reflected in varied responses to follow-up ratings of
therapeutic success. Theoretically, the impact of these varied responses could
account for the lack of relationship found between congruence of perception
and follow-up ratings of therapeutic success.

Closely related to the other explanations, another factor which might
have contributed to the lack of significant results might be found in a decrease
of a strong positive view towards therapy as time passes. A client having a
positive therapeutic experience, and experiencing relief from symptoms, may be
influenced towards reporting either highly positive outcomes at termination or a
possible halo effect. Even without the occurrence of significant life events, or
feelings of abandonment, there is the potential that the client will reevaluate
his/her therapeutic experience in a different vein during the post-therapy period.
Most likely, the kind of reevaluation would vary from individual to individual and
the effects of this reevaluation could have been reflected in the results obtained
in this study.

From a theoretical perspective, the writer prefers an explanation based
on the combination of the factors discussed. Psychotherapy deals with a wide

variety of events and individuals with a wide variety of experiences. During the

 

37

post-therapy period one can expect that the experiences of the individuals will
be as different and varied as the individuals themselves. In trying to account for
the lack of significance between congruence of perception and client ratings, it
would be shortsighted not to take into consideration how these various factors
work individually and in concert to effect the client’s experience, and thus the
client’s ratings of therapy.

In addition to the correlational analysis, t-tests were conducted in order
to gain some insight into the characteristics of subjects who were measured
high in congruence and rated therapy as successful. These analyses,
conducted on client ratings of therapeutic success from termination to the
follow-up point, yielded significant results suggesting that clients reported that
they were functioning better at the follow-up point, were less satisfied with their
psychotherapy experience, perceived less benefit from therapy than previously
stated by them, and saw their therapists as less skilled then at termination.

On the one hand, these results indicate that clients experienced a
significant increase in their level of functioning during the post-therapy period;
on the other, the results indicate that the clients did not retrospectively attribute
this increase in functioning to their therapeutic experience. Due to the lack of
correlation between congruence of perception and follow-up ratings, these
results can not be interpreted with regard to congruence of perception theory.
Yet, the value of these results is found in the information they provide as to the

characteristics of the population of this sample and potential sample biases.

38
Hypothesis II and hypothesis III were based on the probability of the

establishment of a positive relationship between congruence of perception and
follow-up results. The lack of this relationship precludes interpretation of data
with regard to congruence of perception theory. As with the t-tests conducted
on the group in hypothesis I, the analysis of results from these two hypotheses
yielded some useful information when it comes to exploring the characteristics
of the subjects and in exploring the methodological problems found in this
study.

The second hypothesis predicted a deterioration in client ratings of
therapeutic success from termination to follow-up time due to a lack of
congruence of perception at the time of termination. The analysis of the data
failed to support this hypothesis. As analysis from hypothesis l demonstrated, a
significant relationship between congruence of perception and follow-up ratings
failed to be established. With this severe limitation kept in mind, t-tests that
were conducted also failed to reveal any significant changes in client ratings of
therapeutic success from termination to the six month follow-up point.

The third hypothesis sought to explore the possibility of delayed effects
of therapy. Contrary to the predicted hypothesis t—tests conducted for
hypotheses Ill yielded a significant decline in the level of client satisfaction with
therapy, and statistically nonsignificant trends suggesting a decline in clients’
rating of therapy related change (t = - 1.73, p < .20, df = 3) and a decline in

rating their therapists’ experience level (t = 1.73, p < .20, df = 3).

39

Statistically nonsignificant positive trends for better functioning (t = 1.41, p <
.25, df = 3) and improvement on symptom change (t = 1.44, p < .25, df = 3)
were also discovered.

Similar to the first hypothesis, these results suggest that clients
experienced greater symptom relief and were functioning at a higher level at the
time of the six month follow-up than at termination. Yet, the results also
suggest that the clients were less likely to attribute these changes to factors
related to their therapeutic experience, as reflected in the significantly lower
client rating of satisfaction with therapy and the trends in which clients rated
less change resulting from therapy and their therapists as less skilled then
previously rated. Once again these trends can not be interpreted in relation to
congruence of perception theory, but should be viewed as useful only in light of
the information they provide about the characteristics of this sample.

One could speculate that this sample of clients might have attributed their
present status to factors other than their prior therapy. Perhaps this was a
function of client initiated change. Clients previously in therapy might have
viewed post-therapy changes as improvements that they themselves have
initiated. It is certainly possible that successful therapy potentiates self initiated
changes. Naturally, this is clearly speculative and warrants further empirical
examination.

