Title Page
 Table of Contents
 Review of the literature
 Summary, discussion, and recommendations...
 Reference notes
 Biographical sketch

Title: Client's perception of counselor trustworthiness, expertness, and attractiveness as a function of counselor race and dialect /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00099099/00001
 Material Information
Title: Client's perception of counselor trustworthiness, expertness, and attractiveness as a function of counselor race and dialect /
Physical Description: viii, 117 leaves : ; 28 cm.
Language: English
Creator: Stein, Steven Jeffrey, 1953-
Publication Date: 1981
Copyright Date: 1981
Subject: Cross-cultural counseling   ( lcsh )
Psycholinguistics   ( lcsh )
Ethnicity   ( lcsh )
Counselor Education thesis Ph. D
Dissertations, Academic -- Counselor Education -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1981.
Bibliography: Bibliography: leaves 109-116.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Steven Jeffrey Stein.
 Record Information
Bibliographic ID: UF00099099
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 - 000296471
oclc - 08125423
notis - ABS2836


This item has the following downloads:

PDF ( 5 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
        Page viii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
    Review of the literature
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
    Summary, discussion, and recommendations for future study
        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
    Reference notes
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
    Biographical sketch
        Page 117
        Page 118
        Page 119
Full Text






Copyright 19^1


Steven Jeffrey Stein


The author wishes to thank the members of his dissertation com-

mittee, Dr. Rod McDavis, Dr. Jerrie Scott, and Dr. Joe Wittmer for

their guidance and encouragement during all the phases of this proj-

ect. Special thanks are extended to the chairperson, Joe Wittmer,

whose continual support and nonjudgemental reflections make learn-

ing a rich and enjoyable experience.

Additional thanks are given to the two actors, King Morrison and

Tony Russo, who role-played the counselors: to Marty Feuerman and

Dr. Ron Cody, who assisted with the computer analysis of the data;

and to Dr. Robert J. Powers and Rene Kaufman of the East Orange

Veterans Administration Medical Center, who each offered help during

various phases of the project.

Last, but certainly not least, the author offers a very special

thanks to his parents, Esther and Irving; and to his sisters, Risa

and Joanne. Their support has ranged from moral to financial and has

always been unfailing. Finally, to a special source of inspiration,

Amv, his wife and friend, the author expresses his love.




ACKNO-ELEDGEMENTS ................... ............ ........... iii

ABSTRACT ...................... .................................... vii


ONE INTRODUCTION .............................. ..............1

Need for the Study..................... .............. 3
Purpose of the Study.................... .............. 3
Significance of the Study............................... 4
Definition of Terns .................................... 6

TWO RETV EW OF THE LITERATURE ............... ................. 9

Racial Variables in the Counseling Process............11
Same-Race Versus Interracial Counseling Dyads.......11
Special Techniques for Special Populations? .........13
Premature Termination of Therapy and the
Minority Client .................................. 18
Theories of Cultural Awareness for Counselors.......20
Practice of Cultural Awareness for Counselors.......26
A Brief Account of Sociolinguistic and Language
Attitude Research ................. .................. 29
Introduction to Research on Social Dialects.........29
A Short History of "Disadvantaged" Languate..........30
Listeners' Attitudes Toward English
Dialect Speakers ............... ... ..... ..........34
Attitudes Toward Dialect Speakers of
Non-English Languages .............................40
Language Variables in Clinical and Counseling
Settings. ... ................................... .... 42
Formal Psycholinguistic Investigations of
Counseling and Conversational Discourse....... .....43
Language Factors in the Counseling Process..........45
Language Factors in the Classroom....................50

THREE METHODOLOGY............... ..............................54

Population and Sample.... .......... .... ..............55
Experimental Hypotheses............................... 57

Procedure and Methodology. ... ..........................58
Construction of the Video Tapes......................60
Development of the Monologues ........................60
Presentation of the Monologues.......................61
Instrumentation. .................... .................. 61
The Counselor Rating Form.............................61
The Recall Questionnaire............................. 64
Design and Data Analysis ...............................66
Limitations ................ ............................68

FOUR RESULTS ............................ .. ........... ........ 69

FOR FUTURE STUDY. ...................................... 78

Summary................... ............................. 78
Discussion.. ........................................... 80
Sociolinguistic and Cross-Cultural
Counseling Implications............................84
A Serendipitous Finding................. .............89
Present Limitations and Future Research
Recommendations .............. ........................ 91




C TAPESCRIPT OF RESPECT MONOLOGUE..............................95


E COUNSELOR RATING FORM. .................. .................. .97

F RECALL QUESTIONNAIRE.......................................101

RESPECT MONOLOGUE. ....................................... 104

RESPECT MONOLOGUE ............. ..........................105

OF RESPECT MONOLOGUE.. .................................... 106

OF RESPECT MONOLOGUE ...................................... 107

REFERENCE NOTES... ............... ............... .................. 108

REFERENCES.. ............. ...................... .....................109

BIOGRAPHICAL SKETCH ...... ....................... ........... ....... 117


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



Steven Jeffrey Stein

June 1981

Chairperson: Dr. Joe Wittmer
Major Department: Counselor Education

This study investigated how a counselor's speech style affected

racially different clients' judgements of a counselor's trustworthi-

ness, expertness, and attractiveness. Additionally, the study exam-

ined the question of whether certain counselor speech styles enhanced

the amount of information which was retained by clients who had con-

tact with a particular counselor. This study also attempted to de-

termine whether subjects changed their ratings of a counselor over

the course of three contacts with a counselor due to the counselor's

speech style. Finally, this investigation raised the question of

whether black clients and white clients would indicate a preference

for a particular counselor style based upon their own (client) race.

A counselor's vocal accent has been identified as one of several

important aspects of his or her total cultural and/or racial identity.

Sociolinguistic research has demonstrated that listeners, often in-

fluenced by their own backgrounds, have preferential attitudes toward

certain speech styles. This study attempted to integrate these related

themes in an empirical fashion by measuring, with the Counselor

Rating Form, subjects' attitudes toward the following four counselor

conditions: (1) A black counselor speaking with a black English ac-

cent, (2) the same black counselor speaking without an accent, (3) a

white counselor speaking with a southern English accent, and (4) the

same white counselor speaking without an accent. Forty black and forty

white male inpatient clients who were participating in an eight

week alcohol rehabilitation program at the East Orange, New Jersey,

Veterans Administration Medical Center served as subjects. Results

were analyzed by analysis of variance (ANOVA) and revealed that, over-

all, no one of the four different counselor conditions was "superior"

to any of the others. The sole exception to this was for the subset

of black subjects and only when they rated the black counselor.

Under this condition, the dimension of counselor expertness was en-

hanced when the counselor spoke in unaccented English. The dimensions of

counselor trustworthiness and attractiveness were not affected.

The amount of information recalled by subjects was not differentially

influenced by any of the four different counselor conditions. One

serendipitous finding emerged: universally, the black subjects gave

higher ratings to the counselors than did the white subjects. All

the findings are discussed in light of previous, related research

from the areas of cross-cultural counseling and sociolinguistics.

Implications for the practitioner are discussed and future research

questions are recommended.


Over the past three decades many different client and counselor

variables have been investigated as possible sources of influence on

the therapeutic process (Parloff, Waskow & Wolfe, 1978). And,

during the last several years several empirical and theoretical

articles have focused specifically on ethnicity as one of many impor-

tant sources of influence (Harrison, 1975). However, within the con-

text of the therapeutic relationship, only a few writers have con-

sidered sociolinguistic influences as part of the ethnicity question

(Conville & Ivey, 1974). This present study examined social speech

dialects of both black and white counselors and the manner in which

these dialects affected clients' perceptions of various counselor


Since therapy is, more than anything else, a verbal, linguistic

and conversational activity, it is somewhat surprising that more

attention has not been given to therapy from a communication per-

spective (Labov & Fanshel, 1977). Those who have investigated the

role that language plays in therapy have tended to take approaches

that vary along several different and only marginally related

dimensions. Linguistic investigations of the therapeutic process

have looked at meanings of single words (e.g. Sprafkin, 1970),

stylistic complexity (e.g. Meara, Shannon & Pepinsky, 1979),

counselor-client compatibility (Schumacher, Banikiotes & Banikiotes,

1972) and dialect (Stein & Walker, 1980; Stein, Note 1). An impor-

tant allied area concerns the study of language attitudes. Research

in this area has been conducted primarily by sociolinguists and

educators outside the context of counseling.

Recently, the role of language attitudes had a significant impact

on ethnic relations in one of our most important social institutions--

the school. The decision in Martin Luther King Junior Elementary

School Children v. Ann Arbor School District Board (Note 2) included

a plan ordered by the court which aimed at reducing teachers' biased

attitudes toward speakers of black English.

For several years, sociolinguists have been studying the role

that different speech qualities play in shaping listeners' attitudes

toward certain speakers. Briefly, these studies have consistently

demonstrated that people make personality and evaluative judgements

about others based on speakers' accents (Anisfeld, Bogo & Lambert,

1962), dialect (Light, Richard & Bell, 1978) and voice quality

(Markel, Meisels & Houck, 1964). For example, Tucker & Lambert

(1969) demonstrated that white speakers of Network English (the mode

of unaccented speaking characteristic of national newscasters) were

rated higher on positive personality traits than educated black

speakers, who were rated higher than black English speakers, who

were rated higher than southern white speakers. Most Americans use

a speech style or dialect having characteristics which listeners

often associate with a geographical region or social class.

Need For the Study

The concern with language attitudes is particularly crucial when

researching the role of ethnicity in counseling. Aside from obvious

ethnic physical characteristics that clients may notice immediately,

they may next be impressed by a counselor's vocal qualities. A major

goal of the present investigation was to study language attitudes

within the counseling context in order to determine whether these

attitudes were influential in shaping a client's perception of a

counselor. If so, were the effects of these attitudes constant over

time or were they primarily a product of an initial contact? Also,

beyond language attitudes alone, did the visual racial factors inter-

act with the ethnic factor of social dialect?

Purpose of the Study

The following brief description of the design introduces the

specific questions this study investigated (See Figure 1). A black

counselor and a white counselor each made two sets of brief instruc-

tional videotape monologues. Each counselor delivered one set in a

relatively accentlesss" speech style (Formal English) and the other

in an "accented" (Informal English) speech style. The white counselor

spoke Network English in one set and southern (accented) English in

the other set. The black counselor spoke black English in one set

and Conventional English in the other set. Thus, there were a total

of four counselor conditions: black or white counselor plus two

social dialects for each. Each set of video taped instructional

monologues contained three separate short talks in which the coun-

selor described three important basic elements of the counseling

relationship. One monologue focused on the need for preparedness,

another was on the importance of mutual respect and a third was on

the topic of client-counselor compatibility. Subjects viewed the

three video taped monologues on three different days over the

course of one week.

More specifically, this study attempted to answer the following


(1) When a client encounters a counselor for the first time,

how important is the counselor's speech style in deter-

mining the client's perception of the counselor?

(2) Does a counselor's speech style continue to influence the

client's perception of the counselor past the first contact?

(3) Does a client indicate any preference for type of counselor

based on the counselor's dialect?

(4) Does a client indicate any preference for counselor dialect

based on his own race?

(5) Do any of the factors--i.e. counselor dialect, counselor

race and client race--have any effects on how much a client

remembers about the content of the instructional monologues?

Significance of the Study

This study contains three related aspects which have an impor-

tant bearing on the area of counseling research. First, although

a rich literature exists in the allied fields of sociolinguistics

and language attitudes, no counseling researchers have attempted to

extend these linguistic findings into a counseling context. As

"ethnicity" becomes an increasingly important factor in both the

society at large and in the counseling profession, language variation

associated with one's culture is often singled out as an important

factor interfering with the communicative process. This study, which

makes explicit use of the sociolinguistic findings regarding language

attitudes, attempted to demonstrate their legitimate place in the

context of counseling.

Secondly, this integration of findings from two related disci-

plines helped to further clarify and empirically define the question

of "ethnicity." This term has various meanings depending on who is

using it and in what context, and it often becomes a meaningless con-

cept. Language variation as a demonstrable part of one's "ethnicity"

helps to further operationally define this elusive, but important


Finally, the implications for counselor training are an impor-

tant consideration of this study. In a general sense, counselors are

already aware of both what they communicate and how they communicate.

Up to this point, much consideration has been given to many general

aspects of vocal qualities. These include variables such as silence,

raising and lowering one's voice, and paralinguistic variables such

as body language. This study focuses on a more implicit quality of

the communication itself, yet one which is under the counselor's

control. A counselor's voice is one of his or her main tools and is

multi-faceted and more complicated than is generally recognized. At

this stage it would be premature to suggest that counselors should

receive formal voice training as part of their curricular require-

ments. It is certainly not premature, however, for professionals

in the counseling field to consider dialect variation and how it may

be affecting clients' perceptions of counselors.

Definition of Terms

Network English. This term has been used by Tucker and Lambert

(1969) to describe a style of speech which is characteristic of

national newscasters. The important quality of this style of speech

is that it is, in theory, devoid of regional or social accents. For

two reasons this is actually a hypothetical form of speech. First,

no speech style is completely void of some identifying qualities and,

secondly, it is a matter of opinion as to what constitutes an accent.

In this study, Network English is also referred to as "unaccented" or

"formal" English.

Conventional English. Like the other forms of English being

used in this study, Conventional English must be defined operationally.

For purposes of this study, Conventional English is analogous to

Network English. The reason for using the two different terms is to

distinguish the two counselors from one another. Since this term is

analogous to Network English, it may also be referred to as

"unaccented" or "formal" English.

Black English. Dillard (1972) has written an exhaustive text

on this subject. Black English is a dialect of English that has

historical roots in pre-slavery Africa. It is a systematic, rule

governed and quite logical form of English; and it has been estimated

that about eighty percent of American blacks speak some black

English. Because of racism and misunderstanding, many negative

attributions have been made about this form of English. Bell (1979)

gives an example of how a Black English rule differs from the

Conventional English rule:

Rule: The 's' or 'es' is omitted from the present tense verb
when used with third person singular noun or pronoun.

"He go home through the field." Black English
"He goes home through the field." Conventional English

At times, black English may be referred to as "accented" or

"informal" English when writing about the black counselor.

