Title Page
 Table of Contents
 List of Tables
 List of Figures
 Review of the literature
 Design of the study
 Data analysis and results
 Appendix A: Scripts
 Appendix B: Instruments
 Appendix C: Tables
 Biographical sketch

Group Title: influence of selected factors on the social orientation of identified adolescent delinquents toward counselors /
Title: The Influence of selected factors on the social orientation of identified adolescent delinquents toward counselors /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098108/00001
 Material Information
Title: The Influence of selected factors on the social orientation of identified adolescent delinquents toward counselors /
Physical Description: xii, 172 leaves : ill. ; 29 cm.
Language: English
Creator: Tomblin, James Gray, 1944-
Publication Date: 1977
Copyright Date: 1977
Subject: Counselor and client   ( lcsh )
Adolescent psychotherapy   ( lcsh )
Counselor Education thesis Ph. D
Dissertations, Academic -- Counselor Education -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1977.
Bibliography: Includes bibliographical references (leaves 164-170).
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by James Gary Tomblin.
 Record Information
Bibliographic ID: UF00098108
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 - 001481663
oclc - 31319839
notis - AGZ3659


This item has the following downloads:

PDF ( 7 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Tables
        Page vi
        Page vii
        Page viii
    List of Figures
        Page ix
        Page x
        Page xi
        Page xii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Review of the literature
        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
    Design of the study
        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
    Data analysis and results
        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
    Appendix A: Scripts
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
    Appendix B: Instruments
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
    Appendix C: Tables
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
    Biographical sketch
        Page 171
        Page 172
        Page 173
        Page 174
Full Text










The following persons have each made a unique contri-

bution toward the completion of this dissertation:

Dr. Robert Stripling, Committee Chairman;

Dr. William Ware, Committee Member;

Dr. Robert Ziller, Committee Member;

Lorraine Thomas, typist;

The students at Lancaster Youth Development Center

who participated in the study.

During the long period of time that I fluctuated between

completing and not completing this dissertation, I was

supported and encouraged by a number of friends. Most

important among them were Claudia, Julianne and Jeannie.

The person who has given me the strongest incentive during

this period has been my daughter, Sheri.



ACKNOWLEDGMENTS . . . . . . . . . . .iii

LIST OF TABLES . . . . . . . . . . vi

LIST OF FIGURES . . . . . . . . . . ix

ABSTRACT . . . . .. . . . . . . . x


I INTRODUCTION .. . . . . . . ... 1
Statement of Problem . . . . . 3
Rationale . . . . . . . . 4
Purpose . . . . . . . . . 9

Role . . . . . . . . .. . 14
Theoretical Literature . . . ... 14
Role Expectations of Clients . . .. .22
Role Conflict in Counseling . . .. .29
Counselor Sex . . . . . . .. 41
Criterion Variables . . . . . .. 45

III DESIGN OF THE STUDY . . . . . .. 49
Subjects . . . . . . . .. 49
Counseling Tapes . . . . . .51
Instruments . . . . . . . .. 54
Psychotherapy Expectancy Inventory-
Revision . .. . .. .. .. .. 54
Social Orientation Tasks ...... . 61
Data Sheet . . . . ... . . 66
Procedure . . . . . . . ... 66
Grouping . . . . . . . . 66
Introduction to Subjects . . . .. 67
Administration of the PEI-R . . .. .68
Experimental Manipulation . . .. .69

Data Analysis . . . . . . ... 72
Results . . . . . . . ... 79
Identification as Criterion . . .. .80
Counselor role behavior . . .. .80



Counselor sex . . . . . .
Client role expectation by counselor

role behavior . . .
Subject sex . . . .
Inclusion as Criterion . .
Counselor role behavior .
Counselor sex . . .
Client role expectation by
role behavior . . .


Subject sex . . . .
Summary . . . . . .

Implications . .
Counseling Practice
Research . . . .
Limitations of the Study



. . 119

B INSTRUMENTS ... . . . . .... 135

C TABLES . . . . . . . . .

REFERENCES . . . . . . . . . .



. 84
. 85
. 93
. 93
. 96

. 98
. 101
. 106

. . 159

. 164

. . 171


Table Page

3.1 Number and Percent of Sample by Race and
Sex . . . . . . . . .. .. . 52

3.2 Number of Sample by Age and Sex . . .. 52

4.1 Means and Standard Deviations, APPROVAL-
ADVICE Scores, Groups Classified by Role
Expectation . . . . . . . ... 75

4.2 Analysis of Variance, APPROVAL-ADVICE Scores,
Groups Classified by Role Expectation . . 75

4.3 Means and Standard Deviations, AUDIENCE-
RELATIONSHIP Scores, Groups Classified by
Role Expectation . . . . . ... 76

4.4 Analysis of Variance, AUDIENCE-RELATIONSHIP
Scores, Groups Classified by Role Expectation 76

4.5 Paired t-Test, Approval-Advice and Audience-
Relationship Scores . . . . . .. 78

4.6 Four-Way Factorial Analysis of Variance,
Identification as Criterion . . . ... 81

4.7 Counselor Role Behavior, Means and Standard
Deviations--Identification Scores . . .. 83

4.8 Counselor Sex, Means and Standard Deviations--
Identification Scores . . . . ... 83

4.9 Client Role Expectation, Means and Standard
Deviations--Identification Scores . . .. 86

4.10 Client Role Expectation by Counselor Role
Behavior, Means and Standard Deviations--
Identification Scores . . . . . .. 86

4.11 Test of Simple Main Effects--Identification
Client Role Expectation by Counselor Role
Behavior . . . . . . . .. 87

LIST OF TABLES (Continued)

Table Page

4.12 Subject Sex, Means and Standard Deviations--
Identification Scores . . . . ... 90

4.13 Counselor Sex by Subject Sex, Means and
Standard Deviations--Identification Scores. 90

4.14 Client Role Expectation by Counselor Sex,
Means and Standard Deviations--Identification
Scores . . . . . . . . ... 91

4.15 Client Role Expectation by Subject Sex, Means
and Standard Deviations--Identification Scores 91

4.16 Counselor Role Behavior by Counselor Sex,
Means and Standard Deviations--Identification
Scores ... . . . . . . . 92

4.17 Counselor Role Behavior by Subject Sex, Means
and Standard Deviations--Identification Scores 92

4.18 Client Role Expectation by Counselor Role
Behavior by Counselor Sex, Means and Standard
Deviations--Identification Scores . . .. 94

4.19 Four-Way Factorial Analysis of Variance
Inclusion as Criterion . . . . . 95

4.20 Counselor Role Behavior, Means and Standard
Deviations--Inclusion Scores . . . .. 97

4.21 Counselor Sex, Means and Standard Deviations--
Inclusion Scores . . . . .. 97

4.22 Client Role Expectation, Means and Standard
Deviations--Inclusion Scores . . . .. 99

4.23 Client Role Expectation by Counselor Role
Behavior . . . . . . . . 99

4.24 Test of Simple Main Effects--Inclusion
Client Role Expectation by Counselor Role
Behavior . . . . . . . ... 102

4.25 Subject Sex, Means and Standard Deviations--
Inclusion Scores . . . . . ... 103

4.26 Counselor Sex by Subject Sex, Means and
Standard Deviations--Inclusion Scores . . 103

LIST OF TABLES (Continued)

Table Page

4.27 Test of Simple Main Effects--Inclusion
Counselor Sex by Subject Sex . . . .. .105



Figure Page

4.1 Interaction: Client Role Expectation by
Counselor Role Behavior Identification
Scores . . . . . . . . ... 88

4.2 Interaction: Client Role Expectation by
Counselor Role Behavior Inclusion Scores 100

4.3 Interaction: Counselor Sex by Subject Sex -
Inclusion Scores . . . . . . .. 104

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



James Gary Tomblin

December 1977

Chairman: Dr. Robert O. Stripling
Major Department: Counselor Education

Within the theoretical framework of this study the

counseling relationship was described as a social system

which would be influenced by a dynamic developmental process.

Using this perspective, the initial counselor-client

interaction was designated as the first stage of this develop-

ing social system and therefore predictive of the future

existence of the system.

A review of the literature demonstrated that among the

factors which may influence the counseling relationship

during its initial stage are client role expectation, coun-

selor role behavior and counselor sex. It was suggested

that the interaction of the factors, client role expectation

and counselor role behavior, would be important during the

initial counselor-client communication and that the develop-

ment of a counseling social system would be influenced by

this interaction. It was proposed that the development of

the counseling social system would be reflected by the degree

of client identification with the counselor and the degree

to which the client included himself or herself in the

social field of the counselor.

This research was conducted within an experimental

design which was devised to approximate an initial coun-

seling interaction. The initial counseling role expectations

of 104 male and 23 female adolescent residents of a school

operated by the Florida Division of Youth Services were

measured on the Psychotherapy Expectancy Inventory-Revision


RELATIONSHIP expectations. Later, subjects listened to one

of four audio tapes of a role played counseling session.

The content of the tapes was manipulated in order to portray

either directive or nondirective role behavior and to pre-

sent either a male or female counselor. Immediately after

listening to the tapes, subjects completed Social Orienta-

tion Tasks (SOT) which measured the degree to which each

subject identified with the counselor on the tape and the

degree to which each subject included himself or herself in

the social field of the counselor.

The influence of the factors, client role expectation,

counselor role behavior, counselor sex and subject sex, on

the subjects' identification and inclusion scores was

analyzed by two four-way factorial analyses of variance.

It was found that, as a group, subjects in the study


role expectations. As a group, subjects exposed to

directive counselor role behavior had higher inclusion and

identification scores than subjects exposed to nondirective

counselor role behavior. An interaction effect was found

between client role expectation and counselor role behavior

with both identification and inclusion scores. The male

counselor evoked higher levels of identification than the

female counselor. At the same time, there was a tendency

for identification and inclusion scores to be higher among

subjects exposed to an opposite-sex counselor than a like-

sex counselor. This interaction was significant statistically

when inclusion was the criterion.

Several conclusions were drawn from the data and some

implications for counseling practice and future research

were discussed. The limitations of the study were delin-




One way of conceptualizing the counselor-client rela-

tionship is with regard to the social system of which it is

a part. Such a conceptualization views the counseling inter-

change in light of natural social processes and applies the

disciplines of social psychology and sociology to explain

the dynamics of the counseling interaction.

There are at least two ways of describing the counsel-

ing dyad in a social system framework. Bentley (1968)

identified one approach when he depicted counseling as a

dynamic interaction of systems and noted the counselor's

relationship to members of various social systems. This

approach places the counseling relationship in the context

of larger social systems such as the school or community.

The second approach portrays the counseling dyad alone as a

social system and examines the counseling process in light

of the dynamics of such a dyadic relationship. In the

present study, using the second definition, the counseling

relationship is conceptualized as a developing social

system subject to specific dynamic processes.

In a dynamic social system, the action of factors

during one period of the system's existence is related to

interaction in the social system at a later stage of its

development. Lennard and Bernstein (1960) have referred to

the process determinants which create this phenomenon as

"developmental" or "phase" properties of social groups.

They observed that in order to understand or predict the

behavior of individuals who are interacting "one needs to

know when within the life of the system the interaction is

taking place . ." (p. 27). Raush (1965) found support for

this assumption regarding system development in a study of

disturbed and delinquent adolescents. By studying a specific

role relationship over a limited time span, he found that

knowing what had transpired early in a relationship was a

better predictor of future interaction than the setting or

the personality characteristics of the participants.

Perhaps the most significant phase of the counseling

social system is the initial interview. Baum and Fczler

(1964) and Kell and Mueller (1966), who view the initial

interview as a microcosm of the whole course of counseling,

stress the importance of the initial contact in affecting

the resulting reaction of both the client and the counselor.

