 

Tut-{.23

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled

RELATIONSHIP BETWEEN PREMATURE TERMINATION AND
CLIENT PERSONALITY VARIABLES

presented by

KENNETH N. MURRAY

has been accepted towards fulfillment
of the requirements for

M.A.

degree in

 

 

Department of Psychology

AMA/A

Major professor

Date ~9qu 3 2002

 

0-7639 MSUis an Afﬁrmative Action/Equal Opportunity Institution

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DaleDue.p65-p.15

RELATIONSHIP BETWEEN PREMATURE TERMINATION AND
CLIENT PERSONALITY VARIABLES

By

Kenneth N. Murray

A THESIS
Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of
MASTERS OF ARTS
Department of Psychology

2002

ABSTRACT

RELATIONSHIP BETWEEN PREMATURE TERMINATION AND
CLIENT PERSONALITY VARIABLES

By

Kenneth N. Murray

It is common for Psychotherapists to experience a rate of client attrition that
exceeds 50%. The negative implications for those individuals who have terminated
prematurely include their future psychological well being, increased need for medical
services, and lower economic potential. This study examines the relationship of the
interpersonal constructs of dominance and submission with premature termination.
Demographic variables and psychiatric symptomatology are also examined. Clients from
a Psychological training clinic were divided into an Early Dropout group (aﬁer l to 3
sessions; N = 35), a Late Dropout group (after 8-12 sessions; N = 35), and a Persister
group (remaining in therapy at least 18 sessions; N = 45). A signiﬁcant curvilinear
interaction between level of dominance and gender was discovered (F(2, 107) = 4.04,
p<.05). Men who terminated early on in therapy revealed higher levels of dominance
than men who terminated later in therapy and men who persisted. Women displayed a
curvilinear pattern of termination in which those women who terminated therapy at an
early stage as well as those women who persisted in therapy displayed lower levels of
dominance than those who terminated after 8 to 12 sessions. Differences in gender

socialization are discussed as a possible explanation for these results.

ACKNOWLEDGEMENTS

I would like to express my gratitude to all of the members of my committee
whose encouragement, expertise, and patience made this thesis possible. Dr. Norman
Abeles’ enthusiasm, support, and expertise in psychotherapy research was the
cornerstone upon which this thesis was complete. He generously gave of his time and
extended a patient hand whenever I found myself ﬂoundering. His guidance and the
respect he affords his students is an inspiration. Dr. Bertram Karon provided
immeasurable support and a critical eye that proved invaluable to the completion of this
work. Dr. Alex VonEye brought to bear a statistical acumen administered with a gentle
touch that kept me on track when all those numbers threatened to overwhelm. I cannot
imagine how anyone could have made this project more enjoyable to work on than these

three people have.

iii

TABLE OF CONTENTS

Page
LIST OF TABLES ....................................................................... v
LIST OF FIGURES ..................................................................... vi
INTRODUCTION ........................................................................ 1
Consequences of Attrition ...................................................... 3
Factors Involved in Premature Termination .................................. 6
The Interpersonal Circumplex .................................................. 10
Studies Involving Leary’s Model .............................................. 14
Summary .......................................................................... 15
HYPOTHESES ........................................................................... 16
METHODOLOGY ....................................................................... 19
Participants ....................................................................... l9
Therapists ......................................................................... 20
Procedures ........................................................................ 20
Measures .......................................................................... 21
Interpersonal Checklist ................................................ 21
The Symptom Checklist 90 Revised ................................. 22
Demographics ........................................................... 25
RESULTS ................................................................................. 25
DISCUSSION ............................................................................. 44
Further Research ................................................................. 50
APPENDIX A ............................................................................. 53
REFERENCES ........................................................................... 71

iv

LIST OF TABLES

Table Page
1. Means and Standard Deviations for Termination Point X Gender

Interaction with Respect to Level of Dominance .............................. 27
2. Means and standard Deviation of Octant and Sixteenth Levels

By Termination Point .............................................................. 29
3. Means and Standard Deviations of Years of Education by Termination

Point .................................................................................. 36

4. Varimax item Loadings on Factor 1 of the SCL-9 90—R as Obtained in
Current Study ....................................................................... 39

5. Varimax item Loadings on Factor 2 of the SCL—90~R as Obtained in
Current Study ....................................................................... 40

6. Varimax item Loadings on Factor 3 of the SCL-90-R as Obtained in
Current Study ....................................................................... 41

7. Varimax item Loadings on Factor 4 of the SCL-90-R as Obtained in
Current Study ....................................................................... 42

8. Varimax item Loadings on Factor 5 of the SCL—90-R as Obtained in
Current Study ....................................................................... 42

9. Varimax item Loadings on Factor 6 of the SCL-90-R as Obtained in
Current Study ....................................................................... 43

10. Items Loading less than .40 in any of the Factors of the SCL-90-R as
Obtained ‘in Current Study ........................................................ 43

LIST OF FIGURES

Figure Page
1. Axes of Interpersonal Domains (Laforge, et al, 1954) ..................... ll
2 The Interpersonal Circumplex (Leary, 195 7) ............. . .................. 12
3. Mean for Termination Point X Genderlnteraction with Respect to

Level of Dominance .............................................................. 28
4. Mean of Autocratic/Managerial Levels By Termination Point ............ 31
5. Mean of Self-Eﬂ‘acing/Masochistic level by Termination Point .......... 31
6. Mean of Managerial Level by Termination Point ........................... 33
7. Mean of Autocratic Level by Termination Point ............................ 33
8. Mean of Self-Effacing level by Termination Point .......................... 34
9. Mean of Masochistic Levels by Termination Point ......................... 35
10. Mean of Years of Education by Termination Point ......................... 37

vi

INTRODUCTION

Premature termination, dropout, and client attrition are all phrases used to identify
clients who terminate psychotherapy prior to the therapist’s expectations of treatment
duration necessary for a therapeutic gain. It is common for psychotherapists in
independent practice as well as outpatient mental health facilities to experience a rate of
premature termination of services, without prior therapist knowledge, that exceeds 50%.
Pekan'k (1983) found that 30% - 60% of outpatient psychotherapy clients terminate
prematurely. Twenty percent of clients seen in Community mental health settings and
forty percent of private practice clients terminate by the second session (Pekarik &
Wierzbicki, 1986), and 20% to 25% of all people who show up for an intake session do
not show up for any follow-up services ﬂipperson, Bushway, & Warman, 1983). This
premature termination has a devastating effects upon an individual’s ﬁrture psychological
well being, need for medical services, and economic potential (Pekarik, 1992). This is
particularly disturbing considering only 25% of the people in the US. with a diagnosable
mental disorder seek treatment of any kind (Regier, Narrow, Rae, Manderscheid, Locke,
& Goodwin, 1993). Greater understanding of the root causes of these enormous attrition
rates is necessary in order to provide therapists with information from which more
effective therapeutic strategies may be fashioned.

Until very recently, most studies on psychotherapy dropout have focused on the
easily obtainable demographic characteristics of clients and therapists. Among the
variables that have been consistently found to be linked to attrition are race, education
and socio-economic status (SES). In a 1993 article, Wierzbicki and Pekarik called for a

more complex analysis of the variables involved in premature termination. This call has

been heeded by a number of studies (Beckham, 1994; Dahlsgaard, Beck, & Brown, 1998;
Guze & Robins, 1970; Tryon, 1990; Tryon I985; Frayn, 1992; Levinson, McMurray,
Podell, & Weiner, I 978; Hilsenroth, et al., 1995; Piper, Joyce, Azim, & Rosie, 1994;
Chisholm, Crowther, & Ben-Porath, 1997; Vaughn & Nowicki, 1999; Lorr, 1991). While
these studies vary greatly with regard to treatment setting and the measures employed, in
total they point towards the possibility of an underlying personality style present within
the client that may be implicated in early termination from psychotherapy. They also
indicate that the level of awareness of others personality styles as well as the degree of
psychological impairment are also potential predictors of premature termination. While
these studies do suggest that an underlying personality style might be one of the causative
factors involved in premature termination this idea has not been thoroughly explored
within the framework of a cohesive theory of personality. In furtherance of the work put
forward in these studies a comprehensive personality theory that encompasses both
pathological and healthy aspects of human beings should be utilized to explore the
formation of the therapeutic alliance.

Harry Stack Sullivan’s (1953) Interpersonal Theory of Personality as
operationalized in Timothy Leary’s (1957) Interpersonal Circumplex provides a format
within which this exploration may take place. The foundation of Sullivan’s work is
rooted in the belief that human beings “become themselves in relation to others,” such
that, “growth, motivation, adjustment, and disturbances can be understood only within
their social inter-relationships (Wolman, 1973).” Leary used this understanding provided
by Suilivan to create a Circumplex model of personality. Leary’s Circumplex is primarily

used to investigate how traits and emotions are structurally similar. The underlying

assumption crucial to this view is that a relatively seamless circular ordering, or
Circumplex, is the most economical description of the relations among human traits and
emotions.

In addition to demographic factors such as race, education and SES, it is posited
that interpersonal style, interpersonal sensitivity, level of distress, as well as levels of
anxiety and depression are all good predictors of client attrition. Information on the
interpersonal styles that facilitates treatment and those that inhibit attendance to treatment
is vital in devising new methods of intervention. It is hOped that this information will
prove useful in developing new intake procedures and alliance building strategies that
will foster adherence to therapeutic treatment. This in turn should decrease the general

distress of those who feel the need for help but are not being adequately served.

Consequences of Attrition:

In the largest survey on psychosocial treatment ever conducted, Seligman (1995)
found that 7,000 people who responded to a survey published by Consumer Reports
reported incidences of psychological problems for which they sought help. Two
thousand nine hundred of these individuals reported seeking some form of professional
help. Of these respondents, Seligman reported that as the length of treatment increased
client satisfaction with problem resolution increased. While this study has been criticized
for its use of a broad, self-report survey sent out to the middle class subscribers of
“Consumer Reports” of which only a fraction chose to respond, these results would seem
to indicate that the length of treatment has a signiﬁcant impact on the perceived

effectiveness of that treatment.

High rates of attrition have devastating effects upon an individual’s ﬁiture
psychological well being, need for medical services, and economic potential. Pekarik
(1992) found that individuals who terminated early on in therapy (after one or two 'visits)
reported signiﬁcantly higher levels of distress on the Brief Symptom Inventory (BSI;
Derogatis & Spencer, 1982) than clients who dropped out later in the course of treatment
(three or more visits) and those who completed treatment. It is probable that this effect of
premature termination may have greater negative effects upon those individuals who are
experiencing a high level of distress at the time of termination.

In their meta-analytic review of 18 studies investigating the economic impact of
therapy conducted between 1984 and 1994, Gabbard et al. (1997) found that psychosocial
interventions have a signiﬁcant impact on costs associated with clients such that inpatient
treatments were signiﬁcantly reduced as were work related impairments. The
populations found to beneﬁt most from this early intervention were those individuals
experiencing schizophrenic disorders, bipolar affective disorder, and borderline
personality disorder. Not only were future hospitalizations substantially decreased but
job functioning increased as well, enhancing economic potential for these individuals.

Chiles, Lambert, and Hatch (1999) reported on a meta-analysis of 91 studies
conducted between 1967 and 1997 that looked at the economic impact of psychosocial
treatment on future medical costs. They found that the implementation of a wide array of
psychosocial treatments decreased medical expenditures by an average of 20%. These
cost savings were greater than the cost of the psychosocial intervention implemented in a

large number of the studies analyzed.