Whether or not clients attribute their change to therapy is, of course, of

interest to therapists. However, from an epidemiological point of View it is

40

positive to see that client evaluations continue to indicate that they are able to
maintain progress regardless to what they attribute that progress. After all,
many therapeutic approaches are devised to encourage the client to do most of
the work. Therefore, it would seem reasonable that the clients credit
themselves rather than their therapists for enduring change.

Even though clients did not attribute change to therapy, it is important to
note that they did maintain reasonable therapeutic gains. The fact that clients
were able to maintain therapeutic gains corresponds with, and lends further
support to, Nicholson and Berman’s (1983) meta-analytic study which found
that psychotherapy was effective and that therapeutic gains were durable during
the posttherapy period. As with Nicholson and Berman’s (1983) study, the
current study demonstrates that gains made in therapy are persistent and that
these changes continue to develop during the posttherapy period.

Mth I i lExlantin

The lack of significance between congruence of perception and follow-up
ratings of therapeutic success proved to be contrary to congruence of
perception theory. Yet, in order to have a more complete understanding of the
results and their possible meaning, the methodological difficulties found in this
study need to be considered.

One methodological explanation for the lack of significant results in the
current study can be found in the sample size. Due to any number of reasons

(lack of Incentive to return questionnaires, change of address, client repression,

41

etc.) there was a high attrition rate from the general data pool, which resulted in
a relatively small sample (n=32). The small sample size may have played a role
in affecting the sensitivity of the experiment in that even a small deviation in the
data might have had a profound impact on the findings and contributed to the
potential for a type II error.

Similarly, the high attrition rate and resulting small sample size might also
reflect a return bias in a negative direction. It might be that clients who returned
the materials reflect the extremes in both directions; in the analysis of the data,
these responses would cancel each other out resulting in yielding nonsignificant
results. Such a response bias would have a great impact on the sensitivity of
the experiment and possibly contribute to a type II error.

An examination of the raw data in this study revealed considerable
scatter of scores. This examination suggests that is unlikely that sample size
played a role in influencing in the obtained results and that the hypotheses were
not supported by the data.

A second area in which small sample size might have contributed to the
results is in the analyses of the t-tests. The small number of subjects,
especially in hypotheses II and III, could have been a contributing factor to the
lack of significance found between p'ost-therapy ratings and follow-up ratings. A
larger sample might have had the potential to influence the data in such a way
as to bring out significant results in areas which are currently presented as

trends.

42

Another methodological factor that should be considered when
interpreting the results is the effects of subject attrition. The population from
which this study drew its subjects consisted of over 100 participants over a
three year period. Of this population only 32 subjects completed their
commitment through the follow-up period. With this type of attrition serious
consideration should be given to the characteristics of the subjects and the
possibility for sample bias.

An examination of the raw data indicated a significant sample bias found
within this population. The raw data indicated that subjects who returned the
six-month follow-up questionnaires demonstrated characteristics biased towards
high levels of congruence at the time of termination (24 out of 32 subjects). In
addition to this, examination of data also revealed a bias towards high ratings of
therapeutic success at the time of termination (27 out of 32 subjects). While it
is difficult to speculate as to the influence of these biases on the results and
their meaning in relation to congruence of perception theory, the fact that these
biases occurred limit the ability to interpret these results as well as to generalize
them to the population as a whole.

In general, the methodological difficulties in the present study play a
significant role in be able to understand and interpret the inability of the data to
support the results. Attrition, sample size, and sample bias might have had a
tremendous impact on the data and can confound the results of analysis. While

it is difficult to separate the amount of influence each of these methodological

43

factors may have had in this study, the writer believes that combination of these
factors had a significant impact on the data and these problems pose serious
impediment in interpreting and understanding the results of this study.
n l i n

The current study took the ambitious approach of entering two areas that
are lacking a firm research foundation and have received criticism for yielding
conflicting results. These areas are: congruence of perception as a contributing
variable in therapy outcome and long-term follow-up research. This study was
designed to explore the relationship between congruence of perception and the
long-term maintenance and persistence of client ratings of therapeutic success.

The results of data analysis failed to establish a relationship between
congruence of perception and long-term follow-up ratings. Yet, it is difficult to
provide a meaningful interpretation of these results of this study from a
theoretical perspective due to the confounding nature of methodological
problems found in this study. It may be that these methodological difficulties
did not play a role and that the role of congruence of perception was confined
to the more immediate therapeutic environment and therapy outcomes. On the
other hand, it may be that congruence of perception did indeed play some sort
of role in the post-therapy period and this relationship has been obstructed by
the design flaws in the study. The writer prefers to use great caution when
interpreting these findings; the Significance and extent of the methodological

difficulties seem to preclude any discounting of congruence of perception

44

theory at this time. Future research in the post-therapy period should be
carefully planned controlling for as many of the methodological difficulties as
practically possible.