In a study of "failure" cases in psychotherapy, Mendelsohn

(1966) reports findings which substantiate the significance

of the initial interview. He found that of clients who

dropped out permanently, 65 percent did so between the first

and second interviews and that 31 percent dropped out

between the second and third interviews. During the initial

interview, factors are introduced and processes are begun

which affect the subsequent development of the counseling

social system. In some cases, the effect of these variables

will be a failure of the system to develop.

There are numerous factors present during the initial

interview which potentially affect the development of the

counseling social system. Three of these factors are client

role expectation, counselor role behavior and counselor sex.

When viewed in the theoretical framework of the present study,

the variables of client role expectation, counselor role

behavior and counselor sex are significant because of their

ability to influence the initial response of both counselor

and client and therefore affect the subsequent process of

the dyadic interaction. In this sense they are predisposing

factors which are activated within the counseling dyad and

which affect the development of the social system.

Statement of Problem

In designing the present research the basic problem to

be addressed was: what influence will client role expecta-

tion, counselor role behavior and sex of the counselor have

on the development of the counseling social system? More

specifically, in a group of institutionalized adolescent

delinquents, what will be the influence of client role

expectation, counselor role behavior and counselor sex, during

the initial interview, on the client's social orientation

in regard to the counselor?


A concept which is crucial to a social system descrip-

tion of the counseling dyad is role. Bentley (1968), in

applying the concept of role to the counseling relationship,

suggested that role is "an inclusive term consisting of

role performance, role expectations, role conceptions and

role acceptance" (p. 74). Perhaps among these subsidiary

role constructs, the most important to an interactive de-

scription of counseling is role expectations. Cottrell

(1933) emphasized the significance of role expectations

when he stated:

Frequently we fail to recognize early enough
what might be called expectations entertained
by the subject as to the actions or responses
which are to come from other persons ..
There is no conception of one's role,
conscious or unconscious, without a reference
to what action is expected in the situation
of which the role is a part. (p. 110)

The construct of role assumes that another person or group

of persons will hold expectations toward the role position

or toward a specific person occupying a role position. In

the social system delimited by the counseling dyad, one can

not make reference to counselor role without consideration

of the role expectations of the client. Conversely, the

role of client can not be conceptualized without reference

to expectations of the counselor.

A number of studies (Clemes and D'Andrea, 1965; Heine

and Trosman, 1960; Overall and Aronson, 1963; Price and

Iverson, 1969; Severinsen, 1966) have observed initial

client role expectations in relation to the counselor's

perception of his or her role or in relation to actual role

behavior of the counselor during the initial stages of

counseling. It has been demonstrated that incongruence

between counselor and client perceptions of counselor role

exerts a significant influence on subsequent client behavior.

For example, it has been shown that, when such incongruence

exists, clients are more likely to terminate prematurely

(Heine and Trosman, 1960) and be dissatisfied with counsel-

ing (Severinsen, 1966) and less likely to be attracted to

the counselor (Price and Iverson, 1969). Conversely, when

client role expectations and counselor role behavior are

congruent, favorable outcomes are more likely to occur.

These studies suggest that although it may be proposed

that client role expectations independently influence the

client's initial attitude toward his or her counselor, the

effects of client role expectations should be studied with-

in the context of their interaction with counselor role

behavior. Similarly, whereas there may be a relationship

between counselor role behavior in the initial interview and

subsequent client attitude toward the counselor, the

influence of counselor role behavior should be interpreted

in relation to client role expectation.

Most generally accepted definitions of counselor role

and resulting role performance are based on middle class

experiences on the part of the role definers (supervisors,

counselor educators, counselors, professional organizations)

and in the context of counseling experiences with middle

class clients. Hunt (1960), in a discussion of the relation-

ship of the efficacy of psychotherapy to social class, con-

cluded that present psychotherapeutic techniques have been

drawn from treatment and research on a biased sample, upper-

middle and upper class persons. He further suggested the

possibility that efficacy of psychotherapy as it is presently

constituted is confined to middle and higher class patients.

Schofield (1964), in summarizing therapists' descrip-

tions of the typical and desirable psychotherapy client, has

depicted him as young, attractive, verbal, intelligent and

social (YAVIS). The YAVIS client is more likely to remain

in therapy until mutual discontinuance while the non-YAVIS

client more frequently unilaterally drops out of therapy

and is subsequently described by the therapist as "unmoti-

vated" and "not a good candidate for psychotherapy." It can

be inferred that middle class, educated individuals whose

background and psychosocial environment are similar to that

of the counselor and who are likely to have had contact with

professional helpers and thus with normative role behavior

will have expectations which are more congruent with the

counseling role expectations and behavior of professionals

than are the expectations of lower class, less educated,

socially different individuals.

Hollingshcad and Redlich (1958), in their survey of

class and mental illness, found that only a few persons in

the two lower classes entered therapy voluntarily. In

general, those clients who did voluntarily begin therapy

expected the therapist to be authoritarian. According to

their research, upon entering therapy lower class clients

presented somatic symptoms and continued to express an

organic orientation ("asking for 'shots' and 'pills'")

throughout therapy.

In an analysis of the failure of traditional counseling

approaches with disadvantaged clients, Calia (1966) has

delineated several differences in this group of clients and

the usual counseling contact. He considered traditional

assumptions about counseling in light of the personal makeup

and cultural background of disadvantaged clients and proposed

alternative roles for counselors working with this popula-

tion. These suggested roles included active counseling

styles, use of environmental resources in counseling and

counselor help in improving the counselee's environment.

In summary, the foregoing literature suggests that

lower class persons tend to seek counseling less frequently

and expect counselor behavior which is, in many instances,

contrary to traditional counseling roles. In response to

counseling behavior in the initial interview which is incon-

gruent with their role expectations, they often terminate or

otherwise resist. Interpreted from the perspective of the

present study, these findings suggest that in counseling

social systems involving lower class clients the development

of the system is retarded and often terminated due to incon-

gruent role expectations.

The particular lower class group of interest in the

present study are institutionalized adolescent delinquents.

Although there is not a clear relationship between class or

economic status and delinquent behavior, there does seem to

be a connection between class and being legally classified

as a delinquent. According to Sutherland (1939):

First the administrative processes are more
favorable to persons in economic comfort than
to those in poverty. So that if two persons
on different economic levels are equally
guilty of the same offense, the one on the
lower level is more likely to be arrested,
convicted and committed to an institution.
Second, the laws are written, administered and
implemented primarily with reference to the
types of crime committed by people of lower
economic levels. (p. 179)

In addition, adolescents in the higher socioeconomic level

are more likely to be identified as having emotional prob-

lems and be referred for counseling or psychotherapy while

adolescents of lower class families are more likely to be

identified as delinquent or criminal. The role expectations

of adolescents identified as delinquent should therefore be

similar to those of other lower class groups.

Another important variable when defining the counseling

interaction as a developing social system is the sex of the

counselor. In general, counseling research has suggested

that, regardless of the sex of the client, male counselors

are preferred more frequently than are female counselors.

One study, reported by Fuller (1964), revealed that univer-

sity students of both sexes preferred male counselors more

frequently than female counselors and that preferences for

male counselors were most consistent than preferences for

female counselors. Fuller related her findings to the

cultural preference for the masculine role and the attribu-

tion of more highly valued characteristics to males.

In addition to the usual cultural preferences for the

male role, there would seem to be support for the existence

of an even stronger preference among institutionalized

delinquents. In the delinquent subculture potency is valued,

and the male role is esteemed because of cultural sanctions

on aggression by the male and because of the dominant posi-

tion that the male occupies. The potential exists for the

delinquent adolescent to initially identify more favorably

with a male counselor.


During the initial phase of the social system of coun-

seling, factors are introduced which are significant in

affecting the later development of the system. As suggested

by the literature, these factors include client role expec-

tations, counselor role behavior and counselor sex. The

purpose of the present research was to study the influence of

these factors on client social orientation toward the coun-

selor during the initial phase of the social system involving

a counselor and an institutionalized adolescent delinquent


The present research study was carried out within an

experimental design which was devised to approximate an

initial counseling contact. The initial counseling role

expectations of subjects (institutionalized adolescent de-

linquents) were measured on the Psychotherapy Expectancy

Inventory-Revision which yielded Approval-Advice and

Audience-Relationship expectations. Later, subjects listened

to one of four audio tapes of role played counseling ses-

sions. The content of the tapes was manipulated in order

to portray either directive or nondirective role behavior

and to present either a male or female counselor. Immedi-

ately after listening to the tapes, subjects completed

social orientation tasks which measured the degree to which

each subject identified with the counselor on the tape and

the degree to which each subject included the counselor in

his or her social field.

Based on a review of pertinent literature, the follow-

ing research hypotheses were generated:

(1) The group which is exposed to directive counselor

role behavior will have higher identification

scores than the group which-is exposed to non-

directive counselor role behavior.

(2) The group which is exposed to directive counselor

role behavior will have higher inclusion scores

than the group which is exposed to nondirective

counselor role behavior.

(3) The group which listens to tapes presenting a

male counselor will have higher identification

scores than the group which listens to tapes

presenting a female counselor.

(4) The group which listens to tapes presenting a

male counselor will have higher inclusion scores

than the group which listens to tapes presenting

a female counselor.

(5) There will be a client role expectation by coun-

selor role behavior interaction when identifica-

tion is the dependent variable.

(6) There will be a client role expectation by coun-

selor role behavior interaction when inclusion

is the dependent variable.



Much of the literature which is pertinent to the

present research has resulted from studies of psychotherapy

in clinic settings. In this study, except when referring

to specific psychotherapy research or when quoting directly,

the terms "counselor," "counseling" and "client" have been




In the first chapter a theoretical framework was

presented in which the counseling interaction was described

as a developing social system for which the initial inter-

view is the first phase. It was also suggested that three

important variables which influence the development of a

counseling social system during its initial stage are client

role expectation, counselor role behavior and counselor sex.

A research study designed to observe the relationship of

these three variables to subsequent development of the

counseling social system in a group of institutionalized

adolescent delinquents was outlined.

In this chapter, literature which supports the theo-

retical concepts and proposed hypotheses of the present

study is reviewed. This review consists of three parts.

The first section is an examination of literature related to

the concept of role, including (1) theoretical literature

from social psychology and sociology which has discussed role

and related concepts, (2) research which has defined the role

expectations of clients in counseling and psychotherapy, and

(3) research which has examined the effects of role conflict

in the counseling social system. The second section reviews

research which has studied client preferences for sex of the

counselor. Section three is a review of the criterion

variables of interest in this research.


Theoretical Literature

Although the term role is frequently used and numerous

theoretical papers have discussed role and its related

constructs, the term has diverse meanings. After reviewing

over eighty sources in which the term role was used, Neiman

and Hughes (1951) suggested that "the concept role is at

present still rather vague, nebulous and nondefinitive"

(p. 141). They divided the definitions of role into three

categories: (1) those which use role to describe the

dynamic process of personality development: becoming accul-

turated, development of the self or personality; (2) defini-

tions in terms of society as a whole: normative definitions,

role as a synonym for behavior; and (3) functional defini-

tions in terms of specific groups: role as a status, taking

a role. In a similar manner, Bentley (1968) has differen-

tiated normative definitions (standards or norms expected of

occupants of positions), individual definitions (behavior

that is perceived as being appropriate by an individual in

a social setting), and behavior definitions (definitions

which focus upon the behaviors of individuals occupying

social positions).

One consistency among these definitions is that role is

defined in the context of social relationships and with

attention to the interactional nature of the concept. Carson

(1969) reflected this concern with social interaction in his


Every person has several roles--at least one
for every system . in which he participates
Smooth social interaction also requires
that there be role reciprocity between persons
as they interact. One cannot effectively enact
the behaviors of the teacher's role, for
example, unless the other person simultaneously
enacts the behaviors normatively associated
with the role of student. (p. 181)

In discussing role in an interaction context, Cottrell

(1942) suggested the use of the term role:

. to refer to an internally consistent
series of conditioned responses by one member
of a social situation which represents the
stimulus pattern for a similarly internally
consistent series of conditioned responses of
the others) in that situation. Dealing with
human behavior in terms of roles therefore
requires that any item of behavior must always
be placed in some specified self-other
context. (p. 617)

Sarbin (1954) summarized his definition of role as "a

patterned sequence of learned actions or deeds performed by

a person in an interaction situation" (p. 225). The con-

struct of role is functional only in the context of an

interaction, whether actual or symbolic.