More important than cost is the relationship between dropping out of therapy
prematurely and the effect that has on clients who are prone to suicide. Few studies have
examined the speciﬁc relationship between premature termination and suicide. Guze and
Robins (1970) estimated a 15% lifetime probability for suicide among psychiatric
patients with affective disorder. This suggests that a high number of at-risk patients who
have received treatment did not remain in treatment long enough to receive beneﬁt.
Dahlsgaard, Beck, and Brown (1998) investigated response to cognitive therapy as a
predictor of suicide. Their study include 17 outpatients who committed suicide and 17
matched outpatients suffering from similar mood disorders that did not commit suicide.
Those who did commit suicide attended signiﬁcantly fewer therapy sessions when
compared to those who didn’t commit suicide. In addition, more individuals who
committed suicide were rated by their therapists as terminating prematurely from therapy
(88%) as compared to those who did not commit suicide (53%). This study suggests that
an inadequate response to therapy coupled with premature termination has an
“unfavorable prognostic signiﬁcance for eventual suicide (p. 197).”

Up to 40% of people seeking professional help for psychological difﬁculties drop
out of treatment after two sessions (Pekarik & Wierzbicki, 1986). These ﬁrst sessions are
the point at which the therapeutic alliance forms and it is upon this alliance that treatment
depends. While many studies have been undertaken to better understand the variables
involved in premature termination the vast majority have been limited to demographic

variables that account for dropout rates.

ﬂictors Involved in Premature Termination

Until very recently, most studies on psychotherapy dropout have focused on the
easily obtainable demographic characteristics of clients and therapists. While securing
this datum is relatively easy, the information that it presents can hardly be called
deﬁnitive. Increased risk for dropout has been found to be associated with minority race,
less education, and low socio—economic status (SES). It should be noted that effect sizes
for most of these studies have been relatively small.

Wierzbicki and Pekarik (1993) conducted a meta-analysis of 125 psychotherapy
dropout studies. In this analysis the mean dropout rate was 46.86%, falling within the
previously cited range of premature termination (30% - 60%). They found that only two
variables were consistently related to dropout in these studies. The ﬁrst was the
deﬁnition of psychotherapeutic dropout that was used. Dropout was deﬁned by either
therapist judgment, number of sessions attended, or the failure to attend a scheduled
session. Failure to attend a scheduled session was the deﬁnition that recorded the least
amount of dropout presumably because it required a client to schedule an appointment
and then miss the appointment to be classed a dropout from psychotherapy. If the client
did not schedule an appointment, even if highly symptomatic, the person was not
considered a dropout. Wierzbicki and Pekarik concluded that the most ﬂexible measure
of dropout, therapist judgment, is the superior method because the therapist has the best
idea of where the client is in the treatment course. The psychometric difﬁculty with this
position is one of internal as well as external validity. The subjectivity of individual
therapist’s judgment does not allow for a consistency between therapists or even between

clients within one therapist’s practice. It would seem that number of sessions attended

provides a rough compromise between these two deﬁnitions. This deﬁnition takes the
subjective element of both the client and the therapist out of the equation and allows for a
precise operationalization of dropout criterion. While the cut-off points for this deﬁnition
might be somewhat arbitrary, the advantage to internal consistency is pronounced.

The other class of variable that accounted for higher dropout rates in the
Wierzbicki and Pekarik study were those variables related to minority racial status, lower
education, and low SES. They conclude that this information does not shed much light
on dropout rates and that more complex psychological variables should be examined.

Beckham (1994) found that a negative impression of the therapist by the patient
was an essential predictor of dropout in his study of attrition at a medical school
outpatient mental health clinic. Suggested in this article was a sensitivity factor in which
people who dropout may be more sensitive to the therapist’s personality and style then
are those who remain in therapy. Missed and canceled appointments early in treatment
were also negatively prognostic for staying in treatment.

Tryon (1990) speculates that the factors involved in premature termination vary
with respect to the point at which termination occurs. Premature termination was found
to be inﬂuenced by the level of disturbance, such that people who were experiencing
greater levels of distress were more likely to remain in therapy, as well as those who had
a high motivation to change. In a 1985 paper, Tryon postulates that the ﬁrst few sessions
are a period in which the client and therapist become engaged with one another. Ifthis
engagement process is successful, a therapeutic alliance is formed and the client is likely

to continue treatment.

The strength of the therapeutic alliance was found to be a signiﬁcant factor in
client dropout in a study conducted by Frayn (1992). In this article he states:

. It appears that patients who tend to do poorly, bring negative aynannc
factors to the treatment situation that persist across hours of psychotherapy
rather than just being episodic misunderstandings. These resistance factors and
nonfacilitating qualities may be intransigent to the therapist ’s eﬂorts to alter the
impending termination or resolve the stalemate (p. 251).

That same study found clinical ﬁrnctioning, levels of introspection, ﬁ'ustration
tolerance, and impulse control to be signiﬁcant predictors of client attrition. While this
study was small and the measures not in widespread use, the results would seem to
indicate that there are personality traits predictive of premature termination.

In a descriptive study of thirty private practice patients who dropped out
prematurely, Levinson, McMurray, Podell, and Weiner (1978) found that reactive factors
were common (e. g., fear of loss of defenses, fear of dependence, fear of aggression,
transference) as well as intrinsic factors (e. g., masochism, negativism, paranoia).
Reactive factors were found in 87% of the patients that dropped out, while intrinsic
factors were inﬂuential 57% of the time.

Hilsenroth, et a1. (1995) used the Rorschach in an effort to ﬁnd indices of
premature termination. They found that individuals who dropped out of therapy before
the eighth session tended to provide responses that included more cooperative movement
responses, fewer texture responses, and fewer aggressive movement responses compared
with patients staying in treatment. The authors suggest that this pattern of response is
indicative of an individual who has less need for close interpersonal contact. Piper,

Joyce, Azim, and Rosie (1994) found that in a day treatment setting for psychiatric

outpatients, psychological mindedness and quality of object relations (which

characterizes the patient’s lifelong pattern of relationships) were the best predictors of
patients remaining in and beneﬁting from the program. They ﬁirther reported that
patients who were married and who had a more mature history of relationships were more
likely to stay in treatment.

In their article on the utility of the Minnesota Multiphasic Personality Inventory-2
(MMPI-2; Butcher, et al,, 1989), Chisholm, Crowther, and Ben-Porath (1997) found the
anxiety scale to be of some utility in predicting dropout. This limited utility may be
explained by the diagnostic orientation of the MMPI-2 (Laforge, Freedman, & Wiggins,
1985). This orientation does not provide an opportunity to test for the ﬁll range of an
individual’s interpersonal orientation.

While these studies vary greatly with regard to the measures employed as well as
the treatment setting, in total they point towards the possible impact of an underlying
personality style that may be implicated in early termination from psychotherapy. Some
studies also indicate that the degree of awareness of the therapists personality style that a
client possesses (Beckham, 1994) as well as the degree of clinical impairment present in
a client are also potential predictors of premature termination (Tryon, 1990). What seems
called for is a more comprehensive personality theory on interpersonal factors that has
been operationalized in such a way as to allow measurement of the factors that may be
implicated in premature termination. Consequently the Leary Interpersonal Circumplex
Model was utilized to investigate possible personality factors implicit in premature

termination.

The Interpersonal Circumplex

Harry Stack Sullivan’s Interpersonal Theory of Psychiatry (1953) introduced a
theory of personality development that expanded the investigation beyond the internal
workings of the individual to include the social interactions of the individual. He deﬁned
psychiatry as “the ﬁeld of study of interpersonal relations, emphasis being placed upon
the interaction of the participants in a social situation, rather than being centered
exclusively on the private economy of either one of those participants (p. ix, Sullivan,
1954)”. It was his belief that an understanding of an individual’s motivation, growth,
adjustment, and disturbances can be fully realized only within the social inter-
relationships that the individual participates in. Sullivan theorized that there exists within
man two basic needs. These are the need for security and the need for satisfaction.
Emotional disturbances occur when a conﬂict arises between these two needs. This being
the case, Sullivan felt that the unit of clinical study of most import was that of a client’s
interpersonal relationships at a given point in time (Klerrnan, et al., 1984).

Within Sullivan’s theory, relationships are characterized by complementary
personality styles. The conceptual framework underlying his theory suggests that
individuals interact with other people in an attempt to reduce anxiety. This anxiety
reduction is accomplished through seeking affirmation of self-concept within the contact
established with others. When these goals are attained, the interaction is complementary.
However, he did not offer ways to measure and categorize the types of interactions that
produced complementarity.

The Interpersonal Circumplex was devised by Leary (1957) as a means of

operationalizing Sullivan’s (1953) theory. This Circumplex model of personality is

10

primarily used to investigate how traits and emotions are structurally similar. The
underlying assumption crucial to this view is that a relatively seamless circular ordering,
or circumplex, is the most economical description of the relations among human traits
and emotions. “The circumplex model allows a broader view of personality and of the
interpersonal relationships integral to understanding just what makes a personality the
way it is” (p. 3, Plutchik & Conte, 1997). In Leary’s circumplex model, interpersonal
behaviors are laid out on two separate axes (see ﬁg. 1). One of these axes is used to
represent a status dimension that is indicative of the level of dominance or
submissiveness present in a relationship. The other axis is used to represent an afﬁliation
dimension. This afﬁliation axis affords a measurement on a continuum ﬁ'om hostility to
ﬁiendliness. These two orthogonal dimensions allow personality styles to be arranged

into 4, 8, 16, and 32 divisions that are differentiated from one another by greater or lesser

 

 

 

 

 

 

 

 

 

 

degrees of
Status
Dominance
Afﬁliation
Hostility Friendliness
Submission
Fig. 1

Axes of Interpersonal Domains (Laforge, et al, 1 95 4)

ll

status and affiliation (ﬁg. 2). Leary held that 16 divisions (sixteenths) was the optimal

number of categories needed to represent the interpersonal schemas of clinicians. Much

work has been completed in searching for the optimal number of categorical divisions

which provide the maximum amount of information while still providing the most

categorically distinct information (Wiggins, 1985). Lorr and McNair (1965) found that

sixteenths held the factor structure put forth by Leary.

i Dominant

Managerial/
Autocratic

 

 

 

 

A

 

 

 

Submissive

Self-Effacing
Masochistic

The Interpersonal Circumplex (Leary, 195 7)

Fig. 2

12

 

  
 

 

 

Cooperative!
Overconven-
tionnl

 

 

Truckenmil'ler and Shaie (1979) found that the factor structure as posited by Leary held
better when divided into octants, or the division of the circumplex into eight divisions,
and stronger still when divided into the four quadrants which represent the original'two
orthogonal dimensions. The Interpersonal Circumplex Model, developed by Leary
(1957) and others, provides a theoretical viewpoint through which a greater range of
interpersonal orientations may be examined (Vaughn & Nowicki, 1999).

Within a circumplex model, complementarity is deﬁned by personality styles that
are opposite on the status dimension and similar on the afﬁliation dimension. Therefore, a
friendly-dominant personality style would be the complement of a friendly-submissive
style, and a hostile dominant style would serve the same ﬁrnction for a hostile-submissive
style. In contrast, anticomplementarity is deﬁned by personality styles that are similar on
the status dimension and different on the afﬁliation continuum. According to
interpersonal theory, complementary interpersonal styles maximize the reduction of
anxiety and conﬁrm self-concepts, whereas anticomplementary personality styles
increase anxiety and do not conﬁrm self-concepts (Carson, 1969).

