Title Page
 Table of Contents
 Literature review
 Biographical sketch

Group Title: role of subject maturity in the determination of preferred essential therapist characteristics /
Title: The role of subject maturity in the determination of preferred essential therapist characteristics /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098632/00001
 Material Information
Title: The role of subject maturity in the determination of preferred essential therapist characteristics /
Physical Description: viii, 127 leaves : ; 28 cm.
Language: English
Creator: Stone, David Reed, 1947-
Publication Date: 1980
Copyright Date: 1980
Subject: Psychotherapist and patient   ( lcsh )
Clinical Psychology thesis Ph. D   ( lcsh )
Dissertations, Academic -- Clinical Psychology -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 122-126.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by David R. Stone.
 Record Information
Bibliographic ID: UF00098632
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000097613
oclc - 06587965
notis - AAL3054


This item has the following downloads:

PDF ( 5 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
        Page v
        Page vi
        Page vii
        Page viii
        Page 1
        Page 2
        Page 3
    Literature review
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
    Biographical sketch
        Page 127
        Page 128
        Page 129
        Page 130
Full Text







The names of a great many kind and tolerant people should be

placed next to the investigator's on this dissertation. The members

of my doctoral committee deserve top billing, especially Dr. Audrey

Schumacher, whose patient, guidance and continued support of my

work have finally borne fruit. Dr. Larry Severy also has spent a

great deal of time and energy in providing technical assistance and

encouragement on this project, without which it might never have

been completed. The rest of the committee--Dr. Louis Cohen,

Dr. Hugh Davis, and Dr. Marilyn Holly--also have been quite tolerant

and helpful throughout the duration of this research.

Special thanks are extended to Ms. Diane Peters, Dr. Stan Jones,

and Ms. Lore Nielsen, for the many hours they contributed in rating

the therapist tapes, and for also being very supportive of my work.

My appreciation is offered to another friend, Mr. Michael Miller,

who gave a great deal of time to the portrayal of the "client"

role on the therapist tapes. Finally, a sincere debt of gratitude

is owed to a very special friend, Ms. Cheryl Boggess, who ably

met the challenge of translating and typing a very rough draft

into an acceptable final copy.

Most of all, I would like to thank a very special young man,

to whom this and all my work is dedicated: my son, Bryan. He makes

all of my efforts seem worthwhile.







Background Research: Mechanisms of Preference Development 4

Theoretical Formulations 4

Empirical Research 8

The "transference hypothesis" 8

Attitudinal analysis: essential elements of
client attitudes 12

Patient Preferences 17

Basic Therapeutic Roles 17

Modalities of therapist providing behavior 19

Modalities of therapist modeling behavior 21

Dimensions of Therapist Providing Behavior 23

Therapist activity: directiveness vs.
nondirectiveness 23

Therapist affect: warmth vs. detachment 32

Therapist cognition: awareness vs. non-awareness 40

Dimensions of Therapist Modeling Behavior 42

Competence 42

Autonomy 44

Self-confidence 45


Independent Variables: Client Maturity, Dependency
and Sex 47

Maturity and dependency 47

Sex 49

Summary and Implications 51

Summary 51

Implications 55

Hypotheses 58

Therapist Preferences 58

Mediating Variables: Subject Anxiety and Dependency 62

Early Development of Therapist Preferences 64

Operational Definitions 65


Subjects 68

Instruments 68

Design 70

Procedures 72



General Therapist Preferences 96

Subject Variables Related to Maturity: Anxiety and
Au tonomly 100

Developmental Origins of Therapist Preferences 112







Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



David R. Stone

March, 1980

Chairman: Audrey S. Schumacher, Ph.D.

Major Department: Department of Clinical Psychology

The present investigation addressed several problems pertaining

to preferences of therapist behavior. A theoretical framework was

proposed to facilitate a comprehensive understanding of therapist

behaviors. A distinction was made between expectations and preferences

of therapists, and between measures of hypothetical and actual thera-

pist qualities. A primary goal was to explore relationships between

subject personality factors, such as anxiety and autonomy, and the

choice of characteristics for therapist and father figures, as a step

toward understanding the process of development of therapist pre-

ferences, and predicting those choices for individuals.

Eighty-eight psychology undergraduates completed a personal data

sheet, the 16 PF, and the Ideal-Self/Others Rating Scale (ISORS). The

first two measures provided information about the subjects' background

and personality, while the third indicated their idealistic preferences

for four figures, including a therapist and a father. Subjects then

listened to two five-minute audiotaped versions of analogue therapy

sessions. The "model" therapist tape was constant across all eight

conditions, and thus was heard by every subject. The "provider"

therapist tapes varied with each condition, with a total of eight

such tapes being utilized. The model therapist was scripted to exhibit

high levels of personally-oriented qualities, i.e. self-awareness,

self-confidence and independence, but moderate levels of interpersonally-

oriented characteristics, i.e. intuition, warmth and advice/guidance.

The eight provider therapists were scripted for moderate levels of

the personal qualities, but offered varying (high-low) levels of the

interpersonal characteristics, resulting in a 2X2X2 configuration.

After listening to the provider therapist tape assigned to their

condition and that standard model tape, subjects completed the Client

Reaction Questionnaire (CRQ), which rated satisfaction with the

therapists' performance.

Data analysis offered mixed results with respect to the hypotheses.

As a group, the subjects stated a significant preference for a thera-

pist who was guiding, intuitive, self-confident, self-aware and

warm (p<.001). Paired t-tests indicated that of the three basic

qualities of interpersonal behavior--warmth, directiveness and

awareness--the latter played the largest role in determining over-

all subject preferences (p.001). Grouping subjects into a 2X2

matrix according to their high vs. low scores on 16 PF Anxiety and

Independence factors did not substantially alter their ideal-therapist

ratings, except that those subjects with high anxiety scores preferred

a wanm therapist significantly more often than did low anxiety scorers

(p<.001). Whether high or low in independence, low anxiety subjects

cxpressred qreater satisfaction with the model therapist than with the

group of provider therapists (p,.001). The same was true of high

anxiety/low independence subjects, although to a somewhat lesser

extent (p<.01). However, selected provider therapists tended to be

rated at least equally as satisfactory as the model by several subject

subgroups. Neither subject anxiety nor independence was found to be

significantly related to ideal-father preferences. However, subjects

generally exhibited substantial agreement that the ideal father and

therapist should be confident and intuitive leaders, and somewhat warm.

The findings suggest several conclusions. First, therapist

awareness appears to be essential in determining males' preferences

of therapist qualities. Second, there is a discrepancy between

preferred therapist qualities to which males admit and those they

find most satisfactory, with the individual's anxiety level playing

a mediating role in his satisfaction ratings. Third, therapist pre-

ferences are related to subjects' attitudes toward father figures,

and thus might have their roots in childhood. Further investigation

into the relationship between various client personality characteristics

and satisfaction measures of therapist behavior could help elucidate

this theoretical issue and aid in dealing with the practical problem

of premature termination from therapy by male clients.


As a relatively recent entry into the competition for the

clinical investigator's interest, the subject of the kinds of

behavior the client seeks in a therapist has only been a contender

for two decades. Nevertheless, in that brief period of time, the

study of client preferences of therapeutic behavior has attracted

substantial empirical attention. No doubt part of this attention

is due to the fact that measurement of such preferences can be made

comparatively simple and straightforward, a major advantage to any

problem in psychotherapy research. However, a more important factor

probably has been the rising interest in premature client termination,

a phenomenon which appears to be related to the client's pre-existing

attitudes toward the anticipated therapist (Heine and Trosman, 1960;

Heller and Goldstein, 1961; Kumler, 1968). Nonetheless, little

success has been achieved in elucidating some of the more basic

variables which contribute to the determination of these pre-

therapy attitudes. The present investigation will examine the role

that two variables related to maturity, i.e. the subject's levels of

anxiety and dependency, play in his adoption of a particular

attitudinal posture toward an unknown and unseen therapist.

The bulk of the applicable literature tends to be theoretical

and, therefore, little attention has been given to the mode of

development of one's preferences of therapist behavior. Yet,

preliminary consideration of how such attitudes might typically

originate is useful in understanding the rationale behind one's

adoption of attitudes that are seen in adulthood. Especially

intriguing to the present investigator is the question of a possible

linkage between a prospective client's initial attitudes toward an

unknown therapist and his idealistic perceptions of his parents.

That such a connection might exist, at least for selected individuals,

is suggested by the relevant literature. Therefore, Section I of the

following research review will consist of a cursory exploration of

possible mechanisms of development of these therapist preferences,

and will emphasize Freud's notion of transference, which has stimu-

lated a significant empirical response. This discussion should supply

background information useful in studying the issue of client

preferences within some semblance of an historical perspective, as

well as facilitate consideration of initial client attitudes toward


From the discussion of the development of therapist preferences,

the review will then proceed to their analysis and evaluation. It

will commence on a broad scale by attempting to specify the basic

types of behavioral orientations available to the therapist, termed

"interpersonal" and "personal." These orientations will be examined

in the context of two illustrative therapist roles, which will be

termed "provider" and "model." It will be suggested that the

"provider" role revolves around the therapist-client interaction,

while the role of "model" is more a reflection of the therapist's

feelings about himself.

Next, several different modalities of therapist behavior will be

explored which seem to be illustrative of the above two general roles.

It will be suggested that there exists three such modalities ("cogni-

tion," "affect" and "activity") which can account for most, if not

all, therapist behaviors, and that these modalities apply to both

the interpersonal orientation of the "providing" therapist and the

"personal" orientation of the "modeling" therapist as well.

The discussion will then turn to the delineation of the major

dimensional characteristics of therapist behavior which exemplify the

more general modalities mentioned above. It will be these character-

istics which will be measured directly in the present investigation.

Consideration finally will be given to the subject variables of

anxiety and dependency, which are implicated by the available research

as playing an essential role in the determination of an individual's

specific pre-disposing attitudes toward, and post-therapy satisfaction

with, a therapist. It is believed that by studying such attitudes on

this level and in light of existing data pertinent to these variables,

inferences can be made relevant to the basic issue of premature client

termination in male clients.


Background Research: Mechanisms of Preference Development

Theoretical Formulations

Freud (1935) was the first clinician to describe a pattern of

patient attitudes which did not seem to correlate to the therapist's

behavior. He labeled the phenomenon "transference," and portrayed

it as a capricious yet essentially passive entity, often insidious

but at times painfully obvious in its manifestations. The term

itself derived from Freud's belief that the treatment situation itself

could not account for the origin of the powerful emotions often

directed at the therapist. Rather, he felt that transference repre-

sented a pre-existing tendency to respond to the therapist in

patterns originally adopted as a means of resolving conflicts with

parents. Freud viewed this behavior disposition as essentially

benign, and even helpful, so long as its effect was favorable to

the work in which the two persons were cooperating. However, he

believed that when it becomes transferred into a resistance,

attention must be paid to it (Freud, 1935).

A summary of the psychodynamic theory's conception of the origin

and characteristics of transference is offered by Crisp (1964b). The

patient's initial attitude toward the therapist is typically a

composite of feelings termed "positive transference." This reaction

is based upon an attitude of idealization, in which the therapist

is viewed as excessively benign, dependable and concerned for the

client's welfare. Often associated with this idealization is a

dutiful, submissive, hopeful and dependent attitude on the client's

behalf. Idealization is seen as being first developed in relation

to earlier authoritarian figures, usually the father and/or mother.

This attitude is partially the result of a defensive, socializing

adjustment by the young child when the expression of his feelings

of hostility toward the restraining parent proves unacceptable, and

he is punished. Such punishment and criticism, or its anticipation,

leads to anxiety associated with suppression or repression of the

capacity to respond aggressively. The emerging conscious attitude

is idealization, associated with dependency, submission, and a

desire to please, which is viewed as a reaction formation against

aggressive impulses.

As Crisp (1964b) indicates, such an individual's attitude toward

authority figures, including the therapist, is viewed as neurotic

and potentially bipolar. If these figures do not respond to his

submissive needs and dependent demands, or fulfill the idealized

role in some way, then his basic, subconscious feelings of anger may

be evoked, leading to the conscious experience of considerable

anxiety. When the subconscious hostility does finally erupt in

the therapeutic situation, it is termed "negative transference"

and serves as a primary target of the therapist's efforts.

Thus, the psychoanalytic conceptualization of transference

involves idealization, hostility and dependency on the part of the

client. Another element which often looms as an obstacle to

progress in therapy is the patient's sexual feelings for the therapist.

However, such feelings usually remain below consciousness until the

therapeutic relationship has existed for some time, and therefore are

believed to have little effect on the client's initial attitudes.

The views of contemporary writers toward the transference

phenomenon have tended to stray from that of traditional psycho-

analytical thought, as evidenced by Rollo May's experiential con-

ceptualization. May (1958) believes that the neurotic individual

does not transfer feelings he once felt toward his parents to his

spouse or therapist. Rather, because of a lack of development

beyond the restricted forms of experiencing of infancy, such an

individual later comes to perceive spouse or therapist through the

same distorted "spectacles" as he once perceived father or mother.

May calls for an understanding of the problem in terms of perception

and one's relationship to the world, which he believes obviates the

necessity of viewing transference in the sense of a displacement of

detachable feelings from one object to another.

Although he retains the term "transference" to apply to

strongly-held interpersonal attitudes, Apfelbaum (1958) presents

an experiential formulation of the development of initial attitudes

toward a therapist that is quite similar to that of May (1958).