Merton (1957) used the concept "role-set" to describe

the "complement of role relationships in which persons are

involved by virtue of occupying a particular social status"

(p. 110). He was thus noting the various relationships

which a person may hold due to being in a distinct status

position. For example, due to occupying the status position

of school counselor, a person may have role relationships

with teachers, principals, parents, students and peer pro-

fessionals. The role-set of counselor encompasses these

role relationships. In addition, the same person may occupy

the status position of father and therefore maintain a

number of role relationships within the role-set of father.

Merton further observed that the potential always exists for

differing and conflicting expectations regarding the status

occupant among these in the role-set.

In relating role concepts to behavior pathology,

Cameron (1950) suggested that ". . the member of any

organized society must develop more than a single role or

role behavior if he is to reciprocate and cooperate effec-

tively with his fellows" (p. 465). He continued this idea

by suggesting that the person who has developed a number of

realistic social roles is better equipped to meet "new and

critical" situations than a person whose "repertory" of

social roles is "meager, relatively unpractical and socially

unrealistic" (p.465). For example, the client who has had

limited contact with counselors is less likely to have

developed client role behavior and counselor role expecta-

tions which are congruent with the counselor's role.

Sarbin and Allen (1968) have spoken of the probabilistic

nature of interaction and proposed that since a role actor's

"audience" provides discriminant cues-and reinforcement

cues, he must maintain a semblance of flexibility. They

further asserted that "the recognition of this fact (the

probabilistic nature of interaction) renders role theory

continuous with an interactional and functionalistic frame-

work" (p. 491). Later, in stressing the interactional

nature of role learning, Sarbin and Allen suggested that:

. the implication of this point is that
elements of role expectation to be learned
consist not only of specifications of the
individual's own role but also of the other
complementary roles. What must be learned are
the expectations for a specific role and its
complementary roles, that is, the interlocking
system of rights and obligations of a role and
complementary roles. (p. 546)

In the counseling dyad, the role behavior of each partici-

pant must be considered in relation to the role behavior of

his partner in the interaction. Each behavioral component

affects both the subsequent behavior of the actor and the

behavior of the partner.

As implied by Sarbin and Allen (1968), the important

concept in considering the reciprocal or complementary

characteristic of role relationships is role expectations.

In their theoretical framework, the concept of role expec-

tation is "the conceptual bridge between social structure

and role behavior . (p. 497). As stated by them:

Role expectations . are collections of
cognitions, beliefs, subjective probabilities,
and elements of knowledge which specify in
relation to complementary roles the rights
and duties, the appropriate-conduct, for
persons occupying a particular position. (p. 493)

Gross, Mason and McEachern (1958) have defined role in

a normative fashion. In this context they suggested that

"A role is a set of expectations . it is a set of

evaluative standards applied to an incumbent of a particular

position" (p. 60). The authors further noted that roles are

a means of organizing expectations by reference to a social

structure. Expectations delineate for the role actors the

parameters of a particular social interaction; they set the

limits and define the structure of the relationship.

In summarizing some of the theoretical material of

other authors, Bentley (1968) distinguished between two

general kinds of expectations, obligations or duties, and

rights. Obligations or duties are the actions expected of

the role actor, while rights are the actions which the role

actor expects to be able to perform. In applying these con-

cepts to a role relationship, it is clear that there are

reciprocal obligations and rights in each interaction. Both

members of a dyadic relationship may define obligations and

rights for themselves and for their partner.

Again, it is apparent that the role expectations of

each member of the counseling dyad must include a descrip-

tion of both the role of the member and the role of the

partner. When the counselor constructs his role to include

questioning the client, he constructs the client's role to

include giving answers. If the client constructs the coun-

selor's role to include asking questions, he constructs his

own role to include giving answers. In the present study,

the role expectations (reciprocal obligations and rights)

of clients are of primary importance. The client, acting as

a role definer, conceptualizes the role of counselor and the

reciprocal role of client.

An important auxiliary role construct is role enactment.

In their discussion of role, Sarbin and Allen (1968) con-

sidered role enactment as a dependent variable resulting

from the action of the independent variables of role expec-

tations and other primary constructs. In speaking of role

enactment, they focused on overt social conduct and asserted

that the "study of the isolated individual per se has no

place in role theory" (p. 490). Thus, just as with the

broader construct of role and with role expectations, the

variable of role enactment must be considered with respect to

role relationships and reciprocal behavior on the part of

complementary actors. The counselor initially enacts his

role according to his role conceptions, but he can not con-

tinue to enact the role independent of the client's behav-


When there is a difficulty in fulfilling the expecta-

tions of a role relationship or in assuming and enacting a

role, conflict may occur. In discussing role conflict, Ivey

and Robin (1966) listed and defined four types:

(1) Role conflict stemming from role definers--

This type of conflict occurs when persons in a

position to suggest role behavior do not agree

on appropriate conduct for the role actor.

(2) Role conflict internal to the role--Although the

definers of the role agree on certain behaviors,

the role occupant may not be able to maintain all

of the behaviors expected of him. An example

would be a counselor who must be a confidant and

an administrator concerned about rules.

(3) Role conflict stemming from the role in inter-

action with the social system--Ivey and Robin

gave two subsidiary definitions:

a) ". . the normative prescriptions of a role

are not sufficient to allow the role taker to

perform the functions expected of his role in

the larger social system in which it is situ-

ated" (p. 30). An example might be when the

counselor is expected to be available for

personal counseling but in addition is respon-

sible for the periodic testing program of the

school and thus does not have the time to ful-

fill the first function.

b) Role conflict arises from the multiplicity

of roles a person must assume--Consider the

small school where a person may be a teacher,

a clerk and a counselor. Merton's (1957)

description of role-set, which has been dis-

cussed previously, would suggest that

numerous role relationships are associated

with each of these roles.

(4) Role conflict stemming from the interaction of the

individual and his role--The particular role

occupant can not meet the demands of the role de-

finers. For example, the counselor may not be

psychologically secure enough to undertake personal

counseling with clients.

In their discussion of counselor role, Ivey and Robin

viewed the counselor as a member of larger social systems

such as the school, agency or community and they were there-

fore concerned with the effect of role conflict on counselor

role behavior. In addition, internal conflict which has

resulted from role incompatibilities of the counselor and

client may occur in the counseling social system. These

conflicts must be resolved by the members of the social

system if the relationship is to progress.

In summary, role is a relational concept which is

valuable in the analysis of social interaction. The dynam-

ics of role relationships are dependent upon role expecta-

tions, the reciprocally constructed definitions of the

appropriate conduct of the role actors in any relationship.

The result of role expectations and other related variables

is role enactment or role behavior. When role expectations

can not be fulfilled, when incongruent expectations occur,

or when role behavior can not be maintained, conflict may

occur in the relationship.

The literature reviewed thus far suggests three state-

ments in regard to the development of a counseling social

system: (1) the client will come to counseling with precon-

ceived role expectations (his or her conceptions of the

reciprocal obligations and rights of the relationship); (2)

the counselor will initially enact his or her role based on

his or her role expectations; (3) if the client's role expec-

tations and the counselor's role behavior are incongruent,

conflict will probably occur.

Role Expectations of Clients

Based on his personal construct theory, Kelly (1955)

has suggested that there are an infinite number of ways that

a client may conceptualize the counseling relationship. He

theorized: "From the client's conceptualizations of psycho-

therapy comes the role he expects to play and the role he

expects the therapist to play. His behavior as a patient

should be seen in this light" (p. 575). The client brings

a set of expectations to counseling which include concep-

tualizations of appropriate behavior for both the client and

the counselor. This section of the literature review

examines research which has contributed to an understanding

of the nature of these counseling role expectations.

Using a questionnaire, Garfield and Wolpin (1963)

measured the attitudes and expectations regarding psycho-

therapy of seventy applicants for outpatient psychotherapy.

They found that:

All told, about one-third emphasized the
importance of a directive, export role on
the part of the therapist, while two-thirds
of the subjects emphasized the patient's
active effort to help himself. At the same
time, however, if given their choice, most
patients would prefer to be given advice
rather than to be helped in developing under-
standing of their problems. Thus, while
these individuals did not rate advice and
suggestion as being most important in
therapy, they still would prefer to receive
advice. (p. 358)

Of the seventy applicants only half continued in therapy

past the intake interview; however, no attempt was made to

relate termination to the findings on the questionnaire.

In a study by Frankel and Perlman (1969), student per-

ceptions of a college counseling service were surveyed with

a questionnaire similar to the one used by Garfield and

Wolpin. Although most students perceived the student coun-

seling service as helping individuals with personal problems,

the majority of students denied a need for such help them-

selves and judged themselves as having vocational educa-

tional problems. It can be assumed that if these students

sought help from a counselor, they would have expectations

for vocational educational counseling.

Heilbrun (1961, 1964, 1965, 1966, 1968 and 1971) has

conducted a series of studies designed to distinguish

between terminators and continues in-counseling. In one

study (Heilbrun, 1970), using a therapy transcript on which

female subjects demonstrated preference for directive or

nondirective responses, he found 29 "potential terminators"

more likely to prefer directive counselor responses than

14 "potential continuess" Overall, he found a general

tendency for the 43 undergraduate females to choose direc-

tive responses.

As noted earlier, Hollingshead and Redlich (1958) found

that lower class patients expect the therapist to be author-

itarian. In addition, they found that for the most part

lower class patients were oriented externally, emphasized

physical symptomatology, did not see the value of talking

and insight and preferred organic treatment. Overall and

Aronson (1963) reported findings similar to those of

Hollingshead and Redlich. Their study of the role expecta-

tions of 40 lower class psychiatric outpatients demonstrated

a preference for an active but permissive therapist.

Reissman (1964), in delineating the merits of role

playing with lower socioeconomic clients, discussed some of

the personal characteristics of lower class persons which

contribute to the effectiveness of this technique. He sug-

gested that the life style of the poor encourages doing

versus talking, is physically oriented and characterized by

external rather than internal concerns. Their preference

for counseling styles is similarly defined with expecta-

tions for an active, concrete problem solving approach.

Pellegrine (1971) reported a study with 444 university

counseling center clients in which he attempted to differen-

tiate between repressors and sensitizors based on severity

of problem and type of client approach. The repressor-

sensitizor scale measures idiosyncratic responses to

perceived threat; sensitizors respond with awareness and

recognition while repressors respond with repression or

denial. Among the results, he found that sensitizors were

more likely to be seeking personal-social counseling than

repressors while more repressors than sensitizors tended to

be classified as having severe problems. These paradoxical

findings were interpreted as reflecting the unique problem

solving style of each group. The author speculated that

repressors are reticent, withdrawn individuals who deny

feelings. They seek solutions for specific problems, de-

manding action and answers and shy away from deep personal

involvement. On the other hand, sensitizors respond to

internal clues, seek insight when faced with personal prob-

lems and value feelings as guideposts.

When the results of Pellegrine's study are interpreted

in relationship to role expectations, three propositions

emerge: (1) repressors expect advice for specific external

problems and expect the counselor to deal with "facts" rather

than feelings; (2) sensitizors expect a reflective approach

to counseling which includes a concern with feelings; and (3)

role conflict often occurs when repressors become involved

in counseling because most counselors-are oriented toward

"insight" approaches and an emphasis on feelings.