The failure to establish a significant relationship between congruence of
perception and long-term ratings of therapeutic success also raised some
interesting questions concerning client changes in the post-therapy period.
What is the relationship between psychotherapy and client changes in the post-
therapy period? To what do clients attribute changes that might occur in the
post-therapy period? To what extent does client initiated change account for
changes in client level of functioning during the post-therapy period? It might
be useful for future research to explore these areas. Empirical investigations
examining both changes that occur in client status during the post-therapy
period and client attributions of these changes would not only elucidate
variables that contribute to post-therapy change; but also provide the
researcher with an explanation of what may account for these changes. As
such, research which addresses these issues could provide clinical researchers
with a more comprehensible understanding of change occurring in the post-

therapy period.

LIST OF REFERENCES

REFERENCES

Bergin, A. E., & Lambert, M. J. (1978). The evaluation of therapeutic outcomes.

In A.E. Bergin & S.L. Garfield (Eds), Hangpppk pf Payphptherapy
and Bahavipr Changa. (2nd ed.). New York: John Wiley.

Cartwright, R. D., & Lerner, B. (1963). Empathy, need to change, and

improvement with psychotherapy. Journal pf Qpnaulting Paypholpgy,
21(2), 138-144.

Cohen, L. H. (1979). Clinical psychologists’judgments of the scientific merit and
clinical relevance of psychotherapy outcome research. Jpprnal pf

anaulting and Qlinigal Paypholggy, _4_7_ (2), 421-423.

Cooley, E. J., & Lajoy, R. (1980). Therapeutic relationship and improvement as

perceived by clients and therapists. Joprnal pf Clinical Payphplpgy,
36(2), 562-570.

Cowen, E. L., & Combs, A. W. (1950). Follow-up study of 32 cases treated by

nondirective psychotherapy. Jparnal pf Apnprmal ang Sapial Payphplpgy,
55, 232-257.

Derogatis, LR. (1977). Th L- M n all: rin Administration n

Prpcedpraa Far tha SOL-90. Baltimore, MD: John Hopkins University
School Of Medicine, Clinical Psychometrics Unit.

Derogatis, LR, Cleary, PA (1977). Confirmation of the dimensional
structure of the SOL-90: A study in construct validation. Jgurnal pf

Clinical Payphplggy. 33(4), 981-989.

Eysneck, H.J. (1952). The effects of psychotherapy: An evaluation. qurnal pf
Qpnsalting Payphplpgy, 16, 319-324.

Filak, J. & Abeles, N. (1984). Posttherapy congruence on client symptoms and
therapy outcome. Prof i n lP h I Re arch n Pr
15(6), 846-855.

Fiske, D. W., & Goodman, G. (1965). The posttherapy period. Journal of
Apnprmal Payphplogy,10(3), 169-179.

Fiske, D. W., Hunt, H. F., Luborsky, L., Orne, M. T., Parloff, M. B., Reiser, M.
F., & Tuma, A. H. (1977). Planning of research on effectiveness of ~

psychotherapy. Arphivaa pf Genaral Payphiatry, 22, 22-32.

45

45

Garfield, S. L, (1978). Research on client variables in psychotherapy. In A.E.
Bergin & S.L. Garfield (Eds.) Hand k fP h th r n B havior
Qhanga. (2nd Ed.) New York: John WIley.

Garfield, S. L, & Bergin, A. E. (1971). Therapeutic conditions and outcome.
Jparnal pf Apnprmal Payphplpgy, 11, 108-114.

Gossett, J. T., Lewis, S.B., Lewis, J. M., & Phillips, V. A. (1973). Follow-up of
adolescents treated in a psychiatric hospital: 1. A review of studies.

Amaripan Jparnal pf Qrthppayphiatg, 453(4), 602—610.

Gossett, J. T., Barnhart, D., Lewis, J. M., & Phillips, V. A. (1977). Follow-up of
adolescents treated in a psychiatric hospital: Predictors of outcome.

Arphivaa pf Ganeral Payphiatg, 35, 1037-1042.

Gottman, J. M., & Markman, H. J. (1978). Experimental designs in
psychotherapy research. In A.E. Bergin & S.L Garfield (Eds),
H n k fP h th r n B havi r h n (2nd ed.), New York:
John Wiley.

Gurman, A. S. (1977). The patient’s perception of the therapeutic relationship.
In Alan S. Gurman & Andrew M. Razin (Eds), Effegiva Payphptharapy: A
Handpopk pf Raaaarph. Oxford: Pergamon Press.

lmber, S. D., Nash, E. H., Hoehn-Saric, R., Stone, A. R., & Frank, J. D. (1968).
A ten-year follow-up study of treated psychiatric outpatients. In Stanely
Lesse M.D., Med.ScD. An Eval ti n Th Re
Payphptharapies. Illinois: Charles Thomas.