This idea of complimentary and anti-complimentary styles is important to Leary’s
model because he suggested that they are expressed reﬂexively in an involuntary,
automatic, spontaneous manner. Furthermore, he posited that psychologically healthy, or
adaptive individuals, have more conscious control of their interpersonal style allowing
them to regulate their feelings of anxiety. Maladaptive individuals exhibit less ﬂexibility
in their interpersonal style, remaining wedded to one style that is used within almost all
of their interpersonal situations. The interactions engendered by these maladaptive

individuals tend to elicit responses that reinforce the use of that style. In other words,

13

“interpersonal reﬂexes tend to initiate or invite interpersonal responses from the other
person in the interaction that lead to a repetition of the original reﬂex (p. 123; Leary,

1957)”

Studies Involving Leary’s Model

In a study of 24 male and 24 female college students, Vaughn and Nowicki,
(1999) used the Interpersonal Adjective Scale-Revised (IAS-R; Wiggins, 1991) to assess
‘ interpersonal style. This scale is based upon Leary’s Circumplex Model and includes 64
interpersonal adjectives to assess individual interpersonal styles. They found that women
described close, same-gender relationships as being different from themselves in status.
This was particularly evident when comparisons were made with relationships that were
characterized as being “not close”. In contrast, men did not report a signiﬁcant
relationship between the closeness of their relationships and the degree of affrliation or
status.

Lorr (1991) found that the submissiveness quadrants of the Circumplex
compliment the dominant end of the continuum when the individual is in a relationship
with a superordinate (e. g. parent or employer). As the therapeutic alliance forms the
perception of the relationship by the client moves from one of submission to an authority
ﬁgure to that of an egalitarian relationship. Lorr goes on to explain that as a relationship
moves towards this egalitarian stance, submissive behaviors transform themselves into
non-directive behaviors. The heart of some views of the therapeutic alliance, in a host of
theoretical clinical orientations, revolves around this idea of establishing a non-directive

relationship.

14

Sum

Premature termination, dropout, and client attrition are all phrases used to identify
clients who terminate psychotherapy prior to the therapist’s expectations of duration
necessary for a therapeutic gain. Pekarik (1983) found that 30% - 60% of outpatient
psychotherapy clients terminate prematurely. High rates of attrition have devastating
effects upon an individual’s future psychological well being, need for medical services,
and economic potential (Pekarik, 1992).

Until very recently, most studies on psychotherapy dropout have focused on the
easily obtainable demographic characteristics of clients and therapists. In a 1993 article,
Wierzbicki and Pekarik called for a deeper analysis of the variables involved in
premature termination. This call has been heeded by a number of studies which have
found that aside ﬁ'om race, education and SES, other variables point towards a underlying
personality style that is more prone to premature termination. These factors include
increased interpersonal sensitivity and anxiety, decreased levels of distress and need for
close interpersonal contact, as well as the degree to which the individual is dominant or
submissive.

Harry Stack Sullivan’s (1953) Interpersonal Theory of Personality as
operationalized in Timothy Leary’s (195 7) Interpersonal Circumplex in conjunction with
the Symptom Check List of Derogatis (1983) provides an excellent format within which
this exploration may take place. Given the previous literature the following hypotheses

are proffered.

IS

HYPOTHESIS

1. There will be a statistically signiﬁcant difference in interpersonal style between. the
Early Dropout Group, the Late Dropout Group and the Persisters such that those
remaining in therapy longer will have begun therapy in a more submissive stance than
those who dropped out at an earlier point in therapy (as measured by the Status
dimension of the ICL).

Rationale:

As discussed by Lorr (1991) the client begins the therapeutic process by
projecting onto the therapist the role of an authoritarian ﬁgure. This representation
gradually becomes replaced with an egalitarian relationship as the therapeutic alliance
strengthens. Individuals shape relationships in ways that reduce anxiety. The
complimentarity portion of the Interpersonal Circumplex model posits that individuals
who are submissive will experience anxiety reduction when relating to another who is
dominant such as an authority ﬁgure. Individuals who are dominant will experience
elevated levels of anxiety when relating to a dominant authority ﬁgure. This elevated

level of anxiety is aversive and will precipitate premature termination.

2. Exploratory analysis will be employed to examine how both octants and sixteenths
from the ICL will bear out the general conclusion from hypotheses 1.
Rationale:
There is some question as to the optimal number of categorical divisions that
provide the maximum amount of information while still providing the most categorically

distinct information. This being the case, it seems prudent to examine not only

16

respondents with respect to interpersonal quadrants but to also look at the relationships

that appear within the octants and sixteenths as well.

3. For those people who rate themselves as being more submissive on the ICL, distress
level (as measured by the GSI scale of the SCL-90-R) will be higher in the Persisters
than the Late Dropout Group, and both groups will be experiencing a higher level of
distress than the Early Dropout Group. This relationship will be reversed for those
who rate themselves as dominant such that initial distress levels will be lower for
individuals who remain in therapy longer.

Rationale:

Leary posited that maladaptive individuals exhibit less ﬂexibility in their
interpersonal style, remaining wedded to one style that is used within almost all of their
interpersonal situations. More adaptive, psychologically healthier individuals have more
control of their interpersonal style. Tryon (1990) found premature termination to be
inﬂuenced by the level of disturbance, such that people who were experiencing greater
levels of distress were more likely to remain in therapy. Those individuals who are more
submissive will ﬁnd the initial stages of therapy to be anxiety reducing and so will tend to
remain, especially if their level of distress outside of therapy is great. Those people who
are more dominant with respect to interpersonal style will be more likely to remain in
therapy if they are experiencing lower levels of distress because they will be better able

to modulate their interpersonal style to ﬁt the therapeutic situation.

4. This study will examine the factor structure of the SCL-90-R.

l7

Rationale:

The SCL-90-R was developed to elicit client distress within nine symptom factors.
This factor structure has been questioned in the literature and so will be examined in this
study. It has been noted that as distress level increases the factor structure of the SCL-90-
R deteriorates. Should the nine-factor model originally posited by Derogatis (1983) hold

then all of the subscales will be examined for correlations to client attrition.

5. It is hypothesized that the Interpersonal Sensitivity subscale, Anxiety subscale, and
Depression subscale of the SCL-90-R will all be signiﬁcant predictors of premature
termination as will minority racial status, lower education, and low SES

Rationale:

This hypothesis is raised to replicate the ﬁndings of Beckham (1994), Chisholm,
Crowther, and Ben-Porath (1997), and Wierzbicki and Pekarik (1993). Beckham
suggested the importance of a sensitivity factor in which people who dropout may be
more sensitive to the therapist’s personality and style then are those who remain in
therapy. Chisholm, Crowther, and Ben-Porath found the anxiety scale of the MMPI-2 to
be of utility in predicting premature termination. In their meta-analysis of 125 studies,
Wierzbicki and Pekarik (1993) found that the most common variables found to be
responsible for premature termination were race, education and socio-economic status

(SES).

18

METHODS

Participants:

Data for this study were obtained ﬁom the Michigan State University (MSU)
Psychological Clinic, a teaching and research facility associated with the Psychology
Department at MSU. The Michigan State University (MSU) Psychological Clinic is an
income adjusted, fee for service, non-proﬁt serving the general population of the greater
Lansing area. The clients seen are predominantly working and middle class people who
present with a wide variety of psychopathology including, but not limited to moderate to
severe depression and anxiety, eating disorders, anger control difﬁculties, and
interpersonal relationship difficulties.

One Hundred and Fifteen participants participated in this study, 35 of whom were
in the Early Dropout Group (premature termination aﬁer 1 to 3 therapy sessions), 35 of
whom were in the Late Dropout Group (premature termination after 8 to 13 therapy
sessions) and 45 of whom were in the Persister Group (completing at least 18 therapy
sessions). Of those who participated 35 were males and 80 were females. The mean age
of the participants was 30.5 years of age (SD = 9.42) with a mean education level of
14.46 years (SD = 2.40) and a mean income level of $13,602 (SD = $10,851). The vast
majority of this sample was Caucasian (N = 98), with 6 African American, 5 Hispanic, 1
Asian, participants. Three participants classiﬁed themselves as Other with respect to

race.

19

Therapists:

The therapists utilized for‘this study were graduate students working towards a
degree in clinical psychology within the Michigan State University (MSU) Psychology
Department. All student therapists included in this study were working in the MSU
Psychological Clinic at the time data were collected. Second to fourth year trainees as
well as advanced trainees with several years of post-masters experience were be utilized
for this study. The therapists had all completed graduate courses in assessment
procedures as well as in theories of psychotherapy. The majority of therapists had a
psychodynamic orientation with other treatment orientations being represented to a lesser
extent. Since this study was be conducted after therapy had been completed, both

therapists and clients were blind to the hypotheses and purposes of this study.

Procedures

During the initial intake interview at the MSU Psychological Clinic, potential
participants were asked to take part in the ongoing MSU Psychotherapy Research Project.
They were informed that their choice to participate or not participate will not affect the
services they were to receive. A small incentive was offered. Ifthey chose to participate
and complete the necessary pre- and post-therapy forms, they were given a refund equal
to ten percent of their fees paid for therapy up to a maximum of eighty dollars. Clients
who agreed to participate were asked to complete a consent form, the Interpersonal
Check List (ICL), the Symptom Checklist 90 Revised (SCL90-R) as well as a

demographic questionnaire prior to their ﬁrst therapy session.

20

All clients were assigned student clinicians with experience ranging from one to
three years of formal training in psychological assessment and intervention. Participants
who completed the battery of questionnaires, returned them at their ﬁrst session of '
psychotherapy, and then dropped out of treatment prior to a third therapy session, were
entered into the “Early Dropout” group. Those who dropped out of therapy after 8 to 12
sessions were entered into the “Late Dropout” group. Research participant who persisted
in treatment beyond eighteen sessions of therapy were entered in a “Persister”

comparison group.

Measures:

Interpersonal Check List (ICL): The ICL (La Forge & Suczek, 1955) is one of
the most widely used instrument to operationalize interpersonal theory. Developed by
Leary and colleagues (Laforge, Leary Naboisek, Coffey, & Freedman, 1954; Laforge &
Suczek, 1955; Leary & Coffey, 1955) it has been used in well over 200 studies (LaForge,
1977; Paddock, 1982). The Leary (1957) circumplex model is designed to measure
interpersonal style on two major domains of interpersonal behavior. The ﬁrst of these
domains varies along a continuum between dominance and submissiveness. The second
domain varies along a continuum of afﬁliation and hostility. When scored, it yields two
summary indices representing the two major dimensions of interpersonal style: status
(dominance-submissiveness, or Dom.) and affiliation (hostility-friendliness, or Lov.).
These may be further divided to form 16 different interpersonal styles. The questionnaire
examines each of these styles by injecting eight different adjectives into questions

designed to focus on each domain.

21

The ICL is comprised of 128 items constructed of 8 unique adjectives or adjective
phrases which are keyed to 16 interpersonal variables (Paddock & Nowicki, 1986). Each
variable has, “one intensity 1 item which reﬂects a mild or necessary amount of the'trait
... three items refer to intensity 2, a moderate or appropriate amount of the trait three
words reﬂect intensity 3, a marked or inappropriate amount of the trait, and one word
expresses intesnsity 4, and extreme amount of the trait” (Leary, 1957, p.455).