Berzins, Friedman and Siedman (1969) examined the

relationship between client scores on the A-B Scale and

client role expectations in a study involving 60 male

college clinic patients. The A-B Scale was devised from

several items of the Strong Vocational Interest Blank which

describe activities of a manual, technical or mechanical

nature. It has been demonstrated in a number of studies

that A's (persons who demonstrated a dislike of the activi-

ties from the Strong) achieve better results with schizo-

phrenic clients than B's (persons who like the activities

described in the Strong) and that B's achieve better out-

comes with neurotic outpatients. The experimenters adminis-

trated the A-B Scale to clients, along with a measure of role

expectation. A checklist of client complaints was completed

following the initial interview. A's demonstrated a turning

against self and intropunitive symptoms along with expecta-

tions to involve themselves actively in an unburdening type

of therapy. B's were found to be extrapunitive with expec-

tations to receive guidance from an analytical, teacher-like

figure. In this research, as in previously cited studies,

a dichotomy of expectations emerged. The clients held

either expectations for an "insight" approach, with responsi-

bility shared by the partic: ;nts or expectations for a

"guidance" approach, with the therapist being primarily

responsible for the course of therapy.

In order to explore the basic characteristics of the

role expectations of psychotherapy clients, Abfelbaum (1958)

administered a Q sort of therapy expectations to 100 clients

of a university outpatient psychiatric clinic. Cluster

analysis of the client responses fell into three dimensions

of role expectations:

(1) Nurturant These clients expect a guiding,

giving protective counselor.

(2) Model These clients expect the counselor to be

a permissive listener who is neither protective

nor critical.

(3) Critic These clients have expectations that

the counselor will be critical and analytical

and that the client will take much of the

responsibility for the counseling.

Using seniors in high school as subjects, Grant (1954)

had the students complete a forced choice questionnaire

which asked to whom they would go for help with various

problems. Grouping the answers as school helpers (counselor,

teacher) and out of school helpers (family, friends, family

doctor), the students responded that they would take 62

percent of specified educational problems to school helpers

and 36 percent to out of school helpers. They would choose

to take 50 percent of vocational problems to school helpers

and 46 percent to out of school helpers. Of personal

problems, they would take only 4 percent to school helpers

and 75 percent to out of school helpers. The author pro-

posed that these data reflected the students' perceptions of

the role of the high school counselor as a person who pro-

vides primarily educational guidance; and as the problems

become more personalized, the student is less likely to

perceive the counselor as a source of help.

Tinsley and Harris (1976) undertook a study intended

to define the counseling expectations of 282 male and female

undergraduate students. The students were asked to indicate

degree of expectation for 82 items on a questionnaire which

yielded scores on seven scales. Students demonstrated

strongest expectations for counselor behavior defined as

expertness, genuineness and acceptance and for client be-

havior defined as trust. Expectations for counselor behavior

defined as understanding and directiveness were less likely

to be indicated. Significant sex differences were found on

two scales with female students having higher expectations

for accepting behavior and male students higher expectations

for directive behavior.

Using 160 students in an introductory psychology course

as subjects, Reisman and Yamokoski (1974) studied preferences

for type of helping behavior. They had subjects indicate

strength of preference for several types of helping behavior

from friends and therapists. There was a significantly

stronger preference for expository (expert analysis or

explanation of a problem) type behavior than for empathic or

interrogative behavior.

To recapitulate, clients come to counseling with

diverse role expectations. The clearest dichotomy seems to

be between expectations for guiding, directive or structured

counselor behavior and expectations for reflective, non-

directive and nonstructured counselor behavior. While a

number of studies have demonstrated that clients with a

variety of backgrounds have initial preferences for leading

or directive counselor behavior, preferences for a directive,

guiding counselor are especially strong among lower class


Based on the findings presented in this section, it was

assumed that a majority of the persons in the lower class

group (adolescent delinquents), from which the subjects in

the present research were drawn, would have expectations for

a leading or directive approach to counseling. Therefore,

it was suggested that, as a group, the subjects would have

predominantly Approval-Advice role expectations since such

expectations constitute leading or directive types.

Role Conflict in Counseling

Bordin (1955) has discussed the client discomfort pro-

duced when counselors probe for emotional material from

clients who have come to the counseling center for vocational

or educational guidance. He observed that clients seeking

guidance for decision making have little concern about the

personal characteristics of the counselor, while clients

who expect to discuss personal problems are interested in

the counselor's personal characteristics and the inter-

personal relationship established by the counselor.

Goldstein (1966), in reviewing the literature regarding

role expectations in counseling, noted that clients often

respond negatively to counselors whose interview behavior

is not congruent with their expectations. It has been dem-

onstrated that incongruence between counselor and client

expectations has a significant impact on the initial

response of the client to the counselor. Evidence of this

impact is presented in the following group of research in

which the effects of compatible and incompatible role inter-

actions on the counseling process were studied.

Clemes and D'Andrea (1965) had nine psychiatric resi-

dents each conduct five structured and five nonstructured

initial interviews with applicants to an adult outpatient

clinic. Before the interview, they measured patient expec-

tations for psychotherapy on an instrument which categorized

patients as having guidance or participant expectations.

They found that patients whose interviews were compatible

with their expectations rated their anxiety significantly

lower after the initial interviews. In addition, the resi-

dents rated the interviews which were incompatible with

client expectations as more

Severinsen (1966) reported the results of investiga-

tions of client expectations in two separate groups of

clients. Using a questionnaire, he had one group of college

freshman clients (234 students) indicate, prior to the

initial interview, the degree of lead they expected from

their counselors (14 practicum counselors) in the interview.

Similarly he had another group of students (314) indicate

the degree of empathy that they expected from their coun-

selors (13). After the interview, using another form of the

questionnaire, the clients indicated their perception of the

degree of lead or empathy which their counselors demonstrated.

In addition, each client completed a measure of satisfaction.

After comparing satisfaction scores with the differences

between initial expectations for empathy and subsequent per-

ceptions of counselor empathy, the author concluded "that

satisfaction seems to be a function of how closely the

counselor approximates the client's expectation, and not

whether he is 'fact' or 'feeling' oriented in the interview"

(p. 111). Differences in expectation and perception of

counselor lead were not significantly related to satisfaction.

Following a study designed to cast light on the dif-

ferences between terminators and continues in psychotherapy,

Heine and Trosman (1960) concluded that "the variable which

appears to be significant for continuance is that of

mutuality of expectation between patient and therapist" (p.

278). In this study, conducted in a medical setting,

continues in psychotherapy were more likely to have con-

ceptualized their experience in a manner congruent with the

therapist's role image. The authors suggested that thera-

pists "may sometimes not take patients' initial expectations

sufficiently into account in the crucial early hours of

therapy, and thereby may 'lose' the patient before a pattern

of mutual collaboration can be firmly established" (p. 278).

Borghi (1968), like Heine and Trosman, attempted to

discover the reason for termination of psychotherapy

patients by conducting home interviews with 29 terminators.

In the interviews, terminators consistently voiced expecta-

tions which were incongruent with those of the therapists.

In a recent study (Martin et al., 1976) findings in

contrast to other research were reported. The researchers

measured the role expectations of 144 patients and 77

psychotherapists in an inpatient setting. Expectations \.re

classified as nurturant or critical and patients and thera-

pist were designated as having high or low scores for both

scales. For purposes of analysis, each patient-therapist

dyad was classified as mutual high, nonmutual or mutual low

on both types of expectations. At the end of therapy, eval-

uations of the therapy were obtained from therapists and


When dyads were classified according to mutuality of

expectations, only the differences between mutual low critical,

mutual high critical and nonmutual critical dyads were signif-

icant. The patient evaluatic scores of the mutual low

critical group were higher than mutual high critical or

nonmutual critical groups. When subjects were divided into

groups based on quantity of mutual expectations (mutuality

on both scales, one scale or no scales), no significant

differences were found. A further analysis of the patterns

of nurturant and critical expectations revealed that the

group with the highest evaluation scores consisted of dyads

with mutually high nurturant and mutually low critical ex-


In their report of a long term investigation of the

interrelationship of role expectations and communication in

psychotherapy, Lennard and Bernstein (1960) have revealed

some relevant findings. Over 500 therapy sessions between

eight patients and four psychoanalytically oriented thera-

pists were recorded. Content analyses of expectational,

interactional and communicational variables were compared

with questionnaire data gathered at periodic intervals

during the course of treatment. Among other conclusions

the authors stated:

When both members of a dyad are in agreement
regarding their reciprocal obligations and
returns, there is consensus or similarity of
expectations, and harmony or stability occurs
in their interpersonal relations. The system
is then said to be in equilibrium. But when
there is any degree of discrepancy or lack of
consensus between the participants, and their
expectations are dissimilar, the role system
is disequilibriated and manifestations of
strain appear in their interpersonal relations.
If expectations are too dissimilar, the

social system disintegrates unless the
differences can be reconciled. (p. 153)

In a study involving 120 students in introductory

psychology courses, Price and Iverson (1969) found a

positive relationship between confirmation of expectation

and attribution of positive characteristics to the counselor.

The experimenters did not measure the actual expectations

of the 120 subjects but instead deduced, previous to con-

ducting the study, that certain counselor behaviors would be

expected by the subjects. The students listened to taped

excerpts from an initial counseling session with a college

freshman as the client. Five types of counselor verbal role

behavior were manipulated to reflect confirmation or dis-

confirmation of five types of expectations which the experi-

menters assumed were held by the subject group. The role

behaviors manipulated were: (1) high or low commitment to

the client, (2) being problem centered or focusing on

topics irrelevant to the client's problem, (3) self-

confidence in helping the client or personal insecurity in

dealing with the client's problem, (4) reflective or direc-

tive orientation, and (5) confirmation or disconfirmation

of the client in a favorable self image. The findings

demonstrated a positive relationship between confirmation of

subject expectations and favorable subject perception of the

counselor for all of the verbal behaviors except reflective

versus directive orientation.

Since the researchers did not measure the actual

expectations of the subjects in the study, they could only

assume that their opinion of what is expected counselor

verbal behavior was correct. Rather than suggest that

students would expect reflective behavior (as the authors

did), it would be consistent with the results of other

research to assume that students vary individually and that

they would rate their counselors based on confirmation of

their particular expectations. Since other studies have

shown that many clients have directive expectations, it

follows that many of the subjects in the study conducted by

Price and Iverson would have directive expectations which

would be confirmed by directive role behavior and discon-

firmed by reflective role behavior.

Three studies by Heilbrun (1961, 1968 and 1970) may

contribute to an understanding of role conflict in the

counseling interaction. In a study involving 78 female

counseling center clients and 18 counselors, Heilbrun (1961)

found a counselor dominance by client autonomy interaction

when continuance was the dependent variable. Autonomous

female clients were more likely than less autonomous clients

to remain with counselors with high dominance scores on the

Edwards Personal Preference Schedule. The author suggested

that the dominant counselor lets the female clients know

early in the relationship that he will not take responsi-

bility for maintaining the interaction. Autonomous clients

are willing to accept this condition, while clients who are

not autonomous will not and may terminate.

In research designed to explain the meaning

of premature termination, lleilbrun (1968) again used a

college psychological center sample. Counselor ratings of

client problem solving behavior during early counseling con-

tacts were obtained on 32 male and 19 female clients.

Counselor ratings were classified as dependency problem

solving behavior or independency problem solving behavior

and compared to client scores on the Counseling Readiness

Scale of the Gough Adjective Checklist which had been found

to discriminate between continues (high counseling readi-

ness clients) and terminators (low counseling readiness

clients) in past research. The author found a sex by

problem solving interaction such that male dependent problem

solving and female independent problem solving clients were

classified as continues most often on the Counseling

Readiness Scale.

The findings of a previously cited Heilbrun study (1970)

may also be explained in terms of role conflict. In this

study, female subjects classified as terminators on the

Counseling Readiness Scale were more likely than continues

to prefer directive counselor responses. Relying on Parker's

(1967) findings that male counselors showed a significant

tendency to make more nondirective than directive statements

to female clients in initial contacts, Heilbrun explained

that this nondirective style would complement the independent

problem solving female and reinforce continuation. Con-

versely, this nondirective style would frustrate the needs

and expectations of the dependent problem solving terminator.