Southern English. Again, this speech style must be defined in

relative terms. There are certain phonetic characteristics that are,

for example, often identified as being particularly associated with

someone who has lived in the southern United States. For example,

the expression "y'all" is readily associated by most listeners as

typically and exclusively a southern expression. For purposes of

this investigation, what was most important was that obvious and

typical phonetic differences existed between the white counselor's

version of Network English and his version of Southern English. This

term may also be referred to as "accented" or "informal" speech when

referring to the white counselor. It is not analogous to black


Accent. An accent refers to those phonetic variations of

sounds within a given language that can vary without changing the

meaning of the word in which a given sound is found. For example,

a native French speaker may pronounce the word "see" as "zee" as in:

"I'll zee you later." He may be said to have spoken in a French

accent because he used one of the sounds of his native dialect in a

context that made sense to the listener of the English (See Dialect).

Dialect. A dialect is a variety of a language distinguished

according to the user (Halliday, McIntosh & Strevens, 1973). Differ-

ent groups of people within a language community may speak different

dialects of the same language. That is, two dialects of the same

language are mutually intelligible. When a speaker learns a second

dialect, it is often spoken with the accent of his native dialect.

Register. While this term has not been used thus far, it is

important to understand its meaning and how it differs from dialect

and accent. Register refers to a category of language that is dis-

tinguished according to use. This category accounts for what people

do with their language. While dialects differ with respect to sub-

stance, registers differ primarily in form. Lexical items are the

most obvious: "cleanse" puts us into the language of advertising,

"probe" puts us into the newspaper headline language. A given person

may switch registers depending upon the social situation or depending

upon the intent of the communication.

Ethnicity. In this investigation, ethnicity referred to whether

a patient or counselor was black or white. It is understood that

this criterion of color is just one aspect of a larger context which

really includes a person's entire cultural background.

Client. This study utilized subjects who were inpatients in a

Veterans Administration hospital alcohol program. The terms client,

patient and subject all refer to these individuals unless otherwise


Counseling. The terms counseling and therapy as used in this

study refer to the same process. Similarly, for counselor and



In recent years there has been a growing interest in the area

of cross-racial counseling. From reviews of writing in this area

(e.g. Harrison, 1975; Griffith, 1977) several considerations emerge

for counselors and allied professionals to consider in their work

with ethnically different clients.

Considerable controversy exists regarding whether to include

ethnicity as a relevant factor in counseling research. The three

sides to this controversial question are represented, philosophi-

cally, by the following Chinese proverb:

Each person is like every other person, each person
shares certain qualities with every other person,
each person is like no other person.

For those who do consider ethnicity a relevant variable (e.g.

Vontress, 1970; Sue, 1978), the task of defining the factors which

contribute to one's ethnicity becomes challenging when just a few

of the most salient aspects are considered: ancestral heritage,

social class, nationality, political ideology, language, religion,

education and value systems. When these factors are considered

together as possibly interacting process or outcome variables, it

is a complex matter to parcel apart the important variables from

the incidental ones. Furthermore, it is risky to make generaliza-

tions based on single research projects or to overemphasize the

importance of one of the above factors. In addition, there are the

questions of whether the best counselor-client match is necessarily

one where both share similar racial or ethnic backgrounds and

whether therapists should use different counseling techniques depend-

ing on a client's ethnic background (Smith, 1977). With respect to

theory, research and practice, the state of the art of ethnic con-

cerns in counseling portrays a new and uncertain profile.

It seems important to recognize that often unknown to one

another, many researchers have been engaged in efforts to solve

different though related questions; all of which are all relevant to

the area of cross-racial counseling. Several characteristics of this

diverse yet related body of research have been cited. While Griffith

(1977) observed an overrepresentation of anecdotal accounts and

uncontrolled observations, Sattler (1970) found fault with method-

ological inadequacies and inconsistencies in the literature. Further,

Harrison (1975) clearly stated the need for further empirically based

research. Abramowitz and Dokecki (1977) noted an advantageous use of

the results from this broadly based body of research. They noted

that the validity of new theory is supported when convergent findings

emerge from this widely based research. Simultaneously, however, what

can be said about divergent data or areas of research related in

fundamental ways but which remain unintegrated? With these questions

in mind, this chapter contains reviews of several of these related

areas under the following major headings: (1) Racial Variables in

the Counseling Process, (2) A Brief Account of Sociolinguistic and

Language Attitude Research, and (3) Language Variables in Clinical

and Counseling Settings.

Racial Variables in the Counseling Process

In the last several years there have been at least several

hundred articles of either a theoretical or empirical nature which

fell into this category. As a result, not only have there been many

review articles summarizing the trends and findings (Gardner, 1971;

Harrison, 1975; Griffith, 1977; Sattler, 1970, 1977; Jones, 1977),

but there have also been reviews of these review articles (Parloff,

Waskow & Wolfe, 1978; Garfield, 1978). Although many thoughtful

questions have been raised, the answers to many remain inconclusive

and controversial. The remainder of this section explores the work

that has already been directed at some of these important issues.

Same-Race Versus Interracial Counseling Dyads

Most of the writing in this area has centered on dyads in which

the counselor is white and the counselee is black. A general trend

reported by some reviewers indicates a wide range of clients--high

school students, college students and adults--have predictable

preferences for counselors. Harrison (1975), Griffith (1977) and

Parloff et al. (1978) all concurred that clients tended to prefer

counselors of the same race. Griffith (1977) stressed the fact that

same-race dyads were not, however, a prerequisite for successful

therapy. It is important to note that the support for this position

was not unanimous. Sattler's (1977) position was that the thera-

pist's race was for the most part, not a significant factor in

interracial counseling. Parloff et al. (1978) attempted to shed

some light on these differing interpretations of what were, gener-

ally, conclusions based on the same body of research. First,

Sattler's review included a number of studies that demonstrated no

difference due to racial matching (e.g. Bryson & Cody, 1973; Ewing,

1974). Secondly, Griffith relied primarily on a few studies even

though he recognized their shortcomings. Finally, the different

reviewers seemed to place different emphases on what they thought

was the most important measure of client preference: these included

self-disclosure, self-exploration and noted client satisfaction.

There were several other problems noted in this cross-racial

research that concern methodology.

Jones (1978) questioned the generalizability of findings based

solely on one counseling interview. Parloff et al. (1978) suggested

that the prevalence of racially biased counselors may be under-

represented because of the self-selection aspect of therapists for

much of this research. More systematic research is needed to deter-

mine the effects of various racial pairings on the outcome of therapy

(Parloff, et al., 1978; Harrison, 1975). Many variables have often

been mentioned as possibly interacting with racial factors, but they

have not been studied systematically. Most importantly, they include

sex and social class. Additionally, therapists' attitudes and back-

grounds have not been systematically evaluated in the context of

cross-racial counseling.

In summary, it appears from the literature, that the question of

racial matching in therapy is not a very potent question when it is

asked in the abstract. Further, when other measurable factors are

excluded from consideration, or when certain variables are easily

confounded with race, this state of affairs only tends to cloud the

basic issue. Each of the writers reviewed has suggested some new

questions that not only make sense from a theoretical point of view,

but are also formulated with a more critical understanding of the

complex nature of counseling research. Parloff et al. (1978)

suggested that what are probably even more important than studies

of therapist race are studies of the effects of different therapist

attitudes toward and understanding of people with ethnically differ-

ent backgrounds. Griffith (1977) recognized that the research in

the area of racial influences is only in the beginning stages.

Although he recognized race as unquestionably an important factor,

the crucial factor is not whether race influences the therapeutic

process, but in what way and under what circumstances. The thera-

pist's personal attributes are mentioned as one of the most crucial

factors in this question.

Special Techniques for Special Populations?

A frequently asked question in the area of counseling ethnic

minorities concerns the notion of using special counseling tech-

niques with ethnically different clients. The answer to this ques-

tion is as controversial and uncertain as the issue of racial


Vontress (1970) described several special problems which a black

individual may bring to therapy. These include, especially for the

black male, a resistance to self-disclosing to a white therapist.

In addition, the white therapist is likely to have countertransference

reactions that may very well interfere with the therapeutic process.

Although Vontress (1970) did not advocate a special "technique" for

counseling blacks, he offered some very specific suggestions aimed at

increasing counselor effectiveness. First, he argued that there are

not enough professionally trained counselors to meet present and

future needs. Second, the ones already in the field need additional

awareness training to help them become more effective in their work

with culturally different clients. Retraining methods include

in-service workshops aimed at sensitizing the middle-class therapist

to the needs and value systems of the culturally different client.

Gunnings and Simpkins (1972) presented a systemic approach to

counseling the disadvantaged client. This approach stresses the

important role counselors can play in influencing factors outside the

counseling office. The assumption of this approach is that the

community (or "system") is responsible for many of the problems of

the disadvantaged. The system often places the culturally different

minorities at a social, economic and political disadvantage. Thus,

a client is not really "culturally deprived"; but a counselor who is

unaware of the culturally different client's background is prone to

make this type of stereotypical judgement. From the systemic point

of view, a counselor's role is expanded from being just a coping

agent for the client to include being an agent of change. The coun-

selor should function in the community in several roles that attempt

to change the system in ways which will directly or indirectly help

his client and others. The counselor must be an advocate of his

client, build bridges between culturally disparate factions in the

community and act as an appropriate role model. An additional role

of the counselor is one of a consultant--to teachers and others in

influential positions.

An important aspect of the Gunnings and Simpkins (1972) model

is the stress they gave to changes that they advocated in graduate

education. Students should not be protectors of the status quo,

but rather should be actively involved in making improvements in

both their graduate programs and in the community.

Tucker (1973) presented a model for counseling the oppressed

which stressed the importance of action and accountability on the

counselor's part. In this model counselors must take active steps

to make themselves seen. This involves outreach into the relevant

client setting and the development of trust between the therapist

and potential clients. The accountability aspect of this model

involves careful delineation of both the short term and long term

goals of the intervention. Further, proper priorities and appro-

priate strategies must be developed that have special relevancy for

the oppressed. Finally, the results and outcomes nust be periodically

evaluated to insure that the actions of the counselor have been mean-

ingful to the special population in question.

Several authors have presented models which attempt, either

explicitly or implicitly, to transcend race. Harper and Stone

(1974) presented a model of counseling that, although was formu-

lated with the black client in mind, went beyond race per se and

is appropriate for most clients. Basically, any sound counseling

therapy must have a sound basic philosophy and a range of useful

strategies. A counselor's orientation, knowledge and experience

must transcend racism. This means that while one's race can and

should be acknowledged by counselor and client alike, neither should

become fixated on this issue. This transcendent model is rooted by

some sound strategies employable by any counselor. First, basic

needs should be assessed and the appropriate actions taken so the

client can grow both in terms of the therapeutic relationship and

as an individual. Second, the client's current life style should

be assessed and appropriate action should be taken to alleviate any

extreme difficulties. Third, specific behaviors must be evaluated

and the appropriate confrontations must be made by the clients as

they change specific aspects of daily living. Finally, the coun-

selor can help the client assess the dynamics which may have con-

tributed to the problem from the start. This is an educative process

as well as a change process. It involves growth on the part of both

the client and the counselor. Although race may be an issue in the

beginning, the process ends at a point well beyond the mere consider-

ation of a person's race as the crucial part of the therapeutic


Smith (1977) presented the pitfalls of focusing upon racial

differences not only in the counseling context, but in the society

at large. She pointed out that the continuation of describing the

differences between racial groups only serves to reinforce the

predominantly negative attributes already held by many majority

group members about minority group members. Too often, special

counseling models are forwarded describing the culturally different

as "disadvantaged" or "deprived." With such a label there is the

implicit assumption that to be black or culturally different is to

be handicapped in many aspects of one's life.

Smith (1977) described several trends of the counseling

literature that have tended to stress racial differences between

blacks and whites but which have done little to promote a better

understanding or increase mutual tolerance. Much of the literature

has tended to stress that most blacks have negative self-concepts

and that counselors should work toward alleviating such a view.

Similarly, many characterizations of the black family have been

negative and have been seen as contributing to the "black problem."

There is also the myth of the verbally restricted black client. All

these erroneous assumptions have been seen by Smith (1977) as per-

petuating the racist problem that exists in our society. Counselors

are not immune to racism. She attributed much of this bias in the

literature to the manner in which research is conceptualized. The

bottom line for Smith is that despite all the research that has been

conducted on black (and other) groups, very little has been done as

a result to improve their generally poor life situation. When black

clients are seen in therapy first as blacks and secondly as individ-

uals, Smith (1977) contended that all the stereotypes and biases

inevitably become an important factor to overcome. Hypotheses turn

into conclusions and experiments are constructed to validate per-

sonal feelings rather than to investigate uncertainties.

McDavis's (1977) eclectic model also attempted to transcend the

issue of race without ignoring the possibility that certain racial

aspects may have to be addressed during the process. Since each

person is an individual and comes to therapy with different problems

and different perspectives of the world, an eclectic position offers

the maximum range of flexibility for both the counselor and the

client. Behavioral, existential, client-centered, reality and

Gestalt procedures may be used in an effective blend by a skillful

therapist to help any client. If a minority client brings a special

concern to therapy which is a result of his culturally different

development, this is not seen as a "deprivation" or "deficit." The

eclectic position is functional in both a theoretical and practical

sense. Every client is different and will naturally have different

needs in therapy. At the same time, all clients share certain basic

needs. When a well-trained counselor uses the techniques from a

variety of theories in a creative manner, it is not unreasonable to

expect that the range of clients which that counselor will be able

to effectively help will increase both with respect to clients'

backgrounds and their problems.

Premature Termination of Therapy
and the Minority Client

Mental health care delivery in the United States takes place in

a wide variety of settings. School counselors, community mental

health centers, governmental agencies, private and public hopsitals

and private psychiatrists and therapists are all involved in pro-

viding the bulk of (formal) mental health services to the American

public. Many of the minorities who avail themselves of these ser-

vices are restricted by income to those agencies providing services

either at a reduced fee or that participate in some type of insurance

program. Much has been written about the implications of such a

system. Some authors have focused on the discrepant training expe-

riences of individual counselors as they work with clients across

the socioeconomic spectrum (e.g. Lewis & Lewis, 1970). Others have

examined the expectations of counseling from the viewpoint of the

minority client (e.g. Edwards,Greene, Abramowitz & Davidson, 1979).

Still others have attempted to look at the interactions of variables

that necessarily take place when two culturally different people work

together in therapy (e.g. Jones & Seagull, 1977).

A number of assumptions associated with this literature will be

addressed below. One is that we usually assume that a minority

client is being seen by a "white middle-class" therapist. This is

due primarily to practical factors; namely that most therapists fit

the above description. As more minority members enter the mental

health field this ratio can be expected to change accordingly. A

second set of more crucial assumptions have been prevalent for some

time which concern attitudes about the efficacy of counseling for

minorities and especially the poor. These assumptions have been

forwarded by many authors (e.g. Albce,1977; Crowell, 1977; and

McSweeny, 1977) and include the following: (1) Attraction Hypothesis:

Mental health professionals cannot create clinical settings to attract

the culturally different and the poor; and if they could, these

clients would not use them. (2) Duration Hypothesis: Even if the

poor sought help, they would receive less treatment than more affluent

groups. (3) Elitism Hypothesis: Even if duration of treatment of

the minority clients was on a par with the mainstream client, the

quality of treatment would be less. (4) Effectiveness Hypothesis:

All else held equal, even if the culturally different received

treatment equal to the mainstream client, they would profit less.

The accuracy of each of these views has been equally widely

questioned. For example, disputing hypothesis (1), Scharfstein,

Taube and Goldberg (1977) reported that, nationally, 84% of the

federally funded community mental health center clientele were

persons who had annual incomes less than $10,000. With respect to

hypothesis (2), Pettit, Pettit and Welkowitz (1974) failed to find

any overall connection between length of treatment and social

class. Similarly, with respect to the elitism hypothesis, Stern

(1977) found no connection between level of patient education and

type of treatment received. Lastly, it is clinical lore that

lower-class and minority patients don't benefit as much as others

from therapy (Lorion, 1973).

From the above, that which seems most outstanding is the

variety, rather than consensus, of opinion and research findings

regarding mental health efficacy among the poor and minority groups.

The terms "poor" and "minority" mean different things to different

people. As used here, the two terms are not meant to necessarily

be interchangeable. Despite the fact that there is often a large

overlap between the two for some groups (e.g. Native Americans), it

is the intention of this writer to consider both poor people and

ethnic minorities as often culturally different from the counselor.

What does the literature indicate that counselors should know about

the culturally different and what can counselors do for themselves

to increase their effectiveness with clients who are culturally

different from themselves? The next section addresses this question.

Theories of Cultural Awareness for Counselors

Wrenn (1962) described "cultural encapsulation" as a very real

threat to counselor effectiveness. This encapsulation takes the

form of stagnation with respect to the counselor's outlook as a pro-

fessional and can naturally be expected to contaminate his counseling

role. The aspect of change is central to Wrenn's formulation. He

was concerned with the rapid pace of cultural change occurring and

how this may bias the process of therapy. Several formulations were

forwarded intended to be both descriptive as well as educative for

the counselors.

The first of these is the "tendency to be surprised or even

unbelieving regarding changes in truth." The truths, or values a

counselor holds are culture bound. They are also bound by time.

Every person is, to some extent, reared to enter a different world

and hold the values associated with that world. An unquestioning

loyalty to the walls of the world may ultimately betray any person.

The counselor must recognize that his values are for the "now" and

for "him"--not for all time or all people (Wrenn, 1962).

A second encapsulation which may occur is the cushioning of the

"counselor in some academic cocoon that may have little reference to

our total culture." Some of the dangers of this myopia include

notions like: grades are the most important thing to a student, tests

should be given to everyone, and learning is primarily from books.

Therapists are academic people and their academic cocoons may be most

unrealistic in terms of either the future scope of human behavior or

of the reality of the world outside the academic setting. Related

to this is the idea of tentativeness. Whatever the counselor may

know firmly--is it fixed as such as a law? Above all, warns Wrenn,

let the counselor avoid dogmatism--especially where the evidence is


A third caveat is the danger of the "assumption that the coun-

selor may safely draw upon his own education and career experiences

in counseling the client." The danger of this is that the counselor

is using his "yesterday" to help the client with his "today" or

"tomorrow." Although it is most human to draw upon one's own expe-

riences, the counselor must be alert to differences in both the time

frame as well as personal perspectives. Although these suggestions

are relevant to counselors generally, they are especially relevant

to those engaged in delivering services to the minority client. What

are some ways to avoid these pitfalls? Wrenn suggests some prophy-

lactic measures that are described below.

First, counselors should constantly be engaged in unlearning

certain things. Each day should witness some personal "fact" being

removed from the counselor's warehouse of stored information. It is

a process of questioning that leads to such rejection. The process

is characterized by both openness and courage. A second suggestion

for counselors is that they check their counseling methods not only

for the accuracy of their information, but for the direction and rate

of change. The "what" of the message is only part of the story.

Where is it going to lead the client and how long will it take? Next,

the counselor must be obliged to encourage even the client who thinks

very differently from himself. The counselor's comfort should not be

his gauge of success. It may be that these persons who think differ-

ently will be the hope of our world. Finally, counselors must fight

the tendency to be self-righteous. What is "right" for the counselor

may be at variance with what is "right" for the client (Wrenn, 1962).

These admonitions laid down by Wrenn serve well as guidelines, it

seems, for the description of a more systematic theory of eliminating

cultural oppressions in counseling put forth by Sue (1978).

Sue (1978) began his analysis by presenting four possible world

views that any person (i.e. client or counselor) may bring to ther-

apy: (1) Internal locus of control (IC)--Internal locus of responsi-

bility (IR), (2) External locus of control (EC)--Internal locus of

responsibility (IR), (3) External locus of control (EC)--External

locus of responsibility (ER) and (4) Internal locus of control (IC)--

External locus of responsibility (ER). It was proposed that the

IC-IR world view is most characteristic of Western counseling

approaches and assumptions. Cultural oppression occurs when this

world view is blindly imposed upon the culturally different client.

The loci of control and responsibility can be determined by paper and

pencil tests which, to describe adequately, would take us beyond the

scope of this review. Rather, let us turn to a brief description of

each of these four possible world views and consider the implications

of them as they pertain to a counseling relationship.

According to the theory, the Western counselor is likely to bring

the IC-IR world view to counseling. That is, this counselor is likely

to place a high value on personal resources for problem solving; self-

reliance; pragmatism; individualism; status achievement through one's

own effort; and power over others. Many democratic (some would argue,

capitalistic) ideals are reflected in this world view--"God helps

those who help themselves," for example. (Evidence for the prevalence

of this view may be gleaned from a stroll through a local bookstore,

where one can find a plethora of self-help guides.) The effects of

such an approach, if carried through and applied to a minority group

who has a different world view, is that this minority group is often

labelled as somehow deviant.

People who fall into the EC-IR realm are most likely to accept

the dominant culture's definition for self-responsibility but have

very little real control over how they are seen by others. The key

issue here is the dominant-subordinate relationship between two

different cultures. It is reasonable to expect members of one cul-

tural group to adjust to the group that possesses greater prestige

and power and to avoid inferiority feelings. The result of this

type of adjustment for the minority member is ambivalence. The

pressures for acculturation and assimilation are strong and the net

effect has sometimes been referred to as marginality. For an EC-IR

client to focus on his feelings in a therapy session may be very

threatening because of the possible self-hate which may be uncovered

and the realization that they cannot really escape from their own

racial and cultural heritage. A counselor who is culturally encap-

sulated, in Wrenn's sense of the word, may unwittingly perpetuate

such conflicting feelings.

The EC-ER client is one who is likely to blame the system and

feels there is very little he can do to reverse the inevitable pre-

judice and discrimination. In its extreme form, this view of the

world can be conceptualized as akin to Seligman's (1975) learned

helplessness. This mode of adjustment may be characterized by an

individual who doesn't want to "rock the boat" and is mainly con-

cerned with surviving. The black experience of slavery can be seen

as a possible historical contribution to the development of this

world view. In a current situation, an EC-ER client may likely see

the white counselor as symbolic of any other black-white relationship.

A culturally aware counselor realizes these possibilities and makes

the appropriate intervention.

The IC-ER clients believe in their ability to control events in

their own lives if given the opportunity. These persons are not

likely to accept their own present state as an indication of personal

weaknesses. Rather, these persons may realistically see the external

barriers of prejudice and discrimination as formidable barriers.

These individuals, because they understand themselves to be able to

control events, would be more likely to participate in civil rights

activities and to stress racial identity and perhaps militancy. This

theory predicts that the IC-IR counselor and IC-ER client will not

only define problems differently, but would also go about solving

them with different strategies. Furthermore, counselors who are

themselves IC-ER oriented are most likely to use action oriented

approaches in contrast to IC-IR counselors, who would rely more upon

non-directive and verbal strategies.

This theoretical model has implications that naturally extend to

counselor training and research questions. If certain techniques

or skills are more appropriate with one group over another, then it

becomes a training imperative for these techniques to be covered

during a counselor's training. Some of these considerations will be

covered in another section of this chapter. Before moving on, some

of the possible limitations of this model should be mentioned. First,

this model is currently in a speculative form, although future

research may very well validate many aspects of it. Secondly, the

precise behavioral expectancies associated with each of the four world

views haven't been fully described as yet. Lastly, the four

categories represent, to a large extent, conceptual categories. As

such, overlap and variations are expected; both within and across

different cultural groups. Now that some theoretical issues have

been presented, what are the implications of some of the ethnic and

cultural variables considered thus far with respect to counselor

education and delivery of service to the culturally different client?

Practice of Cultural Awareness for Counselors

In a very general way, Schlossberg (1977) addressed the question

of bias as it affects the counselor. Her thesis was that to some

extent, all people are both guilty of some type of bias as well as

being victims. In addition to the familiar ethnic, class and sex

biases, she mentioned age and physical handicaps as other less

common, but no less equally disturbing sources of distortion. She

warned the counselor of dangers of attaching labels to people based

on stereotypical categories. Stereotyping serves to eliminate infor-

mation rather than expand the therapeutic relationship. Counselors

must be aware of their biases. They must take care not to hide behind

the trappings of professionalism. It is suggested that counselors

undergo some group workshop experience as part of their training

program to help them reduce or at least obtain an awareness of their


Lewis and Lewis (1970) presented a training model for educating

inner-city counselors. First of all, the role of the counselor in

this setting takes on several dimensions. Not only must the coun-

selors provide individual and group therapy for the client population,

but their role naturally extends into the general structure of the

agency in which they work. School counselors' responsibilities

include efforts to make the school a more productive place for the

students to learn and grow. As an expert in human relations,

assuming this role should be a natural extension and application of

basic skills. Furthermore, the counselor can come to play a role

outside the school, helping to both coordinate communication between

the institution and the community and trying to change what can be

changed in an unhealthy environment. Some specific training

approaches were presented (Lewis & Lewis, 1970) to facilitate a

meaningful learning experience. Experienced counselors could be

teamed with beginning students, and each team placed in an inner-

city setting. The teams could work as full-time counselors, dis-

covering and attempting to meet the needs of their specific service

population. The didactic components of a relevant training program

must include both specific intervention skills as well as some

multidisciplinary components.

Sikes (1971) and Bell (1971) made some specific recommendations

regarding the curriculum requirements of counseling psychology

programs for doctoral students. These curriculum changes were aimed

at including some of the following topics in the psychologist's

formal training: racism and discrimination from a historical and

sociological perspective; black values and psychology; critical

examination of the detrimental role which standardized tests can

play in counseling and education; and finally, relevant practicum

and internship experiences. Both authors stressed the importance

of these subjects, especially for the minority graduate student.

Minority members who receive advanced degrees and then go back to

their home communities are often perceived as alien figures whose

expertise is not relevant to serving the minority clients' needs.

A systematic program to deal with combating racism has been

developed by Katz and Ivey (1977). They outlined a program of

re-education intended primarily for white counselors (and others)

aimed at raising both consciousness about the problems and ending

at a stage where some action steps are clearly defined. The program,

which was conceived as a two weekend experience, consists of stages

that take the participant through various levels of understanding

the problem. It begins with cognitive aspects of racism and proceeds

to identify, for each individual, a personal awareness of his own

racial attitudes, especially the affective component. Finally,

participants develop specific action strategies to combat personal

and institutional racism and define further steps in their search

to become antiracist.

A model of cultural expertise counseling has been forwarded by

Ivey (1977). He defined a culturally effective counselor as one

who is able to relate with self, others and society within a multi-

cultural framework. One who is culturally effective has developed

skills in a wide range of areas and, further, the counselor uses

flexibility in the application of these skills. The skills were

presented as a taxonomy of behavioral skills: (1) Basic skills of

a culture. These include critical aspects of body language, eye

contact, vocal tone and knowledge of what may be subject matter

regarded as taboo for a given culture. (2) Communication skills.

This involves both attending skills such as listening, paraphrasing,

reflecting, summarizing; and influencing skills such as directing,

interpreting, expressing content and advising. (3) Qualitative

skills include aspects that are appropriately brought to any area

of the taxonomy. They are dimensions intended to enhance the quality

of the relationship: respect, genuineness, concreteness, and

immediacy. (4) Focus skills help the counselor in targeting specific

people or objectives as central to planning action. Foci may include

client, others, specific topics or cultural issues. The ultimate

objective of Ivey's (1977) model is cultural expertise. The model

was intended to cover an array of variables all germane to the com-

plexity of cultural relativity.

The next major section of this chapter reviews some of the major

investigations that have occurred in the field of sociolinguistics in

recent years. Many of the following articles involved investigations

of listeners' attitudes toward speakers of accented English. These

investigations are relevant to the previous discussions because of

the intimate connection existing between one's culture and one's


A Brief Account of Sociolinguistic
and Language Attitude Research

Introduction to Research
on Social Dialects

Williams (1971) described a range of research assumptions that

are currently prevalent in the field of sociolinguistics. It seems

important to recognize that no one discipline has a monopoly on

social dialect research. Even within the language scientist's

community itself, several disciplines are represented. One of these

areas is represented mainly by speech pathologists, speech scientists,

audiologists and speech therapists. Another area is represented

primarily by persons concerned with the teaching of speech ideas in

the tradition of rhetorical theory, the history of public discourse,

and the psychological study of communicative behavior. As with any

young science, there are discrepancies between what the researchers

are saying about social dialects and what the speech teacher is

doing in everyday practice. This type of gap, between research and

practice, is commonly found in most emerging disciplines. In this

case it is exacerbated by the fact that not only have social

dialects recently received research attention, but this attention

has been spread throughout an array of related speech fields.

Williams (1971) concludes that researchers need to know more about

the nature of social dialects and their implications for clinical

and educational practice.

A Short History of "Disadvantaged" Language

Bereiter and Engleman (1966) proposed that speakers of non-

standard English dialects are "deficient" and that those who

speak them are "deprived." This point of view was prevalent during

the 1960's. In fact, this view inspired headstart programs to

employ remedial language drills in an attempt to eliminate dialect

differences of black and other minority students and bring them in

line with the mainstream culture. This deficiency point of view was

based largely on Bernstein's (1961) work. He proposed that

languages fell into two categories. "Elaborated" codes were used by

middle-class speakers and were characterized by the use of abstract

terms, rational thinking and relative comparisons. "Restricted"

codes on the other hand, were used primarily by lower-class children

and handicapped the speaker because of the concrete, personal, and

here and now quality of this speech code. Conceptualizing dialect

differences in this way played easily into the hands of those who

held views similar to Jensen (1969). From here, it was one small

step to the notion that nothing could be done by the teachers--the

child must be naturally "deficient" and his language was merely more

proof of it.

Fortunately, the 1970's introduced a series of writers (e.g.

Labov, 1970; Burling, 1973; Ginsburg, 1972; Dillard, 1972) who

strongly attacked the deficit theory with some powerful evidence.

Linguistically, there is no such thing as a "deficient language."

Grammars are relative and no given system is inherently "better" than

another. Standard English is one of several social dialects spoken

in this country and it seems more than coincidental that this form

is preferred by middle-class educators and others. After all, they

are the ones controlling educational and other social institutions.

Dillard's (1972) classic work, entitled Black English, traces the

history of black English to pre-slavery Africa. He explains the

rules of black English and how they systematically vary somewhat

from standard English. The difficulty for students arises because

the dialect is close enough to standard English to be understood

by others, but is seen as substandard rather than merely different.

In a systematic way, Labov (1970) challenged Bernstein's (1961)

"deficit" position. While Bernstein (1961) interviewed lower-class

children after taking them out of their home community and bringing

them to the alien academic laboratory, Labov (1970) utilized black

researchers who were familiar with both black English and black

culture. Labov's investigations led him to conclusions quite

different from Bernstein's: Labov (1970) discovered that verbal

fluency is an important skill which reflects status within the

young black peer group. Standard English may be rejected by these

individuals because it suffers from a low status.

Edwards (1979) pointed out that this conflict over language

preference can also extend to related paralinguistic cues: gestures,

intonations, and the use of silence may also vary across cultures.

Further, these types of differences extend beyond the black culture.

Abrahams and Troike (1972) described the Navajo custom of speaking

softly to show respect, only to have the teacher exhort Navajo chil-

dren to speak louder in class so they can be heard.

Several suggestions have been forwarded by Edwards (1979) that

addressed some of the above issues. First, many educators and others

must recognize that textbooks, testing programs and many implicitly

advanced values are predominantly middle-class in nature. As soon

as culturally different children set foot in a middle-class institu-

tion, they are expected to change and conform by rejecting what they

have already learned at home. Many teachers and allied professionals

have themselves only recently ascended the class ladder and reached

middle-class status. It is not unnatural for many to feel some

ambiguity when they find themselves in a position of authority. Having

rejected many of their own cultural values in favor of the middle-class

system, they may feel compelled to encourage others to do the same.

Second, all who deal with dialect speakers must recognize the

importance of attitudes. Rosenthal and Jacobsen (1968) showed that

when teachers expect children to either succeed or fail, based on

certain irrelevant observable behaviors like verbal expertise, these

actual behaviors often comply with expectancies.

Edwards (1979) describes the dialect dilemma as two-sided. If

children are taught a dialect other than their native one, the impli-

cation is that the native dialect is somehow inferior. On the other

hand, if children's language is left alone and they are encouraged

to develop their own dialect, these individuals will likely face a

biased society which penalizes them for this. Traditionally, the

first approach has been favored; it has generally failed. Edwards

(1979) suggested that the second alternative, although not tried yet,

holds most hope. However, to leave children's dialects alone does

not mean to leave the study of language alone. She suggested that

teachers (and other professionals) should become familiar with a

second dialect to gain empathy for the hard task that we ask the

speaker of nonstandard English to do. Several socioeducational

alternatives are suggested to deal with this issue. One is to change

completely to the mainstream dialect. A second is to completely

avoid the mainstream culture and develop the native dialect. A

third is a method called "bidialectalism"--learning to speak more

than one dialect fluently so one can appropriately code switch

depending on the social setting. Sledd (1969) and O'Neil (1972)

have criticized this last alternative as merely a veiled attempt of

reform inspired by the same middle values that have already had a

negative influence on the way dialect speakers are perceived.

Listener's Attitudes Toward English
Dialect Speakers

During the last two decades, several empirical investigations

have examined the various ways in which listeners are influenced by

the style of English used by different speakers. One of the impor-

tant methodological techniques used in many of these studies is a

manipulation called the "matched guise" (Lambert, Frankel & Tucker,

1966). Briefly, the procedure involves the use of taped readings of

a bilingual or bidialectical speaker delivering a message at one time

in one of their speech styles (e.g. English) and later, in their

second style (e.g. accented English), the same passage. Groups of

listeners judge the different "guises"; their task often involves

assessing the personality characteristics of each "speaker," using

voice cues only. The procedure has been used with contrasting

languages, such as English and French; dialect variations, such as

Parisian versus Canadian style French; and accent differences, such

as southern accented and unaccented English. In addition, only

brief speech samples have been needed to obtain reliable ratings.

In one case, speech samples were as short as one sentence--or about

ten seconds (Cremona & Bates, 1977). This section describes several

studies in which this technique (or some variation of it) has been

used to assess listeners' attitudes toward various styles of spoken


Listeners' attitudes toward speakers of accented English seem

to form at a young age. Light, Richard and Bell (1978) examined

differences perceived by 92 eight and nine year olds between an

educated southern black woman (standard speaker) and an uneducated

southern black woman (nonstandard speaker) who each delivered one

and a half minute speech samples. In general, the results indicated

more positive qualities (smart, pretty, rich, nice) were attributed

to the standard speaker. Even though both speakers were black, half

of the subjects judged the standard speaker as white and only 6%

judged the nonstandard speaker as white. Although this study

provided evidence that listeners' attitudes about speech styles form

at an early age, the study had some flaws. First, neither sex nor

race of subjects was controlled. Furthermore, although the speech

samples were similar in length and title ("My daily routine")--they

were extemporaneous and so the specific monologue content was also

uncontrolled and may have influenced the subjects' ratings.

James (1976) did a study involving 43 black first graders as

subjects in which she investigated the perceived differences between

black English and standard English content, while holding style con-

stant. The aim was to investigate whether black children perceive a

difference between black English and standard English and, if they

did, was the difference identified in terms of style, referring to

how something was said (e.g. intonation) or in terms of what was said

(e.g. lexicon), or both? Results indicated that the black children

in this study were able to perceive a difference between black

English and standard English style more frequently than between

black English and standard English content. The author suggested

that, since the subjects were able to perceive differences between

style more frequently than content, that phonology is more impor-

tant than syntax, at least in the discrimination of language


In a frequently cited article, Tucker and Lambert (1969) inves-

tigated listeners' reactions to six American English dialect groups:

Network (white), Educated White SoLthern (white), Educated Black

Southern (black), Mississippi (black), Howard University (black), New

York (black). Four representative speakers for each speech style were

chosen by the experimenters. Three groups of college students lis-

tened to recordings of speech samples and evaluated certain personality

characteristics of the speakers on an adjective checklist. One group

of listeners was northern white, another southern white, and a third

was southern black. It was found that both northern white and southern

black groups rated the Network speakers most favorably and the Edu-

cated Black Southern speakers next. The southern white students also

evaluated the Network speakers most favorably. However, in contrast

to the other two groups, the white southern listeners rated the

Educated White Southern style second most favorably. On the other

hand, both groups of white judges rated the Mississippi speakers least

favorably, while the black judges rated the Educated White Southern

speakers least favorably. These results were noteworthy in more than

one respect. First, all subjects were able to reliably differentiate

the dialect groups and they clearly favored the Network style of

English over the other styles. A second revealing finding involved

the different perspectives of the black and white subjects regarding

the least favorable of the dialects. It seemed, the authors con-

cluded, that affectively-toned attitudes were responsible for these

differential ratings. It was obvious that some speech styles were

pleasing to one ethnic group and not to others.

Two recent doctoral dissertations in the field of speech and

language have examined the standard and nonstandard English dialects

of black speakers. In one dissertation, Thomas (1978) studied the

effects of standard and nonstandard English speech styles of a black

male speaker. The speaker delivered audiovisual monologues to male

and female undergraduates and they rated the speaker on trustworthi-

ness and expertness. The subjects were also tested for message

retention. Subjects perceived greater speaker trustworthiness when

the message was delivered in standard English. No significant differ-

ences were found in differences in the analysis of message retention.

It appeared that the race and sex of subjects in this study were not

adequately controlled.

In another recent doctoral dissertation (Yost, 1978) a bidialec-

tical black speaker delivered one informational tape in black English

and another in standard English to a group of black and white college

and business school students. Greater comprehension of content

occurred with both black and white subjects who heard the standard

English version of the tape. In contrast to a finding by Thomas

(1978), no differences were found in the dependent measures due to

subject sex. Greater credibility of the speaker was perceived by

both black and white subjects when the speaker spoke standard English.

This study provided an interesting contrasting finding to some of

the previously mentioned results. There seemed to be no difference

between black and white subjects on any ratings. It was hard to

ascertain whether this was due to the subject sample, who were all

college students, or to some aspect of the speaker.

In a study not involving black speakers, Anisfeld, Bogo and

Lambert (1962) investigated the attitudes held by listeners toward

speakers of Jewish accented English. Using the matched guise tech-

nique described earlier, four speakers each made two short tape

recordings of a passage. One reading was in unaccented English and

a second reading was in Jewish Accented English. The four speakers

used were selected by a panel of judges from a larger group of

speakers because of their ability to speak both accents in an

uncaricatured fashion. The results indicated the accented guises

were comparatively devalued on height, good looks, and leadership by

both Jewish (64) and non-Jewish (114) subjects. Further, this result

was independent of the subjects' perceptions of the voice as Jewish

or non-Jewish. The non-Jewish subjects did not consider the

accented guise as more favorable on any trait while the Jewish sub-

jects evaluated the accented guise more favorably. Like previously

cited studies, the results of this study indicated ethnicity and

language are intimately connected and listeners are prone to make

judgments based upon a combination of their own cultural identity

and the speaker's.

In still another non-black, but accented English language study,

the French-Canadian accent was investigated. Webster and Kramer

(1968) examined listeners' ratings of matched guises of French-

Canadian accented English and unaccented English. Rather than the

typical introductory psychology student subject population normally

used in these investigations, this study used 30 English speaking

students who were enrolled in an evening extension course in an

attempt to tap a more diverse group. Results revealed the

French-Canadian accent was evaluated more unfavorably than the

unaccented speech. The authors ascribed this to the presence of

community-wide stereotypes regarding attitudes toward French-


In an impressive series of investigations, Markel and associates

have investigated the relationship between voice quality and various

personality variables (Markel, Meisels & Houck, 1964; Markel &

Roblin, 1965; Markel, 1969; Costanzo, Markel & Costanzo, 1969; Markel,

Prebor & Brandt, 1972; Markel, Phillis, Vargas & Howard, 1972). These

studies demonstrated several vocal qualities such as pitch, loudness

and tempo were all important cues for listeners who were evaluating

speakers. For example, Costanzo, Markel and Costanzo (1969) found

that peak pitch (high levels of pitch) was associated with the emo-

tion of grief; peak loudness was associated with anger and contempt;

and peak tempo was associated with indifference.

An additional study examined the relationship between dialect

and personality. Markel, Eisler and Reese (1967) examined the

effect of regional dialect on judgment of personality from voice.

College students in Buffalo, New York evaluated one group of speakers

who had a regional Buffalo dialect and another group who had a New

York City dialect. Ratings on the Osgood Semantic Differential

revealed New York City speakers were rated higher by listeners on

the evaluative scale and Buffalo speakers were rated higher on the

activity scale. The authors interpreted the results as an indication

of the role of regional dialect in eliciting stereotypes toward

certain speakers.

Attitudes Toward Dialect Speakers
of Non-English Languages

Research of listeners' attitudes toward various speech styles

has not been limited to English language studies. In Quebec,

Canada, cultural and ethnic identities have been intimately connected

to the languages of French and English. All the major social institu-

tions have been affected by the cultural and sociolinguistic struggles

which have been occurring in the region. Lambert, Frankel and Tucker

(1966) and Lambert, Hodgson and Fillenbaum (1960) have looked at the

different attitudes that listeners have toward French and English.

In these two studies, the matched guise technique described previously

was used to examine two contrasting languages. Perfectly bilingual

speakers were recruited and passages were recorded in both French and

English. In both studies, the subjects doing the ratings were all

students. Ethnically, some were French-Canadian and some were

English-Canadian; and their language styles reflected this background.

In the Lambert et al. (1960) study, the most significant finding was

that the French-Canadian students showed a marked bias in favor of

French spoken with an English-Canadian accent--as opposed to their

native or French-Canadian style. The authors entertained some

possible explanations for this relatively unfavorable view that stu-

dents had of their own ethnic-linguistic group. One possibility was

that the students came from homes where it was felt that increased

family status depended partly upon taking the English-Canadian style

of speech as a model to be emulated. Additionally, it was found

that French-Canadian students most biased in favor of French spoken

with an English-Canadian accent were those from upper and upper-

middle class homes.

In the Lambert et al. (1960) study it was found that English

speaking subjects showed more favorableness to members of their own

linguistic group. Additionally, it was found that French subjects

not only evaluated the English guises more favorably than French

guises, but their evaluations of French guises were reliably less

favorable than those of the English subjects. The authors inter-

preted this as evidence for a minority group reactions on the part

of the French sample.