Several studies have reported an adequate level of test-retest reliability for self-
report ratings of the DOM and LOV summary scores (r > .70) (Laforge, 1977; Laforge &
Suczek, 1955). Internal consistency has not been reported in studies for either octants or
sixteenths. Concurrent validity has been established in studies utilizing mothers ratings of
360 normal children and 65 severely disturbed children (Guemey, Meininger, & Stover,
1973). In a study by Paddock, et al. (1984) in which normal high school students were
compared to severely emotionally disturbed high school students revealed that the use of
the ICL correctly classiﬁed 65 and 75% of the normal and severely emotionally disturbed
students respectively. Several studies have related ICL scores to other psychometric
instruments such as the California Personality Inventory, Cattell’s 16 Personality Factor
Questionnaire, and the MMPI in order to establish convergent and discriminant validity
(Gynther, Miller, & Davis, 1962; Hamilton, 1971; Laforge, 1977; Leary & Coffey, 1955).
While accounting for a relatively small proportion of the variance in these studies, the

predicted relationships were statistically signiﬁcant.

Symptom Checklist 90 Revised (SCL-90-R): The SCL-90—R (Derogatis, 1983) is

a self-report measure of the test taker’s psychological distress and associated symptoms

22

that will be used to measure participants’ level of psychopathology at the time of their
entering therapy. It is a 90 item self-administered questionnaire composed of nine
subscales measuring nine symptom dimensions: somatization, obsessive-compulsive,
interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation,
and psychoticism. Participants are asked questions about the symptoms they are
experiencing. These questions are designed to elicit the degree of distress they are
experiencing over this range of symptoms (e.g., headaches, nervousness, shakiness, etc.)
by having them rate each item on a Likert type scale that ranges from 0, not at all, to 4,
extremely. Means are then computed for each of the nine subscales. A Global Severity
Index (GSI) is computed as the sum of all item responses divided by 90. According to
Derogatis (1983), the G81 is the best single indication of the current level of
psychological distress.

Validity and reliability of the SCL-90-R are discussed at length in the
Administration, Scoring and Procedures Manual (Derogatis, 1983). Internal Consistency
and test-retest reliability were the methods employed to ascertain the instruments
reliability. Internal consistency was established using a sample of 219 symptomatic
individuals. Coefﬁcient alphas obtained from this sample ranged from a low of .77 for
psychoticism to a high of .90 for depression. All coefﬁcients were within satisfactory
limits (Derogatis, Rickels & Rock, 1976). A test-retest reliability study was performed
utilizing 94 psychiatric outpatients. Participants were administered the instrument twice
with a one week time span between testing. Coefﬁcients ranged from .78 for hostility to

.90 for phobic anxiety (Derogatis, Rickels & Rock, 1976).

23

Criterion and construct validity of the SCL-90-R has also received attention.
Criterion validity has been established in a number of studies (Prusoff, Weissman,
Klerman & Rounsaville, 1980; Derogatis, Meyer & King, 1981; Derogatis, 1977). These
studies have found the SCL-90-R capable of measuring the type and severity of
psychological symptomatology experienced by the individuals under study. Construct
validity has also received signiﬁcant attention. Derogatis, Rickels and Rock (1976)
compared the dimension scores of 119 symptomatic individuals on the SCL—90 with their
scale scores on the MMPI. This study found that each dimension of the SCL-9O
correlated to the greatest degree with a like construct on the MMPI. Similar results have
been found in other studies (Prusoff, Weissman, Klerrnan & Rounsaville, 1980;
Derogatis, Meyer & King, 1981; Derogatis, 1977).

Derogatis and Cleary (197 7) performed a conﬁrmatory factor analysis utilizing a
stratiﬁed sample of 1002 outpatient psychiatric patients from a diverse community in a
large eastern state. The results conﬁrmed the hypothesized structure of the instrument.
This factor structure has been questioned in more recent literature. Vassend and Skrondal
(1999) have found that the factor structure of the instrument collapses as distress
increases such that at high levels of distress the only factor remaining is a negative-affect
factor. This factor structure will also be examined in this study.

The instrument has been well accepted in the ﬁeld of clinical assessment. An
earlier version of the SCL-90-R (the SCL-90) was, at one time included in an NIMH
standard psychotherapy outcome research test battery (W askow & Parloff, 1975). The
instrument has proven itself valid and reliable with subject comparable to the current

study. While there is some concern regarding its ability to discriminate between

24

symptom domains in people at high levels of distress, the G81 scale is an excellent
measure of over all distress. Towards this end, the primary scale used for this study will
be the G81. A factor analysis will be run and if the factor structure holds within

reasonable bounds then the nine subscales will be utilized.

Demographics (Derogatis, 1977). Demographics were collected using the intake-
rating portion of the SCL-90-R. Demographic information collected include age, sex,
ethnicity, marital status, source of income, income level, educational level, and

occupational level.

RESULTS

One Hundred and F iﬁeen participants participated in this study, 35 of whom were
in the Early Dropout Group (premature termination after 1 to 3 therapy sessions), 35 of
whom were in the Late Dropout Group (premature termination after 8 to 13 therapy
sessions) and 45 of whom were in the Persister Group (completing at least 18 therapy
sessions). Of those who participated 35 were males and 80 were females. The mean age
of the participants was 30.5 years of age (SD = 9.42) with a mean education level of
14.46 years (SD = 2.40) and a mean income level of $13,602 (SD = $10,851). The vast
majority of this sample was Caucasian (N = 98), with 6 African American, 5 Hispanic, 1
Asian, participants. Three participants classiﬁed themselves as Other with respect to
race.

The ﬁrst hypothesis stated that there would be a statistically signiﬁcant difference

in interpersonal style between the Early Dropout Group, the Late Dropout Group and the

25

Persisters such that those remaining in therapy longer would have begun therapy in a
more submissive stance than those who dropped out at an earlier point in therapy (as
measured by the Status dimension of the ICL). A One-way ANOVA with trend analysis
was run to test this hypothesis. While the main effect (F(2,110) = 2.29, p > .05) and the
linear trend (F(1,110) = .237, p > .05) were not signiﬁcant the quadratic trend was
signiﬁcant, F(1,110) = 4.22, p < .05 (see Table 1). This hypothesis was partially
supported by the signiﬁcant quadratic trend such that the Late Dropout Group displayed
greater levels of dominance (M = -1.64) than the Persisters (M = -11.21). The curvilinear
trend was revealed in the results of the Early Dropout Group whose level of dominance
(M = -9.47) was almost as low as the Persisters.

The relationship between level of dominance and point of termination was ﬁrrther
explored with respect to gender with a Post Hoc test. A Two-way ANOVA (point of
termination X Gender) was run. While no main effect was found for either point of
termination (F(2, 107) = 1.60, p > .05) or gender (F(1,107) = .035, p > .05), a signiﬁcant
interaction was found between these two variables, F(2, 107) = 4.04, p < .05 (see Table 1).
This interaction revealed that men tended to follow the predicted relationship. The Early
Dropout Group showed the highest level of dominance (M = 2.48), the Late Dropout
Group displayed less dominance (M = -10.63), and the Persisters displayed the least
amount of dominance (M: -14.66). Women displayed a pattern of dominance with
respect to termination point that mirrors the results of the One-Way ANOVA trend
analysis. Those in the Early Dropout Group showing relatively low levels of dominance

(M = -1 3.61), those in the Late Dropout Group showing relatively high levels of

26

dominance (M = 3.05), and the Persisters revealing relatively low levels of dominance
(M = -9.88) (see table 1).

The second Hypothesis called for an exploration of the Octants and Sixteenths
contained within the Circumplex to see how these variables bare out the general
conclusions from the ﬁrst hypothesis. The ﬁrst of the analyses regarding this hypothesis
was employed to discover the relationship between the circumplex octants and point of
termination using a One-way AN OVA with trend analysis. The only two octants found
to be signiﬁcant were the two that correspond most closely to the compiled status
dimension of the circumplex. The managerial/autocratic octant corresponds to the
dominant half of the status dimension and was a signiﬁcant predictor of termination,

F(2,]10) = 3.22, p <.05 (see table 2). The self-effacing/masochistic octant

Table 1. Means and Standard Errors for Termination Point X Gender interaction with
respect to level of Dominance

 

 

COND# Sex Mean Std.
Error

Early Male 2.48 6.767
Dropout Female -1 3 .61 3 .981
Combined -9.47 4.24

Late Male -10.63 5.861
Dropout Female 3.05 4.233
Combined -l .64 2.98

Persisters Male -14.658 5 . 861
Female -9. 877 3 .646

Combined -l 1.21 2.93

 

27

Figure 3. Means of Level of Dominance for Termination Point X Gender interaction

Estimated Marginal Means of DOM

10

 

 

 

 

 

m

r:

‘s"

E

.E

R

2 SEX

g a

g male

13 -20 r 0 female
recessions 8—13sessions 180rmore

COND#

corresponds to the submissive half of the status dimension and was a signiﬁcant predictor
of termination, F (2,1 10) = 3 .76, p <.05 (see table 2). As with the status dimension, both
of these octants were found to have signiﬁcant quadratic trends. The
managerial/autocratic octant had a signiﬁcant quadratic trend (F(1,10) = 6.43, p < .05)
such that the mean of the Early Dropout Group was relatively low (M = 9.77), the mean
of the Late Dropout Group was relatively high (M: 13.86), and the mean of the Persisters
was roughly equivalent to the Early Dropout Group (M = 9.84). The self-
effacing/masochistic octant had a signiﬁcant quadratic trend (F (1, 10) = 6.90, p < .05)
such that the mean of the Early Dropout Group was relatively high (M = 18.14), the mean
of the Late Dropout Group was relatively low (M= 13.66), and the mean of the Persisters

was roughly equivalent to the Early Dropout Group (M = 19.53).

28

 

 

owe u Om 3% n Om owe u Om 26 n Gm 3w n Gm 02w n Gm
mow u 2 Ex: n 2 3: u E oo.v u 2 3.2 u 2 vwd u 2 983.23
Se u Om mad .1. Om mm...» H Om tow n Om 21. u Om mew n Cm 3389
8% ME mow n2 ooﬁHE ONGHE 8.2 ME 3.2 NS 8:1—
2 .o u Om So u Om EM n Om 5% u Om RNM u Gm 000 n Dm 323.5
mwsnz - 3.2 HZ own": KANE 3.2.1.2 Rd NE 38m
EEooEmv £2.8ng 36085.5 @6823 0:303 0588
6.35832 3.8352
3356832 wﬁoatm—éom 9:55:34... . aroma—.52 \w=_uatm..=om 359883.. 5.50

 

86m gouging. E 2054 ﬁeoogm can “580 :83:ch mo sonar/0O @8985 use 282 .m 033.

29

Figure 4. Means of Autocratic/Managerial Levels by Termination Point

15

 

14:

13'

121

11-1

10-

 

Mean of AP

9

1-3seesions 8—133ésions 18ormore

 

 

COND#

Figure 5. Means of Self-Effacing/Masochistic Level by Termination Point
20

 

19.
185
17-
16.
15.