In a different type of approach, Biddle (1957) had

subjects view skits portraying parts of an initial counsel-

ing session. He found subjects more likely to feel "uneasy"

and be resistant when the emphasis of the play did not

conform to their expectations which were classified as ex-

pectations for problem solving or for relationship building.

Biddle concluded that nonconformity of the counselor to the

client's norms leads to less progress in all phases of the

initial interview than does conformity.

Several studies have investigated the effectiveness of

precounseling preparation in altering the role expectations

of clients. Hoehn-Saric, Frank, Imber, Nash, Stone and

Battle (1964) used a Role Induction Interview (RII) to pre-

pare 20 neurotic outpatients for participation in therapy

with psychiatric residents. This group was compared with

another group of twenty patients who did not participate in

the RII. The experimental group exhibited better in-therapy

behavior, had a better attendance rate and were rated more

favorably by therapists with respect to establishing and

maintaining a therapeutic relationship.

Orne and Wender (1968) used the phrase "anticipatory

socialization for psychotherapy" to describe the process of

teaching clients about psychotherapy-and the roles of

therapist and client. They noted that some clients have

had little experience, either direct or indirect, with

psychotherapy and little idea of what to expect. Basing his

or her expectation on the only model (medical doctor) in his

or her experience, the client expects to be questioned and

provided with concrete help. The authors continued by

describing the frustrations of both the client and the thera-

pist as they maintain their initial perception of their re-

spective roles, each assuming that the other is at fault.

The process of anticipatory socialization was suggested by

Orne and Wender as a preventative measure to reduce the role

conflict created by incompatible role expectations.

Strupp and Bloxom (1971) tested the effectiveness of a

video tape which they developed for use in preparing lower

class clients for group psychotherapy. One hundred twenty-

two clients were equally divided among three conditions

and assigned to therapy groups of eight to twelve members.

In one condition, the clients viewed a video tape por-

traying a lower class person's experience with an

emotional problem and his resulting positive involvement in

a psychotherapy group. In the second condition, clients

participated in a group role induction interview designed

to explain the process and goals of group psychotherapy.

Clients assigned to the third condition received no treat-

ment prior to entering psychotherapy. Clients in both role

induction conditions demonstrated more favorable results

on numerous process and outcome measures. In summary, the

authors suggested: "We have presented consistent evidence

that a role-induction procedure facilitates a favorable

therapy experience for lower class clients" (p. 66).

Although both procedures were effective, the video tape

seemed to be slightly superior to the interview.

In summary, the data from numerous studies establish

that when counselor role expectations and/or counselor

behavior during the initial stage of counseling are not

congruent with client role expectations, the probability

that the client will react negatively to the counselor is

increased. Counseling relationships in which role conflict

is low are more likely to be successful than are counseling

relationships in which the role conflict is high. Clearly,

role conflict during the initial counselor-client interaction

affects the subsequent relationship.

As noted earlier, there was evidence to suggest that

the population of interest in this research would have pre-

dominantly Approval-Advice counseling role expectations.

The literature in the previous section verifies a premise

that directive counselor role behavior would confirm these

Approval-Advice expectations. Therefore, since the litera-

ture has also demonstrated that confirmation of role

expectations is related to the presence of favorable client

attitudes toward the counselor, it was hypothesized that,

in the present study, the group which'listened to a tape por-

traying directive counselor role behavior would have higher

identification and inclusion scores than the group which

listened to a tape portraying nondirective counselor be-

havior. In addition, the literature summarized in the

previous section provides support for an hypothesis relative

to the interaction of client role expectations and counselor

role behavior. It has been shown that although counselor

role behavior may influence client attitudes toward the

counselor, such results are related primarily to whether the

counselor's role behavior confirms or disconfirms the client's

preconceived role expectations. Consequently, it was

hypothesized that groups where role expectations were con-

firmed by counselor role behavior would show higher levels

of identification and inclusion than groups whose role ex-

pectations were disconfirmed by counselor behavior.

It may be noted at this time that no hypotheses rela-

tive to a direct relationship between client role expectation

and the dependent variables were stated. The basis for this

omission is found in both the theoretical and research

literature contained in the present chapter. According to

role theory, role expectations are functional only in an

interaction context. This assumption is corroborated by re-

search which as demonstrated that, for the most part, it

is not the client's initial role expectation which is signif-

icant in affecting the client's attitudes toward the

counselor but confirmation or disconf-irnation of the client's

role expectations.

Counselor Sex

Another variable which was introduced in this study was

sex of the counselor. There has been limited research in

counseling which has contributed to an understanding of the

effect of counselor sex on the counseling relationship.

Koile and Bird (1956) found that, although freshman college

students with both personal and vocational problems pre-

ferred a counselor of their own sex,

.the proportionate number of problems on
which women were willing to consult a man
counselor . was considerably greater than
the proportionate number on which men were
willing to consult a woman counselor. (p. 104)

These findings are consistent with research with children

(Brown, 1956, 1957) in which it was demonstrated that the

sex role preferences of boys are strongly masculine while

the preferences of girls are variable. Boys establish

masculine role preferences much earlier than girls establish

feminine role preferences; and after preferences of girls

become more feminine, they are still more variable than

those of boys.

In a previously cited study, Fuller (1964) found that

among college students of both sexes, there was a preference

for male counselors in discussing vocational and personal

problems with the exception that females preferred a female

counselor when discussing personal problems. The preferences

of women were more variable than the preferences of men.

Boulware and Iolmes (1970) described findings similar

to those of Fuller. Using 60 male and 60 female under-

graduate students as subjects, the experimenters exposed

them to four pictures of potential therapists which were

varied for age and sex. They found that older males were

the preferred therapists for both sexes with the exception

that females tended to prefer older female therapists when

discussing personal problems. Boulware and Holmes have also

cited unpublished research (Levy and Iscoe, 1963) which was

consistent with their study. Using a design similar to

Boulware and Iolmes, Levy and Iscoe found a preference for

male therapists by both sexes.

Mezzano (1971) in a study of student concerns and

sexual preference found a pattern similar to other studies.

In general, students (1495) of both sexes in grades seven

through 12 preferred male counselors for the problems which

they ranked as most important. In examining preferences by

grade classification, he found that students in grades seven

and eight were more likely to prefer counselors of their own

sex while students in grades nine through 12 had mixed pref-

erences. In grades nine through 12 students of both sexes

were more likely to indicate a preference for male counselors

when discussing concerns for the future, concerns about

school and vocational concerns. Female students in all

grades indicated a stronger preference for female counselors

when discussing personal and social problems.

Using an approach different from other research on

sexual preference, Simons and Helms (1976) found a consist-

ent preference for female counselors in two different groups

They presented stimulus pictures of four males and four

females from four age categories to 32 female undergraduates

and 32 noncollege females and had them answer questions

which measured their expectations of counseling climate,

willingness to disclose, counselor competence and counselor

preference. In addition to being given a standard intro-

duction to each picture, the women were told that the

counselors were specialists in counseling women. The re-

searchers found a significantly higher preference for

female counselors in both groups as measured on all four

scales. They suggested that their findings were an indica-

tion of the changing attitudes of women and a result of

their introduction of the counselors as specialists in

counseling women.

There has been no research relating counselor sex to

the dependent variables of this study (identification and

inclusion) and no research confirming a relationship between

preference for counselor sex and the dependent variables.

There does, however, seem to be a basis for the assumption

of a relationship between counselor sex and identification

and inclusion.

Identification occurs in situations where the individual

has a need to choose an appropriate model (Ziller, 1973).

The criteria for choosing a model include the value attrib-

uted to the person or to the position the person occupies

and the degree of social reinforcement associated with

identifying with the person. It can be inferred from past

research that there is a cultural preference associated with

the male role and that more highly valued characteristics

are attributed to the male role (Fuller, 1964). It has also

been suggested that because of the unique cultural experi-

ences of the adolescent delinquent, the male position is

more highly esteemed among this group than in the primary

culture. Furthermore, because most authorities in the life

space of the delinquent are male, it may be assumed that

males would be viewed as being more capable of providing

reinforcement. Therefore, it was consistent with these

premises to hypothesize that the group which listened to

tapes presenting male counselors would have higher identifi-

cation scores than the group which listened to tapes pre-

senting a female counselor.

Inclusion denotes a perception that the included person

is in the same social field as oneself and a willingness to

allow oneself to be influenced by the other person. As in

identification, inclusion is mediated by attractiveness and

the anticipation of reinforcement. Consequently, it was

hypothesized that the group which listened to tapes pre-

senting male counselors would have higher inclusion scores

than the group which listened to tapes presenting a female


It is relevant to note that no hypothesis regarding a

counselor role behavior by counselor sex interaction has

been stated. It is expected that the result of grouping

subjects across these two variables will be additive such

that male directive counselors will receive the highest

scores, female directive the next highest scores, male non-

directive the third highest and female nondirective the low-

est. Additionally, no hypothetical statement has been

offered regarding an interaction of counselor sex and client

role expectation.

Criterion Variables

In keeping with an interest in the process of the

counseling social system, criterion variables which reflect

the client's perception of the social relationship between

counselor and client were chosen. These variables included

client identification with the counselor and client percep-

tion of whether he or she is within the same social field as

the counselor (inclusion). Since these concepts were taken

from a social psychological theory of personality developed

by Ziller (1973), his discussion of the self-social

components is used extensively.

Identification occurs during the'process of socializa-

tion through the selection of appropriate models of human

behavior. Adaption of the individual to his social environ-

ment is facilitated through identification with significant

others who provide such models.

In his discussion of identification, Ziller (1973) has

noted the relationship between identification and the

psychoanalytic concept of introjection of the generalized

other. Introjection of the generalized other is the process

of incorporating the perceptions and responses of others

into the concept of self. Ziller assumed that identifica-

tion involves introjection. He further stated:

More directly, however, identification may
be understood as modeling behavior. Through
the selection of an appropriate model of
human behavior and through the process of
imitation, socialization is facilitated.
Appropriate or adaptive behaviors and
attitudes for given situations may not be
within the given repertoire of the individual.
A convenient way to learn the appropriate
behaviors and attitudes is to observe the
behavior of a selected or available model,
a person or group of persons who may be
observed emitting the crucial responses.
If the person is similar to the self or a
person whom one is pleased to imitate, a
minimum of adaptation for differences be-
tween self and other (the model) is required.

By modeling others or identifying with others,
the individual can expect with high relia-
bility that his behavior will be supported by
others. Thus, social interest and expecta-
tion of social support are allied concepts.

The parents serve as convenient models in the
process of socialization, and retardation of
identification with parent is usually assumed
to retard the socialization process. The
expanding social environment, however, pre-
sents an array of individuals who may serve
as models, such as friends and teachers. The
ability to identify with more than one group
of significant others including parents,
friends, and teachers is assumed co be
associated with higher expectancy of social
reinforcement. (p. 34)

Identification of the client with the counselor is

significant, both as a basis for effective communication and

in the sense of the counselor being a model in the sociali-

zation of the client. For example, in both behavioral and

relationship approaches to counseling, modeling has been

used to prepare clients for counseling, to induce higher

levels of information seeking and to develop more adaptive

social behavior (Krumboltz, Varenhorst and Thorensen, 1967;

Heller, 1969). In addition, changes in client values in

the direction of counselor values (Rosenthal, 1955) and the

differential response of clients to counselor similarity

and dissimilarity (Cook, 1966) may be interpreted as reflect-

ing modeling on the part of the client. At a minimum,

identification serves to reduce client resistance, increase

trust and allow the client to be more receptive to counselor


Inclusion denotes the existence of a relationship be-

tween self and others. The individual who perceives self as

located within the same social field as a significant other

is establishing the basis of interaction between himself and


the other person. Exclusion implies movement away from the

person and limitation of the interaction by the boundaries

of the social field.