Apfelbaum reasons that when transference is seen as a unitary

phenomenon, ever present yet unfolding as a relationship progresses,

it then becomes a small step to identify transference with the

interpersonal facet of character. Earlier interpersonal attitudes

can persist and influence the perception of others in the present,

not because of a compulsion to repeat, as implied in the analytic

interpretation, but because they have become an integrated part of

the individual's life pattern.

While the essential nature of transference for Freud was

passive, for Apfolbaum (1958) it is a complex of attitudes which

exhibit a very active character:

S. 'transference' here refers to relatively
inflexibly-held beliefs regarding human nature,
that is, the behavior and motives of other
persons. These attitudes mould interpersonal
experience, creating misperceptions and in-
appropriate responses to others. (p. 4)

Apfelbaum believes that transference operates within the patient

upon his initial contact with the clinic, and does not develop

only as therapy proceeds. In this framework, therefore, the

patient's "transference attitudes" mediate his initial experiences

with the therapist, causing him to misperceive the therapist's

behavior in accordance with the nature and strength of the

"transference attitudes."

This association between one's attitudes toward parents and

therapist was incorporated into an interpersonal theory of

personality development proposed by Schutz (1958). This theory

hypothesizes the existence of three fundamental, universal needs,

known as "inclusion," "control," and "affection." The second major

postulate of theory is of special relevance to this study:

An individual's expressed interpersonal behavior
will be similar to the behavior he experienced in
his earliest interpersonal relations, usually
with his parents, in the following way:
Principle of Constancy: When he perceives his
adult position in an interpersonal situation to

be similar to his own position in his parent-child
relation, his adult behavior positively covaries
with his childhood behavior toward his parents (or
significant others). (p. 1967)

Schutz effectively avoids the transference issue completely, choosing

instead to explain this perceived similarity between parent and

therapist in terms of a general developmental task of dealing with

adult authority figures. This avoidance of the transference notion

illustrates its lack of utility in describing such an attitudinal

and behavioral similarity, should one actually exist.

In the presentation of the above theoretical formulations,

complexity and completeness have been forsaken for brevity. It

is apparent that all share one similarity, in that they hypothesize

a relationship between one's attitudes toward parents and those toward

other significant adults, particularly a psychotherapist. In the

section to follow, it will be observed that even this basic hypothesis

is open to question, and that there is little sound empirical

evidence to support any of the above theoretical views.

Empirical Research

The "transference hypothesis"

As one might expect, the experimental investigation of client's

attitudes toward parents and therapists has proceeded along several

lines of study. One of these has had as its focus the evaluation

of the question of the existence of transference in the therapeutic

domain, although as indicated above, proof of the phenomenon's

existence always requires an inference from the data. Perhaps the

earliest study of this question, which is often termed the "trans-

ference hypothesis," was conducted by Chance (1952). She began by

securing the attitudes expressed toward her as therapist of eight

members of a therapy group during the first twelve sessions. These

patients' statements were then rated by graduate students as

friendly, hostile or ambivalent. Later, the patients, who allegedly

no longer remembered having made the statements, rated each one

twice: once as it applied to the relationship with their therapist,

and once for that with their "significant parent," who was identified

prior to the study by examination of individual case information.

Chance found significant correlations between the two ratings for

five of the eight patients. These data lead the investigator to

conclude that many patients assume similar conscious attitudes toward

their significant parent and their therapist, resulting in the

inference that the patients "transfer" their attitudes from their

parents onto the therapist. Of course, this inference was never

directly tested.

There is a host of glaring weaknesses in the methodology of

Chance's (1952) investigation, three of which are noted by Meltzoff

and Kornreich (1970): (1) the proceedings of the group sessions were

recorded from memory, (b) the majority of three judges ruled in the

final selection of patient statements from an originally larger pool,

and (c) the ratings of the therapist were always made prior to the

ratings of the patients, leading to the possibility of an order

effect. In addition to these problems, two which are more obvious

and possibly more important are the small sample size and a

possibility of a lack of patient candor in owning their earlier

statements, which were incorporated into the principal instrument

of the investigation.

A somewhat different approach to investigating the "transference

hypothesis" was taken by Sechrest (1962), who utilized Kelly's Role

Construct Repertory Test (Kelly, 1955) to compare patients'

attitudes toward their therapist with their views of 15 other

significant persons in their lives. The investigator administered

the instrument to 28 hospitalized and 11 clinic patients once

around the time of the second interview and again around the eleventh

interview. Sechrest found that at both points during treatment the

therapist was likely to be described as being most similar to the

physician, the minister and the liked teacher, and was far less

likely to be described as similar to either parent or to a member of

the patient's family. Therefore, the patients' reactions to other

professional persons whom the therapist was seen as being most like

and their reactions to the therapist were very similar. Sechrest

concluded that patients' responses to a psychotherapist are often

elicited by his most obvious stimulus characteristics, rather than

by the patients' early experiences with parents.

The study by Sechrest (1962) has its difficulties, the most

problematic of which is the question of experimental demand

characteristics. It is not clear whether the subjects were very

aware of the range of characteristics they had with which to compare

the various persons on the Repertory. Instead, the qualities rated

were left to the subjects' discretion. It seems little wonder that

they chose the most obvious ones rather than the potentially more

sensitive personality characteristics which might be more closely

related to the transference question.

Crisp (1964a, b) adopted a method similar to that of Sechrest

(1962), but eliminated the weakness mentioned above. He added 36

personality constructs to a Kelly grid and administered the instru-

ment to several groups of patients and nonpatients varying in social

class. Crisp found a modicum of evidence supporting both the

existence of a similarity in attitudes toward parent and therapist,

as well as the grid method of maintaining a record of changes in

patients' attitudes. However, his claim of support for the trans-

ference phenomenon, like those of investigators using similar

designs, was unfounded in that it was purely inferential.

The final study pertinent to the issue of a possible relation-

ship between one's attitudes toward parent and therapist was con-

ducted by Berzon (1962), and based upon Schutz's interpersonal

theory of personality (Schutz, 1958). Berzon used two therapy

groups, one of which was predominately female (N = 13), the other

mostly male (N = 12). Both tended to be college-educated and of

upper socioeconomic class. Measures included the Perception-of-

Parents Behavior Questionnaire, a Perception-of-Others sort to

describe other group members, and a Similarity-to-Parents sort.

Berzon found that clients did not initiate interaction with fellow

clients on the basis of perceived similarity to their more threatening

parent, but rather on the basis of their own initiation into the

group. The investigator concluded that the ongoing dynamics of

group process provided a source of insight into group interaction

not usually equalled by the transference perspective.

In summary, the results of the few studies conducted along this

line of endeavor indicate that the possibility of an association

between the attitudes of an individual toward parents and an unknown

therapist is still open to question. Even in the studies reporting

positive findings (Chance, 1952; Crisp, 1964a, b), the modest

results do not merit the inferential leap necessary to suggest

causality, and to thereby support only the mechanism theorized by

the "transference hypothesis." Any of the other positions (May,

1958; Schutz, 1958; Apfelbaum, 1958) could also account for the


Attitudinal analysis: essential elements of client attitudes

The second major line of study of client attitudes toward parent

and therapist consists of attempts to discover the essential

components of attitudes held by the client toward the therapist prior

to the first therapeutic encounter. One of the most comprehensive

studies along this line of thought is that of Apfelbaum (1958), who

secured pre-therapy Q-sorts from 100 college clients indicating

their expectations of the therapist. Cluster analysis yielded five

clusters, designated by the investigator as A (N = 26), B (N = 18),

C (N = 14), AB (N = 12), and D (N = 5). Twenty-five subjects'

attitudes did not fall into a cluster, and groups AB and D were

dropped, resulting in a loss of 42% of the subjects. Thus, little

can be said for the generality of the results of this study.

Nevertheless, a more detailed look at the findings might provide a

backdrop useful for the examination of other studies.

In his item analysis, Apfelbaum labeled the largest three

clusters Nurturant (A), Model (B) and Critic (C). Subjects falling

into group A expected a guiding, giving and protective therapist.

Subjects in cluster B expected a therapist to be tolerant, accepting,

nondirective, and in general, a model of personal adjustment. Those

subjects whose responses fell into category C tended to expect the

prospective therapist's behavior to be primarily guiding, analytical,

aloof and critical. Study of a variety of data (e.g., MMPI profiles,

case material, measures of stayingness) indicated that the expectation

clusters were apparently associated with strongly-held interpersonal

expectations, which were in turn related to general personality

functioning and consequently to the subsequent character of the client-

therapist relationship.

Further analysis of the data yielded two general dimensions

underlying these three types of attitudes. The first was that of

warmth-nurturance-protectiveness vs. coldness-indifference-detachment.

The other dimension was labelled directiveness vs. nondirectiveness.

Apfelbaum viewed these dimensions collectively as basic to the

therapist attitudes of at least those subjects of his study who fell

into his three highest frequency categories.

The data specifically relevant to the development of similar

attitudes toward parent and therapist are gleaned from the personality

measures of Apfelbaum's subjects. In brief, the results of these

measures and other data on the undergraduate clients utilized in

this study were as follows: Subjects whose expectations of the

therapist fell into clusters A (Nurturant) or C (Critic) tended to

present lists of problems to the counselor, rather passively antici-

pating such action to stimulate the counselor's nurturance (group A)

or critical appraisal (group C). Such action by group A subjects

seemed to occur out of a feeling of weakness, those of group C out

of a feeling of wrongness. On the MMPI, C-group subjects shared

many of the characteristics and a similar overall level of maladjust-

ment as A-group clients, but were not as self derogating. Neither

group expected sympathy or understanding from others. However, the

A-group would relish such a response, while the C-group would

probably reject it outright. Finally, members of the C-group (mostly

males), who expected a critical, detached male therapist, gave

indications of having had difficulty in relationships with their

fathers during formative years.

On the other hand, those clients who expected a model type of

therapist (group B) tended to be generally better adjusted than the

other two groups. They felt in much less need of the counselor's

help, were less dissatisfied with themselves and others, and had

higher scores on ego strength than did members of group A. However,

relative to both other groups, group B clients (who tended to be

mostly female) tended to be more detached from the counseling

relationship and less concerned with their problems. Finally, they

openly indicated feelings of closeness with their fathers.

In general, then, both A and C groups were quite concerned

about their problems, with those in group A seeking a solution

through nurturance and those of group C through criticism. Both

groups appeared to have a jaundiced view of the world and were

distrustful of the motives of others. The A clients felt that

the counselor would, for reasons of his own, wish to reassure and

support them, while the C clients expected that he would, for

equally egocentric reasons, wish to tell them what was wrong with

them. While these group differences in expectations of therapist

behavior were, as Apfelbaum suggested, quite possibly related to

the client's relationships with early significant others, the only

real clue given was the difficulty C clients reported with their


While Apfelbaum's data serve as a rich source of information

regarding the essential elements of clients' attitudes toward

therapists, new and intriguing questions seem to rise Phoenix-like

from the ashes of those that have found explanations. For example,

if the C-group clients (1) tended to react counteractively to the

dependent role of the client, (2) had a history of difficulty with

their fathers during childhood, and (3) as adults anticipated an

unknown male authority figure to behave critically and insensitively

toward them, then the temptation is irresistible to speculate that

the attitudes of these adults toward male authority figures is in

some way associated with the clients' early experiences with the

familial counterpart of that figure, the father. Only the nature

of the association would appear to be open to genuine controversy.

Those theorists with a stake in proving the transference hypothesis,

of course, would conclude from Apfelbaum's evidence that this is a

clear case of the transference of relatively hostile feelings from

the father to the unknown therapist. However, such a conclusion

is clearly not merited for the elementary reason that association does

not prove causality.

A somewhat more conservative view would be taken by theorists

such as May (1958) and Apfelbaum (1958) himself, who might conclude

that the clients with this kind of anticipatory attitude toward an

unseen therapist were basing it upon their presumably unsatisfactory

experiences with a similar authority figure at an earlier time.

That is, they expect as adults only what they have learned to expect

in the past: that a male in a position of power is inclined to

be critical, aloof and insensitive toward them. These experientially-

inclined theorists would assign responsibility for such attitudes

to a kind of developmental insufficiency in interpersonal relation-

ships which continues to influence the individual's attitudes and

behavior toward authority figures in the client's adulthood. Thus,

once again it is clear that no single theoretical framework accounts

for similarities between client perceptions of parents and therapists,

to the exclusion of all the others.

A second question arising from Apfelbaum's data which is equally

as fascinating as the first is as follows: If the members of the

C-group are characterized by the above attributes and attitudes

toward their fathers, what about the members of the A-group? Those

individuals expected a therapist to be warm and nurturant toward them,

much like an over-protective mother. Did these clients, who

evidently did not run into major conflicts with their fathers as

children, instead have a mother who held the position of authority

in the family, and a father who tended to be benign (or perhaps

absent) more than anything else? In order to even begin to answer

this question, it would be necessary to obtain some measure of the

subjects' feelings toward their mothers, which Apfelbaum does not


By contrast, Apfelbaum's B-group, which was largely composed

of well-adjusted females, expressed no dissatisfaction with their

fathers. Neither did they have unrealistic expectations of the

therapist, and the indications were that they were much more detached

from treatment than the other two groups. Dependency would not

seem to be a major issue for such clients, in light of the evidence,

thus raising a final question: Is this group of clients the only

one of the three which is not in some way searching for a more

satisfying relationship with a parent-figure than they had as

children? Is the reason that their anticipated therapist is a

model, after whom they can pattern their adult lives, rather than

someone who will give them something (either warmth or criticism),

due to the fact that they have surmounted that milestone of develop-

ment which requires severing the cord that bonds the young child

to his parents? Apfelbaum's data provide the basic groundwork

necessary to ask these questions. The present investigation will

seek evidence which may help in find their answers.