141

13
-3$essions 8—13sessions 180rmore

 

 

Mean of HI

 

uni

COND#

These same relationships were then explored with respect to the circumplex sixteenths
and point of termination using a One-way AN OVA with trend analysis. Four of the
sixteenths were found to be signiﬁcant. These four sixteenths combined to make up the
two octants that were found to be signiﬁcant and so shared the same correspondence to
the status dimension of the circumplex as did the octants. The managerial sixteenth
corresponds to the dominant half of the status dimension did not reveal a signiﬁcant main
effect (F(2,]10) = 2.77, p >05). It did show a signiﬁcant curvilinear trend when a
quadratic term was added (F (1,1 10) = 5.23, p < .05) such that the Early Dropout Group
showed relatively low levels on the managerial sixteenth (M = 3.97), the Late Dropout
Group showed relatively high levels (M = 6.20), and the Persisters revealed a roughly
equivalent level on the managerial sixteenth to the Early Dropout Group (M = 4.00) (see
table 2). The autocratic sixteenth corresponds to the dominant half of the status
dimension as well and did not reveal a signiﬁcant main effect (F(2,110) = 2.15, p >.05). It
did show a signiﬁcant curvilinear trend when a quadratic term was added (F (1,1 10) =
4.29, p < .05) such that the Early Dropout Group showed relatively low levels on the
autocratic sixteenth (M = 5.80), the Late Dropout Group showed relatively high levels (M
= 7.66), and the Persisters revealed a roughly equivalent level on the autocratic sixteenth

to the Early Dropout Group (M = 5.84) (see table 2).

31

Figure 6. Means of Managerial Level by Termination Point

6.5

 

6.0 1
5.5 1
5.0 l
4.5 r

4.0 C

 

 

Mean of A

3.5

1-33essions 8-13sgssions 18ormore

 

COND#

Figure 7. Means of Autocratic Level by Termination Point

8.0

 

7.5 I

7.0 1

6.5 11

6.0 I

5.5
1-3sessions 8-13se'ssions 18ormore

 

 

Mean of P

 

COND#

The Self-effacing sixteenth corresponds to the submissive half of the status dimension did
not reveal a signiﬁcant main effect (F (2,1 10) = 2.84, p >05), but did show a signiﬁcant
curvilinear trend when a quadratic term was added (F (1,1 10) = 4.519, p < .05 such that
the Early Dropout Group showed relatively high levels on the self-effacing sixteenth (M
= 10.31), the Late Dropout Group showed relatively low levels (M = 8.63), and the
Persisters revealed relatively high levels on the self-effacing sixteenth (M = 11.44) (see
table 2). The masochistic sixteenth corresponds to the submissive half of the status
dimension as well and did reveal a signiﬁcant main effect (F(2,110) = 4.061, p <.05), as
well as revealing a signiﬁcant curvilinear trend when a quadratic term was added

(F(1,110) = 7.97, p < .01) such that the Early Dropout Group showed relatively high

Figure 8. Means of Self-Effacing Level by Termination Point

 

12.0
11.5' I)
11.0I
10.5-
10.0-
9.5 I
9.0 I

8.5 I

 

Mean of H

8.0
1-3sessions 8—135essions 180rmore

 

 

COND#

33

Figure 9. Means of Masochistic Levels by Termination Point

8.5

 

8.0 I J
E
7.5 I
7.0 I
6.5 I
6.0 T

5.5 I

5.0 I

 

Mean of I

4.5
1-35essions 8-13se'ssions 18umore

 

 

COND#

levels on the masochistic sixteenth (M = 7.83), the Late Dropout’Group showed relatively
low levels (M = 5.03), and the Persisters revealed a roughly equivalent level on the
masochistic sixteenth to the Early Dropout Group (M = 8.09) (see table 2).

The third hypothesis stated that for those people who rate themselves as being
more submissive on the ICL, distress level (as measured by the G81 scale of the SCL-90-
R) will be higher in the Persisters than the Late Dropout Group, and both groups will be
experiencing a higher level of distress than the Early Dropout Group. This relationship
will be reversed for those who rate themselves as dominant such that initial distress levels
will be lower for individuals who remain in therapy longer. These hypotheses were not
supported by the data. Two One-way AN OVAs were utilized to analyze the effect of
distress level on the point of termination for both the group characterizing itself as

submissive and the group characterizing itself as dominant. There was no signiﬁcant

34

relationship found for either the Submissive group (F(2,67) = 1.67, p > .05), or the
dominant group (F(2,39) = .665, p > .05).

The fourth hypothesis stated that the Interpersonal Sensitivity subscale, Anxiety
subscale, and Depression subscale of the SCL-90-R will all be signiﬁcant predictors of
premature termination as will minority racial status, lower education, and low SES.
Seven One-way AN OVAs were utilized to analyze the relationship between these
variables and premature termination from therapy. The only above-mentioned variable
that supported this hypothesis was that of education. A statistically signiﬁcant
relationship was found to exist between level of education and point of termination
(F (2,107) = 6.39, p < .01) such that the Early Dropout Group had completed less years of
education (M = 13.27) than both the Late Dropout Group (M = 14.94) and the Persisters

(M = 15.00) (See table 3).

Table 3. Means and Standard Deviations of Years of Education by Termination Point

 

 

COND Level of Education

(in years)
Early M= 13.27
Dropout SD = 2.52
Late M = 14.94
Dropout SD = 2.31
Persisters M = 15.00

SD = 2.09

 

35

Figure 10. Means of Years of Education by Termination Point

15.5

 

15.0 I #43

14.51

14.0 I

13.5-

 

Mean of EDLEVEL

13.0
1-3sessions 8-13Qsions 180rmore

 

 

COND#

Not found to be signiﬁcant were Level of Interpersonal Sensitivity (F (2,1 1 1) = 1.608, p >
.05), Level of Depression (F(2,111) = .804, p > .05), Level of Anxiety (F(2,111) = .647, p
> .05), Racial Status (F(2,110) = .349, p > .05), Income (F(2,102) = 1.48, p > .05), and
Occupation Level (F(2,107) = 1.25, p> .05).

The ﬁfth hypothesis involved an examination of the factor structure of the SCL-
90-R. An unrestricted, principle axis factor analysis yielded 20 factors with eigenvalues
of greater than 1. The percentage of variance explained by each was as follows: 30.40%,
6.24%, 4.45%, 3.80%, 3.07%, 2.73%, 2.46%, 2.32%, 2.08%, 2.00%, 1.95%, 1.78%,
1.65%, 1.65%, 1.59%, 1.53%, 1.45%, 1.33%, 1.24%. Six factors were retained using the
scree test criteria (Cattell, 1978). As a result, only six factors were rotated using

orthogonal, varimax rotation. These six factors explained 50.96% of the variance. Only

36

variables that had factor loadings of at least .40 were selected to represent factors (see
Tables 4 — 9). Eleven items did not meet this criterion and were excluded (see Table 10).
Factor 1, Depressed Interpersonal Sensitivity, explained the largest portion of the
variance explained (30.40%). The items comprising this factor were primarily
characterized by depressive thoughts and actions as well as items reﬂecting a withdrawal
from interpersonal situations. Factor 2, Phobic Interpersonal Sensitivity, explained
6.24% of the variance and was comprised of a combination of phobic responses and
items reﬂecting a withdrawal from interpersonal situations. Factor 3, Anxious Phobia,
explained 4.45% of the variance and included a number of typically anxious and phobic
responses. Factor 4, Somatization, explained 3.8% of the variance and contained only
items that described physical complaints. Factor 5, Hostility, explained 3.07% of the
variance and contained items that were characteristic of hostile, aggressive thoughts and
actions. Factor 6, Obsessive/Compulsive, explained 3.01% of the variance and was

characterized by items that had an obsessive/compulsive quality to them.

37

Table 4. Varimax Item Loadings on Factor 1 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

 

Study.

Item Original
Number Item Description Loading Factor

Factor 1: Depressed Interpersonal Sensitivity

3. Repeated thoughts that won’t stop .56 Obsess/Comp
5. Loss of sexual interest or pleasure .45 Depression
11. Feeling easily annoyed .47 Hostility
14. Feeling low in energy or slowed down .42 Depression
15. Thoughts of ending your life .48 Depression
18. Feeling that most people cannot be trusted .53 Paranoid
20. Crying easily .42 Depression
21. Feeling shy or uneasy with the opposite sex .41 Interpersonal
26. Blaming yourself for things .48 Depression
29. Feeling lonely .74 Depression
30. Feeling blue .75 Depression
31. Worrying too much about things .59 Depression
32. Feeling no interest in things .56 Depression
34. Feelings being easily hurt .49 Interpersonal
36. Feeling others do not understand you .58 Interpersonal
37. Feeling that people are unfriendly .52 Interpersonal
41. Feeling inferior to others .63 Interpersonal
46. Difficulty making decisions .48 Interpersonal
54. Feeling hopeless about the future .51 Depression
55. Trouble concentrating .52 Obsess/Comp
57. Feeling tense or keyed up .41 Anxiety
61. Feeling uneasy when people are watching you .41 Interpersonal
69. Feeling very self-conscious with others .57 Interpersonal
71. Feeling everything is an effort .55 Depression
77. Feeling lonely even when you are with people .67 Psychoticism
79. Feeling worthless .71 Depression
80. Feeling something bad will to happen to you .42 Anxiety
83. Feeling that people will take advantage of you .52 Paranoid
88. Never feeling close to another person .49 Pychoticism

 

38

 

Table 5. Varimax Item Loadings on Factor 2 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

 

Study. '

Item Original
Number Item Description Loading Factor

Factor 2: Phobic Interpersonal Sensitivity

13. Feeling afraid in open spaces or on the street .54 Phobic
21. Feeling shy or uneasy with the opposite sex .41 Interpersonal
25. Feeling aﬁ'aid to go out of your house alone .54 Phobic
35. Others being aware of your private thoughts .45 Psychoticism
37. Feeling that people are unfriendly .60 Interpersonal
43. Feeling that you are watched or talked about .62 Paranoid
47. Feeling afraid to travel .40 Phobic
50. Avoiding things, places, activities from fear .53 Phobic
52. Numbness or tingling in parts of your body .45 Somatization
61. Feeling uneasy when watched or talked about .59 Interpersonal
62. Having thoughts that are not your own .43 Psychoticism
65. Having to repeat the same action .54 Obsess/Comp
68. Having ideas or beliefs that others don’t share .53 Paranoid
69. Feeling very self-conscious with others .43 Interpersonal
70. Feeling uneasy in crowds .57 Phobic
73. Feeling leery of eating or drinking in public .43 Interpersonal
76. Not getting credit for your achievements .55 Paranoid
84. Having thoughts about sex that bother you .41 Psychoticism

 

39

 

Table 6. Varimax Item Loadings on Factor 3 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

 

Study.

Item Original
Number Item Description Loading Factor

Factor 3: Anxious Phobia

2. Nervousness or shakiness inside .60 Anxiety
4. Faintness or dizziness .51 Somatization
13. Feeling afraid in open spaces or on streets .46 Phobic
17. Trembling .62 Anxiety
23. Suddenly scared for no reason .61 Anxiety
25. Feeling afraid to go out of the house alone .42 Phobic
33. Feeling fearﬁJl .52 Anxiety
39. Heart pounding or racing .63 Anxiety
40. Nausea or upset stomach .53 Somatization
48. Trouble getting your breath .45 Somatization
50. Avoiding people, places, things for fear .45 Phobic
57. Feeling tense or keyed up .50 Anxiety
70. Feeling uneasy in crowds .45 Phobic
72. Spells of terror or panic .73 Anxiety
73. Feeling leery of eating or drinking in public .43 Interpersonal
75. Feeling nervous when left alone .52 Phobic
82. Feeling afraid you will faint in public .52 Phobic

 

4O

 

Table 7. Varimax Item Loadings on Factor 4 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

Study.