In the present study, inclusion of self within the

social field of the counselor is viewed as important to the

development of a productive counseling relationship. If

the client does not perceive his interaction with the coun-

selor as occurring within the same social field, the potential

exists for the client to block the development of a

counseling social system.



A review of the literature pertinent to the present

research has demonstrated that client directive and non-

directive role expectations, counselor directive and non-

directive role behavior and counselor sex are factors

present during the initial interview which exert an influ-

ence on the later development of the counseling relation-

ship. It was proposed that, during the initial interview,

these factors affect the client's identification with the

counselor and the client's inclusion of self within the

same social field as the counselor. In relating this

proposal to the present research, several hypotheses were

offered. In this chapter the experimental design constructed

to test these hypotheses is described.


The subjects were students at Lancaster Youth Develop-

ment Center, one of several institutions operated by the

Florida Division of Youth Services. The center is a

regional facility serving primarily youthful offenders whose

home residences are in North Central Florida. Students come

from both urban and rural areas and are representative of

the total population under supervision by the Division of

Youth Services. Although the students' backgrounds are

varied as to past legal offenses, most are under supervision

because of repeated minor offenses and disruptive home and

school behavior. At the time that this research was con-

ducted there were approximately 160 students in residence

at the center.

Each student at the center is assigned to one of nine

cottages. An attempt was made to include all students from

the nine cottages in the research. Students in the orienta-

tion (new students arriving at the center) and security

(students held temporarily for violation of rules) units

were not included because they were not participating in the

regular program of the institution. There were approximately

10 students in these programs during the week that the

research was conducted. Other students who did not partici-

pate were temporarily excused from their usual schedule (in

the hospital; appointments with dentists, doctors, etc.;

special work assignments; or special school programs).

The final subject group included 104 male and 23 female

students. They ranged in age from 13 to 18 years and con-

sisted of 43 black and 84 white students. Modal age was

seventeen, and mean age sixteen. Since subjects

only reported age in years, the mode seems to be a more

representative measure. A further breakdown of subjects by

race and sex is presented in Table 3.1. Subject age and

sex are compared in Table 3.2

In order to check for representativeness of the

sample, data were obtained on the racial and sexual com-

positions of several programs operated by the Florida

Division of Youth Services. According to a report in which

the 1973 population of training schools in Florida was

surveyed, approximately 79 percent of that population were

male and 21 percent were female (State of Florida, 1977c).

In a study which used a sample of 100 students furloughed in

1973 from three types of small, community-based residential

programs, 4 percent of the sample were female, 96 percent

were male, 34 percent were black and 64 percent were white

(State of Florida, 1977b). Of the 190 students included in

a study of the Division of Youth Services family group home

program during 1975 and 1976, 37 percent were female, 63

percent were male, 36 percent were black and 64 percent were

white (State of Florida, 1977a).

Counseling Tapes

Four audio tapes were used to introduce and control for

counselor role behavior and counselor sex. As preparation

for making the tapes, a transcript was made from a counsel-

ing interview with an adolescent client. The problem dis-

cussed by the client was general enough that any student in

the subject group could conceivably identify with the

Table 3.1
Number and Percent of Sample
N = 127





( 6%)

by Race and Sex




Table 3.2
Number of Sample by Age and Sex
N = 127

13 14 15 16 17 18

127 5 13 24 30 42 13

104 4 7 18 25 38 12

23 1 6 6 5 4 1

situation. Specifically, a problem situation was chosen

which might involve either a male or female client.

From this typescript, two scripts were fabricated, one

to reflect a directive counseling style (Appendix A, p. 119)

and the other to reflect a nondirective counseling style

(Appendix A, p. 127). Parker's (1967) descriptions of

directive and nondirective counselor responses were used.

According to Parker, directive counselor responses are de-

fined as those which tend clearly to lead, direct or control

the verbal activity during the counseling interview. In his

study, direct responses included asking direct questions

(DQ), approval and encouragement (AE), giving information

(IN), forcing the topic (FT), reassurance (RS), and per-

suasion (PS).

Nondirective counselor remarks are defined by Parker

"as those which would tend to give responsibility of decis-

ion for choice of area and direction of verbal activity

largely to the client as well as those responses which re-

flect or clarify the client's affect." Nondirective re-

sponses included mm-hmm (m), simple acceptance (SA), main-

taining initiative for the discussion with the client (FN),

the traditional nondirective lead (ND), restatement of all

or part of what the client has just said (RC), and clarifi-

cation of feeling (CF).

Client responses were consistent between the two types

of transcripts and varied only to allow for an integration

of counselor and client res' 'nses. In the directive script

approximately 85 percent of the counselor responses were

directive and in the nondirective script approximately 85

percent were nondirective.

Two fifteen minute tapes were made from each transcript.

The part of counselor was acted directly from the transcript

by two paraprofessional counselors (one male, one female).

Each actor played the role based on both transcripts; thus,

the male and female counselor each portrayed the nondirec-

tive and the directive style. The resulting four tapes

included two with a male counselor and two with a female

counselor. The client part was read verbatim by the experi-

menter in all conditions. This procedure was used to con-

trol for variability due to sex of the client on the tape

and other personal characteristics which may have influenced

subject reaction during the experimental procedure.


Psychotherapy Expectancy Inventory-Revision

The Psychotherapy Expectancy Inventory-Revision (PEI-R)

(Appendix B, p. 135) is a revision of an instrument origi-

nally designed to measure patients' role expectancies

(Berzins, 1966). The original Psychotherapy Expectancy

Inventory (PEI) was written to include four role expectancy

categories, Nurturant (patient expects to be guided and

protected by a benevolent figure), Critical (patient expects

to receive rational guidance, md correction, Self-reliant

(patient expects to help himself via verbal initiative) and

Cooperation (patient expects emotional give and take in an

egalitarian context).

The PEI-R was developed by factor analysis of the

responses of 271 male and 237 female college clients to the

original PEI which consisted of 60 questions (Berzins, 1971).

Four six item subscales were formed from the analysis.

Although the subscales corroborated the original four cate-

gory model, the constructs were named to make them more

consistent with item content. These new categories were

Approval-seeking, Advice-seeking, Audience-seeking, and

Relationship-seeking. The resulting PEI-R is a 30 item

questionnaire consisting of 24 keyed items and six fillers.

Each of the four subscales has six items. Clients respond

to each item on a seven point scale with a response of "1"

meaning the client doesn't expect the behavior described in

the item, a response of "4" meaning it is moderately ex-

pected and a response of "7" meaning it is strongly expected.

Each subscale is scored by summing the responses to the six

items and dividing by six to indicate mean level of response.

Normative data were obtained on the PEI-R after admin-

istration to 637 male and 604 female outpatients of a

university clinic. There was a significant positive rela-

tionship (r = .34) between the approval and advice subscales

and the audience and relationship subscales (r = .52). The

Approval-Advice (APPROV-ADVICE) subscale pair has been termed

the "dependency upon others" role orientation and the

Audience-Relationship (AUDIEN-RELAT) subscale pair has been

termed the "turning toward others and self" role orientation.

According to Berzins (1971) "these role orientations are con-

sistent with the theoretical notions that guided the initial

construction of the PEI and can be regarded as useful

'second order' conceptualizations ... ." (p. 4).

Internal consistency estimates for the 1241 patient

sample were .75 for APPROV, .83 for ADVICE, .86 for AUDIEN,

and .87 for RELAT. Although no test-retest coefficients

are available for the PEI-R, test-retest coefficients for

PEI scores obtained from psychotherapy clients ranged from

.54 to .68 with a one week interval and from .56 to .76 with

a four week interval.

Few tests of validity have been applied to the PEI-R;

however, some research relevant to the validity of this

instrument has been reported (Berzins, 1971). Using the

Behavioral Correlates (BC) Scale (an instrument to measure

client therapy behavior), Berzins had four therapists rate

the interview behavior of 63 clients who previously had

completed the PEI. Behaviors included in the BC are com-

parable to the behaviors described in the PEI and the PEI-R.

Correlations between the four scales of the PEI and com-

parable scales of the BC ranged from .24 to .43 (p .10 or

better, two tailed test). Using a larger sample of clients

(n = 96), PEI scores were found to be uncorrelated with the

Marlowe-Crowne Social Desirability Scales.

Although no comparable studies have been conducted with

the PEI-R, the convergent/discriminant validity of the PEI-R

has been examined in relation to patient concurrent scores

on the Personality Research Form, patient self reported

symptomatology, therapist reported clinical judgements and

decisions, selected background indices and outcome of brief

psychotherapy. Using a sample of 117 male and 102 female

clients of a college mental health clinic, correlations be-

tween PEI-R scores and scores from the Personality Research

Form (PRF) were obtained. In discussing his findings, when

comparing scores on the two instruments, Berzins has stated:

. the data offer better support for the two broader role

orientations (dependency upon others: APPROV and ADVICE;

turning toward others and self: AUDIEN and RELAT) than for

each of the specific subscales" (p. 5).

The APPROV and ADVICE scales were both correlated

(p < .05) with several scales of the PRF. The resulting

composite profile suggests that a person with APPROV-ADVICE

role expectations is approval dependent, help seeking, struc-

ture seeking, dependent and cautious. This profile is con-

gruent with previous theoretical notions about clients with

directive role expectations.

Although less support was found for a general AUDIEN-

RELAT profile, the AUDIEN and RELAT scales were correlated

with several scales of the PRF. The composite profile for

the AUDIEN and RELAT scales describes a person who is domi-

nant, friendly, affiliative id open. This description

supports previous assumptions about a nondirective role


In summarizing the data contributed by correlating

patient self-reported symptomatology and the PEI-R, Berzins


Rather than denoting a high degree of
association between specific expectancies
and symptoms, the present data suggest that
the 'dependent' role orientation generally
covaries with greater (and the 'turning
toward others and self' with lesser)
degrees of felt symptomatic distress. (p. 6)

Additional support for validity is contributed by the fact

that several background indices were found to be related to

the ADVICE and AUDIEN scales. Clients with prior therapy

experience had high AUDIEN scores and low ADVICE scores.

Older clients with higher academic status had the highest

AUDIEN scores, while younger clients with lower academic

status had the highest ADVICE scores. Finally, it was found

that when family background was compared to the PEI-R,

persons with highly educated parents had the lowest ADVICE

scores. These results are in agreement with research that

has demonstrated that lower class, less educated persons

tend to have directive role expectations while higher class,

better educated persons tend to have nondirective role ex-


As previously noted, the APPROV and ADVICE subscales

and the AUDIEN and RELAT subscales are consistently related.

The resulting two role expectation orientations are compara-

ble to Parker's (1967) descr' tion of directive and

nondirective counselor behavior. The APPROV-ADVICE orienta-

tion indicates interest in obtaining the counselor's support

and emotional guidance and expectancies that the therapist

will provide cognitive guidance and evaluation. Similarly,

Parker's description of directive counselor behavior in-

cluded approval and encouragement, giving information,

reassurance and persuasion. The AUDIEN-RELAT orientation

indicates that the client anticipates engaging in verbal

initiative and spontaneous disclosure in the context of a

comfortable, egalitarian relationship with the counselor.

Parker's description of nondirective behavior included

maintaining initiative for discussion with client, clarifi-

cation of feelings and other responses which give responsi-

bility for content and direction to the client.

In scoring the PEI-R for purposes of this research the

subscales were paired, thus forming two scales (APPROV-

ADVICE and AUDIEN-RELAT). Scores for each subscale pair

were obtained by summing the scores for the two original

subscales. The range of possible scores for each scale was

12 to 84. Each subject's predominant role orientation was

determined by his or her higher score on the two scales.