Patient Preferences

Basic Therapeutic Roles

In general, analysis of available research seems to support

the existence of two primary behavioral postures open to a therapist

in fulfilling his function as a facilitator of growth for the client.

These postures, or orientations, may be termed most simply (a) inter-

personal and (b) personal, and each reflects to a great extent a

different facet of the therapeutic process. Each orientation is

frequently illustrated by a therapist who plays a characteristic

role, designated below as "provider" (interpersonal) and "model"


The role of "provider" is that which is frequently brought to

mind when one considers the task of the therapist, and its examina-

tion necessitates focusing upon the interaction between therapist and

client. As will be seen below, the bulk of the research to date

deals with various aspects of this therapeutic function.

The role of "model" is somewhat different. Although it might

be argued the therapist who engages strictly in modeling behavior is

providing the client with a worthwhile experience, the link of

giving that is involved in such behavior appears substantially

more tenuous than in other ways in which he might give something to,

or do something for the client, Indeed, to classify modeling as

"providing" behavior seems accurate only to the extent that it is

therapeutic, and therefore beneficial to the client. It would

appear that the client would have to invest a good deal more effort

in receiving the benefits from a therapist who merely models

appropriate, mature behavior than from one who actively offers him

reinforcement, e.g. warmth, guidance, etc. Thus, the focus in

studying the role of model is much more upon the therapist himself,

rather than upon the client-therapist interaction, although clearly

the client is necessarily involved to some extent. The discussion

that follows should elucidate this distinction.

Modalities of therapist providing behavior

Utilizing a quasi cross-cultural approach, Berrick (1970)

sampled the therapist expectancies of the heterogeneous population

in Hawaii. A Q-sort describing a wide range of possible therapist

behaviors was given to 77 subjects representing three "ethnic"

groups (Caucasian, Japanese and part-Hawaiian), three levels of

social functioning (inpatient, outpatient and nonpatient), and both

sexes. Factor analysis of the Q-sort data yielded eight factors

which described expected therapist types. One of these was very

similar to the nurturantt" type found by Apfelbaum (1958), while

two others seemed to be like, but not equivalent to, his "model"

and "critic" types. Three more factors appeared to be combinations

of two or more of the above, while the last two seemed to be original

to Berrick's study.

Lorr (1965) utilized Schutz's principle of constancy (Schutz,

1958) as the theoretical impetus for an investigation of the post-

therapy attitudes of 523 VA outpatients. The subjects were asked

to rate their former therapists' major personality characteristics

and also to judge their own progress in treatment. Analysis of

the ratings yielded five factors; "understanding," "accepting"

(includes nurturantt"), "authoritarian" (includes "advising,"

"directive"), "independence-encouraging," and "critical-hostile."

The patients' improvement ratings were found to be positively

correlated with their perception of the therapist as understanding,

accepting, and independence-encouraging, but negatively related

to his beinq seen as authoritarian and critical-hostile.

A comparison of Lorr's factors with those of Apfelbaum (1958)

at this point is obligatory. It will be recalled that Apfelbaum

also derived five factors from his data on the expectations of his

patients. Two of these factors were discarded, possibly for lack of

subscription by subjects since they proved to be the least popular

factors. Nevertheless, one factor (AB) was subscribed to by only

two fewer subjects than the next factor, which was retained. It is

assumed that factor AB was labelled such because it shared qualities

of factor A (Nurturance) and B (Model), and therefore might actually

have been discarded for its lack of discriminative ability relative

to the other three major factors. If this reasoning is correct,

then a case can be made for the existence of a high degree of

congruence between Apfelbaum's factors, derived from patients'

expectations of therapists, and Lorr's factors, which described

post-treatment patient ratings. More specifically, the description

of Apfelbaum's factors A, B, and it is presumed AB, seems to relate

closely to that of Lorr's factors of understanding, accepting, and

independence-encouraging. Conversely, Apfelbaum's factor C (Critic)

can be assumed to be very similar to Lorr's factor of "critical-

hostile." Thus, it seems reasonable to conclude that even the

dissimilar populations of VA patients and prospective undergraduate

counseling clients conceive of psychotherapists from similar


More compelling than the similarities between the two sets of

specific factors is the congruence in the more basic dimensions

upon which all the factors appear to be founded; for closer

scrutiny of the factors of Lorr and Apfelbaum compels the conclusion

that those of Apfelbaum reflect two different modalities of therapist

behavior, and that all but one of Lorr's five factors logically find

a home within one or the other dimension. That is, Apfelbaum's

warmth-nurturance-protectiveness vs. coldness-indifference-detachment

dimension clearly reflects affective behavior on the part of the

therapist, while the directiveness vs. nondirectiveness dimension

exemplifies the therapist's level of activity. The factors which

Lorr terms "accepting" and "critical-hostile" seem to fit nicely

at opposite ends of the affective continuum, while his "authoritarian"

and "independence-encouraging" factors fall roughly at opposing

ends of the activity continuum.

At this point, one can only wonder about the patterns of trans-

ference attitudes given by the 42% of Apfelbaum's subjects that

were discarded. Nevertheless, a third basic modality of therapist

behavior can be hypothesized which would account for Lorr's fifth

factor. This dimension encompasses what might be termed the "cogni-

tive" behavior of the therapist, of which, intellectual understanding

of the client's dilemma is a primary facet. Of course, communication

of such understanding is a very complex task involving a variety of

therapist behaviors; but the ability to comprehend a wide range of

client problems, particularly in the same perspective as the client

himself, seems largely a function of this modality.

Modalities of therapist modeling behavior

It is apparent from the discussion above that those few

studies which attempted to analyze the essential elements of client

preferences also reported findings that were consistent with the

notion that another important role of the therapist was that of

model. Both Apfelbaum (1958) and Berrick (1970) found evidence

to suggest that some clients preferred a therapist who did not act as

a provider, but instead, who simply stood as an example of certain

behaviors, feelings, values, etc., which appealed to them. Even

the findings of Lorr (1965) could be interpreted to provide a small

amount of support for this view, to the extent that his factor of

"independence-encouraging" is not seen as being actively guiding;

for it would seem that independence is a behavior which would be

instilled more readily in a client's repertoire by a therapist who

modeled independence than by one who chose to actively direct the

client toward behaving in an independent manner. Certainly the

data of Hutcherson (1967) suggest this hypothesis.

Apfelbaum's study, when compared with those of Lorr (1965) and

Berrick (1970), offers the most lucid evidence suggesting a client

preference for a modeling therapist as completely separate from a

providing one, and only Apfelbaum's data offer a glimpse of the

qualities constituting such a model. Apfelbaum states that this

kind of therapist is preferred by clients who wish him to be

tolerant, accepting, nondirective, and in general, a model of

personal adjustment (Apfelbaum, 1958). It is almost as if they

want little more than a picture of maturity with which to compare

themselves. Implied in this conceptualization of the well-adjusted

modeling therapist is a non-defensive, self-secure individual who

will neither guide nor direct, and who will be emotionally involved

only as an interested fellow human being, rather than having any

semblance of an intense emotional investment in the client. If

this list of inferred qualities is scrutinized, as it will be below,

it becomes evident that the basic topic under discussion is the

therapist as person, i.e. his own self-conception, as reflected in

his level of maturity. Thus, it is not really his active nor

his affective behavior to which these clients are attracted, but

the overall image he projects by simply being himself which they

view as being most important. Nevertheless, it will be suggested

later that even this modeling behavior, despite its relatively

passive nature, also can be analyzed in terms of cognitive, affective

and active behaviors.

Dimensions of Therapist Providing Behavior

Utilizing the hypothetical triad of activity, affect and cogni-

tion* as its operating framework, the discussion now proceeds to a

closer consideration of the major dimensions which appear to have the

most empirical support, and which seem to be logically described

by these three modalities of therapist providing behavior.

Therapist activity: directiveness vs. nondirectiveness

Because most forms of psychotherapy rely heavily upon the

therapist's verbal techniques and abilities, a reasonable measure

*It should be acknowledged at this point that this behavioral
triad shares terminology with that used by Krech, Crutchfield and
Ballachey (1962) to describe the essential components of attitudes.
while it seems reasonable that similar elements should apply to
attitudes as well as more general forms of behavior, the usage of
identical terminology in this study was coincidental.

of his activity would seem to be his level of verbal directiveness.

However, a major difficulty in using this dimension is the large

variability in definition of the term. Most investigators appear

content to assume there is a general consensus as to which kinds

of therapist behaviors constitute directiveness and nondirectiveness,

and therefore see no need in specifying these behaviors. The writer

has found no such consensus that would meet an acceptable level of

reliability. Thus, for the purpose of categorization under the

label of "activity," directiveness will be generally gauged by

the extent to which the therapist attempts to impose his own

structure upon the course and direction of the session, such as

through the use of an inquisitive, or perhaps a guiding, approach.

Similarly, nondirectiveness will refer to the degree to which he is

content to follow rather than lead, and listen rather than inquire

or enlighten.

Overall and Aronson (1963) investigated the expectations of

therapist behavior held by a sample of 40 outpatients lower in

socioeconomic status and educational level than the national average.

As predicted, patients who failed to return for the second session

were found to have significantly greater discrepancies between their

pre-therapy expectations and their view of the initial interview.

Most of the discrepancy was accounted for by the therapist being

less active, medically-oriented and supportive than they anticipated.

The premature termination phenomenon was also the subject of an

earlier study by Heine and Trosman (1960), who administered a patient

expectation questionnaire to all new outpatients referred to a clinic

over an eight-week period (N = 46). Six weeks were arbitrarily

defined as the cutoff in separating continues from terminators.

When length of continuance data were combined with that from the

questionnaires, it was found that continues expected to be active

collaborators in their treatment, while terminators tended to

expect to passively cooperate with the therapist. In addition,

continues anticipated advice or help from the therapist in changing

their behavior, while terminators expected medicine or diagnostic


Although it is not necessarily standard practice among writers

in this area of study, the present investigation will distinguish

between expectations of patients and their preferences, in regard to

therapist characteristics. There is some evidence to suggest that

such a distinction may be helpful (Klepac, 1969). The studies

reviewed thus far in this section have been confined to client

expectations. Those that follow will deal also with client preferences,

which represent the topic under study.

Garfield and Wolpin (1963) utilized a questionnaire to measure

the attitudes toward, and expectations of therapy held by 70

applicants for treatment to a psychiatric institute. The findings of

these investigators were similar to those of Heine and Trosman (1960),

in that, while a majority of these clients believed that an under-

standing of oneself was important for therapeutic change, they

nevertheless preferred the therapist to spend a substantial amount

of time providing them with advice.

The question of the universality of this tendency to expect the

prospective therapist to be advice-giving and directive has been the

subject of a number of studies other than those discussed above.

Ilutcherson (1967) had a therapist role-play four therapeutic approaches

on audio tape; two represented the high and low conditions of

patient responsibility, while the other two represented high and low

independence. These conditions reflected the degree to which the

therapist encouraged the patient's responsible behavior and whether

he assumed a directive or nondirective role. Sixty-four undergraduate

male and female students with no therapy experience completed the

I-E Scale and the Autonomy Scale of the Omnibus Personality Inventory.

They then listened to all four tapes, rating their reactions to the

therapist in each. Hutcherson found that the two approaches empha-

sizing patient responsibility were preferred significantly more than

the other two. Therefore, since one of these tapes was directive

and the other nondirective, and both were equally preferred, he

concluded that responsibility more than independence appeared to be

the primary factor determining these subjects' preference; however,

it is apparent that therapist directiveness rather than client

independence was evaluated.

In an investigation of a similar design, Kumler (1968) obtained

the expectations of therapist characteristics from 120 college

students and compiled them into an "expectation Q-sort." He then

had subjects view eight brief videotapes of role-played therapy in the

eight conditions of a 2X2X2 design, in which the therapist's age,

directiveness and warmth were manipulated. The results indicated

that the subjects' reactions to the therapist were more favorable

if he conformed to their expectations of age and warmth than of

directiveness. Like Hutcherson (1967), Kumler found no support for

any pervasive effect of the therapist's level of directiveness

among a nonclient group of college subjects.

In his investigation of the role of authoritarianism as a

patient's personality trait in the therapy situation, Vogel (1961)

asked 62 clinic patients to complete the California F Scale and an

instrument describing their ideal therapeutic relationship. The

investigator found that the patient who possessed authoritarian

personality characteristics tended to describe his preferred therapist

as directive, paternalistic and nurturant.

Plummer (1966) secured the reactions of 42 male college students

waiting to enter into a counseling relationship to films of three

different therapists interacting with the same client. Also obtained

were measures of the would-be patients' need structures through

their responses to the Picture Identification Test. Plummer found

that significantly more subjects preferred the therapist who appeared

to be independent and controlling, was not nurturant, and who did

not attempt to solve the client's problems for him. The investigator

concluded that this therapist seemed to fulfill the needs and desires

of a majority of the subjects, who were young adult males struggling

with the issue of dependency and seeking acceptance as mature

individuals. The implication is that the subjects viewed this

therapist as a prospective model. Unfortunately, a shadow of doubt

is cast upon these findings by a significant order effect of thera-

pist presentation.

Klepac (1969) obtained the Q-sort therapist expectations from

48 college subjects, 24 of whom were expecting a directive therapist,

while the other 24 were anticipating a nondirective one. Half of

the subjects from each qroup were interviewed via closed circuit

television by a directive therapist, the other half by a nondirective

one. All subjects were forced to press a switch at least once per

second to continue viewing the therapist on the monitor. This switch-

pressing was considered an index of the reinforcing value of the

therapist. Subjects also rated their reactions to the therapist by

the Barrett-Lennard Relationship Inventory and an original scale

consisting of six items.