Item Original
Number Item Description Loading Factor

Factor 4: Somatization

1 . Headaches .47 Somatization
12. Pains in heart or chest .47 Somatization
17. Trembling .45 Anxiety
27. Pains in lower back .61 Somatization
40. Nausea or upset stomach .50 Somatization
42. Soreness of your muscles .64 Somatization
48. Trouble getting your breath .52 Somatization
49. Hot or cold spells .64 Somatization
52. Numbness or tingling in parts of your body .62 Somatization
56. Feeling weak in parts of your body .65 Somatization
58. Heavy feelings in your arms or legs .73 Somatization
87. The idea that something is seriously wrong .43 Psychoticism

with your body

 

 

Table 8. Varimax Item Loadings on Factor 5 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

 

Study.

Item Original
Number Item Description Loading Factor

Factor 5: Hostility

24. Temper outbursts that you could not control .67 Hostility
63. Having urges to beat, injure, or harm someone .56 Hostility
67. Having urges to break or smash things .76 Hostility
74. Getting into frequent arguments .51 Hostility
81. Shouting or throwing things .65 Hostility

 

41

 

 

Table 9. Varimax Item Loadings on Factor 6 of the SCL-90-R as Obtained in the Current

 

 

 

 

 

 

Study.
Item Original
Number Item Description Loading Factor
Factor 6: Obsessive/Compulsive

10. Worried about sloppiness or carelessness .48 Obsess/Comp
28. Feeling blocked in getting things done .40 Obsess/Comp
38. Having to do ﬁrings slow to insure correctness .62 Obsess/Comp
45. Having to check and re—check what you do .57 Obsess/Comp
46. Difficulty making decisions .41 Obsess/Comp
51. Your mind goes blank .47 Obsess/Comp
55. Trouble concentrating .46 Obsess/Comp

 

Table 10. Items Loading Less than .40 in any of the Factors of the SCL-90-R as Obtained

in the

Current Study.

 

 

 

 

 

Item Original

Number Item Description Factor
Items Loading Less than .40 in any of the
Factors

6. Feeling critical of others Interpersonal
7. The idea that others control your thoughts Psychoticism
8. Blaming others for most of your troubles Paranoid
9. Trouble remembering things Obsess/Comp
16. Hearing voices that others don’t hear Psychoticism
22. Feelings of being trapped or caught Depression
53. A lump in your throat Somatization
78. Feeling so restless that you can’t sit still Anxiety
85. The idea that you should be punished for sins Psychoticism
86. Thoughts and images of a frightening nature Anxiety
90. The idea something is wrong with your mind Psychoticism

 

 

42

 

DISCUSSION

The single most common ﬁnding in research on premature termination has been
that those individuals who come from impoverished backgrounds tend to drop out of
therapy prematurely. Among variables found to be signiﬁcant predictors of early
termination have been low socio-economic status, lower education, and minority racial
status. It was hypothesized that this would also be in evidence in this study. Toward this
end Race, Education Level, Income, and Occupation level were all examined with respect
to point of termination. The only variable that showed a statistically signiﬁcant
relationship with point of termination was Education level. Those individuals who
remained in therapy at least 8 sessions had a mean of approximately 15 years of
education as compared to those who terminated after 1 to 3 sessions whose educational
mean was 13.3 years. While it is in keeping with past research to ﬁnd that education has
a signiﬁcant relationship with premature termination, it is somewhat surprising that none
of the other variables were signiﬁcant. The income level within this sample ranged from
$0 to $50,000 with a comparable spread in occupational levels. These wide
representations of income and occupation level provide enough power for hypothesis
testing on this issue. Less surprising were the results testing the relationship between
race and point of termination. Relatively few clients were seen during the duration of this
study and so the power is likely to be insufficient to detect the true nature of this
relationship.

It was ﬁrrther posited that the level of distress experienced by the client as well as

certain types of pathologic symptomatology contributes to the manner in which clients

43

engage in therapy. The SCL-90-R was used to investigate the levels of distress and
pathology observed within the sample population. For the majority of the analyses
performed the factor structure, as originally provided by Derogatis, was used. It was
hypothesized that the Interpersonal Sensitivity subscale, Anxiety subscale, and
Depression subscale of the SCL-90-R would all be signiﬁcant predictors of premature
termination. This was not found to be the case. All of the subscales of the SCL-90-R
were tested and none of them produced a signiﬁcant relationship with the point at which
clients terminated psychotherapy. Additionally, the global measure of distress provided
by the SCL-90-R (GSI) was hypothesized to have a differential effect upon termination
depending upon whether the client was predominantly dominant or submissive.
Individuals rating themselves as predominantly dominant were posited to reveal lower
distress levels for those who remain in therapy longer. Individuals rating themselves as
predominantly submissive were posited to reveal higher distress levels for those
remaining in therapy longer. This hypothesis was not supported by the data.

Past studies have called the factor structure of the SCL-90-R into question.
Several of these studies have noted that the factor structure of the instrument changes
depending upon the population being studied. To gain a better understanding of the
population under study an exploratory factor analysis of the SCL-90-R was performed.
Six Factors were obtained. These six Factors explained 50.96% of the total variance
observed. Of this, 36.64% of the total variance was explained by the ﬁrst two rotated
factors, Depressed Interpersonal Sensitivity and Phobic Interpersonal Sensitivity. This
suggests that interpersonal problems are the complaint most commonly experienced by

individuals seeking therapy in this sample. Of interpersonal complaints reported, those of

44

a depressed nature explained the greatest portion of the variance, followed by those of a
phobic nature. It was the general supposition of this thesis that one of the reasons clients
terminate psychotherapy prematurely is that their preconceived understanding of the
therapeutic relationship is incompatible with their style of interpersonal functioning.

The interpersonal factors hypothesized to have an impact upon the client’s ability
to form an alliance with a therapist were those factors associated with self-perceived level
of dominance/ submission. This impact was postulated to reveal itself in the number of
sessions that the individual attended therapy. As discussed by Lorr (1991), clients begin
the therapeutic process by projecting onto the therapist the role of an authoritarian ﬁgure.
This representation should gradually be replaced by an egalitarian relationship as the
therapeutic alliance strengthens. Sullivan has posited that individuals shape relationships
in ways that reduce anxiety. The complimentarity portion of Leary’s Interpersonal
Circumplex model posits that individuals who are submissive will experience anxiety
reduction when relating to another who is dominant such as an authority ﬁgure.
Dominant individuals will experience elevated levels of anxiety when relating to a
dominant authority ﬁgure. Elevated levels of anxiety are aversive and should precipitate
premature termination.

Neither a signiﬁcant main effect nor a signiﬁcant linear trend was obtained when
this hypothesis was tested. However, a signiﬁcant quadratic trend did lend partial
support to this idea such that the Late Dropout Group displayed greater levels of
dominance than did the Persisters. The curvilinear trend was revealed in the results of the

Early Dropout Group whose level of dominance was almost as low as the Persisters.

45

Post Hoc analyses demonstrated that this relationship did not hold for the entire
sample. A signiﬁcant interaction between level of dominance and gender was
discovered. Men showed a linear trend in support of the hypothesized relationship of
level of dominance and point of termination. Early Dropouts revealed higher levels of
dominance than did Late Dropouts and Persisters evidenced the lowest level of
dominance. Women did not evidence this relationship. The women displayed a
curvilinear pattern of termination in which women who were in the Early Dropout group
were lower in dominance than those in the Late Dropout Group. Women Persisters were
also lower in dominance than the Late Dropout Group. That the pattern held for men and
not women would seem to indicate that dominance has a different meaning for the two
groups. Men would seem to be more consistently effected by an assumed hierarchical
relationship in a therapeutic situation than are women. While it does not seem to be the
case that women are not effected by an assumed hierarchical relationship in therapy, it
does seem as if women’s reactions to this relationship are different.

A possible explanation of this difference might lie in the differential manner in
which men and women have been socialized. Hankin and Abramson (2001) discuss
socialization as a precursor for gender differences in the development of depression.
Within their model, girls experience a greater number of interpersonal negative events
such as peer rejection during adolescence (Davies & Windle, 1997), as well as having
experienced higher levels of sexual abuse (Kaplan, Pelcovitz, & Labruna, 1999). “In
addition to encountering more negative events, girls also encode these events in greater
detail in large associative cognitive networks” (Hankin & Abramson, 2001, p. 785).

These factors, in turn lead to a greater cognitive vulnerability for depression. This

46

cognitive vulnerability is discussed in terms of speciﬁc cognitive attributions made in
speciﬁc situations. If one takes into account research on gender socialization that shows
mothers to be more controlling with their daughters than with their sons (Pomerantz &
Ruble, 1998), it would seem a reasonable step to infer that these cognitive attributions
coupled with gender socialization trends may also effect how relationships in general are
viewed.

Dominance and submission are inherently hierarchical concepts. Ifone assumes
that men have been raised to see themselves as active participants within hierarchies then
it would naturally follow that they would learn speciﬁc rules for participating within
those hierarchies. The necessity for assessing the level and type of participation would
be important to avoid conﬂict. Consistent with the notion of complimentarity, people
immersed in this understanding of hierarchy would be more prone to feeling anxious
when associating with someone with whom they might have to compete for a position
within the hierarchy. If one assumes that women are not socialized to participate within
hierarchies but rather to cope with the ramiﬁcations of those hierarchies then it would
seem evident that their understanding of the hierarchical structure itself would be much
different. Submission can then be seen as a coping strategy that serves to avoid those
dominant individuals who might harm them and please those who might help them.
Dominance could likewise be viewed as a method of achieving independence from a
hierarchical structure. If this is the case then it becomes apparent that submissive women
who view the therapist as a dominating threat would be more likely to avoid that threat
and terminate prematurely. Submissive women who view the therapist as a dominant

person who can help them will court the therapist’s favor and continue with treatment for

47

as long as the dominant therapist deems appropriate. Women who take a dominant
stance under this supposition would seem more likely to take control of their own therapy
and terminate at a point that they deem appropriate without relying solely upon the '
therapist’s judgment. While this explanation mirrors the results found within this study,
the conclusions that it offers are meant only as conjecture. More research is called for to
examine this discrepancy between men and women.

An exploration of the different gradations of the Interpersonal Circumplex with
respect to premature termination was also completed. It must be remembered that Leary
proposed using a circumplex to examine interpersonal relationships because he felt that
the divisions commonly made between personality structures did not capture the uniﬁed
nature of an individuals interpersonal presentation. This being the case the Dominance
measure of the Interpersonal Circumplex is derived from a geometric formula that
includes all of the separate categories contained within the Circumplex. These separate
categories of the Circumplex may be divided up into 8 independent constructs (octants)
or 16 independent constructs (sixteenths). It was Leary’s original intent to allow for each
of these constructs to be studied on it’s own or in concert with the ﬁlll circumplex. This
leads one to question the optimal number of categorical divisions that provide the
maximum amount of information. Toward this end, the individual octants and sixteenths
ﬁom which the circumplex is derived were analyzed.