Kolotkin, R. L, & Johnson, M. (1983). Crisis intervention and measurement
of outcome. In Michael J. Lambert, Edwin R. Christensen, 8. Steven S.

DeJuilo (Eds.) The Assessment Of Psyphotharapy Outcome. New York:
John Wiley and Sons.

Leary, T. (1957). Interperaonal Diagnosia of Parapnality. New York: Ronald

PTBSS.

Liberman, B. L., Frank, J. D., Hoehn-Saric, R., Stone, A. R., lmber, S. D., &
Pande, S. K. (1972). Patterns of change in treated neurotic patients: A
five-year follow-up investigation of the systematic preparation of patients

for psychotherapy. rnl f n ltin an lini IP hl 38(1),
36-41.

47

Liberman, B. L. (1978). The maintenance and persistence of change: Long-
Term follow-up investigations of psychotherapy. In Jerome D. Frank,
Ruldolf Hoehn-Saric, Stanely D. lmber, Bernard L Liberman, & Anthony

R. Stone (Eds.). Effaptiva Ingradienta Of §ucpas§ful Payphptharapy.

New York: Brunner & Mazel Publishers.

Lorr, M. (1965). Client perceptions of therapists: A study of the therapeutic
relation. Jpamal pf Qpnaalting Paypholpgy, _2_9(2), 146-149.

Luborsky, L., Chandler, M., Auerbach, A. H., Cohen, J., & Bachrach, H. M.
(1971). Factors influencing the outcome of psychotherapy: A review of

quantitative research. Payphplpgipal Ballatin, 15(3), 145-185.

Mash, E. J., & Terdal, L G. (1980). Follow-up assessment in behavior therapy.

In Paul Karoly, & John J. Steffen (EdS.) Imprpving Lang-Iarm Effaga pf
Payphptharapy. New York: Gardner Press Inc.

Mitchell, K., Bozarth, J.D., & Krauft, C. C. (1977). A reappraisal of the
therapeutic effectiveness of accurate empathy, nonpossessive warmth,
and genuiness. In Alan S. Gurman, & Andrew M. Razin (Eds.). Effagive

Psyphgtharapy: A Handpppk of Rasaarph. Oxford: Pergamon Press.

May, P. R., Tuma, A. H., & Kraude, W. (1965). Community follow-up of
treatment of schizophrenia-issues and problems. Amaripan Jpprnal pf

Qrthgpayphiatg, 33, 754-763.

May, P. R., Tuma, A. H., & Dixon, W. (1976). Schizophrenia-a follow-up study
of results of treatment. Archives of Ganeral Payphiatgy, 33, 74-478.

McNair, D. M., Lorr, M., Young, H. H., Roth, I., & Boyd, R. W. (1964). A three-

year follow-up of psychotherapy patients. Journal pf Clinipal Paychplpgy,
fl, 258-264.

Nicholson, R. A., & Berman, J. S. (1983). Is follow-up necessary in evaluating
psychotherapy? Payphplogical Bulletin, 93(2), 261-278.

Paul, G. L (1967). Insight versus desensitization in psychotherapy two years
after termination. Jparnal pf Qonaplting Psyphplggy, 3_1(4), 333-348.

Raimy, V. C. (1952), Clinical methods: Psychotherapy. Annpal Raviaw of
Payphplpgy, 3, 321-350.

 

43
Sargent, H. D., (1960). Methodological problems of follow-up studies in

psychotherapy research. Amaripan Jamaal pf Qrthppayphiatry, 33, 495-
506.

Schramski, T. G., Beutler, L F., Lauver, P. J., Arizmedi, T. A., & Shanfield, S.
B. (1984). Factors that contribute to posttherapy persistence of

therapeutic change. Jpamal of Clinical Payphplpgy, 53(1), 78-85.

Smith, M.L, Glass, G.V., & Miller, Ti. (1980). Tha Banafita pf Payphptharapy.
Baltimore: John Hopkins University Press.

Stone, A. R., Frank, J. F., Nash, E. H., & lmber, S. D. (1961). An intensive five-
year study of treated psychiatric outpatients. qurnal pf Narvgaa and
Mantal Diaaaaa, 133, 410-422.

Tobias, R. J. (1987). Predictors of return of follow-up data by former
psychotherapy clients. Unpublished master’s thesis, Michigan State
University, East Lansing, Mi.

Truax, C.B., & Carkhuff, R. R. (1967). prarda Effagiva qunaaling ang
W Chicago: Aldine

"Illllllllllll’llllllllll“