Klepac found that subjects rated themselves more willing to

enter treatment with the therapist who behaved directively than with

the nondirective one. Also, the directive therapist induced more

switch-pressing behavior in subjects than did his nondirective

counterpart. Unfortunately, the sex of the subjects was not

specified. Therefore, it is unknown whether or not these results

are in accord with those of Reiter (1966), Hagebak and Parker (1969),

and Heilbrun (1961b), regarding sex differences in the preferences

of subjects for therapist directiveness,

Helweg (1971) presented 77 college students and 77 inpatients

a film of the directive techniques of Albert Ellis and the nondirectiye

techniques of Carl Rogers, after which the subjects recorded their

preferences. They also completed the Rokeach Dogmatism Scale,

Rotter's I-E Scale, the Taylor Manifest Anxiety Scale, and Gordon's

Survey of Interpersonal Values, It was found that undergraduates and

patients who stated a preference for the directive approach were more

dogmatic and externalized than were those preferring nondirective

techniques. Patients choosing the Ellis approach were more anxious

and less educated than those preferring the approach of Rogers.

Finally, both students and patients preferring the nondirective

approach valued independence as a basis for relating to others, This

finding is in accord with those of Hutcherson (1967), who implies

that patient preferences for independence and for a nondirective

therapist are highly correlated.

The issue of the effect of personality differences on clients'

preferences of therapist behavior was also the subject of an in-

vestigation by Reiter (1966). Two hundred twenty college freshmen

were administered the Therapist Behavior Scale, MMPI, Self-Activity

Inventory, and the I-E Scale. Analysis of the data indicated that

the preference for a directive over a nondirective therapist varied

with a host of patient characteristics, including age, sex, ego

strength and defensiveness. In general, the less directive therapist

was preferred by females who were high in ego strength and in social

desirability needs, and low in self/ideal-self discrepancy--in short,

by females who would typically be viewed as well-adjusted in this

culture. This pattern is reversed for those females preferring the

more directive therapist. On the other hand, males as a group

tended to prefer the more directive therapist. Those who did not

were found to exhibit a higher self/ideal-self discrepancy and less

defensiveness than those who expressed a preference for the directive

therapist, Reiter concluded that therapist preferences of male

undergraduates were related to the individual's overall level of

defensiveness, a conclusion which is supported by Rosen (1967).

Substantial support for Reiter's findings and interpretations

for males can he found in the results of a study by Hagebak and

Parker (1969). These investigators had 40 male undergraduates who

scored high on measures of dominance and 40 scoring low play the

role of clients with an assortment of problems, in the process

choosing between 32 pairs of high and low resistant client responses

following the therapist's statements. The results indicated that

nondirective therapist techniques elicited the most "client"

resistance and that low dominant subjects offered more resistant

responses than high dominant subjects. There was also a significant

interaction between type of client problem and the "client's"

preference. Nondirective therapist statements were followed by

more resistant statements in all client problem areas except

academic, the area viewed by subjects as the least severe of all


The results of these last two studies indicate that nonclient

undergraduate males appear to prefer the directive over the non-

directive approach, especially when their attention is focused upon

a problem of any significant severity. Males with such preferences

rank higher on measures of defensiveness and lower on measures of

self/ideal-self discrepancy than males who prefer less directive

techniques. The impression that one receives from these findings

is that males preferring the directive approach are perhaps more

concerned with appearing adequate and mature, and thereby conforming

to cultural stereotypes, than are those who prefer nondirective

techniques. Additional support for this hypothesis comes from

Heilbrun (1961h), who found that male and female clients conforming

mostly to the cultural model of masculinity or femininity tended to

terminate therapy prematurely, For males, such a model included a

virtual denial of one's dependency upon others in dealing with

personal difficulties, similar to the denial manifested by the

male subjects expecting the Critic for a therapist in the study by

Apfelbaum (1958).

In summary, several statements appear to be warranted by the

literature regarding the role of therapist directiveness in

determining a client's preferences and reaction to treatment. It

seems fair to conclude that the variable of therapist directiveness

has received substantial empirical attention and can be considered the

primary general component of the therapist's activity during the

treatment session. However, the response this behavior evokes in

the client appears to be influenced by several other factors,

especially client sex (Reiter, 1966) and variables related to his/her

maturity (Vogel, 1961; Hagebak and Parker, 1969). In some cases

it is also unclear as to whether the subject's response is due to

the therapist's level of directiveness, or to his degree of emotional

involvement in his work (Heine and Trosman, 1960; Overall and

Aronson, 1963). A confounding of these two variables may be at

least partially responsible for the divergent results of Overall

and Aronson (1963), Hutcherson (1967) and Hagebak and Parker (1969).

Therapist affect: warmth vs. detachment

In scanning the literature, it quickly becomes apparent that,

as intuition would predict, the most popular affective behavior of

the anticipated therapist from the client's perspective is that

which is often labeled nurturance or warmth. However, like the

issue of therapist directiveness, the quality of warmth in a

therapist does not seem to be either universally expected or pre-

ferred by prospective clients in therapy, although some consisten-

cies do seem to exist.

It is common knowledge that Carl Rogers was one of the first

to publicly defend certain types of emotional responsivity on the

part of the therapist toward the client. Indeed, his classic

paper on the "necessary and sufficient conditions of therapeutic

personality change" (Rogers, 1957) posited that significant

positive alteration of client behavior in the absence of the

predominately affective qualities of genuineness, empathy and

unconditional positive regard was quite unlikely to occur. Several

of Rogers' colleagues helped contribute two more factors--non-

possessive warmth and the willingness to be known--which they

believed deserved to be placed on that list (Truax and Carkhuff,

1967). Again, it will be noted that the first of these qualities

is affective in nature (warmth), and, in fact, the one to which most

clients and nonclient subjects alike seem to allude in their

statements of satisfaction with and preferences for therapist


Rogers and his collaborators are by no means alone in their

belief that the effective and desirable therapist is the one who,

all other things beinq equal, adequately communicates his feeling

for the client. Grater, Kell and Morse (1961) assume that anyone

who is genuinely interested in psychotherapy as a profession will

have nurturance as a basic need; otherwise, according to these

writers, that individual will be unlikely to attain true satisfaction

in his clinical work. This view is echoed by Schofield (1970),

who convincingly argues that the warmth and concern which is

communicated by the therapist to the client is of paramount

importance in enabling that individual to engage in meaningful

self-disclosure, and thus to be able to ignore the formalities

of customary social roles. The lack of these therapist qualities

apparently would be conducive to a more business-like atmosphere,

which in Schofield's view, would result in less progress in treatment.

Evidence that clients expect their therapist to be nurturant

and/or warm derives principally from two studies, Since they have

both been described in more detail in the foregoing discussion, only

brief mention need be made of them here. Apfelbaum (1958) found

that his sample of 100 male undergraduates who were awaiting

assignment to counselors were inclined to expect three kinds of

therapists. That type expected by the largest number of clients

(26%) was termed the "Nurturant" therapist, because the descriptions

given him by the clients corresponded most closely to such a label,

In an attempt to replicate this study for a cross-cultural sample

of subjects from Hawaii, Berrick (1970) was only able to find

partial support for all of Apfelbaum's three therapist types.

However, the one type she discovered which was most clearly related

to Apfelbaum's categories was the "Nurturant" type of therapist,

The vast majority of investigations relevant to the subject of

therapist nurturance and warmth are based upon patient preferences

rather than expectations of therapist behavior. Hiler (1958)

conducted one of the earliest of such studies in an attempt to

analyze patient-therapist compatibility. The experimental method

involved relating the number of responses given by a patient on

the Rorschach to the length of his stay in therapy. Hiler's data

indicated that those therapists rated as most warm and friendly by

three staff psychologists tended to retain in treatment a larger

percentage of "unproductive" patients (on the Rorschach) than those

therapists who were rated as least warm and friendly.

It is obvious that the study by Hiler (1958) is open to

criticism from several quarters. First, it appears that there was

a confounding of the variables of therapist sex and warmth, in

that female therapists retained more of the patients who were rated

"unproductive" on the Rorschach than did males. Second, Hiler

chose a rather large, arbitrary number of interviews (20) as

the cutoff for distinguishing between continues and terminators.

Finally, the selection of the total number of responses given on

the Rorschach, although popular during the time period in which

this study was conducted, can certainly be questioned as an adequate

measure of the predicted productivity of a client in therapy. Taken

together, these difficulties substantially weaken Hiler's findings

and inferences.

For all of its weaknesses, however, the early investigation by

Hiler is helpful to the purposes of the present study in that it

provides a glint of a phenomenon of which Hiler himself seemed to be

aware, but one which he did not pursue. The issue in question is a

possible relationship between the status of male client and the

preference for a warm or nurturant therapist. The existence of such

a relationship was suggested by Apfelbaum (1958), although his

experimental design was not really adequate to subject it to viable

test. Nevertheless, it will be recalled that 26% of Apfelbaum's

clients, most of whom were male, reported expectations of being

assigned to a nurturant male therapist, In addition, it was

concluded from a wealth of clinical evidence that it was this same

group of clients which tended to have significant difficulty with

feelings of helplessness, which were accompanied by a yearning

for a role in which they could be dependent in a relatively passive

manner upon the therapist for support, rather than actively become

engaged in finding a solution for their own problems. Finally,

it was noted that these clients eventually remained in counseling

for a protracted length of time, compared to the other two client

groups, in the absence of a corresponding increase in positive

therapeutic movement.

Support for this hypothesis comes from a variety of studies,

several of which have been described more completely in the previous

section. Overall and Aronson (1963), in their investigation of the

expectations and termination rates of patients belonging to the

lower socioeconomic class, found that compared to those patients

who failed to return to the clinic following the initial interview,

continues exhibited less discrepancy between their pre-therapy

expectations and their subsequent rating of the first interview.

One of the categories in which there was most discrepancy for

terminators was that of therapist level of supportiveness. That is,

less emotional support was forthcoming from the therapist than the

patient had anticipated prior to the session. The obvious impli-

cation is that those patients who did not return chose not to do

so because they were disenchanted with the relatively "cold"

therapist they had encountered.

Feifel and Eells (1963) analyzed the responses to an open-

ended questionnaire of 63 ex-patients of a VA clinic at the close

of therapy. The investigators found that these ex-patients emphasized

the importance to them of having had the opportunity to talk over

problems and of the "human" characteristics of the therapist.

Conversely, major detrimental effects noted by these ex-patients

were the therapist's expression of irritation, anger and boredom--

all emotional responses generally not associated with warmth and

nurturance, A follow-up study four years later yielded a

suggestion by these ex-patients that the therapists try to increase

their communications of warmth and interest for the patient.

In a similarly-designed study, Lorr (1965) surveyed the attitudes

of VA outpatients toward their therapists. One of the five

resultant factors was an attitude of "acceptance," which included

expressed feelings of interest and nurturance on the part of the

therapist, This accepting attitude was also viewed by the patients

as being positively related to their improvement in therapy.

One of the major limitations of studies such as those of Feifel

and Eells (1963) and Lorr (1965) is that little is known of the

personality characteristics of the patient respondents. Wallach

(1962) attempted to circumvent this limitation by correlating the

therapist preferences of a large sample (N = 216) of predominately

(90%) male college students with their scores on the F Scale, a

measure of authoritarianism. Wallach presented the subjects with

written descriptions of the major personality qualities of the three

types of therapists found by Apfelbaum (1958), and found that 82%

of the subjects preferred the "Critic," while only 16% chose the

"Nurturant" type. Further, those subjects choosing either the

"Nurturant" or the "Model" therapist scored higher on authoritarian-

ism than those choosing the "Critic."

The findings of Wallach (1962) support the hypothesis that

males with adjustment difficulties tend to prefer nurturant thera-

pists for several reasons. First, the subjects utilized were all

nonpatients; Therefore, relatively few would be expected to be

experiencing significant personal difficulties, and the majority

would not likely be quick to indicate their desires for a nurturant

therapist even if they had them. Second, those who did indicate

such a preference also tended to score high on authoritarianism,

suggesting some problem in dealing with authority figures, and

possibly a tendency to assume a passive-dependent posture toward

them. This explanation also seems to be quite in line with the

findings and conclusions of Apfelbaum (1958). Finally, Wallach's

description of the Critic emphasized client autonomy and personal

responsibility, qualities highly valued by healthy college males

(Hutcherson, 1967).

In another study very similar to that of Wallach, Vogel (1961)

administered the F Scale and an original instrument designed to

gauge the individual's preferred therapeutic relationship to two

samples of outpatients. One group was composed of college students

seeking counseling at a university counseling center, while the

other was a heterogeneous group of outpatients from the surrounding

community. Both sexes were represented equally in each sample.

Like Wallach (1962), Vogel found a significant relationship be-

tween patient authoritarianism and preference for a nurturant

therapist, but for only the community clinic patients. No such

correlation was found for the student group, possibly because of

differential responsiveness of the younger women in the student group

on the measure of authoritarianism compared to their older counter-

parts in the other group (Reiter, 1966).

A study which does not seem to support the hypothesis in

question is that of Plummer (1966), whose male college clients

tended to prefer a counselor high in the qualities of autonomy,

exhibition and rejection, and low in nurturance. In general, these

students apparently preferred a counselor after whom they could

pattern their own behavior, rather than one who would just provide

them with warm support. However, several weaknesses in the

investigative method tend to undermine the validity of these

findings, particularly the observation of an order effect of

therapist presentation.