This more detailed examination of the Circumplex did not bring greater clarity to
the question regarding the main effect of dominance on point of termination. The only
octants and sixteenths that were found to be signiﬁcantly related to point of termination

were those that were aligned with the status poles of the circumplex. The two octants

48

that had signiﬁcant curvilinear trends were the Autocratic/Managerial octant, which
corresponds with the dominant quadrant of the circumplex, and the Self-
Effacing/Masochistic octant, which corresponds with the submissive quadrant of the
circumplex. Four of the sixteenths were found to have signiﬁcant curvilinear trends.
These were the two sixteenths that, when compiled together, form the
Autocratic/Managerial octant as well as the two sixteenths that form the Self-
Effacing/Masochistic octant. The pattern of scores for each of these octants and
sixteenths followed the same pattern that the more global status measure of dominance
had found. This pattern was such that Early Dropouts displayed lower levels of
dominance (higher levels of submissiveness), Late Dropouts displayed higher levels of
dominance (lower levels of submissiveness), and Persisters displayed lower levels of
dominance (higher levels of submissiveness). While examination of the more global
Status dimension did reveal a signiﬁcant interaction with gender, the octants and
sixteenths did not. It must be concluded that the octants and the sixteenths do not add
additional information when assessing whether or not an individual is likely to terminate
psychotherapy prematurely. The status dimension of the Interpersonal Circumplex is the

most appropriate measure to use in this regard.

Further Researcir

This study explored just one of the dimensions within Leary’s interpersonal
Circumplex. His expanded diagnostic scheme includes not simply the individual’s self—
report ratings of interpersonal style (level of conscious communication) within one

situation, in this case therapy. Also included are: 1) level of public communication in

49

which signiﬁcant others provide ratings, 2) level of private perception in which the
Minnesota Multiphasic Personality Inventory (MMPI) is utilized, 3) level of the
unexpressed in which the Thematic Aptitude Test (TAT, Murray, 1943) is utilized with
speciﬁc rules for scoring the instrument, and 4) level of values. These four levels, in
concert with the level of conscious communications used within this study provide a ﬁrll
interpersonal diagnostic picture of a client. Future research on interpersonal factors
involved in premature termination from therapy would beneﬁt ﬁ'om making full use of
the entire diagnostic system.

While a considerable body if research has been devoted to the etiological
differences of gender, much less research has been devoted to understanding gender
differences in the psychotherapeutic process. More work needs to be done with respect
to the differential view that men and women have of the therapeutic situation at the outset
of therapy as well as during the process. Issues that should be investigated include
gender of the therapeutic dyad, interpersonal style of the therapist, as well as client
education about the therapeutic process they are entering prior to therapy. This research
will be important in establishing individualized therapeutic strategies that keep
individuals in therapy long enough to derive therapeutic beneﬁt.

While type of symptomatology was not found to be a signiﬁcant predictor of
premature termination, questions regarding the factor structure used to measure this
variable have been raised. Further research investigating this factor discrepancy is
needed. Additionally, this portion of the study should be replicated with an instrument

that provides greater symptomalogic detail.

50

APPENDIX A

INSTRUIVIENTS

51

Respondent #

THE INTERPERSONAL CHECKLIST

Instructions: Below is a lisot of descriptive words and phrases which you will use to
describe yourself. Read the items quickly and ﬁll in the circles of each item you consider
to be generally descriptive of yourself at the present time. Leave the circle blank when
an item does not describe you. In the example below, the subject has indicated that Item
A is true, and Item B is false as applied to him. That is, he says he is generally well-
behaved but not suspicious.

Example:
A. well-behaved .
B. suspicious 0

Your ﬁrst impression is generally the best so work quickly and don’t be concerned about
duplications, contradictions, or being exact. If you feel much doubt whether an item
applies, leave it blank.
ITEMS:

1. well thought of

2. makes a good impression
3. able to give orders

4. forceful

5. self-respecting

6. independent

7. able to take care of self

8. can be indifferent to others
9. can be strict if necessary
10. ﬁrm but just

11. canbe frank

12. critical of others

0000000000000

13. can complain if necessary

52

Respondent #
ICL Page 2 — if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank.
14. often gloomy
15. able to doubt others
16. frequently disappointed
17. able to criticize self

18. apologetic

19. can be obedient

20. usually gives in

21. grateful

22. admires and imitates others
23. appreciative

24. very anxious to be approved of
25. cooperative

26. eager to get along with others
27. friendly

28. affectionate and understanding
29. considerate

30. encourage others

31. helpful

32. big-hearted and unselﬁsh

33. often admired

OOOOOOOOOOOOOOOOOOOOO

34. respected by others

53

Respondent #

ICL Page 3 — if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank.
35. good leader

36. likes responsibility

37 . self-conﬁdent

38. self-reliance and assertive

39. businesslike

40. likes to compete with others
41. hard-boiled when necessary
42. stern but fair

43. irritable

44. straightforward and direct

45. resents being bossed

46. skeptical

47 . hard to impress

48. touchy and easily hurt

49. easily embarrassed

50. lacks self-conﬁdence

51. easily led

52. modest

53. often helped by others

54. very respectful of authority

OOOOOOOOOOOOOOOOOOOOO

55. accepts advice readily

54

Respondent #

ICL Page 4 - if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank.
56. trusting and eager to please

57. always pleasant and agreeable
58. wants everyone to like him/her
59. sociable and neighborly

60. warm

61. kind and reassuring

62. tender and soft-hearted

63. enjoys taking care of others

64. gives freely of self

65. always giving advice

66. acts important

67 . bossy

68. dominating

69. boastful

70. proud and self-satisﬁed

71. thinks only of him/herself

72. shrewd and calculating

73. impatient with others’ mistakes

74. self-seeking

75. outspoken

OO OOOOOOOOOOOOOOOOOOO

76. often unfriendly

55

Respondent #

ICL Page 5 — if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank
77. bitter

78. complaining

79. jealous

80. slow to forgive a wrong

81. self-punishing

82.shy

83. passive and unaggressive

84. meek

85. dependent

86. wants to be led

87. lets others make decisions

88. easily fooled

89. to easily inﬂuenced by friends
90. will conﬁde in anyone

91. fond of everyone

92. likes everybody

93. forgives anything

94. over-sympathetic

95. generous to a fault

96. overprotective of others

0 0 000 O 0 000 OO 00 CO OO O O O

97. tries to be too successful

56

Respondent #

ICL Page 6 — if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank

98. expects everyone to admire
him/her

99. manages others

100. dictatorial

101. somewhat snobbish

102. egotistical and conceited
103. selﬁsh

104. cold and unfeeling

105. sarcastic

106. cruel and unkind

107. frequently angry

108. hard-hearted

109. resentﬁrl

110. rebels against everything
111. stubborn

112. distrusts everybody

113. timid

114. always ashamed of self
1 15. obeys too willingly

116. spineless

OOOOOOOOOOOOOOOOOOOO

117. hardly ever talks

57

Respondent #

ICL Page 7 — if the item generally describes you
Fill in the circle. If the item does not generally
Describe you, leave the circle blank
118. clinging vine

1 19. likes to be taken care of

120. will believe anyone

121. wants everyone’s love

122. agrees with everyone

123. ﬁiendly all the time

124. loves everyone

125. too lenient with others

126. tries to comfort everyone

127. too willing to give to others

O 0000000000

128. spoils pe0ple with kindness

58

Respondent #
SCL-90-R

Sex: Male 0 Female 0

Instructions: Below is a list of problems people sometimes have. Please read each one
careﬁrlly, and circle the number to the right that best describes HOW MUCH THAT
PROBLEM HAS DISTRESSED OR BOTHERED YOU DURING THE PAST 7 DAYS
INCLUDING TODAY. Circle only one number for each problem and do not skip any
items. If you change your mind, erase your ﬁrst mark carefully. If you have any
questions please ask about them.

HOW MUCH WERE YOU DISTRESSED BY:

1. Headaches
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

2. Nervousness of shakiness inside
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

3. Repeated unpleasant thought that won’t leave your mind
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

4. Faintness of dizziness
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

5. Loss of sexual interest
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

6. Feeling critical of others
0 1 2 3 4

Not at all A little bit Moderately Quite a bit Extremely

7. The idea that someone else can control your thoughts
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

8. Feeling others are to blame for most of your troubles

0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

59

Respondent #
SCL-90-R: Page 2

HOW MUCH WERE YOU DISTRESSED BY:

9. Trouble remembering things
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

10. Worried about sloppiness or carelessness
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

11. Feeling easily annoyed or irritated
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

12. Pains in heart or chest
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

13. Feeling afraid in open spaces or on streets
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

14. Feeling low in energy or slowed down
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

15. Thoughts of ending your life
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

16. Hearing voices that other people do not hear
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

17. Trembling
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

18. Feeling that most people cannot be trusted

0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

60

SCL—90-R: page 3
HOW MUCH WERE YOU DISTRESSED BY:

19. Poor appetite

0 1 2 3
Not at all A little bit Moderately Quite a bit
20. Crying easily
0 1 2 3
Not at all A little bit Moderately Quite a bit
21. Feeling shy or uneasy with the opposite sex
0 1 2 3
Not at all A little bit Moderately Quite a bit
22. Feelings of being trapped or caught
0 1 2 3
Not at all A little bit Moderately Quite a bit

23. Suddenly scared for no reason

0 1 2 3
Not at all A little bit Moderately Quite a bit
24. Temper outbursts that you could not control
0 l 2 3
Not at all A little bit Moderately Quite a bit
25. Feeling afraid to go out of your house alone
0 1 2 3
Not at all A little bit Moderately Quite a bit
26. Blaming yourself for things
0 1 2 3
Not at all A little bit Moderately Quite a bit
27. Pains in lower back
0 1 2 3
Not at all A little bit Moderately Quite a bit
28. Feeling blocked in getting things done
0 1 2 3
Not at all A little bit Moderately Quite a bit

61

Respondent #

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

Respondent #
SCL-90-R: page 4

HOW MUCH WERE YOU DISTRESSED BY:

29. Feeling lonely
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

30. Feeling blue
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

31. Worrying too much about things
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

32. Feeling no interest in things
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

33. Feeling fearful
O l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

34. Your feelings being easily hurt
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

35. Other people being aware of your private thoughts
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

36. Feeling that others do not understand you or are unsympathetic
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

37. Feeling that people are unfriendly or dislike you
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

38. Having to do things very slowly to insure correctness

0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

62

SCL-90—R: page 5
HOW MUCH WERE YOU DISTRESSED BY:
39. Heart pounding or racing
0 1 2 3
Not at all A little bit Moderately Quite a bit

40. Nausea or upset stomach

0 1 2 3
Not at all A little bit Moderately Quite a bit
41. Feeling inferior to others
0 1 2 3
Not at all A little bit Moderately Quite a bit

42. Soreness of your muscles
0 1 2 3
Not at all A little bit Moderately Quite a bit

43. Feeling that you are watched or talked about by others

0 l 2 3
Not at all A little bit Moderately Quite a bit
44. Trouble falling asleep
0 1 2 3
Not at all A little bit Moderately Quite a bit

45. Having to check and double-check what you do

0 1 2 3
Not at all A little bit Moderately Quite a bit
46. Difﬁculty making decisions
0 l 2 3
Not at all A little bit Moderately Quite a bit
47. Feeling afraid to travel on buses, subways, or trains
0 1 2 3
Not at all A little bit Moderately Quite a bit
48. Trouble getting your breath
0 1 2 3
Not at all A little bit Moderately Quite a bit

63

Respondent #

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

4
Extremely

Respondent #
SCL—90—R: page 6

HOW MUCH WERE YOU DISTRESSED BY:

49. Hot or cold spells
O l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

50. Having to avoid certain things, places, or activities because they ﬁ'ighten you
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