The PEI-R was modified in two ways for the use in the

present research. The word counselor was substituted in

every case for the word therapist. In several items idio-

matic phrases which were not common to the population being

studied were deleted and comparable phrases which were under-

standable to this group were substituted. This procedure

was carried out in consultation with a reading specialist in

order to increase the probability that the statements would

have comparable meaning. As an additional check, the

readability level of the test was determined by using a

procedure developed by Fry (1968), which uses length of

sentences and numbers of syllables to arrive at an estimate

of reading level. According to the Fry Readability Formula,

the PEI-R, with the modifications mentioned above, has a

sixth grade reading level.

In order to test this modified form of the PEI-R, it

was administered to a group of ten adolescent, delinquent

males who were residents of a small state operated treatment

facility. The group was comparable in age and socioeconomic

background to the residents of larger state institutions for

delinquents. Each subject was given a test booklet with the

answer grids next to each question. The instructions and

the test items themselves were given orally by the experi-

menter. The subjects were instructed to read along with the

experimenter and to answer each question as it was given.

Following the testing, the experimenter elicited reactions

and feedback from the group.

Reliability for this modified form of the PEI-R was

determined by split-half method. A correlation coefficient

of .625 was obtained for the APPROV-ADVICE subscale pair and

was corrected for length by the Spearman-Brown formula which

yielded a reliability coefficient of .763. For the AUDIEN-

RELAT subscale pair an uncor-rcted coefficient of .724 was

obtained and converted by the Spearman-Brown formula to a

reliability coefficient of .839.

Based on the feedback received from the group, minor

changes were made in several items in order to make them

more understandable. The instructions were not altered as

this group appeared to be able to follow them easily.

Because of the low reading level of the population being

studied, the instructions (Appendix B, p. 135) and the

items from this instrument were read to subjects in the pres-

ent research. As noted in the instructions, time was al-

lowed prior to the testing to clarify the instructions.

Social Orientation Tasks

Social Orientation Tasks (SOT) (Appendix B, p. 140) are

derived from social psychological theories of personality

which stress the inherently social nature of man. By

arranging symbols representing self and others, the subject

demonstrates his perception of his relationship to signifi-

cant others. Ziller (1973) has used SOT extensively in the

development of a cognitive theory of personality related to

social psychological theory. About this theory Ziller has


Social adaptation is presumed to be mediated
by self other concepts. It is proposed that
social stimuli are screened and translated
into personal meaning through crude mappings
of the self in relation to significant
others. (p. 4)

At the present time, the instrument developed by Ziller

has three forms: adult, student and child. The tasks in

the instrument are interpreted as reflecting 10 components

of self other orientation: (1) self esteem, (2) social

interest, (3) self centrality, (4) complexity, (5) majority

identification, (6) identification, (7) power, (8) margin-

ality, (9) inclusion, and (10) openness.

The tasks are primarily nonverbal and the instructions

give no clues to the specific concepts being measured, thus

resulting in low visibility for the instrument. Because of

their low visibility, the tasks are less susceptible to

error due to biased responses. In addition, this reliance

on preverbal communication allows the subject to project his

perceptions rather than have to respond within the limita-

tions of the investigator's frame of reference.

Those components of self other orientation of interest

in the present study were inclusion and identification. Al-

though the tasks were essentially the same as those used in

other studies, they were modified to include the counselor

as a significant person in the subject's life space. Since

explanations of the components have been described in the

previous chapter, they are not included here. The tasks used

to measure each component, data on reliability and validity,

modifications of the tasks and interpretation of the re-

sponses of subjects are presented below. In explaining the

basic components, Ziller's (1971) discussion of the concepts

is relied upon.



Children's Form: This item is only used with
children and students. A horizontal display of
9 circles is presented. A significant "other"
such as "mother" is located in the circle to the
extreme left and later to the extreme right. The
task requires the subject to mark any of the other
circles in the row to represent himself. Distance
in units from the significant other is the meas-
ure of identification intensity. The circle
immediately adjacent to the significant other is
scored 8 and the succeeding circles 7, 6, 5, etc.
The score for identification with one significant
other such as mother is equivalent to the sum of
the scores from the two identification items
using mother as the focal other. The focal
persons include mother, father, and friend.

Reliability: Split-half reliability corrected
for length:

1. Mother: .80 (99 high school students)
.94 (81 fifth grade students)

2. Father: .95 (99 high school students)
.85 (81 fifth grade students)

3. Friend: .78 (99 high school students)
.78 (81 fifth grade students)


1. Girls locate self closer to mother than
do boys; boys locate self closer to father
than do girls.

2. Under conditions where the father is
absent from the family, there is lesser
identification with the father.

3. Institutionalized behavior problem
children in comparison with a control
group are less identified with a friend.

4. Asian Indian students in comparison with
a control group of American students
identified me with mother, father and

5. First born children identify more closely
with the father than do later born

Identification (Range)


Adult's and Student's Forms: In the two tasks of
this type, the subject is required to arrange cir-
cles representing nine significant people including
the self into as few or as many groups as he wishes.
With regard to children, inclusion of the father or
mother within the self grouping is deemed signifi-
cant. ... .The score is simply the number of
social objects included within the self category
for the two items. The self, of course, is not
included in the score.

Reliability: Split-half reliability corrected for
length is .88 (99 high school students), .86 (81
fifth grade students), .89 (207 ninth grade students).


1. Institutionalized behavior problem children
in comparison with controls include fewer
others within the self grouping.

2. Male neuropsychiatric patients in compari-
son with normals tended to include fewer
others within the self grouping.

In the present study, the identification items were

modified to include counselor as a focal person in the in-

tensity items and counselor as one of the significant others

in the grouping tasks. The intensity items were scored just

as in the original tasks. Two items of this type were

included. In the grouping tasks, placing the counselor in

the group with self was given a score of "2" with all other

placements being scored "1". Three grouping items were used.

The range of possible scores for identification was 5 to




Adult's, Student's and Children's Forms: Varying
numbers of small circles are located inside and
outside a symbolic social field. The small cir-
cles represent the self and other persons. The
subject is asked to mark one of the small circles
to stand for himself. Choice of a circle within
as opposed to outside the social field is assumed
to indicate inclusion and receives a score of one.
If an outside circle is selected, a score of zero
is ascribed. Eight items are included which
counterbalance the number of persons represented
inside and outside the social field. The sum of
the scores for the eight items is the total
inclusion score.

Reliability: Split-half reliability corrected for
length was .63 (299 teachers, principals, super-
intendents, and politicians).

The inclusion items were modified in two ways. In each

task there were only two circles in the symbolic social

field. The number of circles outside of the field varied

among items as in the original tasks. The items were also

changed so that in each task one of the circles in the

social field was marked "C" for counselor. The instructions

were altered to explain the placement of counselor to the

subjects. As in the original instructions, subjects were

asked to place a "Y" in one of the small circles to stand

for self. Placement of self within the social field of the

counselor was scored as "2" while placement out of his field

was scored "1". There were six items of this type presented.

The range of possible inclusion scores was 6 to 12.

Identification and inclusion items were interspersed

with self esteem items when presented to subjects. Self

esteem scores were not included in the analysis.

Data Sheet

In order to consider the possible effect of pre-

existing factors on the criterion variables, subjects com-

pleted a data sheet (Appendix B, p. 157) on which several

questions were asked. From these questions scores were

obtained for rating of past counseling, number of counselors,

types of counselors and expressed interest in seeing a

counselor. In addition each subject gave his or her age in




In an attempt to be as unobstrusive as possible,

students were exempted as groups from part of their regular

schedule. In most cases they were tested during a time that

they normally would have been participating in a study labo-

ratory. Since students were assigned to their regular groups

based on their cottage groups, they were called back to their

cottages at prearranged times and exposed to the experimental

procedure. Groups ranged in size from four to ten students

with the exception of one group of fifteen. All groups were

homogeneous for sex. Groups were assigned to listen to one

of the four experimental tapes, the only consideration being

an interest in equal representation among groups.

Introduction to Subjects

Students were told that the experimenter was gathering

information about what students like themselves thought of

counselors. The experimenter stressed that the study would

be important in helping counselors understand the needs and

points of view of students. They were assured that the

experimenter was not connected with the school and that

their answers would be seen only by the experimenter. (This

point was restated during the course of the session.)

Although students were not forced to participate in the

experiment, their participation was assumed. Any questions

about the necessity of participating were either reflected

or answered with a restatement of the experimenter's desire

that they participate. No students refused to participate

although two students did not complete the instruments which

were a part of the research. Three other students were

excused during the experimental procedure and therefore did

not complete all of the instruments.

Because several documents were completed by each

student and the experimenter wished to maintain confiden-

tiality, each student was instructed to place his or her

name beside a number on a sheet of paper. They were then

instructed to place this number on the data sheet and on all

subsequent documents in order to insure that all of their

papers could be placed together.

Initially all students were given the data sheet to

complete. Items from the data sheet were read in order by

the experimenter and any questions were answered and terms

defined. All 127 subjects completed this form.

Administration of the PEI-R

After completing the data sheet, students were reminded

that the experimenter was interested in their points of view

about counselors. Some discussion was facilitated about the

different counselors to whom they had been assigned and they

were asked to direct their attention to the PEI-R which had

been handed out. The experimenter read the instructions

(Appendix B, p. 135) for the PEI-R, emphasizing that he was

interested in their thoughts about an imaginary counselor

and not an actual counselor that they may have had then or

at one time. The PEI-R was administered with the experimenter

reading each item.

Experimental Manipulation

Following the administration of the PEI-R one of the

four audio tapes was introduced. The procedure for con-

struction of the tapes was explained to the students (with

the exception of the alteration of scripts). They were

told that a tape recording had been made of a counseling

session with a counselor and a student like themselves.

Sex of the student was emphasized as male for male groups

and female for female groups. Students were told that a

script similar to one used by actors was then made from the

tape in order to hide the identity of the counselor and the

student. At this time it was explained that a new tape had

been made with another counselor playing the part of the

counselor and the experimenter playing the part of the

student. The purpose of the study was emphasized again and

students were asked to put themselves in the position of the

student, paying close attention to what the counselor said

since they would be asked for their responses after they

heard the tape.

Following the tape the SOT was introduced with the

experimenter stressing that any reference to counselor meant

the counselor on the tape to which they had just listened.

The SOT was administered with the experimenter reading the

instructions for each item.

All documents were collected and subjects were given an

opportunity to discuss their reaction to the experimental


procedure, the tape or any of the instruments. The experi-

menter specifically elicited their opinions of counselors

and counselor roles.



The theoretical notion which guided the development of

the present study was that the counseling relationship is

a social system which is subject to a dynamic developmental

process. Within this framework the initial counselor-

client interaction is designated as the first stage in the

development of a counseling social system and therefore

predictive of the future existence of the system. A review

of the literature has demonstrated that, among the factors

which may influence this developing social system, three

factors are client role expectation, counselor role behavior

and counselor sex.

In order to test these assumptions, a population

(identified adolescent delinquents) in which these factors

seemed especially important was chosen, criterion variables

(identification and inclusion) which were congruent with a

social system model of counseling were selected and an

experimental design was developed and carried out. The

counseling role expectations of 127 institutionalized

adolescent delinquents were measured. Subsequently they

were exposed to one of four audio tapes on which counselor

role behavior and counselor sex had been manipulated. Fol-

lowing their exposure to these tapes, subjects completed

Social Orientation Tasks (SOT) which-measured degree of

identification with the counselor and degree of inclusion

of self within the same social field as the counselor.

In this chapter the procedures used in analyzing the

data from the present study and the results of the study

are presented. The first section is a review of the process

through which the data were organized and a description of

the statistical methods which were applied in analyzing the

data. In the second section the results of these analyses

are examined.

Data Analysis

Previous to performing the study, the following null

hypotheses were posited:

(1) There will be no difference in identification

scores between groups divided according to

counselor role behavior.

(2) There will be no difference in identification

scores between groups divided according to

counselor sex.

(3) There will be no client role expectation by

counselor role behavior interaction when

identification is the dependent variable.

(4) There will be no difference in inclusion

scores between groups divided according to

counselor role behavior.

(5) There will be no difference in inclusion

scores between groups divided according to

counselor sex.

(6) There will be no client role expectation by

counselor role behavior interaction when

inclusion is the dependent variable.

As noted in Chapter II, no hypotheses regarding a main

effect for client role expectation, an expectation by coun-

selor sex interaction, nor a counselor role behavior by

counselor sex interaction were proposed. It was suggested

that when subjects were grouped using counselor role be-

havior and counselor sex there would be an additive effect.

Total N for this research was 127. All subjects com-

pleted the initial data sheet. Two subjects failed to com-

plete the Psychotherapy Expectancy Inventory-Revision (PEI-R)

and the SOT. Three other subjects completed the PEI-R but

did not complete the SOT.

Each subject was classified into one of two groups using

the higher score on the two scales of the PEI-R. In addi-

tion to the two previously mentioned subjects who did not

complete the PEI-R, four subjects were unclassifiable due to

having equal scores on both scales. The range of scores for

the 121 classifiable subjects was 18 to 81 on the APPROVAL-

ADVICE scale and 14 to 77 o, 'he AUDIENCE-RELATIONSHIP scale.

Eighty-one (81) students were classified as APPROVAL-ADVICE

and forty (40) students as AUDIENCE-RELATIONSHIP.

In order to test the validity of-the classification,

the mean score of the two groups on each of the scales was

compared. On the APPROVAL-ADVICE scale the group classi-

fied as APPROVAL-ADVICE had a mean score of 65.67 while the

group classified as AUDIENCE-RELATIONSHIP had a mean score

of 53.48. These data are presented in Table 4.1. An

analysis of variance performed to compare the two groups on

APPROVAL-ADVICE scores yielded an F score of 36.778 which

was significant at the .001 level. A summary of this

analysis is presented in Table 4.2.

On the AUDIENCE-RELATIONSHIP scale the group classified

as APPROVAL-ADVICE had a mean score of 50.80 and the group

classified as AUDIENCE-RELATIONSHIP had a mean score of

63.48. These data are presented in Table 4.3. An F score of

31.122 is produced when these means are compared by analysis

of variance. Table 4.4 summarizes this analysis of the data.

In Chapter II, following a review of relevant literature,

an assumption was made that the counseling role expectations

of adolescent identified as delinquent would be predomi-

nately APPROVAL-ADVICE type expectations. In order to test

this assumption the mean scores of the subject group on the

two scales of the PEI-R were compared. On the APPROVAL-

ADVICE scale the mean score was 60.824 and on the

AUDIENCE-RELATIONSHIP scale the mean score was 54.392.

Table 4.1
Means and Standard Deviations
Groups Classified by Role Expectation

X S.D.

N = 121

N = 81

N = 40





10 .10


Table 4.2
Analysis of Variance
Groups Classified by Role Expectation

Sum of






Residual 12877.793





Table 4.3
Means and Standard Deviations
Groups Classified by Role Expectation

X S.D.

N = 121

N = 81

N = 40







Table 4.4
Analysis of Variance
Groups Classified by Role Expectation

Sum of



Groups 4300.176 1 4300.176 31.122*

Residual 16442.621 119 138.173



These differences were tested by a paired t-Test which

yielded a value for t of 5.04. These data are presented in

Table 4.5. The significant t-score provides support for an

assumption that the subject group had predominately APPROVAL-

ADVICE role expectations.

Scores on the two scales of the PEI-R were examined

in relation to subject sex and subject race. These compari-

sons are found in Tables C.1, C.2, C.3, C.4 (Appendix C, pp.


Subjects were further classified according to the level

of the two experimental factors (counselor role behavior

and counselor sex) to which they were exposed on the audio

tapes. Although subject sex was not a factor in the

original experimental design, its potential influence during

the experimental procedure was considered. As a result,

subject sex was introduced as an additional factor previous

to the statistical analyses. Subjects were therefore

grouped across four factors, client role expectation, coun-

selor role behavior, counselor sex and subject sex, each

represented by two levels.

In Chapter III it was noted that each subject completed

a data sheet, the scores from which were to be considered

as possible covariants during the statistical analyses. In

order to measure the strength of the relationship between

each of these scores and the criterion variables, identifi-

cation and inclusion, a Pearson correlational procedure was


Table 4.5
Paired t-Test
and Audience-Relationship Scores













t = 5.04 (p < .0001, d.f. = 124)






performed. It was determined that, in order to be included

as a covariant, the correlation between any of these scores

(rating of past counseling, number of counselors, types of

counselors, expressed interest in seeing a counselor and

subject age) and one of the criterion variables should be

.31 or above (accounting for 10 percent of the variance).

No coefficient was .31 or above, therefore none of these

scores was used as a covariant. Table C.5 (Appendix C, p.

161) and Table C.6 (Appendix C, p. 161) display these

correlation coefficients.

Two four-way factorial analyses of variance were the

primary statistical tests, one analysis being performed for

each of the two criterion variables, inclusion and identifi-

cation. Because of unequal cell frequencies, a classical

experimental approach was used in the analyses. Allowing

for unclassified subjects on the factor role expectation

and missing identification and inclusion scores, data for

119 subjects were available for analysis. The level of

statistical significance acceptable for these analyses was

set as equal to or less than .05.


For purposes of clarity, the results of each of the two

four-way factorial analyses of variance are presented

separately. The results when identification was the crite-

rion variable are presented first. Following these results

are the results with inclusion as the criterion


Identification as Criterion

A four-way factorial analysis of variance was performed

with the factors client role expectation, counselor role

behavior, counselor sex and subject sex,each at two levels

and using identification as the criterion variable. A

summary of this analysis is found in Table 4.6. In the

presentation of the results of this analysis, data relevant

to three null hypotheses are reviewed first, followed by an

examination of data for the factor subject sex. Finally,

other data will be surveyed in order to explore the rela-

tionship among factors.

Counselor role behavior

Earlier in this chapter it was demonstrated that the

subjects in the present research had predominately APPROVAL-

ADVICE type role expectations. This finding supports a

previous assumption which had led to a prediction that

directive counselor role behavior would result in higher

levels of identification. In order to test this prediction,

the following null nypothesis was stated:

There will be no difference in identification

scores between groups divided according to

counselor role behavior.

Table 4.6
Four-Way Factorial Analysis of Variance
Identification as Criterion

Sum of Mean
Source Squares d.f. Square F p

Main Effects
Role Expectation (RE) 3.352 1 3.352 0.342 n.s.
Role Behavior (RB) 246.938 1 246.938 25.228 0.0001
Counselor Sex (CS) 77.218 1 77.218 7.889 0.006
Subject Sex (SS) 1.815 1 1.815 0.185 n.s.

Two-Way Interactions
RE by RB 939.444 1 939.444 95.978 0.0001
RE by CS 3.473 1 3.473 0.355 n.s.
RE by SS 10.749 1 10.749 1.098 n.s.
RB by CS 21.841 1 21.841 2.231 n.s.
RB by SS 3.916 1 3.916 0.400 n.s.
CS by SS 35.498 1 35.498 3.627 0.060

Three-Way Interactions
RE by RB by CS 7.215 1 7.215 0.737 n.s.
RE by RB by SS 5.109 1 5.109 0.522 n.s.
RE by CS by SS 5.544 1 5.544 0.566 n.s.
RB by CS by SS 4.524 1 4.524 0.462 n.s.

1017.963 104 9.788


Examination of the mean identification scores of groups

differentiated by counselor role behavior which are

presented in Table 4.7 reveals a mean score of 16.0000 for

subjects exposed to directive role behavior and a mean score

of 12.8871 for subjects exposed to nondirective role

behavior. Analysis of these differences yields an F score

of 25.228 which is significant beyond the .001 probability

level. The null hypothesis is therefore rejected.

Counselor sex

It was predicted that male counselors would invoke

higher levels of identification than female counselors among

the subjects in this study. The following null hypothesis

was therefore posited.

There will be no difference in identification

scores between groups divided according to

counselor sex.

Means and standard deviations of identification scores

for groups determined by counselor sex are presented in

Table 4.8. Subjects exposed to a male counselor had a mean

score of 15.3333 while subjects exposed to a female coun-

selor had a mean score of 13.4068. The F score resulting

from an analysis of these differences was 7.889 which was

significant at the .006 probability level. As a result of

these findings the null hypothesis is rejected.

Table 4.7
Counselor Role Behavior
Means and Standard Deviations--Identification Scores

X S.D. N

Directive 16.0000 4.6029 57

Nondirective 12.8871 4.6873 62

Table 4.8
Counselor Sex
Means and Standard Deviations--Identification Scores

X S.D. N

Male 15.3333 4.1156 60

Female 13.4068 4.8329 59

Client role expectation by counselor role behavior

Following a review of relevant research (Chapter II),

it was stated that client role expectations were most

productively studied in relation to counselor role behavior.

As a result no hypothesis regarding a main effect for client

role expectation was presented. An analysis of the data

seems to support this assumption. The mean identification

score for the APPROVAL-ADVICE group was 14.2532 and the

mean score for the AUDIENCE-RELATIONSHIP group was 14.6250.

These data are presented along with their standard devia-

tions in Table 4.9. The test for a main effect for role

expectation supports an assumption of no difference between


Although no main effect for client role expectation was

expected, it was suggested that there would be an inter-

action of the two factors client role expectation and coun-

selor role behavior created by the process of confirmation

or disconfirmation of role expectations by role behavior.

Therefore, the following null hypothesis was tested:

There will be no client role expectation by

counselor role behavior interaction when

identification is the dependent variable.

Referring again to Table 4.6, which summarizes the

analysis of variance, one finds a highly significant F score

for the test of an interaction between client role

expectation and counselor role behavior. In order to

determine the source of the interaction, a test of simple

main effects was performed. This test demonstrated that

there was a significant difference between the cell means

when compared at the two levels of both factors.

In Table 4.10 the means and standard deviations of

identification scores are listed for the four groups

created by crossing the two factors. A summary of the test

of simple main effects is presented in Table 4.11. In

Figure 4.1 the mean scores resulting from the interaction of

the two levels of each factor are plotted. Based on these

analyses the null hypothesis is rejected.

Subject sex

An examination of Table 4.12 reveals that male subjects

had a mean identification score of 14.2784 while female

subjects had a mean score of 14.8182. The test applied to

these means yielded a nonsignificant F ratio. The data do

not suggest a main effect for subject sex.

A further inspection of Table 4.6 shows that a test for

interaction between counselor sex and subject sex produced

an F ratio of 3.627 (p < .06). An inspection of cell means

shows that among female subjects those exposed to a male

counselor had higher identification scores (X = 17.1818)

than those exposed to a female counselor (X = 12.4545).

Table 4.9
Client Role Expectation
Means and Standard Deviations--Identification Scores


Advice 14.2532 4.8713 79

Relationship 14.6250 4.0679 40

Table 4.10
Client Role Expectation by Counselor Role Behavior
Means and Standard Deviations--Identification Scores

Role Expectation
Approval- Audience-
Advice _Relationship
X S.D. X S.D.

Role Behavior

Role Behavior

18.1622 2.2300
(N = 37)

12.0000 3.2444
(N = 20)

10.8095 3.8524 17.2500 2.9890
(N = 42) (N = 20)

Table 4.11
Test of Simple Main Effects--Identification
Client Role Expectation by Counselor Role Behavior


Role Expectation (RE)
RE at Directive RB
RE at Nondirective RB

Role Behavior (RB)
RB at Approval-Advice RE
RB at Audience-Relationship


Sum of








F p



1 246.938 25.228 0.001
1 1063.434 108.647 0.0001

275.625 28.159 0.0001




Client Role

Directive Nondirective

Counselor Role Behavior

Interaction: Client Role Expectation
by Counselor Role Behavior -
Identification Scores



Figure 4.1

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