Two final investigations are relevant to this topic. Kumler

(1968) found that the reactions of a sample of nonclient college

students were more favorable to a therapist if he conformed to their

expectations of age and, more importantly, warmth. Similarly,

Greenberg (1969) reported that male and female student subjects

who were made to believe that a therapist was either warm or

experienced, as opposed to cold and inexperienced, were more attracted

to him, more receptive to his influence, and evaluated his work

more positively. Further, those told that he was warm were more

willing to meet with him and were more easily persuaded by his

communications. Again, it should be noted that in the study by

Kumler (1968), subject sex was unspecified, but presumably males and

females were included.

In summary, another major dimension of therapist behavior which

appears to be of substantial importance to many clients and non-

client subjects alike is that involving therapist affect. For

the general client population, a therapist who manifests qualities

related to interpersonal warmth or nurturance is usually preferred

to one who is lacking in such qualities (Overall and Aronson, 1963;

Feifel and Eells, 1963; Lorr, 1965). The warm, nurturant therapist is

most often preferred by persons, usually males, either engaged in

or preparing to engage in a psychotherapeutic relationship

(Apfelbaum 1958; Feifel and Eells, 1963). In addition, nonclient

college students of both sexes also tended to express a preference

for a warm therapist (Kumler, 1968; Greenberg, 1969). Nevertheless,

there is some evidence to suggest that as a group, nonclient college

students--especially males--would rather have a therapist who

emphasizes the client's autonomy and responsibility, even if it is

in a critical manner, than one who primarily exudes warmth and

positive feelings (Wallach, 1962). As the discussion below will

suggest, the principal mediating variable appears to be related to

the subject's sex and level of maturity, particularly his relative

success in coping with the issues of personal adequacy and dependency.

Therapist cognition: awareness vs. non-awareness

Relatively little research has been conducted in the area of

client expectations and preferences of the therapist's "cognitive"

behavior. A reason for the dearth of such evidence has already

been mentioned, namely the fact that investigators have not

attempted to separate the major dimensions of therapist behavior

into such categories as those used in the present study.

Two notable investigations do not mention evidence of specific

client expectations regarding the therapist's cognitive character-

istics. Apfelbaum's (1958) data from the expectations of male

college clients yielded only two dimensions, involving therapist

warmth and directiveness. Vogel (1961) reported that his authori-

tarian clients preferred a therapist who was directive, paternalistic

and nurturant. None of these qualities seem to have their locus

in strictly cognitive behavior.

The study of Lorr (1965), however, yielded different results.

It will be recalled that he asked clients to complete questionnaires

describing their therapists, and to rate their own improvement in

treatment. One fo the five factors found in the analysis of the

descriptions was "understanding," a quality which can be viewed as

illustrative of an experienced therapist's cognitive abilities.

Just one other study was found in the literature supporting

the notion that clients have preferences regarding a cognitive

modality of therapist behavior. In his investigation of the

preferences and expectancies of therapist age and sex, Boulware

(1969) reported that the most important single determinant of the

preferences of undergraduate students was their expectancy about

the therapist's ability to understand their problem. Taken out

of the context of the present discussion, this statement would

probably merit scant attention by the reader. However, the fact

that only two studies in the area even mention the importance to

the client of a therapist's cognitive characteristics--specifically

the quality of understanding--makes one wonder if perhaps this

sphere of therapist behavior is usually taken for granted, and

consequently has been treated as a non-issue by investigators of the


Clearly, there is a semantic difficulty with the term "under-

standing." The verb "understand" denotes intellectual comprehension

of a problem or event, and therefore, as an intellectual qualtiy,

is void of any affective meaning. However, the adjective and noun

forms of the term, i.e. "understanding," have a connotative meaning

of tolerance or sympathy with another's situation or condition.

Thus, they connote affect. While Lorr (1965) and Boulware (1969) do

not attempt to distinguish between these two meanings in their

discussion of therapist understanding, it is apparent that such a

distinction must be made in the present investigation so that this

therapist characteristic, which is typically ignored despite its

heuristic promise, can be meaningfully evaluated. Because this topic

is vital to the purpose of the present study, it will receive further

elaboration in a later section.

Dimensions of Therapist Modelinq Behavior


In an earlier subsection of this review, it was noted that

there was research which suggested that some prospective clients

prefer a therapist not to behave in a particularly active manner so

much as they just want him to "be." More specifically, they appear

to wish him to be a model of positive adjustment, or personal

maturity (Apfelbaum, 1958). Because maturity is such an elusive,

value-laden construct requiring scrutiny if it is to be useful in

an empirical endeavor, attention will now be given to this subject

within the context of psychotherapeutic behavior. An exhaustive

review of the body of literature devoted to the area of personality

adjustment, psychological maturity, self-actualization, etc., will

not be attempted, however.

There are those who would say that the premier characteristic

of personality maturity is that which is most commonly termed

"competence," a quality which has received a fair amount of attention

among researchers in this area. Havelick and Vane (1974) studied

the imitative behavior of 188 fifth- and sixth-grade children in

response to models of two conditions of competency. The investiga-

tors found that the high competency model was imitated significantly

more than the less competent one, and that children with below-

average achievement records imitated significantly more than their

peers with better records. These findings are supported by those of

Strichart (1974), who also found a main effect of model competence

in a comparison study of retarded and nonretarded adolescents.

In addition, Strichart found that less competent subjects exhibited

more imitative behavior than their more competent peers, which also

follows from the findings of Havelick and Vane (1974). Similar

results attesting to the general importance of competency to the

model's repertoire come from studies by Shepherd-Look (1972),

who used graduate students and professional psychologists as

subjects, and by Schuh (1971), who noted that the effect of model

competency applied only to the females among her group of preschool


Several investigative efforts have been directed at the basic

subject variables associated with a preference for a competent model.

Utilizing undergraduate subjects, Goldman and Olczak (1975) found a

negative relationship between one's overall level of "psychosocial

maturity" and one's fear of appearing incompetent. Similarly,

Melnick and Pierce (1971) found that undergraduates in counseling

relationships who had pervading feelings of weakness and helpless-

ness tended to overestimate their counselor's competence and

strength. Both of these studies, coupled with those of Strichart

(1974), Schuh (1971) and Shepherd-Look (1972), tend to support

White's contention that personality maturity is directly related to

one's feelings of competence (White, 1960).


A personality quality of equivalent importance as competence to

the concept of maturity is independence, perhaps more aptly terms

"autonomy." Indeed, there are those who would measure one's

maturity entirely in terms of one's ability to be self-reliant,

which is to say, autonomous (Wijngaarden, 1968). Nevertheless, most

writers appear to be in agreement with Smitson (1974), who believes

that the ability to take charge of one's life and to arrive at a

relatively successful resolution of the developmental conflict be-

tween dependency and independence is at least a necessary, if not

sufficient, condition of maturity. Lending support to this position

is a study by Tewari and Tewari (1968). These investigators compared

thirteen "highly adjusted" college subjects with thirteen "highly

maladjusted" peers on a variety of personality measures, and found

that the well-adjusted group exhibited signs of a stronger superego,

higher ego strength and more autonomy relative to their poorly

adjusted peers.

Not all writers view autonomy in such simplistic terms, however.

Rosen (1972) contends that neither dependency nor independence are

enviable goals in themselves, but instead represent two poles on a

continuum of interdependence, which is the most adaptive condition.

Indeed, this seems roughly equivalent to Wijngaarden's position,

for his definition of maturity is a state of spiritual self-reliance

within the unavoidable dependency in which one lives (Wijngaarden,

1968). A logical developmental sequence could thus be hypothesized

that would begin with dependency in childhood, progress with a solid

effort at independence in late adolescence, and eventually settle

into an awareness of one's own strengths and a willingness to utilize

them when desirable and appropriate for preservation and enhancement

of the self.


A necessary ingredient to the development of autonomy which is

at least implicitly suggested by most articles reviewed is a

feeling of self-esteem or self-confidence. One such investigation is

that of Tippett and Silber (1966), who analyzed the methods used by

well-adjusted and poorly-adjusted undergraduates to handle informa-

tion discordant with their self-concept. The investigators found

that the healthier subjects responded much more selectively in

their assimilation of information contradictory with their self-

image, but yet were more open to such input within this selective

range, compared to the less healthy subjects. Despite being given

information potentially threatening to their self-concepts, the

healthier subjects were nonetheless able to maintain a high sense

of self-esteem, relative to their less healthy peers. It will be

recalled that Melnick and Pierce (1971) found that clients who

perceived themselves as weak and dependent, not surprisingly had

a low sense of self-esteem, and tended to idealize their therapist

in terms of these qualities they lacked. Similar findings for

nonclients were noted by Cox and Thoreson (1977).

The preceding discussion has attempted to delineate the essential

features of maturity, noting in the process three related yet distinct

characteristics. These have been termed competence, autonomy and

self-confidence. It seems apparent that despite the passive nature

of these qualities when compared to those of a "providing" therapist,

discussed above, the basic behavioral framework would appear to apply

to the personally-oriented "model" therapist as well. Specifically,

competence and autonomy rather easily fit under the heading of the

model therapist's "active" behavior, reflecting characteristic,

though passive, dispositions in his therapeutic, and probably his

everyday, behavior. That is, such a therapist, in order to be

considered mature by the present definition, would model competence

and autonomy in his interaction with the client.

The reader might recall that the second basic modality of an

interpersonally-oriented therapist's behavior was "affect."

Intuitively, one would expect a personally-oriented therapist also

to exhibit some consistency in his affective behavior, although it

would not be provided primarily for the client's benefit. In view

of the foregoing discussion, the mature model therapist's dominant

affect would be self-confidence, self-security or self-esteem.

All of these terms seem to share the same essential meaning, which

is a positive feeling about oneself. This feeling probably is

communicated via a calm, relaxed demeanor as well as in a variety

of other subtle ways; however the affect is communicated, the

message the client receives is that here is a person who feels about

himself the way that he (the client) would like to feel. As a

result of being exposed to such a therapist, the client is probably

inclined to become hopeful that he might eventually come to feel the

same way about himself.

As was the case with the "cognitive" modality of the interpersonal

therapist's behavior, there is a paucity of research attesting to

the importance of this specific form of behavior in the personal

therapist as well. Nevertheless, it is believed that implicit in

most formulations of maturity (e.g. Smitson, 1974; Tippett and

Silber, 1966) is the self-awareness factor, the knowledge of one's

strengths and weaknesses which help form the basis of the capability

and self-confidence projected by the mature individual. Likewise,

for the personally-oriented therapist this capacity for self-

understanding and self-evaluation could be viewed as essential to

his being perceived as mature and effective by the client.

Independent Variables: Client Maturity, Dependency and Sex

The discussion thus far has focused upon the basic forms which

the client's preferences for a therapist tend to assume, namely in

terms of active, affective and cognitive behavior. It is believed

that the major share of the evidence to date supports the hypothe-

sizing of several primary client variables which help to determine

these preferences, and these will now be discussed.

Maturity and dependency

The components of maturity discussed above for therapists

also apply to clients as well as nonclient subjects. The literature

already reviewed suggests that clients seem to have more difficulty

than most in dealing with their dependency needs, which may help

explain why they become clients in the first place (Alexander and

Abeles, 1969); and it seems evident that generally, difficulty in

dealing with dependency belies a more basic concern with one's own

competence as a person, and implies the existence of feelings of

inadequacy and low self-confidence (Tcwari and Tewari, 1968;

Wijngaarden, 1968; Rosen, 1972; Smitson, 1974; White, 1960;

Saccuzzo, 1975; and Cox and Thoreson, 1977).

Helweg (1971) found evidence supporting the notion that a

client's desires to feel independent and his preferences for a

nondirective therapist are highly correlated. Helweg's results

indicated that nonclient college students and psychiatric inpatients

preferring the nondirective approach tended to value independence as

a basis for relating to others. In accord with these findings are

those of Lorr (1965), who reported that his sample of VA patients

viewed their improvement in therapy as being positively related to

the therapist's behaving in an "independence-encouraging" manner,

and negatively related to his behaving authoritatively.

In general, the evidence suggests that clients like to feel

they are becoming independent, even if reality indicates otherwise

(Heller and Goldstein, 1961), and will react positively to a thera-

pist who promotes this feeling (Lorr, 1965). Nevertheless, they

also require a therapist who accepts their dependent behavior, at

least during the early phases of treatment. Support for this

generalization comes from Winder et al. (1962) who found that

clients engaged in therapy were more likely to return if the thera-

pist reinforced their expressions of dependency than if he did not.

It will be recalled that Hutcherson (1967) attempted to assess

the importance of feelings of responsibility and independence to

undergraduates. He hypothesized that for these subjects, the desire

for personal responsibility seems to be more important than

independence in determining therapist preferences. He concluded that

a suhiect prefers a therapist who makes hiim feel responsible for

finding and implementing solutions to his difficulties over one who

assumes that responsibility himself. However, llutcherson's design

and findings were such that he is unable to rule out the possibility

that dependency needs determine one's therapist preferences.


Some rather consistent relationships between personality charac-

teristics and early termination from therapy have been noted within

sexes. Heilbrun (1961b) concluded from his study of 71 college

clients that the individual who conforms most closely to the expected

cultural sexual stereotype tends to terminate early. Thus, the male

who remains is likely to appear more feminine, and the continuing

female more masculine, than their nonstaying peers.

Heilbrun (1961a) also found that males who remained in counseling

for six sessions were more deferent and self-abasing, and lower on

autonomy and dominance, than were those who terminated. This

pattern suggested that staying males were more immature and inadequate,

and indicated a problem with dependency. Females who remained in

counseling also were more dependent than those who did not, but

only if their counselors were average or below average in dominance.