51. Your mind going blank
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

52. Numbness or tingling in parts of your body
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

53. A lump in your throat
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

54. Feeling hopeless about the ﬁlture
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

55. Trouble concentrating
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

56. Feeling weak in parts of your body
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

57. Feeling tense or keyed up
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

58. Heavy feelings in your arms or legs

0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

64

Respondent #
SCL-90-R: page 7

HOW MUCH WERE YOU DISTRESSED BY:

59. Thoughts of death or dying
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

60. Overeating
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

61. Feeling uneasy when people are watching or talking about you
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

62. Having thoughts that are not your own
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

63. Having urges to beat, injure or harm someone
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

64. Awakening in the early morning
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

65. Having to repeat the same actions such as touching, counting, or washing
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

66. Sleep that is restless or disturbed
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

67. Having urges to break or smash things
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

68. Having ideas or beliefs that others do not share

0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

65

Respondent #
SCL-90-R: page 8

HOW MUCH WERE YOU DISTRESSED BY:

69. Feeling very self-conscious with others
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

70. Feeling uneasy in crowds, such as shopping or at a movie
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

71. Feeling everything is an effort
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

72. Spells of terror or panic
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

73. Feeling uncomfortable about eating or drinking in public
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

74. Getting into frequent arguments
O l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

75. Feeling nervous when you are left alone
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

76. Others not giving you proper credit for your achievements
O l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

77. Feeling lonely even when you are with people
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

78. Feeling so restless you couldn’t sit still

0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

66

Respondent #
SCL-90-R: page 9

HOW MUCH WERE YOU DISTRESSED BY:

79. Feelings of worthlessness
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

80. The feeling that something bad is going to happen to you
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

81. Shouting or throwing things
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

82. Feeling afraid you will faint in public
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

83. Feeling that people will take advantage of you if you let them
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

84. Having thoughts about sex that bother you a lot
0 l 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

85. The idea that you should be punished for your sins
O 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

86. Thoughts and images if a frightening nature
0 l 2 3 ‘ 4
Not at all A little bit Moderately Quite a bit Extremely

87. The idea that something serious is wrong with your body
0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

88. Never feeling close to another person

0 1 2 3 4
Not at all A little bit Moderately Quite a bit Extremely

67

SCL-90-R: page 10
HOW MUCH WERE YOU DISTRESSED BY:
89. Feelings of guilty
0 1 2 3
Not at all A little bit Moderately Quite a bit
90. The idea that something is wrong with your mind

0 l 2 3
Not at all A little bit Moderately Quite a bit

68

Respondent #

4
Extremely

4
Extremely

REFERENCES

69

REFERENCES

Beckham, E. (1994). Predicting patient dropout in psychotherapy.
Psychotherapy. 29, 1 77- 1 82.

Carson, R. B. (1969). InteLaction concepts ofpersogrlity. Chicago: Aldine.

Chiles, J. A., Lambert, M., Hatch, A, L. (1999). The impact of psychological
interventions on medical cost offset: A meta-analytic review. Clinical Psychology;

Science and Practice, 6(2), 204 —2 20

 

Chisholm, S., Crowther, J ., & Ben-Porath, Y. (1997). Selected MMPY-2 scales’
ability to predict premature termination and outcome from psychotherapy. Journal of

Personality Assessment. 69(1). 127 — 144.

 

Dahlsgaard, K., Beck, A., & Brown, B. (1998). Inadequate response to therapy
as a predictor of suicide. Suicide and Life-Threateniag Behavior; 28, 197-204.

Derogatis, L., Rickels, K., & Rock, A. (1976). The SCL-90-R and the MMPI: A
step in the validation of a new self-report scale. British Journal of Psychiatry. 128, 280-
289.

Derogatis, L. R. (1977). The SCL-90 Manual 1: Scoring administration grad
procedures for the SCL-90. Baltimore: Johns Hopkins University School of Medicine,
Clinical Psychometrics Unit.

Derogatis, L R., & Cleary, P. (1977). Conﬁrmation of the dimensional structure
of the SCL-90: A study in construct validation. Journal of Clinical Psychologxﬁ, 981-

989.

70

Derogatis, L., Meyer, J ., & King, K. (1981). Psychotherapy in individuals with
sexual dysfunction. Ameriaan Journal of Psychiatry. 138. 757-763.

Derogatis, L. R. (1983). SCL-9O Administration scoringand procedures maaual
for the revised version. Baltimore: Clinical Psychometric Research.

Derogatis, L., & Spencer, P. (1982). The Brief Symptom inventory:
Administratioa, scoring and procedures mama]. 1. Leonard R Derogatis.

Epperson, D., Bushway, D., & Warman, R. (1983). Client self termination from
counseling after one counseling session: Effects of problem recognition, counselor
gender, and counselor experience. Jourrral of Counseling Psychologxlg, 307-315.

Frayn, D. (1992). Assessment factors associated with premature psychotherapy
termination. American Journﬂf Psychotherapy,_4§, 250-261.

Gabbard , G., Lazar, S., Homberger, J ., & Spiegel, D. (1997). The economic
impact of psychotherapy: A review. American Journal of Psychgtry. 154. 147-155.

Guerney, B. G., Meininger, S., & Stover, L. (1973). Normative and validation
studies of the Interpersonal Check List for young children as seen by their mothers.

J oumal of Clinical Psychology. 29(2),_ 219 — 225

Guze, S. B., & Robins, E. (1970). Suicide and primary affective disorders. m
Jougal of Pajachiatry. 19, 437-448.

Gynther, M., Miller, E, & Davis, H. (1962). Relations between needs and
behavior as measured by the Edwards PPS and Interpersonal Check List. Journal of

Social Psycholggy. 57, 445-451.

7]

Hamilton, D. (1971). A comparative study of ﬁve methods of assessing self-

 

esteem, dominance, and dogmatism. Education and Psycholoﬂl Measurement. 31.
441-452.

Hilsenroth, M., Handler, L., Tornan, K., & Padawer, J. (1995). Rorschach and
MMPI-Z indices of early psychotherapy termination. Jourpal of Consultingand Clinical

Psycholagy.63(6): 956-965.

 

Klerman, G., Weissman, M., Roundaville, B., & Chevron, E. (1984).
Interpersch Psychotherapy of Depression. United States: Basic Books.

Laforge, R., Leary, T., Naboisek, H., Coffey, H., & Freedman, M. (1954). The
interpersonal dimensions of personality: II. An objective study of repression. Journal of

Persomrlity. 23. 129-153

 

LaForge, R., & Suczek, R, (1955). The interpersonal dimensions of personality:
111 An interpersonal checklist. Journal of Personlaity. 24. 94-112.

LaForge, R. (1977). Using the ICL. Mill Valley, CA: Author.

 

LaForge, R., Freedman, M. B., & Wiggins, J. (1985). Interpersonal circumplex
models: 1948 — 1983 (Symposium), Journal of Personality Assessment, 42(6), 613 —
621.

Leary, T. (1957). Interpersonﬂiagnosis of Permality. New York: Ronald
Press.

Leary, T., & Coffey, H. (1955). Interpersonal diagnosis: Some problems of

methodology and validation. Journal of Abnormal and Social Psychology, 34, 72-79.

72

Levinson, P., McMurray, L., Podell, P., & Weiner, H. (1978). Causes for
premature interuption of psychotherapy by private practice patients. American Journal of
Psychiatry. 135. 826-829.

Lorr, M. (1991). A redeﬁnition of dominance. Perscmalitv and Individual

 

Differences. 12. 877-879.

Lorr, M., & McNair, D. (1965). Expansion of the interpersonal behavior circle.
Journal of Personalitav and Social Psychology,_2, 823-830.

Paddock, J. (1982). The Interpersonal Check List—2a relialﬂyand validity study.
Unpublished master’s thesis, Emory University, Atlanta, GA.

Paddock, J., Woodruff, D., & Pate, C. (1984). Assessing nonspeciﬁc
relationships factors in the behavioral treatment of severely emotionally disturbed (SED)
adolescents. Journal of Clinical Child Psycholggy,_1_3_, 74-80.

Paddock, J ., & Nowicki, S. (1986). An examination of the Leary Circumplex
through the Interpersonal Check List. Journal of Research in Personality. 20. 107-144.

Pekarik, G. (1983). Follow-up adjustment of outpatient dropouts American
Journal of Orthopsychiatry. 53, 501-511.

Pekarik, G., & Wierzbicki, M. (1986). The relationship between clients’
expected and actual treatment duration. Psychotherapy, 23, 532 — 534.

Pekarik, G. (1992). Posttreatment adjustment of clients who drop oout early vs.
late in treatment. Journal of Clinical Paychology, 48, 379-387.

Piper, W., Joyce, A., Azim, H., & Rosie, J. (1994). Patient characteristics and

success in day treatment. Jourrgl of Nervous mental Disease. 182. 381-3 86.

 

73

Plutchik, R., & Conte, H. (1997). Circumplex Models of Personality.
Washington, DC: American Psychological Association.

Prusoﬁ", B., Weissman, M., Klerman, G., & Rounsaville, B. (1980). Research
diagnostic criteria subtypes of depression: Their role as predictors of differential response
to psychotherapy and drug treatment. Archives-of-General-Psychiatry. 37(7): 796-801.

Regier, D, Narrow, W., Rae, D., Manderscheid, R., Locke, B. Goodwin, F.
(1993). The de facto U.S. mental and addictive disorders service system: Epidemiologic
Catchment Area prospective l-year prevalence rates of disorders and services. Archives
of General Psychgrygy, ﬂ, 85-94.

Seligman, M. (1995). The effectiveness of psychotherapy: The Consumer
Reports study. American-Psychologist. 50(12): 965-974.

Sullivan, H. S. (1953). The Interpersonal Theory of Psychim. New York:
Norton.

Sullivan, H. S. (1954). The Psycﬂrtric Interview. New York: Norton.

Truckenmiller, J ., & Shaie, K. (1979). Multilevel structural validation of Leary’s
interpersonal diagnosis system. Journi of Consulting and Clinical Psychology._4Z,
1030-1045.

Tryon, G. S. (1985). Session depth and smoothness in relation to the concept of
engagement in counseling. Journal of CounselinLPsychology,_3_’_7 (3), 248-253.

Tryon, G. S. (1990) The engagement quotient: One index of a basic counseling

task. Joumﬁ of Collage Student Personnel. 2Q 351-3 54.

74

Vassend, O., & Skrondal, A. (1999). The problem of structural indeterminacy in
multidimensional symptom report instruments. The case of SCL-90-R. Behaviour
Research and TherapyﬂU), 685 — 701.

Van Buren, A., & Nowicki, S. (1997). Awareness of interpersonal style and self-
evaluation. The Journal of Soaial Psychology,1_37_, 429-434.

Vaughn, E., & Nowicki, S. (1999). Close relationships and complementary
interpersonal styles among men and women. The Journalaof Social Psychology. 139.
473-478.

Waskow, I., & Parloff, M. (1975). Psychotherapy change measures.
Washington, DC: DHEW.

Wierzbicki, M., & Pekarik, G. (1993). A meta-analysis of psychotherapy

dropout. Journal of Professional Psychology: Research and motice. 24, 190-195.

 

Wiggins, J. (1985). Symposium: Interpersonal circumplex models: 1948 —1983
(Commentary). Journal of Personality Assessment, 49, 626-631.

Wiggins, J. S. (1991). The Interpersonal Adjective Scale manual. Lutz, FL:
Psychological Assessment Services.

Wolman, B. (1973). Dictionary of Behavioral Science. New York: Van Nostrand

Reinhold.

75

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

ll131lljjlljljyjljjljl1ij