In Tel Aviv, Israel, Lambert, Anisfeld, and Yeni-Komshian (1965)

investigated attitudes toward two different dialects of standard

Hebrew and one dialect of Arabic. The Jewish subjects' ratings of

the matched guise comparisons of Hebrew and Arabic revealed Jewish

subjects held stereotyped views about Arabs. They (Arabs) were con-

sidered less humorous, less friendly, less honest and less desirable

as friends. The authors reported puzzlement over the finding that

Arabs were considered as more wealthy. This did not fit the social

facts of Arab life in Israel, but it might have reflected a wider

view on the part of the Jewish subjects: which in this case may

have been an interpretation of wealth in terms of land possession

and the Jews saw their own as small in comparison to the enormity

of the Arab world. The patterns of findings of the Arab subjects

suggested they considered members of the Jewish group less capable,

less intelligent, less dependable and less desirable for marriage

than people in their own group. Mutual distrust of this sort would

certainly seem to contribute to restricted social interaction since

members of neither group would be likely to initiate a friendly

overture if they expected the other to react in a relatively dis-

honest, unfriendly and selfish manner.

Finally, this section on non-English dialects describes the

results of a study carried out in a rural section of southern

Italy. Cremona and Bates (1977) looked at the attitudes of elementary

school children toward their own dialect, from the beginning of ele-

mentary school through grade five. The authors found a clear ten-

dency for the children to devalue their local dialect. Admittedly,

this has been one of the objectives of early schooling in Italy.

However, the authors found evidence of negative attitudes which had

already been established by many children before they had even begun

the school program. The authors suggested perhaps early exposure to

standard Italian on television had contributed to this process. It

was also suggested that some parents, even though they spoke the

dialect variety of Italian in their homes, may have made an active

effort to "correct" certain dialect features in their children's


The studies in this section on non-English dialect attitudes

have been described to indicate the universal basis of this

phenomenon. Although this investigation will be examining American-

English variations, it shares theoretical and practical aspects with

many of the previously mentioned studies.

Language Variables in Clinical
and Counseling Settings

Language variables have been examined in a number of related

contexts all having at least some relevance to counseling. This

section is divided into three interrelated parts. The first part

describes research which has investigated psychotherapeutic and

conversational discourse from primarily a psycholinguistic per-

spective. The second part discusses articles concerned primarily

with the therapeutic process, but pays close attention to language

factors. The third section discusses a few studies which have

addressed language issues in educational settings.

Formal Psycholinguistic Investigations of
Counseling and Conversational Discourse

Pittenger, Hockett and Daheny (1960) presented a detailed (word

for word) analysis of the first five minutes of an initial psychiatric

interview of an outpatient client. The authors, two psychiatrists

and one linguist, aimed at studying (1) purely linguistic aspects

such as phonemes, (2) paralinguistic variables such as inhalation or

exhalation and (3) the meaning of what was said. This text was an

ambitious undertaking, yet seemed to be limited in the light it has

shed on our knowledge of counseling per se. The heavy linguistic

focus of the analysis has had the effect of reducing the importance

of the therapeutic implications to a secondary, or minor role.

In a similar analysis, Gottschalk (1961) outlined a series of

wide ranging analyses of two psychotherapeutic interviews. The

analysis of language factors in this study was not of a formal psy-

cholinguistic nature, but rather focused on the following aspects

of the interviews: therapist activity, dyadic analysis, expressive

aspects of the patient's speech, and some physiological correlates

of psycholinguistic patterns. Although this investigation exceeded

the relatively pure orientations of the Pittenger et al. (1960) work,

this text seems to have gone too far in the opposite direction. For

example, the section describing physiological correlates of psycho-

linguistic patterns, while interesting, provides information which

is of little relevance to the counseling session itself.

A third text, by Jaffe and Feldstein (1970), investigated the

"rhythms" that characterize dyadic dialogue. This work did not

explicitly address psychotherapeutic interactions, although it did

not exclude them as possibly falling under the generic classification

of dialogue. The orientation of this work was primarily a technical,

information oriented one. Among other topics, it focused on stochas-

tic models of the time patterns of dialogue, speaker switches and

information theory.

Finally Labov and Fanshel (1977) have systematically examined

a fifteen minute segment of a third therapy session of an anorexic

client. Since Labov has made significant contributions to the field

of sociolinguistics, he naturally attempted to incorporate this per-

spective into the dialogue analysis. While the authors did find some

sociolinguistic variables played a role in the therapeutic session,

they turned out to be comparatively minor factors in the overall com-

municative patterns. The authors found that the social dialects used

by the client and therapist had reached a fairly stable state at the

time their analysis was made. This was attributed to the familiarity

which the client and therapist had previously established. On the

other hand, the intonational patterns proved to be of crucial interest

in defining fields of discourse, in identifying patterns of communica-

tion and in clarifying contradictions which would have been unresolved

if only the word meanings themselves had been used.

Overall, these four texts are interesting, but esoteric. They

reveal quite a lot about communication from a technical, linguistic-

analytical perspective. They do not, however, reveal much about

counseling as a special form of communication or how it may be

different from general discourse.

Language Factors in the Counseling Process

Many writers have looked at the different roles that language

factors play in the counseling process. This line of research within

the general area of counseling is relatively new. As such, the work

to date has been characterized by a fragmentary and exploratory

nature. The following descriptions reflect this newness.

Several years ago Bernstein (1964) suggested that because lower-

class clients had "restricted" language codes (i.e. were limited in

terms of what they could elaborate) they were not likely to benefit

as much from therapy as more verbally sophisticated clients. More

recently, Conville and Ivey (1974) have attempted to refute this

notion by describing, in a more enlightened fashion, the appropriate-

ness and value of considering sociolinguistic aspects as part of the

counseling process. Many empirical investigations have focused on

specific language aspects of the counseling process.

Meara, Shannon and Pepinsky (1979) presented data generated from

a computer assisted language analysis system (CALAS) which examined

differences between the three theoretical orientations of counseling

represented by Carl Rogers, Fritz Perls and Albert Ellis; who each

used a very different approach to counseling. A film they made in

1966 was used for the data collection. In the film, each of the

therapists saw a client (Gloria) separately for about twenty minutes

and they engaged in therapeutic discourse. The date were analyzed

according to four measures of what the authors called stylistic

complexity: (1) number of sentences (SEN), (2) average sentence

length (ASL). (3) average block length (ABL), and (4) average clause

depth (ACD). The theoretical underpinning to which these measures

claim relevancy concerned the notion of informative display. The

measurable units of language were defined as part of displayed

information and indicated the complexity of the conceptual levels on

which client and counselor were operating. Thus, studying speech

patterns may be helpful in matching clients and counselors.

The authors found that the four measures demonstrated the

following: (a) Different patterns among the therapists--this was

consistent with the fact that each therapist represented a differ-

ent theoretical counseling position. (b) Differences for the client

(in terms of her SEN, ASL, ABL and ACD)--depending upon which coun-

selor was interviewing her. (c) Evidence for concerted action--

concerted action means activities people do together or in concert.

The assumption is without this concerted action in counseling there

is likely to be little change occurring for the client. Further,

without concerted action, there is likely to be poor communication

taking place.

The implications of this line of research involve some interest-

ing considerations. Measures of linguistic structure might be help-

ful in predicting counseling outcomes or in training counselors to be

aware of the effect of speech on their clients. Counseling is a

highly verbal activity and there is significant information conveyed

in the content of the message as well as in the style.

Sprafkin (1970) examined the extent to which clients would

change their confidence in the meanings of certain words depending

on the (1) level of expertness of a counselor (high or low) who

would (2) either advocate change in certain word meanings or advocate

no change. The author expected to find an interaction between level

of counselor expertness and change of word meaning by the client--

i.e.,the more expert the counselor, the more likely the client would

be to change a word meaning to agree with the counselor and for the

less expert counselor, the less likely the client would be to agree

with the advocated meaning. The findings did not entirely support

the hypotheses. Regardless of the counselor's level of expertise,

clients tended to achieve agreement with the counselor. Some

explanations were offered for these findings. Even those subjects

who assigned low ratings to their counselors may still have perceived

them to be much more expert than the subjects considered themselves.

Age differences between counselors and clients may have contributed

to perceptions of counselors, even in the low expert condition, as

relatively more expert and knowledgeable than subjects. Finally, in

both conditions, the label of "counselor" may have involved certain

social expectations of subjects.

Schumacher, Banikiotes and Banikiotes (1972) looked at language

compatibility of (30 white) counselors and (30 white and 30 black)

high school students in terms of how well they understood words often

used by the other group. One hundred common words corresponding to

the language of the counseling profession were chosen and from this

list twenty words were finally selected as words most often used by

the counselors. Similarly, and with the help of an Afro-American

dictionary, a list of the twenty most common words used by the black

students was constructed. Each group was tested on how well they

understood the common words of the other group. Results were as

follows for the black vocabulary words: black students scored

higher than white students who scored higher than white counselors.

For the (white) counselor vocabulary words the opposite was found.

The counselors scored significantly better than the white students

who scored better than the black students. The results indicated

language compatibility between black students and white counselors

was low in terms of this "vocabulary test." The implications of

these results involve several aspects of the counseling relationship.

Most importantly, it suggested the establishment of a strong working

relationship between a black client and a white counselor may be

difficult to obtain if there exists a high degree of misunderstanding

between the two.

Stein and Walker (1980) conducted a study to determine the

effects of (black) client and counselor race upon certain counseling

conditions. A black actress was recruited from a university drama

department who could speak both black English and standard English.

Speaking each dialect in an identical role playing situation, she

was paired with a black and white (acting) counselor to obtain each

of four conditions. Subjects were forty experienced counselors who

each viewed a videotape of only one session. A counselor rating form

was used by the subjects to rate the counselor on levels of trust-

worthiness, attractiveness and expertness. A second questionnaire

assessed the type of therapy the subject thought would be most

appropriately used by the counselor in the film.

Results revealed that the subjects perceived the counselor as

more trustworthy and more expert when the client spoke black

English. White subjects anticipated the black counselor would

select a behaviorally oriented therapy for a client using black

dialect and an insight oriented therapy for the same client speaking

standard English.

Stein (Note 1) conducted an experiment to determine whether the

speech a counselor used influenced the way potential clients per-

ceived the counselor. Two counselors, one black and one white, each

made two short videotape monologues in which they described their

philosophy of counseling. In each tape, each counselor used a

different voice pattern. The'subjects were sixteen black and sixteen

white students enrolled in an introductory psychology class at a large

southern state university. Dependent measures were the levels of

perceived trustworthiness, expertness and attractiveness as measured

by the Counselor Rating Form (Barak & LaCrosse, 1975). Results

revealed the following: black subjects rated the black counselor

higher when he spoke black English and the white counselor higher

when he spoke unaccented English; white subjects rated the black

counselor higher when he spoke standard English and the white coun-

selor higher when he spoke unaccented English. The results of this

experiment approached but did not reach statistical significance

because of a main effect due to order of tape presentation.

A recent doctoral dissertation (Guy, 1978) examined whether

dialect differences between a therapist and a client would

significantly affect the clinical evaluations of the therapist.

Eighty clinical psychology graduate students listened to one of two

tape recordings, in either black English or standard English, and

then made clinical evaluations of the client they heard. Black

therapists showed a positive bias for the black English speaking

client while white therapists showed a positive bias for the

standard English speaking clients.

Language Factors in the Classroom

The articles reviewed below indicate the importance of language

factors in a field closely related to counseling. The educational

system in this country often mirrors the dilemmas found elsewhere in

this society. Sociolinguistic conflicts are no exception.

McGinnis and Smitherman (1978) described some of the most preva-

lent barriers faced by black children in most of the schools in our

country today. One of the most striking barriers described is the

apparent misunderstanding of the function of language on the part of

most teachers. Apparently many teachers equate the student's acqui-

sition of a concept with the correct word usage which describes the

concept. This is a violation of a basic semantic principle: the

word is not the thing. The authors described the difference between

correct/proper speech (nothing more than social convention and con-

formity), linguistic competence (ability to use and understand rules

of one's native language) and intelligence. Every speaker of a

language naturally acquires linguistic competence as they learn

their native language. The black child encounters difficulty as

school begins and a new environment is encountered. Schools with

predominantly black populations are environments where two language

systems meet. This results in linguistic-cultural conflict because

the socially dominant language of the school system interfaces with

the students' mode of expression. Since communication is funda-

mental to the entire learning process, the conflict has serious

negative consequences for the academic achievement of black students.

McGinnis and Smitherman (1978) suggested a two-pronged attack on the

problems described above.

First, they suggested teachers should abandon the speech-

correctionist approach for a strategy aimed at expanding the stu-

dents' linguistic repertoire. Language experimentation and stimula-

tion should be encouraged in a non-threatening way, such as role

playing. Second, with respect to school policy, educational insti-

tutions should implement mechanisms which attempt to restructure

teacher and student attitudes about language. Teacher language bias

similar to the type described above has been investigated in some

empirical investigations.

Granger, Mathews, Quay and Verner (1977) examined teacher

evaluations of speech performances of lower and middle socioeconomic

(SES) black and white children. Functionally equivalent speech

samples were obtained from the children as they described certain

pictures. The descriptions were equivalent on features described,

but the speech patterns used in the descriptions varied across the

sample of children. The raters of the taped samples were fifty-six

female preschool and primary grade teachers (9 black and 47 white)

who were all enrolled in a graduate education program. Differences

in ratings were found both between races (i.e. black and white,

regardless of SES) and between SES groups (i.e. middle and lower

SES, regardless of race). Black and white middle SES speakers were

rated higher than black and white lower SES speakers. Black and

white lower SES speakers were not rated differently from each other.

However, the middle SES black speakers were rated lower than the

middle SES white speakers. So, bias appeared to be operating both

in terms of race and class. The findings suggested teachers were

attending less to what a child said and more to how he said it. The

authors found the most striking differences between tapes was phono-

logical--e.g. voiceless th becoming f in the final position. Authors

suggested that teachers should be trained to develop an understanding

of their attitudes toward a child's speech and how these attitudes

may hamper their ability to evaluate a child objectively. There

were a few methodological problems in this study. The sex of the

children was not controlled, nor was there any mention of relia-

bility of the raters. Thus, generalizability of the results must

be made with caution.

DeMeis and Turner (1978) assessed the effects of a student's

race, dialect and physical attractiveness on teachers' evaluations

of the students. Sixty-eight white teachers listened to students'

speech samples and rated them in terms of personality, quality of

speech and current and future abilities. Results indicated black

students, black English speaking students and low-attractive

students were rated lower than the others. Results also indicated

the teachers' ratings in the different areas were highly consistent

with one another. The authors discussed the implications of this

last finding in terms of the disastrous consequences these attitudes


could have on student performance. It was described as an illustra-

tion of how teachers unfairly attribute children' failures to their

race and dialect rather than to their actual performance.

A recent doctoral dissertation (Hines, 1978) examined the effects

of ethnic group membership, age, sex, and educational context on recall

of information when stories were told in either black English or

standard English to both black and white elementary school children.

The most significant conclusion drawn by the author based on the

results of the study was black children performed better on recall

tasks if materials were presented in black English.


The major purpose of this investigation was to study the effects

of counselor race and dialect upon the levels of trustworthiness,

expertness and attractiveness perceived by racially different sub-

jects. And further, the purpose was to study the effects of counselor

race and counselor dialect on the amount of information which could be

accurately recalled by a client after three contacts with a particular


This chapter is divided into several sections. First, the

Population and Sample section describes the population setting of the

experiment as well as the particular sample of clients who were the

subjects in this study. The Experimental Hypotheses section

describes the questions under study and presents the experimental

hypotheses of this experiment in a null form. The Procedure and

Methodology section describes how the video tape monologues were

prepared and the instructions that were given to the subjects. Next,

in the Instrumentation section, the Counselor Rating Form is dis-

cussed and the Recall Questionnaire is introduced as an instrument

designed specifically for this investigation. The Design and Data

Analysis section outlines the experimental nature of this research

and the statistical tests that were used. Lastly, Limitations are

discussed in a separate section.

Population and Sample


The sample for this study was 80 patients from a total popula-

tion of approximately 275 patients at the Alcohol Rehabilitation

Unit (ARU) of the East Orange, New Jersey, Veterans Administration

Hospital. The ARU, headed by a counseling psychologist, is a

psychodynamically oriented treatment program for alcoholic patients.

There are approximately thirty-five inpatients, two hundred out-

patients and forty ambulatory care patients at any given time. While

all patients are alcoholic, the group therapy and individual treat-

ment they receive is designed for the person as a whole. Ongoing

research by staff is an integral part of the program.

This study utilized only those patients from the ARU who were

in the inpatient portion of the program. The inpatient program is

an intensive, eight week experience during which each patient par-

ticipates in daily group therapy and individual counseling when

appropriate. A recent survey of previous participants of this

program revealed the following demographic characteristics: They

have all been males. Approximately fifty percent have been white,

forty percent have been black and ten percent other. The range in

age has extended from about 25-65 years with a mean of 38 years.

The range of educational levels has extended from ninth grade to

college graduate and at times Ph.D.s have participated in the pro-

gram. Approximately thirty-three percent of the patients were

married at the time of participation, fifty percent were divorced

or separated, and seventeen percent were never married.

The eighty patients who participated in this study were a repre-

sentative sample of the previously treated alcoholic clients. Sub-

jects in this study had an average age of 43 years; the range extend-

ing from 22 to 68. The sample was comprised of individuals who each

had at least a high school education--including one medical doctor,

a registered nurse and a Ph.D. level chemist. Although no descriptive

socioeconomic data were gathered specifically for the subjects in

this study, general intake interview information indicated that, for

the most part, they could be described as middle or lower-middle

class. Since patients were living in the hospital, they often had

free time to participate in activities of their choice. In the past,

hundreds of patients agreed to participate in research. In this

study, each of the newly admitted patients who entered the program

every week for three months were asked to participate. The forty

black and forty white subjects in this study were simply asked soon

after admission whether they would like to voluntarily participate in

some research which involved "watching a video tape of a counselor

and then filling out a short form to rate the counselor." Those

agreeing to participate were randomly assigned to one of the four

different counselor speech style conditions. Not one client who was

asked to volunteer declined to participate. During the data collec-

tion phase of the experiment, three clients prematurely left the

Alcohol Rehabilitation Program and data from their ratings were

eliminated from the study.

Experimental Hypotheses

In general, the purpose of this investigation was to study how,

and to what extent, a client was influenced by the combination of

both the counselor's race and speech style. Referring to Figure 1,

the experimental hypotheses of this experiment are expressed below

in their null form:

(1) No difference will exist among the subjects' ratings on

the Counselor Rating Form for any of the four counselor

conditions as follows:

--White Counselor/Unaccented English (A+E)
--White Counselor/Accented English (B+F)
--Black Counselor/Unaccented English (C+G)
--Black Counselor/Accented English (D+H)

(2) No difference will exist between the black subjects' ratings

and the white subjects' ratings on the Counselor Rating Form.

(3) No difference will exist between black and white subjects'

perceptions of a given counselor for either speech style

spoken by that counselor. Referring to Figure 1:

For White Subjects
--White Counselor/Unaccented English (A)
--White Counselor/Accented English (B)
--Black Counselor/Unaccented English (C)
--Black Counselor/Accented English (D)

For Black Subjects
--White Counselor/Unaccented English (E)
--White Counselor/Accented English (F)
--Black Counselor/Unaccented English (G)
--Black Counselor/Accented English (H)

(4) No difference will be found on the subjects' ratings on

the Counselor Rating Form over time. (Contact 1 versus

Contact 2 versus Contact 3.)

(5) No difference will exist among subjects' scores on the

Recall Questionnaire for any of the different counselor


All hypotheses were tested at the .05 level of significance.

For a corresponding list of the specific effects which were due to

these hypotheses, refer to the section of this chapter on Design and

Data Analysis.

Procedure and Methodology

This study involved a total of eighty subjects who were divided

into eight groups of ten each. Half of the subjects were black (40)

and half were white (40). (See Figure 1.)

Each subject viewed a total of three video taped monologues

delivered by one of two counselors. Further, the counselor delivered

the monologue in one of two possible speech styles. The black coun-

selor used either Black English (accented) or Conventional English

(unaccented). The white counselor used either Southern English

(accented) or Network English (unaccented). Following each monologue

presentation, the subject rated the counselor on the Counselor Rating

Form. This rating form indicated the levels of trustworthiness,

expertness and attractiveness perceived by the subjects. Each sub-

ject saw only one counselor, speaking the same speech style, at each

of the three contacts. At each contact the counselor discussed one

of three possible topics. After subjects viewed all three monologues

Subject Counselor First Second Third i Recall
Race Condition Contact Contact Contact Measure

h Wite
Counselor 10
No Accent

}H White
Counselor 10

Black I
Counselor 10
No Accent
T (C)

Counselor 10 -
F Accent

S Counselor 10
No Accent t

L white
Counselor 10

A (F)
Counselor 10
No Accent
C (G)

Counselor 10

Figure 1. The mixed design of this experiment is illustrated (Myers,
1966). Note that the ten (10) subjects in each cell remained with
the same counselor condition over all three contacts.

and completed their three Counselor Rating Forms, they completed the

Recall Questionnaire. This instrument determined the amount of

information contained in the monologues which was accurately recalled

by the subjects.

Construction of the Video Tapes

Two graduate level and experienced actors in the drama depart-

ment at Rutgers University, Newark, played the roles of counselors

in the video tapes. The black actor was naturally fluent in two

different speech styles. The white actor had experience in roles

in which he had to use a southern accent. In addition, both actors

had formal training in language and voice classes. Both counselors

rehearsed the scripts several times before the tapings were made.

This insured that they each became comfortable with both styles of

speech and totally familiar with the content. During the tapings,

the tapescripts were posted off camera as an additional aid to them,

if they needed it. Tapescripts of the monologues can be found in

Appendices B, C and D. Samples of all four speech styles, trans-

cribed in the International Phonetic Alphabet, can be found in

Appendices G, H, I and J.

Development of the Monologues

The content selected for inclusion in each of the three mono-

logues was chosen from material described by Egan (1975). The three

topics discussed by the counselors were: (1) The importance of

being prepared for counseling; (2) the importance of mutual respect

in counseling and (3) the issue of client-counselor compatibility.

The actual content of the monologues is available in Appendices B,

C, and D.

Just prior to the subjects' viewing of the first monologue, they

viewed a video tape displaying the written instructions. This video

presentation was accompanied by a recorded male voice simultaneously

reading the instructions. These instructions are found in Appendix A.

The Recall Questionnaire was used to assess the amount of information

retained by subjects after they had seen all three monologues. This

questionnaire is described in the Instrumentation section of this


Presentation of the Monologues

The subjects saw the complete set of three monologues over the

course of five week days. The order of viewing was counterbalanced

to control for effects of content. One topic (i.e. monologue) was

presented on day one (Monday), another on day three (Wednesday) and

a third on day five (Friday). The Recall Questionnaire was given on

day eight (Monday).


The Counselor Rating Form

The Counselor Rating Form (CRF) was developed by Azy Barak and

Michael LaCrosse during 1975-76 at The Ohio State University. The

CRF was developed to measure the perceived counselor dimensions of

trustworthiness, expertness and attractiveness. The CRF itself con-

sists of thirty-six seven-point bipolar adjectives (see Appendix E).

Each dimension of perceived counselor behavior is thus measured by

twelve items. Sometimes each of the three subsets are referred to

as scales. The possible range of scores for any dimension extends

from a minimum of 12 to a maximum of 84. The reliability coeffi-

cients reported by LaCrosse and Barak (1975) are as follows: .87 for

expertness, .85 for attractiveness, and .91 for trustworthiness.

Barak and Dell (1977) reported that the CRF has also been proven

capable of detecting perceived differences among counselors on the

three dimensions as well as being sensitive to differences among the

three dimensions for a given counselor. Detecting the presence of

these qualities in a counselor is one thing, and assuming that they

are important in the process of counseling is a different question.

Several studies (e.g. Strong & Dixon, 1971; Strong & Schmidt, 1970)

show that the greater the perceived levels of the three dimensions,

the more likely a subject is to initiate change. The process of

change and influence is intimately connected to the development of

the CRF. In fact, the CRF was, in large part, inspired by the work

of Stanley Strong's (1968) theory of counseling as an interpersonal

influence process.

Strong's (1968) theory of counseling as an interpersonal influ-

ence process grew out of opinion-change research of social psychology.

Attributes of the communicator, in both opinion-change research and

in counseling paradigms are crucial. In opinion-change, a communi-

cator attempts to move the listeners in a predetermined direction.

A counselor tries to influence the client to reach the goals of coun-

seling. Strong (1968) describes some of the dimensions in operational

terms. Expertness involves objective evidence of advanced training

such as diplomas. Or it can be displayed behaviorally, with rational

and confident presentations by the communicator. Finally, the

communicator may be considered an expert based on his or her reputa-

tion. Trustworthiness involves the communicator's reputation for

honesty and good will; social role, such as doctor or teacher; open-

ness and sincerity; and perceived absence of motivation for personal

gain. Attractiveness of the communicator can be achieved by dis-

playing compatibility with the listener, by reassuring the listener

that the communicator will be likable, by the presence of background

material similar to that of the listener. Strong (1968) describes

an aspect of his theory which is especially crucial to this investi-


. verbal communication is the main technique used by
an opinion changer in influencing his audience; verbal
communication is also the counselor's main means of influ-
encing the client. For both, these communications present
opinions or conceptions different than or discrepant from
the opinions or conceptions of the audience or client.
Finally, characteristics of the communicator as perceived
by the audience, characteristics of the audience, and
characteristics of the communication affect the success
of the influence attempts. (p. 215)

The importance of this point to the present investigation is obvious.

Since the verbal communication is the main tool of the counselor,

perceived trustworthiness, expertness and attractiveness may have

been significantly influenced by the language or speech style, which

was as much a part of the communicator as it was of the communication


In this investigation the CRF was used by the subjects to rate

the counselor after each viewing. The CRF was broken down into three

equivalent parts, so that the subject did not see any of the items on

previous days. Since there were twelve items for each dimension,

four items were randomly selected from each scale to construct a

"one-third" CRF consisting of twelve items. This "one-third" CRF

yielded a measure for each dimension ranging from a minimum of 4 to

a maximum of 28. At the end of the three viewings the three sub-

scores were summed to yield an overall score. The "one-third"

scores served as indicators of changes in subjects' perceptions of

the counselor levels of trustworthiness, expertness and attractive-

ness over time. The intercorrelations between items within each of

the three scales have undergone a preliminary analysis on the Hebrew

language version of the CRF. Results indicated that these within

scale correlations were sufficiently high to permit using the CRF in

three equivalent parts (LaCrosse, Note 3).

The Recall Questionnaire

The Recall Questionnaire (see Appendix F) was constructed to

test how much of the monologue content could be accurately recalled

by the subjects. The questionnaire was in a multiple-choice format

and there were a total of fifteen items. The Recall Questionnaire

was developed in a four-step process.

First, a pool of twenty-eight multiple-choice items were con-

structed that corresponded to the content of the monologues. This

initial pool of items and the monologue contents were inspected by

a panel of experts consisting of five Ph.D. counselors and psycholo-

gists. These judges rated the quality of the questions with respect

to how well they corresponded to the monologue content and how well

they were written (content and face validity). This process resulted

in the elimination of four items which clearly emerged as the very


The second step consisted of item analysis based on a sample of

fifty scores obtained from a group of college undergraduates. Twenty-

five scores were obtained from subjects who had never been exposed to

the monologue content. Another twenty-five scores were obtained from

subjects who had heard tape recordings of the monologues. An index

of discrimination (Sax, 1974) was constructed based on these before

and after scores. As a result of this analysis, eight additional

items were eliminated because they either did not discriminate at

all or they discriminated in a negative direction.

Third, biserial correlation coefficients (Anastasi, 1976;

Greene, 1952) were computed for the remaining sixteen items to insure

that the individual items were measuring the same thing that the test

as a whole purported to measure. Thirteen of these items were highly

significant (p< .001) and two were marginally significant (p< .12).

These fifteen items were retained and a sixteenth was discarded

because it was negatively correlated with the other items.

Finally, a Kuder-Richardson 20 (Cronbach, 1970) reliability was

computed for the fifty scores on the final fifteen items. The Kuder-

Richardson 20 reliability coefficient was computed to be .86.

The Recall Questionnaire has been assessed for its readibility

level by the Fry Readibility Formula (Fry, 1972). It has been deter-

mined that the Recall Questionnaire has a readibility level placing

it at the ninth grade level.

Design and Data Analysis

The type of design used in this experiment has been described

by Myers (1966) as a mixed design (see Figure 1). A mixed design

is one in which measures are repeated across levels of some, but not

all, factors. In this experiment this means that the Counselor

Rating Form score was obtained at each of the three times a subject

had contact with a counselor, but no subject saw more than one level

of counselor condition. Subjects viewed the same counselor, speaking

the same dialect, at each contact. Of course, at each contact a

different topic was presented by the counselor.

The specific effects due to the hypotheses (see Figure 1 and

Purpose and Experimental Hypotheses) were analyzed as follows:

(1) Main effect of condition. Results were collapsed across

subject race and the following comparisons were made:

(A+E) vs (B+F) These compared dialects of different
(C+G) vs (D+H) counselor races, regardless of subject

(2) Main effect in overall ANOVA of subject race.

(3) Linear Contrasts (actually, a series of t-tests) were made:

A vs B
SSubject race = white
C vs D Were any of these four
E vs F Subject race = black comparisons different?
G vs H

(4) Interaction of Contact X Counselor Condition or interaction

of Contact X Counselor Condition X Subject Race.

(5) All of the above, using the Recall Questionnaire score.

(Except #2.)

Figure 2, a flow chart, illustrates the sequence of steps that

was used to answer the above questions and make the appropriate com-




Do contrasts in
#1 for main ef-
fects and do co
trasts in #2 fo
main effect of
Subject Race.

highest order int

S Contact x Subject
Condition signifi

Collapse across con-
tacts and look at
Subject Race x Couns-
elor Condition. Any
significance? (yes &

Do linear
n- in #3.

fraction. Was yes
Race x Counselor I
cant? (no)

Look at Subject
SRace x Counselor
Condition at each

Main effect 1 DC linear
of Counselor I contrasts
Condition and in #3.
contrasts in
#1 Main ef-
fect of Sub-
ject Race in

Figure 2. Flow chart indicating the sequence of data analysis. The
#1, #2, and #3 comparisons refer to procedures which are described
in the Design and Data Analysis section. Double lined arrows indic-
ate direction of decision.


There were three main limitations to this investigation. In

some respects they were all related to one another. (1) One restric-

tion of the study involved the analogue nature of the design.

Although real hospital clients were used as subjects, the contact

with the experimental counselor was somewhat artificial since com-

munication occurred in only one direction. (2) Related to this was

the fact that the communication was a video tape presentation, rather

than a counselor in the flesh. Although the counselor was a real

person,some artificiality was inevitable with the use of video

taping techniques. (3) The third limitation involved the limited

generalizations that can be made because of the restricted nature of

the subject population. All subjects were male and all were veterans

in the same treatment program. In addition, both counselors were


Although the limitations mentioned above all imposed some arti-

ficiality on the investigation, there were also advantages to making

these sacrifices. The main advantage was the amount of control gained

over the very subtle process under study. The video tape technique

allowed manipulation of the language variable, but in a fashion that

permitted measurements of variation to be made which would have been

difficult to obtain with a live counselor. Similarly, even though

the sample was not representative of the total client population in

counseling, the process investigated was assumed to have universal

importance. As such, limitation (3) can actually be seen more as a

measure of control than as a severe limitation of the conclusions to

be drawn from the results.


The major questions of this investigation focused on whether the

vocal accent and race of a particular counselor were important factors

in racially different clients' judgements of the counselor's trust-

worthiness, expertness, and attractiveness. The results, which com-

prise this chapter, generally did not support the position that these

were important variables. Referring to the five hypotheses as they

were presented in the Experimental Hypotheses section of Chapter

Three, the results of each are sequentially presented in this chapter.

Hypothesis One generally stated that differences would emerge

among the four different counselor conditions for all the subjects'

ratings considered together. Table 1 indicates that no main or inter-

action effect emerged for the factor of counselor condition on any of

the three counselor conditions. The F values for trustworthiness (T),

expertness (E), and attractiveness (A) all failed to reach statisti-

cal significance at the .05 level. Table 2 displays the T, E, and A

means for the four different counselor conditions. Thus, Hypothesis

One was not supported by the data: no one counselor condition emerged

as "superior" to any of the others.

In general, Hypothesis Two stated that there would be no differ-

ences between the ratings of the black subjects and those of the white

subjects. Referring to Table 1, a main effect for subject race was

Table 1

Analysis of Variance for the Three Factors: Counselor Condition,
Subject Race, and Contact on the Measures of Trustworthiness (T),
Expertness (E), and Attractiveness (A)

Source df SS(T) SS(E) SS(A) F(T) F(E) F(A)

CC 3 77.71 48.41 69.25 1.04 .68 .67
SR 1 139.53 192.60 192.60 5.59* 8.08** 5.61*
SJ (CC) 72 1795.90 1717.03 2472.03 -- -- --
CC x SR 3 62.31 129.91 23.21 .83 1.82 .23

CT 2 8.23 41.87 42.30 .41 2.53 1.73
SR x CT 2 18.10 7.25 15.40 .90 .44 .63
CC x CT 6 34.40 23.72 28.25 .57 .48 .39
SJ x CT 144 1444.60 1193.26 1755.86 -- -- --
CC x SR x CT 6 10.00 23.87 24.82 .17 .48 .34
CC = Counselor Condition
SR = Subject Race
SJ = Subject
CT = Contact
p< .05
** p< .01

Table 2

Results of Duncan's Multiple Range Test for Significance of Differ-
ence Between Means for the Four Different Counselor Conditions for
all the Subjects.

Counselor Condition
Black White
Counselor Dimension Accent No Accent Accent No Accent

Trustworthiness 71.85 72.65 72.45 76.20

Expertness 69.65 72.65 71.25 73.10

Attractiveness 65.40 66.40 67.30 69.75

Note. Continuous underscoring of the means indicates no significant
differences between the means at the .05 level.

found for ratings on all three of the counselor dimensions. For trust-

worthiness: F (1, 79) = 5.59, p< .05; for expertness: F (1, 79) =

8.08, p< .01; and for attractiveness: F (1, 79) = 5.61, p< .05. Thus,

Hypothesis Two was not supported by the data: Table 3 shows the means

of the black subjects' ratings were significantly higher than the means

of the white subjects' ratings on all three counselor dimensions for all

four counselor conditions. Stated simply, black subjects gave higher

ratings than white subjects regardless of the counselor condition and

regardless of the dimension on which they were rating the counselor.

Hypothesis Three suggested that for each racial subset of subjects

(i.e., black and white) significant differences would emerge among or

between the four different counselor conditions. Table 4 presents the

ANOVA for all the paired contrasts on each dependent measure. The

counselor dimension of expertness emerged as the only significantly

affected dimension. Inspection of the means in Table 5 reveals that

black subjects, rating a black counselor, gave a significantly higher

rating when the counselor spoke without an accent--but only on the

dimension of expertness. Hypothesis Three, then, was partially sup-

ported by the data.

Hypothesis Four, that subjects' ratings of a counselor would

change over the course of the three counselor/subject contacts, was

not supported by the data. Table 1 reveals no main or interaction

effects for the factor of changed ratings over time (contact). Sub-

jects did not tend to significantly change their ratings, regardless

of the counselor condition, on any of the three counselor dimensions.

Thus, Hypothesis Four was not at all supported by the data.

Table 3

Results of Duncan's Multiple Range Test for Significance of Differ-
ence Between Means for Black and White Subjects on the Three Counselor
Dimensions for all Counselor Conditions.

Counselor Dimension




Subject Race (Mean Scores)
Black White







Note. Broken underscoring of the means indicates significant dif
ferences between the means at the .05 level.

Table 4

Results of Analysis of Variance for all Paired Contrasts for Each
Group of Subjects on Each Dimension: Trustworthiness (T), Expert-
ness (E), and Attractiveness (A)

Contrasts a T E A
F(1, 36)
Black Subjects
Black Counselor/Accent vs Black Couns/No Accent 2.59 9.95** .34
White Counselor/Accent vs White Couns/No Accent 1.76 .14 1.63

White Subjects
Black Counselor/Accent vs Black Couns/No Accent .62 .48 .00
White Counselor/Accent vs White Couns/No Accent .55 .34 .00

a The cell to the right produced the higher score.
** p< .01

Table 5

Means and Standard Deviations of Trustworthiness (T), Expertness (E),

and Attractiveness (A) for
Each Group of Subjects

Group and Condition

Each of the Four Counselor Conditions, for

Mean SD Mean SD Mean SD

Black Subjects
Black Counselor/Accent
White Counselor/Accent
Black Counselor/No Accent
White Counselor/No Accent

White Subjects
Black Counselor/Accent
White Counselor/Accent
Black Counselor/No Accent
White Counselor/No Accent



5.24 67.50
7.72 68.80
4.90 69.70
7.86 73.60









Grop ad Cndiio

Hypothesis Five stated that, in general, the four counselor con-

ditions would contribute to significantly different Recall Scores

among the subjects in the four different counselor condition groups.

Referring to Table 6, it can be observed that no one of the four

counselor conditions contributed significantly more to enhance the

scores than any of the other three counselor condition groups.

In summary, the above results were analyzed by using the analysis

of variance (ANOVA) procedure. There were a total of three factors:

counselor condition (four levels), subject race (two levels), and

contact (three levels). Main effects and all interactions were

examined for statistical significance at the .05 level. Overall, the

results supported the argument that no one of the four different coun-

selor conditions was "superior" to any of the others. One exception

to this general rule was found for the specific combination of black

subjects rating the black counselor: under this condition the black'

subjects gave the counselor a higher rating on the dimension of

expertness when he spoke in the unaccented style. The other two coun-

selor dimensions measured in this study, attractiveness and trust-

worthiness, were not affected. The amount of information recalled

by subjects, as measured by the Recall Questionnaire, was not differ-

entially influenced by any of the four different counselor conditions.

One unexpected difference between the black and white subjects'

ratings was found on the dependent measures of trustworthiness,

expertness and attractiveness. Black subjects, for all counselor

conditions, gave statistically higher ratings than did the white


Table 6

Results of Duncan's Multiple Range Test for Significance of Differ-
ence Between Means for the Four Different Counselor Conditions for
all the Subjects' Scores on the Recall Questionnaire

Counselor Condition
Black White
Variable Accent No Accent Accent No Accent

Recall Questionnaire 8.1 9.2 8.3 9.2

Note. Continuous underscoring of the means indicates no significant
differences between the means at the .05 level.


Following the summary, this chapter addresses the meaning of

the results of the specific hypotheses while assessing some of the

particular cross-racial implications of the findings. Secondly,

the results are discussed in the context in which the hypotheses

were raised: as partially an interdisciplinary approach to suggest

directions for both the psychologist interested in counseling

processes and for the sociolinguist concerned with the effects of

different language styles. Finally, as the chapter draws to a

close some limitations of the study are addressed and future ques-

tions are raised which naturally emerged from both the results and

limitations of this study.


One of the major purposes of this study was an attempt to

empirically examine and integrate themes from two related disci-

plines--cross-cultural counseling and sociolinguistics. A

counselor's vocal accent has been identified as one of the many

important aspects of his or her total cultural or racial identity.

Sociolinguistic research has demonstrated that listeners often

demonstrate preferences for certain speech styles or accents.

To recapitulate, this study examined whether a counselor's

vocal accent affected the way in which racially different clients

perceived the counselor's trustworthiness, expertness and attrac-

tiveness. The following four different counselor conditions were

experimentally created to investigate this question: (1) a black

counselor speaking with a black English accent; (2) the same black

counselor speaking without an accent; (3) a white counselor

speaking with a southern English accent; and (4) the same white

counselor speaking without an accent. Another salient question

was whether any of these four conditions significantly contributed

to different amounts of information retained by the subjects.

The results demonstrated that the variable of counselor accent

did affect subjects' ratings--but only in one very specific context

of the experiment. It was found that black subjects, only when

rating the black counselor, gave the counselor significantly higher

ratings on the dimension of expertness when he spoke in unaccented

English. The other counselor dimensions of trustworthiness and

attractiveness measured in the study were not affected for this

group. No significant interactive effects were found for the group

of white subjects with any of the four counselor conditions. One

unexpected result emerged: universally, black subjects' ratings

were higher than white subjects' ratings. This result was inter-

preted from a psychosocial perspective. The other results were

discussed in both a theoretical manner and from a counselor inter-

vention perspective.

It was suggested that if certain counselor-client pairing

occurred; the setting, vocal style and race would all be likely to

contribute to a client's expectation of what might occur in coun-

seling and how the client might perceive the counselor's dimensions

of expertness, trustworthiness and attractiveness. When certain

counselor attributes are enhanced (e.g. expertness) it was

suggested that the client would be more receptive to certain types

of interventions (e.g. information and advice).

Lastly, the study suggested directions for future research

which naturally developed both from a consideration of the limita-

tions of the study and from the results of this investigation.


The first hypothesis suggested that one (or more) of the four

different counselor conditions would emerge as superior on at least

some of the measured counselor dimensions of trustworthiness (T),

expertness (E) and attractiveness (A). Clearly, this did not occur.

No one counselor condition proved to be "superior" to any of the

others. This result has more than one meaning within the parameters

of the experiment. The most obvious and simple explanation is that,

all things being equal in terms of the counselors' identifying

information--training, experience and education--the racial and

vocal qualities were simply not crucial to a client while he was

rating the counselor. This raises the question of the importance

of the identifying introductory information regarding the counselors'

background and training. It should be recalled that in each of the

four counselor conditions the identical information was furnished

to the subjects about the counselors' formal expertise (i.e. degree,

title and experience). The implications of this identifying infor-

mation are discussed further below.

The second hypothesis stated that there would be no overall

difference between the black subjects' ratings and the white

subjects' ratings. What emerged, in fact, was just the opposite:

the black subjects universally gave higher ratings on all of the

three counselor dimensions for all of the four different counselor

conditions. In one respect, this was a serendipitous finding.

Although it was hypothesized that certain (perhaps racial) subsets

of subjects would have shown preference for certain counselor

conditions, it was not expected that, universally, black subjects

would rate counselors higher than did the white subjects. These

results are more difficult to interpret due to both their unexpected

nature and the lack of previous research bearing directly on this

issue. The issue is further complicated if, instead of concep-

tualizing the findings as "higher overall ratings by black

subjects," they are considered as "lower overall ratings by the

white subjects." In either case, a host of relevant questions arise

as to what may have contributed to these results. Response sets

and differing expectations of the counselor (and counseling) are,

initially, two plausible explanations which will be addressed below

in more depth.

The third hypothesis suggested that for each racial subset of

subjects (i.e. black and white) differences would emerge among

their ratings of the four different counselor conditions. This

hypothesis was based on previous research done in the field of

sociolinguistics. Results of several studies have demonstrated

that certain listeners tended to rate accented speakers differently

from unaccented speakers (e.g. Tucker & Lambert, 1969). However,

the results of this study--and this hypothesis in particular--

deviated from this expectancy. Only one aspect of hypothesis three

was supported by the data: black subjects, when rating the black

counselor, gave the counselor significantly higher ratings on the

dimension of expertness when the counselor spoke in unaccented

English. There were no differences on the dimensions of expertness

or trustworthiness for this group. The white subjects showed no

differences in their ratings for any of the counselor conditions.

In light of previous research (e.g. Thomas, 1978), this result alone

can be interpreted as suggesting that the black subjects considered

the black counselor as more of an expert when he spoke in unaccented

English. However, the complete results of this study suggest that

this may be an incomplete interpretation of the data. Why, for

example, did only the black subjects reveal a difference in their

ratings and why only along this dimension of expertness? It is

certainly plausible that the black subjects were more sensitive

to stylistic voice differences within their own racial group.

However, this is problematical. The results revealed that white

subjects were not influenced by the (accented or unaccented) voice

styles of either the black or white counselor. This result contra-

dicts some earlier findings within the sociolinguistic literature

and is discussed in more detail later in this chapter.

Hypothesis four, which stated that subjects' ratings would

change over time, was not supported by the data for any of the four

different groups. It can be stated that in this study all of the

subjects tended to remain fairly consistent in their judgements

about counselor T, E and A over the course of the week during

which the three video tape monologues were presented. Perhaps the

relatively short period of one week was not sufficient time for

all the subjects to experience a meaningful change of their

impressions of the counselors. Some anecdotal data, recorded by

the investigator, revealed that at least a few clients expressed

subjectively changed feelings about their "counselor" over the

week's course. Some examples included: "This guy gets worse

every time I see him"; "At first I wondered about him . but

by now it's O.K."; "Hey, a southerner--they oughtta send him back

where he came from"; and "He's O.K., 1 think I've seen him around

the hospital."

The final hypothesis, that the four different counselor condi-

tions would significantly contribute to differing amounts of infor-

mation recalled by the subjects, was not supported by the data.

Table 3 reveals that when either of the counselors spoke in

unaccented English, the subjects tended to recall more of the mono-

logue information on the average compared to the accented English.

This result may be interpreted in two ways. First, it may mean

exactly that which it suggests: that accented speech isn't really

an important variable influencing information retention. Secondly,

it may be interpreted as the beginning of a trend which might have

reached statistically significant levels if the conditions (i.e.

accents) had been "stronger." This is a notion which should be

addressed in future research.

Sociolinguistic and Cross-Cultural
Counseling Implications

What are some of the more general implications of this study

which are relevant to the area of therapy, but which also involve

the counselors' vocal style? The overall results--that no one of

the four counselor conditions emerged as "superior" to any of the

others, despite obvious vocal and racial differences among the

four conditions--may be explained by some of the other experimental

conditions which were part of the study. In this investigation the

counselors were all identified in an identical manner. Regardless

of their speech style, both counselors were given the same hypo-

thetical Ph.D. in psychology and three years of counseling experi-

ence. Siegel (1980) has found that the presence of objective evi-

dence of expertness (e.g. verbal introduction, posting of diplomas)

enhanced subjects' ratings of counselors on the dimensions of trust

and credibility. Siegel's (1980) finding, for the most part, was

not inconsistent with the results of this study. Generally, with

all things being equal in terms of training and experience, the

race and vocal styles were not very potent factors of influence on

subjects' perceptions of the counselors as reflected by their

ratings. Sociolinguistic theorizing has also advanced some support

for this interpretation of the results of this study.

Williams (1970) has described the stereotype hypothesis as a

listener's evaluational reaction to a speech style, which elicits in

the listener certain attitudes, biases and prejudices about the

speaker of the particular speech style. The best illustration of

these biases for black and white racial groups have been demonstrated

by Tucker and Lambert (1969). The currency of their findings (see

Chapter Three, Listeners' Attitudes Toward English Dialect Speakers)

have been challenged by the results of this study. Where Tucker and

Lambert (1969) essentially found a hierarchy of most favorable to

least favorable styles, the present study did not reveal such a

hierarchy at a statistically significant level. Although an inspec-

tion of the means in Table 2 reveals differences between each speech

style--with the unaccented speech always rated higher--the differ-

ences in no instance reached statistically significant levels. Some

further comments about the concept of context for language studies

will attempt to explain the meaning of this study's results in a

fashion consistent with the sociolinguistic literature concerning

language attitudes and context (e.g. Williams, 1970; Williams, 1973;

Naremore, 1971).

Williams (1970) has raised a central point which bears directly

on the interpretation of the results of this study when he discussed

the variable of "context." Most of the research before the 1970s on

language attitudes took place in a relative vacuum. That is, lis-

teners were usually presented with a voice tape played on a tape

recorder, and then given some type of rating scale on which they evalu-

ated the "speaker" or the "speech of the speaker." The type of results

(e.g. Tucker & Lambert, 1969; Anisfeld, Bogo & Lambert, 1962) which

generally emerged from these studies clearly suggested that certain

styles were preferred over others. But, as Williams (1970) stated:

Many Negroes, particularly in middle-class urban areas, are
clearly bidialectical, and it seems unnecessary to empha-
size how closely tied the choice of dialect is to the con-
text of speech. Negro nonstandard English is expected of

blacks by blacks in the inner city, whereas a Negro's
use of standard English is expected, mainly by whites,
for participation as a member of certain social or
occupational structures (say, as a schoolteacher). For
the speaker to violate such expectations elicits dras-
tically revised stereotypes--the black male being con-
sidered a traitor (or sometimes a homosexual) when speak-
ing standard English in the inner city, or being con-
sidered militant or "uppity" when using black speech in
situations where standard English is expected. The point
here seems obvious: We cannot talk about speech and
stereotypes without including context as a variable.
(p. 390)

This concept of context was a central issue of the purpose of

this study and germane to the interpretation of the results. The

purpose and methodology of this investigation were designed specifi-

cally in an attempt to integrate the earlier "contextless" language

attitude findings into an experimental paradigm which also contained

a very specific context--counseling. And reciprocally, the variable

of speech style has been one of those enigmatic dimensions of the

even larger question of cross-cultural counseling theory. What

empirical support did the results lend to the emerging theory in

both the cross-cultural counseling and sociolinguistic realms?

The results of this study appear to support Williams' (1970)

notion of context as a crucial methodological concern in this type

of research. Additionally, the idea of context also seems to have

contributed to our understanding of how a specific cross-cultural

variable affects (and does not affect) some of the key dimensions

of a counseling relationship. In this study, context was carefully

prepared and controlled. The counselor's role was a very specific,

clearly understood one. Furthermore, the use of audio visual tapes

made it possible to introduce a new (visual) dimension into a

previously established methodological strategy of studying language

attitudes--the "matched guise" technique. The stringent controls

imposed upon the context of this study apparently contributed to

some type of "wash out" effect for the different speech styles--but

not completely. The one very specific context in which a signifi-

cant effect was found was when black subjects rated the black coun-

selor--and rated him lower on the dimension of expertness when he

spoke with the black English accent. Interestingly, the dimensions

of trustworthiness and attractiveness were not affected by the

accent. This might mean different things for the counselor in the

real world.

First of all, the type of counseling setting must be taken into

account. A black client would quite possibly bring different sets

of expectancies (regarding language styles) depending upon whether

the potential counselor was working in a predominantly black or

predominantly white counseling or community setting. In the present

study, the setting was racially mixed client-wise, but the staff was

predominantly white. Replication of this type of study, perhaps with

a different group of subjects and/or with different types of coun-

selor speech styles would shed additional light on this still unclear

issue. Generalizations of the results of this study to, for

example, a predominantly black staffed mental health setting must be

made wtih caution. It should also be recalled that the clients

studied in this experiment were inpatients at a hospital located in

the northeastern section of the country. Perhaps if the study were

replicated in the southern part of the country different findings

would appear. The above caveats should be kept in mind as some

implications for counselor practice and intervention are discussed next.

The three dimensions of counselor attributes measured by the

Counselor Rating Form have implications for the impact a particular

counselor intervention may have on a client. In this study, for

example, it was found that when a black counselor spoke to a black

client in unaccented English it enhanced the counselor's dimension

of expertness for that client. Interventions, then, under this

condition may facilitate and strengthen certain qualities of the

counseling relationship--and not, it seems, at the expense of the

other two important dimensions measured by the Counselor Rating

Form. Under conditions of high perception of counselor expertness,

a client would be likely to experience the counselor as an

experienced and skilled professional; someone who is prepared,

clear and logical in the counseling relationship. The client

would be more prone to accept certain types of counselor feedback

rather than others. Advice and information from someone perceived

as an expert counselor would probably be received as valid and accu-

rate by the client because he sees the counselor as an informed and

intelligent professional. Conditions which enhanced a client's

perception of counselor trustworthiness would contribute to feelings

of genuineness, honesty and sincerity in the counseling relation-

ship. A client experiencing the presence of these qualities and

feeling unthreatened would be prone to disclose more personal and

perhaps more painful feelings. Since this study found only one

counselor-client combination (black subject with black counselor)

which significantly enhanced one of the three measured dimensions

expertnesss), more research is clearly needed to explore how other

possible counselor-client combinations affect the other dimensions

of the counseling relationship.

A Serendipitous Finding

Finally, the serendipitous finding of significantly higher

ratings by the black subjects is discussed here from more of a

sociological position--focusing on some literature which has addressed

the issue of ethnic minorities' perception of mental health providers.

Schneider, Laury and Hughes (1980) reviewed much of the research

on ethnic group perceptions of mental health services and carried

out an empirical study to test whether differences emerged among

different ethnic groups. One of their findings revealed that

different ethnic groups did have different perceptions of mental

health providers. Blacks (and Chicanos) reported that they were more

likely than whites to discuss personal problems with a counselor.

They (Schneider et al., 1980) suggested that white clients were more

likely to have had previous contact with a counselor and this

familiarity could have contributed to the white clients' realization

that counseling was not always helpful. Many writers (e.g. Vontress,

1970) have suggested that minority clients experienced greater and/or

more special difficulties which are subjectively felt as more intense

than the personal problems of white clients. This may have led to

the blacks in this study feeling more concerned than white clients

about their difficulties. Harrison (1975) has suggested that white

society has traditionally encouraged minority role behavior to lack

openness and candidness. In reaction, perhaps, the black subjects

may simply have been motivated to show an increased tendency to

discuss personal issues because of a demand effect to "tell him

(the white experimenter) what he wants to hear." Sue (1978)

reported that minority clients were recently becoming more likely

to discuss personal problems with counselors because they expected

effective action and intervention by a mental health professional.

All of these views are worth considering in an attempt to inter-

pret the meaning of the unexpected finding in this study of uni-

versally higher ratings by the black subjects.

It is plausible that all the viewpoints described above may

have contributed to the result now under discussion. Although no

information was elicited from subjects regarding their previous con-

tacts with counseling experiences, it was assumed that the subjects

were at least partially representative of the total population of

therapy clients. Given this, the white subjects may have given

the counselors lower ratings because of their realization that the

counselor was potentially "not helpful." Following Vontress'

(1970) thesis, perhaps the black clients were experiencing a

greater degree of subjectively felt emotional discomfort. This

would have resulted in increased ratings for the counselor whose

role as a potential helper was seen as more vital. The point

raised by Harrison (1975), plus the fact that the experimenter was

a white counselor himself, may have combined to produce in the black

subjects a need to "tell them what they want to hear"--that coun-

selors are positively regarded. Sue's (1978) point may also have

some validity. The clients in this study were all inpatients in a

voluntary rehabilitation program and had expressed explicit needs

and reasons to be there. Naturally, they expected to deal with

their problems and knew they would be involved with professional

counselors during the course of their hospitalization. Finally, the

reality of individual differences should be considered in relation

to the preceding discussion. Perhaps certain expectations were

stimulated in some clients, but not in others. This is a complex

area to interpret, especially in such a post hoc fashion. Future

research is clearly needed in this area.

Present Limitations and Future
Research Recommendations

The analogue nature of this study resulted in a methodological

issue previously described as having imposed some artificiality on

the counselor-client contacts in this study. Careful presentation

of the audio visual tapes and stimulus materials were essential

steps in capturing as much counselor "realism" as possible without

losing control over the subtle differences among the four counselor

conditions. Future research might try to improve upon this method-

ology in two ways. First, although the actors in this study each

did an excellent job in role playing the two different accents, it

would be desirable in the future to use real, bidialectical coun-

selors. In addition, to transcend the analogue quality of the

research, real-life situations could be developed in which to study

the processes in question. Adequate control of the precise vocal

qualities would be much more difficult to obtain, but it is an area

worth developing.

The setting and subjects used in this study involved only one

possible counseling environment which resulted in a few implicit

limitations. First, the clients were all being treated for

alcoholism; they (and the counselors) were all males; and all were

veterans in the same treatment program. While alcoholism was

certainly a legitimate mental health concern for all the subjects,

the restricted sample naturally imposed some restrictions upon the

generalizability of the results. Caution must be used when dis-

cussing the application of these results to other mental health

settings. Several directions for future research naturally arose

from these considerations.

The factor of client and/or counselor sex was one area certainly

needing further exploration. How does one's gender interact with

the factors of race and voice style? Another variable which should

be studied is the factor of accent from the counselor's point of view.

How do counselors react to clients' vocal qualities? Also, this

society is a multi-ethnic one and there are many other vocal accents

in addition to the ones examined in this study. In addition, it

should be recalled that the location of the treatment program used

in this study was in the northern United States. Future studies

could be conducted in other regions of the country. Finally, there

are many other viable settings in which counseling relationships

could be studied--schools, colleges and community mental health

centers--to name only a few.

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