With highly dominant counselors, more independent females tended to

stay while more dependent ones did not. Heilbrun explained that

independent females warm to acting responsible for their treatment,

while dependent ones wait for their counselor to give them a "cure."

The manner in which the counselor initially responds to these client

expectations and dependency needs determines whether the female client

continues or terminates counseling. For malps, the counselor's

response seems less important than is the strength of the client's

dependency needs, lie will remain in counseling if these needs make

leaving appear intolerable, and quit if they do not.

An explanation for this differential responsivity of males and

females to counselor behavior can be found in several other studies.

It will be recalled that Apfelbaum (1958) reported that the two

predominately male groups of clients in his investigation exhibited

more elevated MMPI profiles and generally poorer patterns of personal

adjustment than did the predominately female group. In addition,

both of the male dominated groups showed evidence of substantial

difficulty in handling their dependency needs. Those tending to

deal with the problem passively scored higher on measures of

stayingness (in treatment) than did those handling it counteractively.

Apfelbaum concluded that mildly troubled women may feel freer to

seek help than do men of the same level of maladjustment, which

accounts for the higher frequency of females in the cluster B

group of his study. It may also help explain why women tend to

outnumber men as outpatients, while men outnumber women as in-


Alexander and Abeles (1969) supported Apfelbaum's contention

with their finding that male clients had more T scores above 70 on

the MMPI than their nonclient norm groups, and also scored higher

on the Mf scale, as Helibrun (1961b) would predict. Van Atta

(1968) may be correct in assuming that, rather than being less

defensive and more open as some have concluded (Reiter, 1966),

males who remain in therapy may be less strongly defended than

early terminators. Somi evidence will be shed on this notion by

the present study.

Summary and Implications


The intent of this review was to begin on a broad scale and

progressively narrow down the discussion to consideration of specific

variables essential to predicting individual therapist preferences.

The first section focused upon background research relevant to the

origin of client preferences of therapist behavior. The treatment of

this topic included a discussion of several hypothetical frameworks

which have implications for the development of these preferences, as

well as a review of the empirical evidence which must serve as the

foundation for study of the prediction of these attitudes. It was

concluded that all four formulations assumed a relationship between

one's early experiences with adult authority figures (usually

parents) and one's later therapist preferences. However, the

scant research available relevant to this issue offered little solid

evidence to support the existence of such a relationship. Neverthe-

less, there were data to suggest that individuals differing on a

variety of characteristics still tended to share basic similarities

in their perceptions and preferences of therapists.

The discussion then turned to a more thorough consideration of

the general perspectives from which clients view therapists. A

basically dyadic framework was proposed that would seem to account

for the primary types of behavior clients tend to value in a thera-

pist. These two behavioral roles were categorized generally as

"providing" and "modeling," and were seen as reflecting an inter-

personal and a personal orientation respectively by the therapist.

These two behavioral roles were further explored, and each

was broken down into three hypothetical modes of behavior labeled

therapist activity, affect and cognition. It was observed that

only the third member of this triad lacked a wealth of empirical

evaluation, and that its heuristic promise merited attention in the

present investigation.

The next section of the review proceeded to the delineation

of the specific dimensions which research seemed to support as

being illustrative of the active, affective and cognitive modes of

therapist "providing" behavior. These dimensions were respectively,

directiveness vs. nondirectiveness, warmth vs. detachment, and

high awareness vs. low awareness. It was concluded that (a) the

degree of directiveness offered by a therapist typically evokes a

response from the client, the type and strength of which often

varies with several other factors, including sex and level of

maturity of the client; (b) male clients in particular tend to

prefer a therapist who exhibits some evidence of affective wannth or

positive regard toward them; however, some nonclient groups--

notably undergraduate males--would rather have the therapist be

fairly detached and even critical of them. a finding which again

suggests the importance of client maturity level in determining

such preferences; and (c) the cognitive mode of therapist behavior,

i.e. awareness or intellectual understanding, has received little

empirical attention.

The succeeding section was addressed to the analysis of maturity

as a therapeutic behavior, and it was concluded that the three pri-

mary modalities of a therapist's repertoire, i.e. activity, affect

and cognition, could be exemplified by four essential qualities:

competence and autonomy (activity), self-confidence (affect), and

Self-awareness (cognition). It is believed that these characteris-

tics comprise the principal therapist qualities desired by those

clients who are inclined to prefer a therapist to act as a model.

The present study will attempt to test the applicability of this

belief to nonclient college subjects.

Because the literature suggested that client preferences

regarding both directiveness and warmth vary with the sex and

maturity of the individual, these factors were discussed as

essential to the determination of the client's preferences. It was

concluded that the sex of the client appears to operate primarily

in helping him/her decide whether to seek therapy, and thus be

compelled to accept his/her inability to cope with personal

difficulties independently. Presumably, the differential cultural

expectations of the sexes regarding dependency are responsible for

the hesitancy by males to seek such help unless or until their

problems seem nearly overwhelming, while females tend to be less

reticent to accept the role of client along with its dependent

overtones. Thus, the importance of the sex variable is viewed


Behavioral Interpersonal Personal

Illustrative Provider Model

Modalities Aff Act Coa Aff Act Cog

tics Warm Direct Aware Self-Con Self-Aware

Comp Auton

Figure 3.1. Conceptual chart: Basic behavioral orientations and
illustrative roles, modalities and characterisitcs of

Legend: Aff = Affect; Act = Activity; Cog Cognition; Warm =

Warmth; Direct = Directiveness; Aware = Awareness; Self-Con =

Self-Confidence; Self-Aware = Self-Awareness; Comp = Competence;

Auto = Autonomy; Indepen = Independence.

as relative to the basic issue of maturity, especially the de-

pendency component.


It will be recalled that in their study of the male under-

graduate's resistance to directive and nondirective therapist

techniques, Hagebak and Parker (1969) found that the nondirective

approach elicited more resistance, particularly from low dominance

subjects, than did the directive approach. If it is assumed that at

least some males high in dominance will defend against their depen-

dency needs in accordance with the cultural sexual mandate, and that

therefore they will not remain clients very long or even become

clients in the first place, then it is a simple matter to fit

these findings in with those of Helibrun (1961a, b) and Reiter

(1966). The resulting implication is that male client and non-

client subjects alike probably can be grouped into three categories,

determined by their level of maturity and the general way in which

they handle the dependent role. The first group includes those

rather inadequate-appearing individuals who respond passively to

such a role. These males would likely be low in both maturity and

autonomy, would remain in a counseling relationship for some time

once engaged, and would be expected to prefer a "providing" type

of therapist, who would act at least moderately nurturant

(Apfelbaum, 1958; Feifel and Eells, 1963). Finally, it is this

group of individuals who would be expected to hold similar attitudes

toward ideal parent and therapist.

The second cluster of males appears to cope with dependency

differently, in a manner to which Apfelbaum refers as "counter-

actively." In a sense, such males seem to rebel at being placed

in a dependent role, and will travel some lengths to avoid it

entirely. This group is likely to seem defensive and high in

autonomy, although the maturity level of these subjects would be

expected to be rather low. It is anticipated that these individuals

would prefer a "providing" therapist whose approach may be

directive and who conducts therapy in a business-like manner

(Apfelbaum, 1958); or they may wish him to be relatively non-

directive, and to emphasize their responsibility and autonomy

(Plummer, 1966). However, they probably would not desire him to

exude warmth, acceptance and nurturance, and may well prefer him

to be critical of them when it is at all appropriate (Apfelbaum,

1958; Plummer, 1966). If there is evidence of a similarity in

these individuals' attitudes toward ideal parent and therapist,

the qualities shared are likely to be largely negative.

The third group of males is likely to be fairly high in both

their level of maturity and autonomy. Probably few such individuals

would likely be found in therapy for any great length of time

because they already possess the basic skills to successfully handle

most problems of living. It is presumed that these persons would

prefer a "modeling" therapist with whom to compare themselves,

as described earlier. It is also hypothesized that there would only

be a moderate similarity in their attitudes toward the ideal

parent and therapist.

In general, nonclient undergraduates are assumed to be moderate-

ly well-adjusted and not to be extremely concerned with the therapist

characteristic of activity, that is, his level of directiveness,

More important to them would seem to he the issues of therapist

affect and cognition. The therapist's dominant affect must be

warmth, in that he must communicate that he at least cares what

happens to them. However, the evidence does not suggest that his

expression of warmth needs to be exorbitant for these subjects.

Although the data are sparse, the indications are that for most

individuals the therapist must communicate that he understands the

problems and feelings they bring to him, and that he has an idea

of how to handle them. Thus, he must exhibit an air of competence,

which for subjects high in maturity and autonomy at least, may be

sufficient to facilitate growth. As noted earlier, understanding

is a complex variable, and to denote true competence in a provider

therapist, it probably should be manifested in an intuitive awareness

of the client's verbal and nonverbal communications. Such awareness

would be expected to be possessed by the model therapist, although

his focus would largely be turned inward, as self-awareness. It

is the present investigator's belief that this cognitive quality

of "knowing," more than any variable related to activity or affect,

defines the therapist's competence in the eyes of the client, and

therefore will heavily influence the therapist's effectiveness

within the therapeutic domain.

Logically, this notion makes sense in terms of the stage of

development in which these individuals tend to fall. Becoming

heavily involved for the first time in roles requiring self-

sufficiency and independent efforts, it is not surprising that for

them, dependency is an issue; for they are struggling to various

degrees to feel comiptent in a number of roles as a means of feeling

more adequate in general. Thus, the basic issue can be inferred

as that of adequacy, or maturity, which entails the confident

feeling of being able to cope well. Thus, a therapist should be

respected and accepted as a positive model to the extent that he

exhibits this maturity (Lorr, 1965; Hutcherson, 1967; Strichart,



Therapist Preferences

The logical first question about which predictions should be

made concerns the stated preferences of the subjects, and simply

asks, what are the important qualities that undergraduate males

admit to seeking in a therapist? It was noted that the literature

supports the thesis that client and nonclient subjects in general

tend to consider important at least two, and perhaps three basic

modalities of therapist behavior, which were termed activity, affect

and cognition. Thus, the three hypotheses related to the question

of the essential qualities valued by subjects will be framed in

this perspective, although in reverse order, and data testing

these hypotheses will derive from the ISORS ratings.

Hypothesis la (1) condition

In general, the "ideal therapist" will be described as being

high in the skills of both an interpersonal and a personal nature,
and thus, he will be seen as "intuitive" and "self-aware."

The issue implicit in this hypothesis is the believed importance

to the prospective client of the therapist's ability to thoroughly

understand the client's problems, which may obtain through a solid

awareness of himself. Although these adjectives reflect the

therapist's understanding of the situation, and despite the fact

that the quality of understanding has some empirical support

(Boulware, 1969; Lorr, 1965), the term itself will not be offered

to the subjects, nor will it be used by the raters of the taped

therapy segments because of the semantic difficulties noted

earlier. In its place will be used the terms "intuitive" and "self-

aware" on the ISORS for the subjects and "awareness" in the instruc-

tions for the raters of the tapes.

Hypothesis la (2) affect

In general, the "ideal therapist" will be described as being

moderate in interpersonal affect ("warm") and high in personal

affect ("self-confidence").

It is assumed that these nonclient subjects will neither

require the high degree of nurturance some clients prefer nor will

they completely reject any semblance of warmth on the part of a

therapist as do some client groups (Kumler, 1968; Greenberg, 1969;

Plummer, 1966; Apfelbaum, 1958). It is anticipated that nonclient

subjects on the whole will be more concerned with a positive

feeling derived from an internal rather than an external source.

Hypothesis la (3) activity

In general, the "ideal therapist" will be described as moderate

in interpersonal activity and high in personal activity, thus

being seen as only somewhat "advising/guiding" but fairly


The data on this topic are neither plentiful nor consistent

for nonclient undergraduates. Nevertheless, some evidence in its

favor can be obtained from the findings of Hagebak and Parker

(1969) and Helweg (1971). The assumption underlying this hypothesis

is that this population of subjects is relatively well-adjusted

compared to that of clients in therapy, and therefore, will value

independence (Helweg, 1971) and responsibility (Hutcherson, 1967),

thus tending to make the nondirective therapist most appealing to


The second question about which predictions will be made is

the companion of the first, and concerns the subjects' satisfaction

ratings of the taped therapists, asking, What are the important

general qualities that nonclient undergraduates actually prefer

in a therapist? Because of a possibility of a discrepancy between

the qualities which subjects say they prefer and those to which

they respond positively on a more unobtrusive measure, it is

necessary to assess the subjects' preferences in more of an indirect

manner. Although it was not possible in the present design to

guarantee in the tapes equally high levels of the specific

characteristics listed on the ISORS, it is believed that the subjects'

ratings of the general dimensions of awareness, warmth and directive-

ness can be inferred from those characteristics on the ISORS and

compared with the satisfaction ratings of the general dimensions

on the CRQ. The hypotheses to be evaluated are as follows:

Hypothesis Ib (I)

The "interpersonally-oriented" therapists rated by judges as

offering high awareness will receive higher satisfaction ratings by

the subjects as a group than will those therapists offering high

warmth or high dirrctiveness.

This reasoning derives primarily from the findings of Lorr

(1965) and Boulware (1969), which suggested the importance of thera-

pist understanding to clients. However, in both studies "under-

standing" was ill-defined, and no attempt was made to distinguish

between the term's denoted intellectual and connoted affective

meanings, as explained earlier. Thus, in the present study "aware-

ness" refers to the therapist's capacity for intellectual and non-

intellectual understanding of the subject's communications and

feelings. Defined in this manner, it is anticipated that therapist

awareness will be the most important of the three qualities evaluated

in this hypothesis, as far as the subjects are concerned.

Hypothesis Ib (2)

The "interpersonally-oriented" therapists offering high warmth

will receive higher satisfaction ratings by the subjects as a group

than will the therapists offering high directiveness.

A high level of therapist warmth, while anticipated to be less

universally important to subjects than therapist awareness, is

nonetheless expected to find more general acceptance among the

subjects than therapist directiveness.

Mediating Variables: Subject Anxiety and Dependency

The next step is to proceed from prediction of group preferences

to prediction of those of the individual. The literature suggests

that among males, prediction of an individual's preferences of

therapist characteristics is related to the subject's general level

of adjustment and his level of dependency. Again, the preferences

will be examined differentially, according to whether they are

made directly, i.e. stated specifically on the ISORS, or indirectly,

via the satisfaction ratings of the taped therapist from the CRQ.

The relevant hypotheses to the state preferences are:

Hypothesis IIa (1)

There will be a positive relationship between subject maturity

(approximated by low anxiety and high independence) and the preference

for a personally-oriented "model" type of therapist, i,e. one high

on self-awareness, self-confidence and independence.

As previously indicated, the underlying assumption in this

hypothesis is that the more mature subjects will be apt to want

little more from the therapist than cognitive and intuitive under-

standing along with a behavioral example of maturity. Thus, this

group of subjects would be expected to prefer a therapist who is

high on the "personally-oriented" qualities: self-confident,

self-aware and independent.

Hypothesis IIa (2)

There will be a positive relationship between low subject

maturity (approximated by high anxiety and low independence) and

the preference for an interpersonally-oriented "provider" therapist,

i.e. one high in the qualities of intuitive, warmth and advising/


The assumption underlying this hypothesis is that those indi-

viduals funcl.ioninq at a diiminished level of personal adjustment

prefer a therapist who is warm and assertive.

Hypothesis IIa (3)

Subjects who score high on anxiety and on independence will

state a preference for a therapist whose dominant style is

"providing," but who is not nurturant, i.e. one who is high on

intuition and advising/guiding, but low on warmth.

This hypothesis is somewhat suggested by the data of Apfelbaum

(1958), Plummer (1966) and Heilbrun (1961b),

The following three hypotheses are companion to those above,

but are based upon the somewhat less obtrusive measure of therapist

preferences, the Client's Reaction Questionnaire (CRQ), where

responses gauge the subject's satisfaction with the taped therapists.

Hypothesis lib (1)

Subjects who exhibit a high level of maturity (i.e. score low

on anxiety and high on independence) will give the highest satisfac-

tion ratings to the "model" therapist, who is high in self-awareness,

self-confidence, and independence.

Studies by Apfelbaum (1958), Plummer (1966) and Heilbrun (1961b)

suggest this and the next hypothesis, with the exception of the

statements regarding the awareness variable.

Hypothesis lib (2)

Subjects who exhibit a low level of maturity (i.e. score high

on anxiety and low on autonomy) will give the highest satisfaction

ratings to the "provider" therapist, who is high in awareness,

wannth and directiveness.

The similarity between the above two hypotheses and the three

previous ones which were based upon the ISORS data is apparent.

Thus, it is expected that for this nonclient sample of subjects,

there will be relatively little difference between the subjects'

therapist preferences which are directly solicited and those obtained

in an indirect manner.

Early Development of Therapist Preferences

As noted earlier, the post hoc exploration of the origin of

basic preferences of therapist behavior must always be viewed with

some suspicion, especially when its primary purpose is the

garnishment of evidence in support of a particular theory of per-

sonality development. Such is not the goal here. The purpose of

offering the following hypotheses is to evaluate the role of

subject maturity in the determination of one's preferences of

parental behavior, and to facilitate assessment of the validity

of the basic assumption made by Freud (1935), Apfelbaum (1958),

May (1958) and Schutz (1958), concerning a relationship between

one's perceptions of parent and therapist. It is acknowledged

that there is little research to warrant these hypotheses, and

that in general, they represent an extension of related data

regarding preferences of therapist behavior.

Hypothesis lia

Subjects who exhibit a high level of maturity (i.e. score low

on anxiety and high on autonomy) will state a preference for a

father who most nearly fits the type of configuration of the "model"

therapist, i.e. one who is high on qualities of "self-aware,"

"self-confident" and "independent."

Hypothesis IlIb

Subjects who score relatively low on measures of maturity (i.e.

score high on anxiety and low on autonomy) will state a preference

for a father who most nearly fits the type of configuration of a

"provider" therapist, i.e. one who is high on qualities of "intuitive,"

"warm" and "advising/guiding."

Just as one would expect the dependent subject to seek more

from a therapist than the independent one (Apfelbaum, 1958), the

same differential in preferences is expected to apply attitudes

of these individuals toward the same-sexed parent. The preferences

for an ideal mother are expected to fall in the same directions

as those for the ideal father, but not to fit into any particular


Operational Definitions

Therapist Activity the therapist's basic, observable behavior

during the treatment session.

Therapist Affect the therapist's basic, predominant feeling

toward the client and/or toward himself.

Therapist Cognition the therapist's intellectual experiencing

during the therapeutic session.

Therapist Directiveness the extent to which the therapist

attempts to impose his own structure upon the course and

direction of the session.

Theprapist WarmLth the therapist's emotional attitude toward the

client, which is based upon genuine concern, respect and

positive regard.

Therapist Awareness the degree to which the therapist communicates

a genuine comprehension of the client's total condition

and of his thoughts and feelings regarding that condition.

Dependenc the experience of habitually seeking resources outside

of the self for satisfaction of certain basic psychological

needs and/or desires.

Maturity that stage of personal development characterized by a

feeling of self-confidence, a facility for self-awareness,

as well as for competent and flexibly autonomous behavior.

Self-Confidence the feeling which reflects self-esteem and one's

basic ability to cope well.

Competence behavior which is appropriate and largely successful

in its function.

Flexible Autonomy the ability to behave dependently or indepen-

dently, according to the reality of the situation.

Self-Awareness the ability to examine and evaluate one's own

thoughts, feelings, and other behavior in a reasonably objec-

tive manner.

Provider Therapist a therapist with an "interpersonal" style, who

generally seeks or feels obliged to give the client something

in therapy, and to assume the large share of responsibility for

the therapeutic process and progress.

Model Therajpist a therapist with a "personal" style, who is

largely concerned with behaving in a genuine manner with the

client, whom he allows to assume much of the responsibility

for the process and progress of therapy.

Interpersonal Orientation a therapeutic style which emphasizes

the mutuality and social interchange aspects of therapy,

and which for the therapist, entails focusing upon the client's

thoughts, feelings and actions.

Personal Orientation a therapeutic style which emphasizes contact

with the therapist's own experiencing of the therapeutic

process. and the reflection of this experiencing back to the




A total of 103 male undergraduates from the subject pool of the

Psychology Department of the University of Florida participated in

the experimental procedure. However, the data from seven of those

were deleted immediately from computation due either to an obvious

misunderstanding of, or an inattention to, the experimental instruc-

tions. The data of eight additional subjects were deleted in order

to facilitate analysis by leaving an equal number of subjects in

each cell. These eight were chosen randomly within the remaining

conditions having more than eleven subjects. Therefore, data from

15 subjects were discarded prior to computation, leaving a total

of 88, or eleven per condition.


The intention was to compare the therapist preferences of

individuals rated both within and between subjects. Two brief

questionnaires original to the present study were used. The first

was the Ideal-Self/Others Rating Scale (ISORS, Appendix B). This

scale assessed the subject's ratings of ideal parents, therapist

and self, in terms of six behavior characteristics related to

the active, affective and cognitive modes of behavior, of both the

personal and interpersonal styles. The importance of this

instrument is that it taps the stated preferences of therapist

qualities via the subject's description of the "ideal therapist."

It also allows for comparison of descriptions of ideal parents and

ideal therapist. Secondarily, the ratings for ideal self might have

proved useful as a comparison with the 16 PF data, but was not used

as a primary measure of maturity.

The second original instrument was the Client's Reaction

Questionnaire (CRQ, Appendix C), which represents a comparatively

unobtrusive measure of the subject's preferences of therapist behavior.

It was designed to be administered after a subject listened to a taped

segment of role-played therapy, and simply recorded his reactions to

the therapist's behavior in the segment.

The principal measure of the subject's level of maturity that was

utilized in this investigation was the Sixteen Personality Factor

Questionnaire (16 PF). It was used primarily to estimate the subject's

personality resources that were related to the complex issue of

maturity, particularly as it involved dependency. The 16 PF is a

self-administered inventory which is based upon sixteen basic

personality dimensions derived from factor analytic studies by Cattell

and his colleague (Cattell and Stice, 1950), This instrument was

selected for use in the present investigation for numerous reasons,

including the empirical manner in which it was developed, its

utility as a measure of autonomy and maturity, the existence of rele-

vant normative data, and its relative brevity in administration.

Although the full 16 PF was administered (Forms A and B), only

certain scales were used in testing the hypotheses of this study.

These scales were those relating to autonomy (scales E, M, Q1 and Q2)

and to maturity (scales, C, 0 and Q4).* The direct validities of

these scales with the underlying second stratum traits are as follows:

Autonomyr Anxiety

E (Dominance) .71 C (Ego Strength) .77

M (Autia) .71 0 (Adequacy) .86

Q1 (Conservatism) .68 Q4 (Ergic Tension) .63

Q2 (Self-
sufficiency) .80

The final questionnaire used was simply the Personal Data Sheet

(Appendix A), which secured demographic and historical information of

relevance to the topic under study. The historical information dealt

specifically with the subject's relationship with his parents.

The basic experimental design was of the repeated measures type.

Each subject completed the Personal Data Sheet, the ISORS, and the 16 PF

(Form A) respectively. Then, following a brief period, each was

assigned to one of the "interpersonal" (I) conditions listed below,

and listened to one of eight audio-taped segments of role-played thera-

py sessions. Each student listened to a standard "personal" (P) style

tape as well. To eliminate a possible order effect, the order of the

two tapes was counterbalanced. The specific conditions used were as


*Cattell et al. termed these scales "independence" and "anxiety."

Condition # Subjects Therapist Qualities

11 11 High A High W High D

12 11 Low A High W High D

13 11 High A Low W High D

14 11 Low A Low W High D

15 11 High A High W Low D

16 11 Low A High W Low D

17 11 High A Low W Low D

18 11 Low A Low W Low D

P 88 High Aut High S.C. High S.A.

(Legend: A = Awareness, W = Warmth, D = Directiveness, Aut = Autonomy,

S.C. = Self-Confidence, S.A. = Self-Awareness.)

The simulated therapy tapes are the product of lengthy development

and frequent revision by the investigator, who also plays the role of

therapist on the tapes. Prior to their use in this study, the tapes

underwent a series of ratings by three colleagues of the investigator,

each of whom held at least a master's degree in one of the helping

professions, and who were carefully instructed as to the criteria being

rated. The tapes were considered reliable when the mean.of thejudges'

ratings fell within the following limits of acceptability for each

characteristic being rated: low = 1-3.3, moderate = 3.4-7.3, and high =

7.4-10. In several instances (13%), where numerous revisions and

ratings failed to produce such an acceptable mean, the tape was con-

sidered reliable when two of the three judges produced ratings falling

within the limits of acceptability. In addition, the final question on

the CRQ was intended to tap the subjects' ratings of the validity of the

tapes. The purpose behind these various ratings, of course, was to help

ensure that all of the combinations did indeed reflect the levels of

therapist characteristics that were intended.

After listening to that "I" tape randomly assigned to them as well

as the standard "P" tape, the subjects were asked to briefly rate their

level of satisfaction with the therapist on the tapes by completing

the CRQ.


The subjects were run through the experimental procedure in small

groups, and contributed approximately two hours of their time. Upon

each group's arrival, the experimenter introduced himself and stated

the purpose of the investigation as being to learn something about the

way college students view themselves and other significant people in

their lives. There were informed that the procedure was a benign one,

but that no further details about the nature of the study could be

given until the end of the session.

The subjects were then seated and asked to complete the Personal

Data Sheet, the ISORS and Form A of the 16 PF. A brief rest period was

then allowed, lasting five minutes, after which the subjects completed

Form B of the 16 PF. They then listened to the two tapes assigned to

them, and evaluated the therapist's performance after each one by com-

pleting the CRQ. In making their preferences via these evaluations,

they were instructed to listen carefully to the tapes, since afterwards

they would be asked to rate the therapist. The rationale given to the


subjects was that different people prefer different kinds of therapists,

and thus that no judgements are correct or incorrect. A debriefing

session was then allowed, when desired by the subjects.


The data analysis began by focusing upon the subject's stated

preferences on basic therapist characteristics, vis a vis the

variables of cognition, affect and activity. In order to check

the accuracy of Hypotheses la (1)-(3), chi-square tests were conducted

on all of the subjects' ratings of their ideal therapists, which were

derived from responses to the Ideal-Self/Others Rating Scale (ISORS).

The purpose of using the chi-square was to compare expected scores

with obtained scores (Smith, 1962). The results of this analysis

as they apply to Hypothesis la (1) are summarized in Table 4.1, and

indicate that subjects subscribe to five of the six therapist

characteristics at a statistically significant level. Only inde-

pendence failed to find strong support as a personal quality of

the ideal therapist.

Hypothesis Ib (1)-(2) focused upon the three characteristics of

the interpersonal therapist orientation, namely, awareness, warmth and

directiveness. The hypotheses dealt with a comparison of these

qualities with each other, as gauged by satisfaction ratings of the

interpersonally-oriented therapists. The ratings were obtained from

the subjects' responses on the Client Reaction Questionnaire (CRQ),

after their exposure to the eight simulated therapy tapes. The

results of t-tests used in making these comparisons indicated that

the awareness variable played a significantly greater role in the


Table 4.1. Summary of preferences for ideal therapist traits,
as rated by ISORS responses for all subjects (N = 88).

Ideal therapist Behavioral Chi-square
characteristics mlodality _value __Significance (3df)

Advising/guidin! Activity 355.86 .001

Intuitive Cognition 348.20 .001

Self-confident Affect 176.32 .001

Self-aware Cognition 94.10 .001

Warm Affect 48.65 .001

Independent Activity 5.54 n.s.

determination of subjects' satisfaction ratings of the interpersonally-

oriented taped therapists than did warmth or directiveness (Table 4.2).

After these data had been collected, it became apparent that

hypotheses were lacking which would offer predictions regarding the

satisfaction of subjects with the behavior of the standard personally-

oriented therapist compared with that of the eight interpersonally-

oriented therapists. Therefore, such a comparison was made on a

post hoc basis.

The first step in this procedure was to conduct an analysis

of variance on the CRQ ratings of all 88 subjects for the standard

personal therapist tape. It will be recalled that this tape was

presented to each subject, along with whichever interpersonal

therapist taDe was assigned to him in his particular experimental

condition. Despite the fact that independent judges had rated the

personal therapist as meeting the necessary criteria specified

earlier, it was deemed important to determine the level of variability

in the satisfaction ratings of this tape across the eight different

conditions as a reliability check. The results of the ANOVA were

reassuringly nonsignificant (F = 1.97). Therefore, the model

therapist tape appears to have been viewed similarly by all eight

groups of subjects, and can be considered reliable as a standard

example of personally-oriented therapist behavior.

The next step in the post hoc analysis involved conducting an

ANOVA on the CRQ ratings of all subjects for the eight inter-

personally-oriented therapist tapes. This procedure was viewed as

necessary to provide certainty that subjects did in fact perceive


Table 4.2. Comparison of interpersonally-oriented therapist
qualities, as gauged by CRQ satisfaction ratings
of all subjects (N = 88).

Interpersonal therapist
S qualities t-value __ Sig nificance

Awareness vs. warmth 8.13 .001

Awareness vs. directiveness 10.00 .001

Warmth vs. directiveness -0.56 n.s.

and react differently to each of these tapes. Again, the ANOVA

provided reassurance (F = 6.00, p<.01). A graphical representation

of the CRQ ratings for both the standard model and the eight inter-

personal therapist tapes found in Figure 4.1.

The final step in the post hoc procedure was to conduct

individual comparisons of the CRQ ratings for the two tapes (one

personal, one interpersonal) heard by subjects in each condition.

These data would then provide a clearer idea of the relative

importance of both the general roles as well as the specific

behaviors of a therapist to the subjects as a group. This analysis

utilized individual t-tests, and the results are summarized in

Table 4.3.

Inspection of Figure 4.1 and Table 4.3 yields several pertinent

observations. First, it is apparent that the personally-oriented

therapist produced broader and more consistent satisfaction among

subjects than did the various interpersonally-oriented therapists.

Indeed, only one tape in the latter category (tape 5) was rated higher

than the personal therapist tape, although the difference was not

significant. Second, and of more potential importance, is the

pattern that emerges when the individual configuration of therapist

traits is examined across conditions. Of the three conditions for

which there were significant differences in subject satisfaction

scores (numbers 4, 6 and 8), all of the interpersonal therapists

involved were characterized by a low level of awareness. In the

fourth condition also characterized by low therapist awareness

(number 2), there was a nonsignificant trend in the same direction.

Figure 4.1. Comparison of subject CRQ satisfaction ratings of
personally-oriented and interpersonally-oriented
therapist taDes for all eight experimental conditions
(N = 88).


40 ----Interpersonally-oriented tapes

Personally-oriented tape

(standard across all

1 2 3 4 5 6 7 8

Tape conditions



Table 4.3. Comparison of subject satisfaction (CRQ) ratings for the personally-oriented
therapist tape paired with each of the interpersonally-oriented therapist tapes,










Comparison tape qualities
*Personal (P) **Interpersonal (I)

High SA,SC,Ind High A,W,D

same Low A, High 'W,D

same High A,D, Low W

same Low A,W, High D

same High A,W, Low D

same Low A,D, High W

same Hioh A, Low W,D

same Low A,W,D

scores (P/I)























n. s.


n.s .


*Personal tape is moderate in the interpersonal qualities (awareness, warmth,

**Interpersonal tapes are moderate in personal qualities (self-awareness, self-
confidence, independence).

FAll tests are two-tailed, d.f. = 10.

Unlike conditions 4, 6 and 8, however, condition 2 involved high

levels of the other independent variables (warmth and directiveness).

In the remaining four conditions, which were all characterized by

hiqh levels of therapist awareness (numbers 1, 3, 5 and 7), there

were no significant differences in subject satisfaction ratings of the

personal and interpersonal therapist tapes.

The second level of data analysis attempted to match preferences

of therapist characteristics with subjects' personality traits. The

findings are based upon subjects' self-ratings on the 16 PF and

their therapist preferences, as assessed by their responses on the

ISORS. The 88 subjects were first categorized according to their

respective scores on the 16 PF second-order scales of Anxiety and

Independence. Their relative scores above or below the group

means resulted in a 2X2 matrix, formed by high vs. low scores on

the two variables. Then, chi-square tests were employed to evaluate

the null hypothesis that the therapist preferences of various

sub-groups of the sample were normally distributed. The results

of these analyses (Tables 4.4 through 4.7) indicate little variability

in preferences between subject groups, and a strong preference

for a therapist high in certain personal qualities (self-confidence

and self-awareness) and interpersonal skills (guidance and


In an attempt to evaluate the possibility of a main effect of

subject anxiety and independence upon therapist preferences via

ISORS ratings, a multivariate analysis of variance was conducted.

The results indicated no significant main or interaction effect


Table 4.4. Preferences of therapist characteristics, as rated by
ISORS responses of subjects scoring high on 16 PF
second-order factor of Independence and low on Anxiety
(N = 28).

Ideal therapist
chariactrise tics







oriental tion







value Siqnrlificance(df-3)

100.75 .001

91.35 .001

52.64 .001

31.05 .001

5.36 n.s.

4 35 n.s.



Table 4. 5. Preferences of therapist characteristics, as rated by
ISORS responses of subjects scoring high on 16 PF second-
order factor of Anxiety and low on Independence (N = 30).

Ideal therapist Therapist Chi-square
characteristics orientation value

Intuitive Interpersonal 133.52

Advising/guiding Interpersonal 122.37

Self-confident Personal 66.00

Self-aware Personal 34.35

Warm Interpersonal 23.03

Independent Personal 3.54

S_ini fic an cdlf=3)__








Table 4.6. Preferences of therapist characteristics, as rated by
ISORS responses of subjects scoring high on 16 PF
second-order factors of Anxiety and Independence (N = 18).

Ideal therapist







Therapist Chi-square
orint nation va_ lvue __Si ificance_(df=3_

Interpersonal 93.72 .001

Interpersonal 70.79 .001

Personal 51.22 .001

Interpersonal 27.89 .001

Personal 20.82 .001

Personal 4,89 n.s.


Table 4.7. Preferences of therapist characteristics, as rated by
ISORS responses of subjects scoring low on 16 PF second-
order traits of Anxiety and Independence (N = 12).

Ideal therapist
cha racte trristics







Therapist Chi-square
ori on ta tion _valu .1c__ _Significance(df=3)

Interpersonal 63.80 .001

Interpersonal 32.02 .001

Personal 17.21 .001

Interpersonal 7.27 n.s.

Personal 5.46 n.s.

Personal 5.16 n.s.

of these two personality variables, a finding consistent with the

above data in indicating that these variables play no significant

role in the "public" selection of therapist characteristics.

As a means of evaluating the possible role played by subject

anxiety and independence in the determination of satisfaction

ratings for the various taped therapists, several procedures were

conducted. First, it was necessary to determine whether or not

a "global" satisfaction score could be derived from the CRQ to

facilitate analysis. This was accomplished by obtaining product-

moment correlations for all 88 subjects' responses on the three

CRQ items gauging therapist satisfaction, as they pertained to

both the personally- and interpersonally-oriented therapists.

As indicated in Table 4.8, the resulting six correlations were

relatively high, and significant at well beyond the .001 level.

Thus, it seems fairly safe to conclude that there is significant

overlap among the three items for each set of data to justify

collapsing the two sets of three individual satisfaction item

scores into one global satisfaction score each.

The second step in this phase of the analysis involved the

categorization of the 88 subjects, according to their respective

scores on the 16 PF second-order scales of Anxiety and Independence.

Having located the subjects in their respective quadrants in the

manner described earlier, the global satisfaction scores for each

of the four subject groups were calculated (Table 4.10).


Table 4.8. Product-moment correlations of items 1 through 3 of
Client Reaction Questionnaire (CRQ) (N = 88).

Comparison Comparison
tapes__ items (CRQ) r-value

Interpersonal 1-2 .85
(tapes 1-8)

(tape 9)

t-value p-value

14.87 .001

10.24 .001

13.33 .001

11.16 .001

9.27 .001

10.53 .001

Table 4.9. Comparison of subjects CRQ satisfaction ratings of personally-oriented vs.
interpersonally-oriented therapists, with subjects grouped according to
their scores on 16 PF second-order factors of Anxiety and Independence (N

Mean satisfaction scores (CRQ)
Personal therapist Interpersonal therapist

t-value Significance

A. Low Anxiety/
low Independence

B. Low Anxiety/
high Independence

C. High Anxiety/
low Independence

D. High Anxiety/
high Independence






12 4.69

28 4.58

30 3.49

18 0.68

Several observations are apparent from Table 4.9. First, the

personally-oriented (model) therapist is consistently rated higher

by all four subject groups than is the collective set of eight

interpersonally-oriented therapists. Three of the four differences

between the groups reach statistical significance. Second, the

highest satisfaction ratings of a therapist by any group were given

to the personally-oriented therapist by the subjects low in both

anxiety and independence. The lowest satisfaction ratings given

this therapist derived from the subjects with the opposite personality

configuration (hiqh in anxiety and independence). However, an

analysis of variance revealed that the difference between these two

sets of ratings was nonsignificant.

It is noteworthy that the highest satisfaction ratings of the

interpersonally-oriented therapists also came from the subjects

high in anxiety and independence. This finding prompted closer

scrutiny of the CRQ data, which revealed that twelve (67%) of the

subjects in this group happened to fall into experimental conditions

that included exposure to interpersonally-oriented therapists who

were high in at least two of the three qualities of awareness,

warmth and directiveness. Conversely, nine (75%) of the subjects

in the former group (low in anxiety and independence) were exposed

to interpersonal therapists who were low in at least two of those

three basic qualities.

Finally, a multivariate analysis was conducted on the CRQ

satisfaction ratings of both the personally- and interpersonally-

oriented therapists, in order to evaluate the relative roles played

by the subjects' levels of anxiety and independence in the determina-

tion of these scores. The results indicated an absence of a main

effect for either anxiety or independence for the interpersonal

therapist ratings, hut the presence of an interaction between the

two variables (r = .31, pc.05). There were no significant effects

for the personal therapist data.

The third level of analysis dealt with the issue of possible

relationships between the maturity of the subjects and their stated

preferences of therapists and fathers. Hypothesis liIa predicted

that subjects scoring high on the 16 PF Independence scale and low

on Anxiety would express a preference for an ideal father figure

most nearly like the personally-oriented ("model") therapist,

i.e. one highly "self-aware," "self-confident" and "independent."

A chi-square analysis conducted on the ISORS data of the 28

subjects falling into this subgroup found that, indeed, those

three ISORS qualities were attributed to the ideal father at a

significant level (Table 4.10). However, the qualities "advising/

guiding" and "intuitive" were also attributed to the ideal father

at a significant level, a pattern very similar to that noted

earlier for the ideal therapist.

Hypothesis IIIb was based upon data from those subjects

presumed to be lower in maturity, with low scores on Independence

and high scores on Anxiety. This hypothesis predicted that such

subjects would state a preference for an ideal father who most

nearly conformed to the configuration of the interpersonally-

oriented "provider" therapist who exhibited a highly "advising/


Table 4.10. Preferences for ideal father characteristics, as rated by
ISORS responses of subjects scoring low on 16 PF second-
order factor of Anxiety and high on Independence (N = 28),

Ideal father







Behavioral Chi-square
modaiti__ty_ value





Cognition 44.05

Cognition 43.74












guiding," "intuitive" and "warm" behavioral style. As before,

a chi-square analysis was conducted, with the results being similar

to those for the subjects presumed to be of a higher maturity level

(Table 4.11). Once again, five of the six ISORS characteristics

were attributed to the ideal father, instead of only three as was


After noting this similarity in choice of characteristics for

an ideal father figure, chi-square analyses were also conducted on

the preferences of the remaining two subject groups, i.e. those

characterized either by low or high levels of both anxiety and

independence. As seen in Tables 4.12 and 4.13, very similar patterns

emerged for these groups compared to the previous two, indicating the

lack of influence apparently held by subject anxiety and independence

in the selection of ideal father figures.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs