Citation
Attribution of diagnostic credibility to clinical supervisors among counselor education students

Material Information

Title:
Attribution of diagnostic credibility to clinical supervisors among counselor education students
Creator:
Carroll, Jane J
Copyright Date:
1995
Language:
English

Subjects

Subjects / Keywords:
Attribution theory ( jstor )
Counselor training ( jstor )
Doctoral students ( jstor )
Educational counseling ( jstor )
Faculty advisers ( jstor )
Mental health ( jstor )
Professional certification ( jstor )
Psychological counseling ( jstor )
Psychology ( jstor )
Women ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jane J. Carroll. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
33815402 ( OCLC )
AKP4880 ( LTUF )
0021951926 ( ALEPH )

Downloads

This item has the following downloads:


Full Text










ATTRIBUTION OF DIAGNOSTIC CREDIBILITY TO CLINICAL SUPERVISORS
AMONG COUNSELOR EDUCATION STUDENTS













By

JANE J. CARROLL


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1995














To my clients and students, those I have known and those I have yet to meet. May my work always reflect the exceptional education and supervision I received as I prepared to be a counselor and counselor educator.















ACKNOWLEDGMENTS

Without the influence of people who encouraged and supported me in this undertaking, I'd have had to sacrifice the pleasure. I learned this is a journey I could not have traveled alone.

The spirit of my father, James Judy, lives on. His powerful example and his belief in me were essential to this effort.

My mother, Mildred Judy, lovingly gave me invaluable guidance and assistance.

James Carroll and Timothy Carroll, my children, endorsed my plan to take a different path. Their advocacy made the striving that much easier.

My grandchildren Kaleigh, Jackson, and those to be born, make the striving that much more meaningful.

I cannot overstate the importance of James Bazan's many

contributions. I am particularly grateful for his belief in my competence, pride in my work, and steadfast presence.

Dr. Eve Hershberger's friendship and unrivaled clinical supervision empowered me to develop personally and professionally.

Rick Sayers and Dr. David Miller brought light to a dark place--my brain on statistics.


iii









Candy Spires, Donna Johnson, and Sharon Lake in 1215 Norman were unselfish with their time and encouraging with their words. Their friendship relieved some of the most stressful moments.






My committee:

I appreciate Dr. Jin-Wen Hsu's eagerness to help me understand concepts in quantitative analysis.

Dr. Rob Sherman introduced me to the world of the qualitative researcher--a professional gift. He nurtured my homesick soul with his familiar and charming Maine-speak.

Dr. Joe Wittmer is truly a gentle person. His confidence in me

helped me learn how it feels to be a counselor educator. His influence will never end.

Dr. Larry Loesch has been the single greatest influence in my

professional life. I hold this man in high esteem not only for the example he sets, but for his superb management and supervision of my academic experiences, reassuring words and presence, and confidence in my ability. He allowed me to learn, but never to suffer. I am proud and grateful that he is my friend and mentor.


iv















TABLE OF CONTENTS
page

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

LIST OF TABLES.............................................. vii

LIST OF FIGURES ..............................................viii

ABSTRACT .............................................. ......ix

CHAPTERS

I INTRODUCTION

Overview 3 Person Perception and Attribution ..........................5
Statement of the Problem .................................7
Need for the Study .........................................8
Purpose of the Study .................................... 14
Rationale for the Approach ...............................15
Hypotheses ............................................ 16
Definitions....................................... 16
Overview of Remainder of Study 18


2 REVIEW OF LITERATURE ..............................19

Application of Counseling Theories and Diagnostic Skills 22 Relevance of Theoretical Framework .......................35
Elaboration on Need for the Study ........................ 41
Research on Perceptions and Attributions of Parties in
Counseling Process ................................ .... 41
Summary................................ ............59

3 METHODOLOGY ..................... ...............63

Delineation of the Variables 63 Population 64









Sampling Procedure .....................................65
Research Procedure .....................................66
Data Analyses 68

4 RESULTS................................... ..........71

Quantitative Analyses ...................................71
Qualitative Analyses .................................... 76

5 DISCUSSION ..........................................88

Limitations of the Study .................................88
Relevance of Attribution Theory .......................... 91
Conclusions and Interpretations ...........................92
Implications .......................................... 103
Recommendations for Future Research ....................107
Summary .............................................110

APPENDICES

A LETTER OF INTRODUCTION .......................... 112
B INSTRUCTIONS FOR ADMINISTERING INSTRUMENT .... 113 C DEMOGRAPHIC INVENTORY 114 D PROFILE OF A FICTIOUS CLIENT 115
E DESCRIPTIONS OF FICTITIOUS
CLINICAL SUPERVISORS............................. 117
F AGREEMENT RATING SCALE
AND QUALITATIVE DATA 119
G GENDER-CREDENTIALS CODE 120 H POST CARD REMINDER 121

REFERENCES 122

BIOGRAPHICAL SKETCH 131


vI















LIST OF TABLES


Table page

4-1 Summaries of Rating by Levels of Educational
Credentials and Proportion of Program Hours Completed ........... 73

4-2 Analysis of Variance of Ratings of Diagnostic
Credibility by Educational Credentials, Gender,
and Proportion of Program Hours Completed............................ 74


Vii














LIST OF FIGURES


Figure page

1 Cell means of educational credentials by
proportion of program hours completed ....................75

2 Reasons for rejecting supervisors' diagnosis ... .............78

3 Group #1 response frequencies 79 4 Group #2 response frequencies 79 5 Group #3 response frequencies 80 6 Group #4 response frequencies ............................80

7 Summary of response frequencies for groups #1 #4 81

8 Reasons for provisionally accepting supervisors'
diagno sis ................ 82

9 Group #5 response frequencies 83 10 Group #6 response frequencies 83 I I Group #7 response frequencies 84 12 Group #8 response frequencies ........................... 84

1 3 Summary of response frequencies for groups #5 #8 85

14 Reasons for accepting supervisors' diagnosis
without question ........................... .. ................. 86

15 Group #9 response frequencies ........ 86 16 Group #10 response frequencies .......... ................ 87

17 Summary of response frequencies for groups #9 and #10 87


viii














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

ATTRIBUTION OF DIAGNOSTIC CREDIBILITY TO CLINICAL SUPERVISORS AMONG COUNSELOR EDUCATION STUDENTS By

Jane J. Carroll

August, 1995

Chairman: Larry C. Loesch, Ph.D. Major Department: Counselor Education

This study investigated the extent to which counselor education

students differentially attribute diagnostic credibility to clinical supervisors based on clinical supervisors' educational credentials and gender, and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation.

Subjects were 191 male and female counselor education students who had completed varying proportions of their entry-level program hours from CACREP-accredited counselor education programs. They rated their perceptions of the diagnostic credibility of male and female: (a) counselor education doctoral student, (b) doctoral-level psychologist, (c) counselor


ix








education department faculty member, and (d) psychiatrist acting as supervisors.

Quantitative and qualitative analyses were conducted. An analysis of variance and post hoc analyses of significant interactions between supervisors' educational credentials and students' proportion of program hours completed required for graduation were done. Qualitative analyses of subjects' narrative replies were done to achieve broader understanding of the subjects' views on clinical supervisors' diagnostic credibility.

Counselor education students who had completed varying

proportions of program hours required for graduation did not attribute diagnostic credibility differentially to their clinical supervisors. Also, gender of supervisors and supervisees was not a significant factor in attribution of diagnostic credibility.

The results have implications for clinical training of counselor

education students, counselor education curricula, staffing practices in mental health facilities and managed care programs, and student supervising practices.


x














CHAPTER 1
INTRODUCTION

In his novel The Prince of Tides, Conroy described a psychiatrist as a "sojourner among wounded, disabused souls" (1986, p. 386). Conroy's description is not peculiar to the fictional doctor. It applies as well to many mental health counselors whose primary professional ambition is to help people who have problems that interfere with normal functioning. The State of Florida defines the practice of mental health counseling as

the use of scientific and applied behavioral science theories,
methods, and techniques for the purpose of describing, preventing,
and treating undesired behavior and enhancing mental health and
human development. Such practice includes the use of methods of a
psychological nature to evaluate, assess, diagnose, and treat
emotional and mental dysfunctions or disorders, whether cognitive,
affective, or behavioral; behavioral disorders; interpersonal
relationships; sexual dysfunction; alcoholism; and substance abuse.
(Department of Business, 1994, p. 2)

When counselors and clients work together effectively they can

identify clients' problems and needs. When the right facilitative action is taken, clients' lives often become more fulfilled and happier than they had been previously.

At the time a client seeks help, the counselor's first task is to determine accurately the nature and range of the client's problem or disorder (Seligman, 1990). This undertaking requires specialized skills,


I







2


the development of which begins early in counselors' professional preparation and continues throughout training.

Need for assessment and diagnostic skills typically is not evident until counselor education students start supervised field placement experiences. At the beginning of supervised field experiences, counselor education students are relatively inexperienced as clinicians and, consequently, are unprepared to recognize and conceptualize easily the meaning of signs and symptoms of mental distress. Therefore, counselor education students are expected to gain proficiency in assessment and diagnosis of clients, primarily through instruction and clinical supervision in field placement experiences.

Cognitive change may lead to behavioral change (Forsterling, 1988). Social learning theory, commonly used to explain cognitive change, suggests that the confidence counselor education students have in their respective supervisors' diagnostic competence would be instrumental in determining the degree to which counselor education students develop effective diagnostic skills (Bandura, 1977). However, do counselor education students conclude that confidence in their supervisors as instructors or models is deserved?

Attribution theory suggests that belief strength varies as a function of source credibility (Secord & Backman, 1974). Thus, the degree of confidence counselor education students have in their supervisors'








diagnostic abilities could vary as a function of their supervisors' educational credentials and gender and of counselor education students' gender and proportion of counselor education program hours completed that are required for graduation. It was unknown, however, the extent to which counselor education students' confidence in their supervisors' diagnostic credibility is a function of these variables.

Overview

Counselor education students in the United States have the

opportunity to become competent professional mental health counselors through more than 90 departments of counselor education that offer programs accredited by the Council for the Accreditation of Counseling and Related Educational Programs (CACREP). In particular, counselor education students enrolled in CACREP-accredited mental health counseling preparation programs must complete supervised practica and internships as curricular requirements. Counselor education students, having been taught counseling theories and techniques in the classroom, are expected to transfer the knowledge gained in classes to field experiences. There, presumably capable professionals supervise student progress and practice as they deal with "client welfare, counseling relationship, assessment, diagnosis, clinical intervention, prognosis, and appropriate referral techniques" (Bradley, 1989, p. 4).







4


Students most likely to be interested in diagnostic credibility are those in CACREP-accredited community or mental health counseling programs. These are people who practice as supervised mental health counselors at sites such as community mental health agencies, inpatient and outpatient treatment facilities, human resource centers, businesses and industries, criminal justice settings, or individual private practices. These students shall be referred to hereafter as counselor education students, with the understanding that they are students in community counseling or mental health counseling programs.

Mental health service providers generally require nationally

recognized and accepted descriptive diagnoses of clients. When based primarily on criteria found in the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (APA, 1994), these diagnoses identify clients' relevant biological, psychological, and sociological conditions. Thus, diagnosis is generally accepted as a prelude to designing appropriate treatment plans.

Supervisors guide counselor education students in practica or

internships as they determine the nature of their clients' problems. This guidance frequently takes the form of advising counselor education students to "accept" (i.e., use and act upon) diagnoses developed by someone other than the students, usually their immediate supervisors. In essence, supervisors typically expect counselor education students to









assume the correctness of others' diagnoses and to employ established treatment plans or to develop treatment plans based on others' diagnoses. Several questions arise in this regard. Do counselor education students assess the diagnostic credibility clinical supervisors differently as a function of the supervisors' credentials? Does the gender of supervisors influence counselor education students' attribution of credibility? How does the proportion of counselor education program hours completed that are required for graduation and the gender of counselor education students relate to attribution of credibility to supervisors?

Person Perception and Attribution

An individual's reality is formed subjectively by his or her

perceptions (Heider, 1958). However, there are a limited number of details to which people can attend simultaneously, and several factors determine which details will be perceived. For example, Postman, Bruner, and McGinnies (1976) wrote that perceivers' personal values, interests, and survival needs play major roles in determining the stimuli individuals select to perceive and those they defend against or reject. Individuals choose information they find personally significant. According to Secord and Backman (1974), one of the main sources of information used in evaluating other people are their actions and characteristics (e.g., credentials attained and gender).







6
Attribution theory addresses how people use information they have collected to answer questions that ask "why?" (Hewstone, 1983; Kelley, 1973). Kelley (1973) admitted that an individual's process of attribution is not free from bias, is based on incomplete evidence, and is scientifically inconclusive. In his ANOVA model of attribution in social interaction, however, Kelley (1973) proposed that causes (independent variables) covary with effects (dependent variables).

Kelley (1973) said that effects are caused by (a) entities that are present in the situation, (b) the situation or circumstances in which the action occurs, or (c) the perceptions or characteristics of observers who are present, or combinations of these three factors. The measure of validity of people's perceptions, judgments, and evaluations can be determined by any of three criteria. Kelley (1973) proposed that the first criterion for validity is consensus. Is a response to a stimulus like that of other people's? May the respondent therefore have confidence in her/his belief? The second criterion is consistency. Is a response consistent over time? This criterion includes the presumption that like responses have taken place prior to the time in question. The third criterion of validity of a perception is distinctiveness. Is a response associated with a particular stimulus?

Attribution of credibility to an entity and the validity of the

attribution can be illustrated by examining a hypothetical example of a







7


counselor education student in a clinical internship (the circumstances) attributing credibility to a supervisor (an entity) by responding (based on student's personal perceptions) to the question, "why do you believe your supervisor, Dr. Doe, arrived at a correct diagnosis for that client?" The counselor education student, using information gained from observations and experience, could answer, "because my colleagues in the office agree (consensus) that s/he is correct, the criteria are present for this diagnosis (distinctive stimulus), and I think Dr. Doe's diagnoses are usually correct (consistency)."

A perceiver attributes credibility to a source of information based on the nature and the strength of her or his perceptions (Kelley, 1973). In addition, as Postman, Bruner, and McGinnies (1976) and Secord and Backman (1974) suggested, the characteristics of the perceiver (e.g., gender and proportion of counselor education program hours completed that are required for graduation) influence the extent to which an impression of credibility is formed.

Statement of the Problem

Although fundamental principles of attribution theory have been examined in many contexts, they have been examined infrequently in the counseling profession. Rarer still is investigation of these principles within the context of clinical supervision of counselor education students. Therefore, the problem addressed in this study was that differences among







8
counselor education students' attribution of diagnostic credibility to psychiatrists, doctorate-holding counselor education program faculty, doctorate-holding psychologists, and doctoral students holding master's degrees in counselor education were unknown. It was also unknown if attribution of credibility varied as a function of diagnosticians' gender or of counselor education students' gender and proportion of counselor education program hours completed that are required for graduation.

Need for the Study

Do clinical supervisors' educational credentials or gender or

supervisees' gender or proportion of counselor education program hours completed that are required for graduation affect the extent to which supervisees attribute diagnostic credibility to supervisors? Among those who most frequently supervise counselor education students in practica and internships are psychiatrists, counselor education department faculty members holding doctoral degrees, doctorate-holding psychologists, and doctoral students who have earned at least a master's degree in counseling. Unfortunately, little research has existed about supervisees' beliefs in the credibility of supervisors' assessments of clients' problems.

Results of this study included implications for the mental health counseling profession and for counselor education departments. Differences among counselor education students' attribution of diagnostic credibility to supervisors with varying educational credentials by students







9


who have completed different proportions of their counselor education programs could affect students' education and treatment of clients. Educational Credentials

Heppner and Handley (1981), Worthington (1984), and Worthington and Stern (1985) found that supervisees' perceptions of counselors' competence are not related to educational credentials. College students perceive psychiatrists, however, to be more credible than other mental health providers at assessing mental health problems (Truatt & Bloom, 1982; Greenberg & Wursten, 1988). Gender

There has been meager research on how gender relates to perceived credibility of clinical supervisors. Supervisors' gender, at least in initial supervisory encounters, is not related to supervisees' perceptions of supervisors' competence (Schiavone & Jessell, 1988). Apparently it has been unknown if gender of supervisors or supervisees affects counselor education students' perceptions of supervisors' diagnostic credibility. Proportion of Program Hours Completed

Evidently there has been no published research on the relationship between attribution of diagnostic credibility by counselor education students to clinical supervisors and proportion of counselor education program hours completed that are required for graduation. However, Stoltenberg and Delworth's (1987) proposed levels of counselor







10
development implies that counselor education students would experience gradually changing perceptions of expertness and trustworthiness of clinical supervisors. Usher and Borders (1993), who examined professional counselors' preferred areas for supervisory concentration, discovered that they favor supervisors who emphasized conceptual skills to those who emphasized professional behaviors, professional practice, and process skills.

The information learned in this study also can be meaningfully

related to (a) research on perceptions and attributions of immediate parties in the counseling process, (b) clinical training of counselor education students in counselor education programs, and (c) relationships of clinical supervisors and counselor education students in practica and internships. Additionally, the information could be applied to investigations related to other groups of people such as (a) counselors who have completed their basic education and have had some professional counseling experience, (b) counselors in different work settings who work with various types of clientele, and (c) counselors who are disinclined to diagnose. The information also could be applied to studies of educational programs that place differing emphases on teaching the art and science of diagnosis, as well as in situations where supervision varies in the extent of cooperation or cohesiveness between supervisors and counselor education students. These results also could be used in the study of relevant personality or







II


attitude variables in the context of supervision. For example, characteristics of supervisors, such as the extent to which they project confidence, might be significant to their being considered credible, as might be the (a) role that counselor education students' expectations play in attribution of credibility, (b) influence of counselor education students' self-efficacy on their perceptions of supervisors' credibility, and (c) personality types of counselor education students and supervisors. Research on Perceptions of Parties in Counseling Process

The early application of attribution theory, wherein it was used to explain how people perceive and interpret others' actions (Heider, 1958; Rotter, 1966), has explained clinicians' inferences about clients' psychological states. "The clinician's conceptual framework--whether behavioral, dynamic, or humanistic--is a set of systematic attributions for the causes of human behavior" (Jordan, Harvey, & Weary, 1988, p. 91). Further, "virtually all clinical endeavors, diagnostic and therapeutic, have attributional processes at their core" (Jordan, Harvey, & Weary, 1988, p. 91).

Clients' and counselors' perceptions of each other result from the fundamental interactive nature of counseling. Counselors make clinical judgments of clients based on knowledge gained through assessment (Smith, 1988), and clients value certain counselor attributes as significant to the success of the counseling experience. Researchers have shown that







12
counselor expertness is a variable clients consider favorable to the counseling process (Holland, Atkinson, & Johnson, 1987, Littrell, Caffrey, & Hopper, 1987; Ponce & Atkinson, 1989). Apparently no research exists, however, that examines counselor education students' perceptions of their supervisors' diagnostic expertise. Therefore, it was unknown whether counselor education students considered their supervisors as credible diagnosticians and how the attribution of credibility, or lack of it, influences the supervisory experience and ultimately client care. Clinical Training of Counselor Education Students

Clinical judgment develops in an ongoing process. The struggle to understand and make sense of others' feelings, thoughts, and behavior originates early in counselor education students' classroom education. Counselor education students refine their ability to diagnose, however, when they are supervised in practical field experiences. Counselor education students are exposed to a number of theories from which to choose viable explanations for clients' problems, as well as suitable approaches to treatment. The ultimate decisions, however, regarding diagnosis and treatment planning must be, for moral, ethical, and legal reasons, a result of considered judgment.

Counselor education students may not receive the level of training in diagnosis that some counselor educators and students think they need (Fong, 1993; Furlong & Hayden, 1993; Hohenshil, 1993; Seligman, 1990;







13
Waldo, Brotherton, & Horswill, 1993). By discovering the factors that counselor education students consider when attributing credibility to diagnosticians, this study may help determine if counselor education programs adequately prepare counselor education students to diagnose clients thoughtfully and knowledgeably. Relationships Among Clinical Supervisors and Supervisees

The practice of professional mental health counseling may benefit from this study. Knowledge of beginning counselors' attitudes toward their supervisors' diagnostic credibility affects two primary areas of concern, client care and facility staffing.

Novice counselors could jeopardize the quality of client care if they

(a) view the staff who assess, diagnose, and suggest treatment modalities as either incompetent or infallible; (b) are unable to accept the judgments of certain diagnosticians because of personal biases; (c) base their clients' treatments on diagnoses other than those on record, or, likewise, compromise the integrity of the system by unquestioningly following the lead of others; (d) philosophically or theoretically differ with particular diagnosticians; and (e) have no philosophical or theoretical basis for agreeing with or disagreeing with particular diagnosticians.

Those who staff facilities that offer mental health counseling could benefit from knowing (a) the likelihood of power differentials among staff based on educational attainment or gender; (b) the likelihood of counselors







14


to accept or reject diagnoses, regardless of source; and (c) which staff member(s) would most likely be respected as having diagnostic credibility.

Mahoney (1988, p. 156) proposed that a counselor will "influence those whom he assesses, and counsels. One would hope that the selection and training of such professionals would assure that their assessment [a prelude to diagnosis] and advice are qualitatively superior to that of their untrained peers." Clearly, counselor educators, clinical supervisors, and staffers of facilities could use the results of this study when planning curricula and supervising students' performances in practica and internships. As Herman (1993) noted, therapist training may be the decisive factor in therapy outcome.

Purpose of the Study

The purpose of this study was to investigate an application of the nature of principles of attribution theory to clinical supervision of counselor education students. The study was conducted to determine the extent to which counselor education students think clinical supervisors' diagnoses of clients are credible, to ascertain whether supervisors' educational credentials and gender influence counselor education students' views of supervisors' credibility, and whether counselor education students' gender and proportion of counselor education program hours completed that are required for graduation influence their beliefs about







15
supervisors' credibility. Data were subjected to quantitative and qualitative analyses and interpretations.

Although other, potentially pertinent variables (e.g., race or

theoretical orientation of the supervisors or students) could have been investigated, their inclusion would have made the design prohibitively complex and beyond the resources available for this study. Therefore, this study focused upon those variables deemed most significant in this initial, exploratory research.

Rationale for the Approach

This experimental research was conducted to determine the extent to which counselor education students differentially attribute credibility to clinical supervisors. There were a variety of potential ways to assess subjects' attitudes. However, one of the most commonly accepted techniques is to use a bogus profile to which subjects can respond. The essential content of the profile of a fictitious client, as well as the diagnosis, were constant across conditions presented to subjects. The credentials of the diagnostician were not constant, but rather reflected potentially different levels of credibility. This approach allowed for variation in responses, anonymity for subjects, time for thinking before responding, administration to many people at one time, and responding to otherwise identical content (Bailey, 1987, Henerson, Morris, & FitzGibbon, 1987).







16


Hypotheses

The following null hypotheses were tested in this study:

1. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials.

2. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' gender.

3. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students.

4. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation.

5. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' gender.

Definitions

The following operational definitions were used throughout this dissertation:

Ajge is a period of human life. In this study, a subject self-reported years

of age.







17
Attribution theory is an explanation of how people explain causes of events

and the actions that follow.

Credibility means to be believable, trustworthy, and worthy of confidence.

The believability of diagnosticians was addressed in this study.

Diagnosis is the judgment reached about a person's mental health status

based on an assessment of the client, including the presence or absence of signs and symptoms as described in the most recent

edition of the Diagnostic and Statistical Manual of Mental Disorders

(APA, 1994).

Gender is defined in this study as self-reported as male or female. Educational credentials are defined operationally as the highest educational

degree acquired (M.D., doctorate, master's).

Signs are objective indications of a mental disorder (e.g., weight loss). Supervisees are counselor education students who work with clients as part

of their practica and internship (i.e., supervised field placement)

experiences.

Supervisors are counselors who are designated to "directly oversee the

professional clinical work of counselors" (Ethical guidelines, 1993). Symptoms are phenomena that accompany a mental disorder (e.g.,

insomnia).







18
Overview of Remainder of the Study

The introduction to the study has been presented in Chapter 1. The review of related literature is presented in Chapter 2, the methodology in Chapter 3, and the results of the study in Chapter 4. Chapter 5 includes a summary of the study, discussion of the results, implications, and recommendations.














CHAPTER 2
REVIEW OF RELATED LITERATURE

It is the same thing which makes us mad or delirious, inspires us
with dread and fear brings sleeplessness, inopportune mistakes,
aimless anxieties, absentmindedness, and acts that are contrary to
habit. These things that we suffer all come from the brain, when it is
not healthy, but becomes abnormally hot, cold, moist, or dry.
(Hippocrates, 460 377, B.C.) (Kellerman & Burry, 1988)

This archaic, biologically based explanation for mental illness

eventually replaced the ancient belief that divine or demonic interventions caused psychopathology. Historically, the etiology of mental disorders and how to codify and treat them has been controversial. "What is behind the symptom?" is a question the developmentalist Menninger posed in 1963 (p. 325), but one which mental health professionals had been debating for centuries.

Currently, researchers and clinicians propose that there are

physiological, psychological and sociological reasons for what is perceived as abnormal behavior and that identifying commonalties in mental disorders is the critical first step toward diagnosis and effective client treatment. Only a little more than a half century ago, however, "lunatics" were still being housed in grim "insane asylums" where, over the years, patients were regarded with varying degrees of insight, empathy, and concern.


19







20


In the early 1900s the organic psychiatrist Kraepelin's conviction that patients' mental problems were physiological (i.e., the consequences of diseased brains) influenced the treatment of people with mental disorders. Abnormal psychological phenomena were thought to be symptoms of those afflicted brains (Weckowicz, 1984).

Although much of what Kraepelin proposed is no longer accepted, his conception is alive in the current belief that biochemical imbalance is a contributing cause of some mental disorders. Several other explanations have been offered for why people think or behave in ways that are not customary in contemporary Western cultures. Alternative theories to Kraepelin's conjecture of the causes of mental disorders have been proposed and are bases upon which clinicians diagnose and treat clients. They include the following theoretical perspectives, briefly presented.

Psychodynamic. Psychodynamics has its basis in Freudian

psychoanalysis. Clinicians who practice psychodynamic therapy view mental disorders as being caused by individuals' desires, motives, and conflicts (Weckowicz, 1984).

Behavioral. Behaviorism was a response to psychodynamics and is rooted in the research related to learning. Theorists, such as Bandura (1977), advocate encouraging positive behavioral change as a means of eliminating maladaptive behaviors. They believe clients' problems are learned behaviors.







21

Cognitive. Therapies based on the concept that people act according to their perceptions are termed cognitive therapies. Cognitive therapists, such as Ellis (1979), theorize that clients' constructs, ideas, and expectations determine their behavior.

Developmental. Developmentalists, such as Erikson (1963) and

Menninger (1963), hold that psychopathology results from regression to a lesser stage of psychosexual, cognitive, or emotional development. They believe that human psychological development is directional and that difficulties develop if individuals are confronted with the demands of stages with which they are not yet equipped to cope (Crain, 1985; Weckowicz, 1984).

Sociocultural. Sociocultural theorists contend that the causes of mental disorders are either socially or biologically influenced and that the form mental disorder takes is determined by social factors. The various sociocultural constructions of mental disorders are based on the premise that individuals are links in society and therefore are part of a problem related to the whole (Weckowicz, 1984). This concept is found in family systems theory, such as that espoused by Minuchin (1974), which stresses the interaction and interdependency of family members.

Humanistic. Humanists assert that peoples' needs and aspirations do not vary across cultures and that all people have a need to move toward self-actualization. Rogers (1961) developed his response to psychoanalytic







22


theory by testing the philosophical hypothesis that clients' potential for positive change is inherent and that with proper facilitation, normal growth and development occur. He proposed that psychological disturbances result from three influences: (a) the clients' need for self-respect, (b) important others' estimations of clients, and (c) the dichotomy between what clients perceive themselves as being and what they need to be in order to have self-respect (Lynn & Garske, 1985; Weckowicz, 1984).

Application of Counseling Theories and Diagnostic Skills

I used to be astonished to see humane physicians going daily
round the wards of asylums, mere spectators of every form of
distressing coercion, without a word of sympathy. The first principle in the treatment of lunatics was fear, and the best
means of producing fear was said to be punishment. Thus torture
became more and more ingenious. (Conolly, 1856, p.13)

Regardless of differences in current mental health providers'

theoretical approaches to etiology and treatment of mental disorders, all mental health counselors, when interacting with clients, make inferences about their clients' relative states of well-being. Mental health counselors are expected to base their therapeutic actions on verifiable theoretical grounds. "[The counselor] will have some underlying rationale for what he does, some hunches about what different clients will do in different situations, and some general ideas about the counseling procedures that are apt to be effective" (Pepinsky & Pepinsky, 1954, p. iii).







23


Although researchers and practitioners are regularly testing

theoretical bases of mental health counseling, the fundamental objectives of counseling are constant. That is, mental health counselors routinely gather data about clients, diagnose the nature of problems, and provide appropriate professional interventions (Gambrill, 1990). Historical Perspective on Diagnosis

Prognostic diagnosis, based on a medical model, was favored by Kraepelin, and is still practiced when clinicians use clients' signs, symptoms, and history of their development to predict the course of events (Weckowicz, 1984). However, from the end of World War II until the mid1970s, a psychosocial (rather than a medical model) of mental disorders prevailed. During that time, the first Diagnostic and Statistical Manual of Mental Disorders (APA, 1952) was published. Because psychopathology resulting from wartime experiences had to be explained, descriptions of mental disorders, largely based on psychodynamic theory and following that language, reflected the influence of the psychosocial model. Former "functional" syndromes became "reactions" (Wilson, 1993).

Biologically oriented psychiatrists who advocated more research and a return to the medical model in the practice of mental health criticized the use of the psychosocial model to describe psychopathology. Academicians and other critics argued that if mental disorders were psychosocial in nature, they lay outside the province of psychiatrists and were more







24


properly in the social, political, and legal domains. Therefore, mental illness was a lie and psychiatric labeling a dangerous practice (Goffman, 1961; Scheff, 1966; Szasz, 1961). However, little empirical support for the argument that labeling alters self-concept has been produced. Morey, Skinner, and Blashfield (1986) pointed out that the behaviors that cause people to seek counseling were there before the diagnostic label, so it is obvious the label did not cause the clients' conditions.

By the 1970s, research dollars, and consequently the amount of research on mental disorders, had dwindled. President Jimmy Carter's commission on mental health bemoaned the lack of research data, while insurance companies reduced allowed services and payments (Wilson, 1993).

In short, a confluence of pressures on the profession [psychiatry]
from outside influences--the major providers of resources for
psychiatric research and treatment and the gathering momentum of
the antipsychiatry critique--threw the issues of diagnostic reliability, treatment outcomes, and accountability for those outcomes into bold
relief. Under these unfavorable professional conditions, the
psychosocial model, as the dominant organizing model of psychiatric knowledge and the source of many of these problems, would have to
be significantly altered, if not jettisoned altogether. (Wilson, 1993,
p. 403)

The DSM-III published in 1980 (APA), was considered by some to be a return to the medical model of diagnosing mental and personality disorders, although the majority of the disorders catalogued have no known organic cause (Wilson, 1993). The DSM-III (APA, 1980) was authored mainly by medical researchers, but because it was atheoretical and









descriptive, no explicit etiologic premise was represented. The American Psychiatric Association's Task Force on Nomenclature and Statistics decided the manual would be a "descriptive manual which would emphasize the assessment of easily observable symptoms" (Wilson, 1993, p. 405).

The DSM-IV (1994) amends and enhances the DSM-III-R ( 1987). Changes in the DSM were made after work groups reviewed published literature, reanalyzed data sets, and conducted field trials. The DSM-IV is not a philosophical departure, however, from the DSM-III-R. While the DSM-IV is coming into widespread use, the DSM-III-R will continue to be a reference for some clinicians. Also, for a while, clinicians will be working with clients diagnosed from criteria from the DSM-III-R.

There are three indications that diagnosis using the DSM-III (APA, 1980), (and likewise DSM-III-R, 1987, and DSM-IV, 1994), requires a more narrow focus than does a psychosocial model. First, the DSM-III, its revised version (DSM-III-R 1987), and DSM-IV do not attend to the unconscious mind; focus is on the visible workings of the mind. Second, symptoms are not placed within the context of individual development; rather, symptoms are viewed at a particular time in the client's life. Finally, clinical concerns are narrowed so that personality development, family dynamics, and social factors are deemphasized (Wilson, 1993).

The American Psychiatric Association acknowledged that the DSMIII (APA, 1980) diagnostic criteria were hypotheses to be tested, however,







26


"reification of the discourse of descriptive psychiatry through daily use of DSM-III and DSM-III-R [and, by extension, the DSM-IV (1994)] seems unavoidable" (Wilson, 1993, p. 408). Although the DSM-III was scrutinized and subjected to field trials by psychiatrists prior to its publication, the scientific quest to arrive at total diagnostic reliability in using the DSM-III was not achieved. Research has validated, however, the overall accuracy of the Diagnostic and Statistical Manual of Mental Disorders (Third Edition Revised) (APA, 1987) when clinicians are welltrained in interviewing processes and application of the DSM-II-R (APA, 1987) (Hohenshil, 1993; Morey, Skinner, & Blashfield, 1986). Current Diagnostic Standards

"Mental disorder" is an imprecise term that has been defined rather loosely as

a clinically significant behavioral or psychological syndrome or
pattern that occurs in a person and that is associated with present
distress (a painful symptom) or disability (impairment in one or more important areas of functioning) or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom.
In addition, this syndrome or pattern must not be merely an
expectable response to a particular event. Whatever its original
cause, it must currently be considered a manifestation of a
behavioral, psychological, or biological dysfunction in the person.
Neither deviant behavior nor conflicts that are primarily
between the individual and society are mental disorders unless the
deviance or conflict is a symptom of a dysfunction in the person, as
described above. (DSM-IV, 1994, p. xxi)

The approach of the DSM-IV (APA, 1994) is generally atheoretical regarding the etiology or pathology of mental disorders, except where the







27


etiology is known and is presented in the definition of the disorder (e.g., Alcohol Withdrawal, Vascular Dementia). It is expected that, over time, there will be an increase in the certainty of the etiology of disorders. Some disorders will be found to be caused by biological factors, and others by psychological, sociological, or interaction of those three elements (DSMIV, APA, 1994).

The DSM-III-R (APA, 1987) and DSM-IV (APA, 1994) present

descriptive criteria which many clinicians and researchers agree are clinical manifestations of the disorders named in the manual. A multiaxial system is used to integrate clients' biological, psychological, and social data and allows for a comprehensive first step in assessing clients' problems. The information on which diagnoses are based in the DSM-III-R, however, are incomplete in important correlate areas such as the outcome of the disorder, clients' family histories, and treatment responses (APA, 1987). A recognition that ethnic and cultural beliefs and practices influence thinking and behavior prompted the writers of DSM-IV to attend to ethnic and cultural considerations when discussing criteria for psychopathology. Counselor Preparation for Diagnosis of Clients

The objective of diagnosis depends on the clinician's purpose and

theoretical commitment to diagnosis. Diagnostic reliability is more likely, however, when clinicians are trained in, and use, universal diagnostic criteria such as those found in the DSM (APA, 1994) (Weckowicz, 1984).







28


Diagnosis provides (a) standard nomenclature that facilitates communication among clinicians and researchers; (b) descriptions of disorders or syndromes so that individuals can be compared, and therefore, allows categorization of disorders; (c) prediction of etiology, prognosis, and responses to treatment; (d) justification for treatment; and (e) a system for retrieving information when descriptive, etiologic, and treatment information is needed (Morey, Skinner, & Blashfield, 1986; Weckowicz, 1984).

Background for development of diagnostic skills

To diagnose clients and to provide them with effective treatment, counselors must see clients' symptoms in meaningful contexts. Mental health counselors are aided by knowledge of social, political, legal, and educational influences on clients. Quantitative and qualitative results of clinical assessments, conclusions reached through subjective observations of behavior, and knowledge of how clients are affected by family and other social dynamics all contribute relevant data about clients' conditions. In addition, familiarity with personality theory, human development theory, and awareness of recent scientific findings related to mental disorders are essential for broadly focused assessment and diagnosis.

CACREP-accredited mental health counselor education programs offer curricula that prepare counselor education students to diagnose clients. These courses are designed to enhance awareness of the forces







29


that may affect diagnosis. For example, graduating counselor education students are required to have successfully completed courses in

(a) Human Growth and Development--the study of individuals at their different maturation levels; (b) Social and Cultural Foundations--the study of individuals' concerns in a multicultural society; (c) The Helping Relationship--the study of counseling and consulting processes; (d) Group Dynamics, Processes, and Counseling--the study of the practices and processes in observing and counseling individuals in groups; (e) Career and Lifestyle Development--the study of individuals' career development concerns; (f) Appraisal--the study of assessment and evaluation; (g) Research and Program Evaluation--the study of scientific inquiry and statistical methods; and (h) Professional Orientation--the study of how counselors function in professional roles (Vacc & Loesch, 1994; Wittmer, 1993).

Counselor education students in CACREP-accredited programs who are preparing to be mental health counselors also learn the procedures for differential diagnosis using criteria in the latest Diagnostic and Statistical Manual of Mental Disorders (APA, 1994). Counselor education students are instructed in the use of the DSM (APA, 1994) because it is the "most widely accepted system for describing and diagnosing mental disorders" (Seligman, 1990, p. 3).







30


Supervised field experiences

Each entry level student in a CACREP-accredited counselor

education program must complete two supervised practica that each total 100 clock hours and a supervised internship of 600 hours (Wittmer, 1993). Experiences in field placements are intended to provide students with the opportunities to develop and refine their diagnostic skills. Supervision in field experiences

Counselor education students engaged in mental health counseling practica and internships are under the charge of clinical supervisors who help supervisees develop assessment, diagnostic, and counseling skills. Counselor education students are relatively inexperienced clinicians, and therefore, in order to function effectively, they work to develop the aforementioned skills as well as the confidence that they can adequately fulfill their other responsibilities as counselors. It is also helpful for supervisors to provide feedback to counselor education students about how personal emotional reactions may affect their ability to diagnose. Borders and Leddick (1987) wrote that the responsibilities of supervisors, after having established supportive relationships with counselor education students, are to facilitate growth through teaching, identifying students' needs, and giving valuable feedback to students.







31


Conditions of supervision

Bradley (1989) reported that "warmth, acceptance, understanding, respect, and trust" (p.37) are qualities supervisees think are important in their supervisors. It was unknown, however, to what extent counselor education students consider their supervisors credible or what factors are associated with variations in attributed credibility. Attribution theory can be used not only to evaluate action (e.g., diagnosis) based on the observer's perceptions, but also to predict (trust) that the action will be repeated. Hovland, Janis, and Kelley (1953) wrote that trustworthiness and expertness are the two components of credibility. Trustworthiness was defined by Strong (1968) as being a function of the communicator's (a) reputation for honesty; (b) social role, such as professional status; (c) sincerity and openness; and (d) perceived lack of reason for personal gain. Strong (1968, p. 216)) wrote that a communicator is perceived as expert (i.e., "a source of valid assertions") if there exists (a) objective evidence of specialized training [e.g., advanced degrees], (b) behavioral indicators of rational and knowledgeable arguments and confident presentations of views, and (c) consensus among others that the individual is an expert.

Studies have shown that trustworthiness and expertness are

characteristics that counselor education students value in supervisors. Beginning counselor education students, in particular, feeling less efficacious than more experienced counselors, indicate they want







32


supervisors whom they can trust to provide reliable (i.e., expert) direction (Carey, Williams, & Wells, 1988; Hillerbrand & Claiborn, 1990; Tracey, Ellickson, & Sherry, 1989; Wiley & Ray, 1986).

Counselor education students' need for structure in supervision is described by Stoltenberg (1981) and Stoltenberg and Delworth (1987) as being related not only to their level of skill and knowledge of theories, but to their level of development (i.e., qualitative differences among their needs, behavioral motivation, and resistance at different stages). Stoltenberg's (1981) theory was tested by Wiley and Ray's (1986) study of 107 dyads of supervisors and supervisees from nine university counseling centers. Wiley and Ray (1986) determined that supervisees' needs differ according to developmental level during training, and that developmental level and amount of training are not interchangeable concepts.

Hillerbrand and Claiborn (1990) studied the needs of supervisees at different training levels. They discovered that, often, novice counselors with limited clinical and educational backgrounds cannot identify the features inherent in clients' problems, so the nature of clients' problems are not easily recognized. Conversely, more experienced and better educated counselors, who felt more confident about their skills, perceived cases more clearly than did novice counselors.

Tracey, Ellickson, and Sherry (1989) studied counseling students' preferences for supervision style. They found that 40 novice counseling







33


students who had completed up to one semester of practicum training preferred a more authoritative and prescriptive supervisory experience than did 38 more advanced counselor education students who had completed two or more semesters of practicum but had not yet enrolled in internships.

Beginning counselor education students not only desire expert

direction in supervision, but appreciate supervisors whom they can trust to help them manage anxiety in clinical situations (Carey, Williams, & Wells, 1988). Uncertainty about case conceptualization and client diagnosis could be a source of trepidation to novice counselors. Counselor education students could benefit from instructional supervision in problem-structuring skills. Knowing how to clarify clients' problems enables counselors to redefine ill-structured problems and therefore be better able to arrive at deliberated diagnoses.

Counselor education students sometimes doubt their ability to be effective counselors. Sipps, Sugden, and Faiver (1988) found that selfefficacy in counseling among counselor education students rose directly as the level of graduate school training, with the exception of the second year. The first year students' efficacy levels were significantly lower than the efficacy levels of third and fourth year students. The second year students' efficacy levels were significantly lower than the efficacy levels of the first, third, and fourth year students. Sipps, Sugden, and Faiver (1988) attributed the second year efficacy score as reaction to counseling







34


students' relatively naive first year attempts at counseling. Johnson, Baker, Kopala, Kiselica, and Thompson (1989) found that counseling students' efficacy increased with training over an eight-week period. In conducting their study, Johnson, Baker, Kopala, Kiselica, and Thompson's (1989) subjects engaged in ten classroom contact hours of case conceptualization that included exploration and clarification of a client's problem.

Meeting counselor education students' need for expert and

trustworthy supervision and instruction in diagnosis appears to be justified. Students' counseling efficacy and effectiveness presumably would improve with attention to diagnostic skill-building. Confidence in supervisors

Principles of person perception and attribution theory suggest that clinicians' reasoning is subjective and imprecise. Additionally, although the DSM-IV (APA, 1994) criteria are useful because they are predicated on observation and research, there is no question that diagnoses based on the DSM-IV descriptive criteria are limited. Also, Faust (1986) warned of the folly of failing to acknowledge that clinical decisions are influenced not only by diagnosticians' perceptual selectivity and the frequency of occurrence of certain mental disorders, but by diagnosticians falsely associating frequently occurring signs with certain conditions with which the signs do not covary. Diagnosticians may also seek confirming evidence







35

for their hypotheses and ignore disconfirming evidence. Faust (1986) also offered the reminder that diagnosticians have cognitive limitations. Again, to what extent do counselor education students have confidence that their supervisors' opinions are credible? To what extent are counselor education students certain that their supervisors will be worthy role models from whom diagnostic skills can be learned? Kelley's (1973) attribution theory, which was built on Heider's (1958) hypotheses, was used to infer that counselor education students perceive degrees of credibility in their supervisors based on levels of educational attainment and genders of the supervisors and proportion of counselor education program hours completed that are required for graduation and genders of counselor education students.

Relevance of Theoretical Framework

This study was an attempt to use the conceptual framework of

attribution theory to determine the extent to which counselor education students attribute credibility to clinical supervisors. Attribution theory is a rather loosely structured set of hypotheses, rather than a tightly organized theory.

In 1958, Heider, one of the pioneers in attribution theory, set forth his ideas about how people perceive each other, how they feel about each other, how they react to each other, and what they expect from each other. Heider (1958) did not believe that explaining the workings of the







36


unconscious mind was essential when describing the tasks of perception and attribution. Rather, he believed that people attribute characteristics to other people by relying on common sense, a key component of which is person perception.

Person Perception

Individuals construct their cognitive reality by starting with what Dewey (1984, p. 249) called the "unanalyzed whole." People allow their interests to fuel new thoughts and new understandings. They then associate those ideas with previous thoughts, interests, or values (i.e., "dispositional properties" ) (Heider, 1958). When original thoughts, interests, and values ally with new thoughts and understandings, new inferences and judgments are formed (Kelly, 1963; Sherman, 1985).

It is in the act of person perception that individuals make inferences or judgments about other people. Observations of others' characteristics are incorporated into one's reality in the form of social cognitive responses to the people observed (Tagiuri,1958).

How are individuals motivated to select those stimuli they perceive from the constant inundation of stimuli to which they are exposed? Tagiuri (1958) contended that three factors, which can be viewed singularly or in interaction, influence person perception: (a) the attributes of the person being perceived, (b) the characteristics of the perceiver, and (c) the nature of the situation in which the association occurs. Tagiuri's (1958) position







37


is consistent with that of Postman, Bruner, and McGinnies (1976) who argued that individuals' values, interests, and survival needs determine which stimuli will be perceived. Tagiuri's (1958) reasoning also is consistent with Heider's (1958) proposal of how individuals form perceptions. Heider (1958) wrote that a perception is the result of mediation between an observer and an object. In the act of person perception, the other person, an individual who cannot be experienced in totality by the observer, is the object (distal stimulus) toward which the observer's perception (proximal stimulus) is directed. Mediation occurs between the distal stimulus and the formation of a percept. Mediation involves psychological functioning in the observer that leads to a response in the form of a perception of the object. "The proximal stimulus, this message from the object, is formed and determined not only by the object, but also by facts that are really independent of the object and that therefore falsify or cloud the message" (Heider, 1980, p. 5). This perception, albeit constructed in part by the observer, becomes a part of the observer's reality.

Common Sense, Common-Sense Psychology, and Attribution

Heider (1958, p. 2) described common sense as an "unformulated" and "vaguely conceived" awareness of a situation that, because individuals understand themselves and others, allows those individuals to interact with


j







38


others. Common sense also could be defined as a system of beliefs that are customarily accepted to be true in everyday life (Hewstone, 1983).

Common-sense psychology makes use of the "unformulated or halfformulated knowledge of interpersonal relations as it is expressed in our everyday language and experience" (Heider, 1958, p. 4). Common-sense psychology explains attributory behavior by asserting that individuals' perceptions of themselves and their environment, and the individuals' resultant reactions concern "perceptions about the causes of behavior and not the actual causes of social behavior" (Snyder, 1976, p. 54).

Heider (1958) wrote that common-sense psychology has scientific value for two reasons. First, it provides the principles used in social perceptions and reactions (i.e., individuals interpret other people's actions, and from there predict what those people will do). Second, common-sense psychology contains subjective truth based on data from individuals' environments. That is, it relies on the obvious. Scientific Thinking and Attribution

Kelley (1973) acknowledged that Heider's (1958) writings were the principle source of the hypotheses in attribution theory; however, Kelley (1973) apparently thought effective understanding of attribution theory involves gathering clarifying information from the attributer and, therefore, that Heider's (1958) theses insufficiently explained many phenomena. Kelley (1973) offered an addition to Heider's (1958) work, which he called







39


his ANOVA model of the attribution process. Kelley's (1973) covariation theory of attribution proposes that people act as social scientists when constructing their reality. Kelley (1973, p.108) stated that it is possible to infer cause of an event from observation of covariance. "An effect [dependent variable; e.g., attribution of credibility] is attributed to the possible cause(s) [independent variables; e.g., characteristics of entities observed, nature of circumstances, characteristics of observers] with which, over time, it covaries."

Kelley (1973) claimed that there are three classes of causes for

attribution: (a) the entities that are present (e.g., clinical supervisors), (b) the time or circumstances in which the action occurs (e.g., diagnostic action by clinical supervisors in practica/internships settings), and (c) the perceptions and characteristics of the observers who are present (e.g., counselor education students).

How does an individual determine if her/his perceptions are valid foundations upon which to base attributions? Kelley (1973) maintained that the test for validity of a perception is (a) consensus of perceptual response among observers of the event, (b) consistency of observers' responses over time or modality, (c) the presence of a distinctive stimulus that initiated the perception, or (d) a combination of these factors.

Various combinations of high and low distinctiveness, high and low consensus, and high and low consistency lead to attributions being granted







40


to a single cause of an event or to a combination of causes. For example, high distinctiveness, high consensus, and high consistency lead to attribution to an entity; whereas, low distinctiveness, high consensus, and high consistency lead to attribution to the observer and the entity (Jones, 1990).

The purpose of this study was to determine, through responses to a bogus profile of a troubled client, the extent to which credibility is granted to clinical supervisors (entities) by counselor education students (observers). The covariation of the causes (independent variables) with the effect (dependent variable) were examined. The independent variables were supervisors' educational credentials, supervisors' gender, students' gender, and proportion of counselor education program hours completed that are required for graduation by students. The dependent variable in this study was the degree of attribution of credibility granted to clinical supervisors by counselor education students. The tests for validity of the attribution were consensus (agreement among subjects regarding level of attributed credibility), reaction to distinctive stimuli (supervisors' diagnosis portrayed in bogus profile), and consistency (agreement among students in various counselor education departments with the assumption that these attributions have been made previously).

Kelley's (1973) ANOVA model of attribution was the basis for explaining the extent to which counselor education students grant







41

credibility to clinical supervisors based on their educational credentials and gender and counselor education students' proportion of counselor education program hours completed that are required for graduation and gender. However, quantitative analyses were not sufficient to explain subjects' relevant thoughts and experiences, the "complex whole" (Dewey, 1984, p. 246) being examined in this study. For that reason, the data were subjected to qualitative analyses and interpretation as well. Fischoff (1976) offered the intriguing suggestion that people are not intuitive scientists as Kelley (1973) wrote, but rather are intuitive historians. He wondered if it might be valuable to produce "an integrated psychology of predictive and explanatory behavior that accommodates the historians' observations, the philosopher's formalizations, and the psychologists' and the sociologists' theories and empirical findings" (Fischoff, 1976, p. 445).

Elaboration on Need for the Study

Research on Perceptions and Attributions of Parties in Counseling Process

The act of counseling is not done in a vacuum with interaction only between clients and counselors who are unaffected by extraneous elements. The proceedings are mediated by biological, psychological, and social forces emanating from and acting on clients and counselors who are frequently supervised. For this reason, parties immediately involved in a counseling experience (i.e., clients and counselors) form perceptual inferences about those with whom they are interacting.







42


Clients' view of clinicians

Fundamentals of attribution theory, although not always identified as such, have been used to describe the circumstances in which clients attribute credibility to counselors. The impact of distinctive stimuli is apparent in a study in which Holland, Atkinson, and Johnson (1987) tested 209 undergraduates and found that counselors who expressed opinions about sexual behavior that were similar to those of students were perceived as credible. There were no findings that likenesses in gender between counselors and respondents affected the results. Ponce and Atkinson (1989) found in a study of 169 Mexican-Americans that counselors of the same ethnicity as the subjects were perceived as more credible than those who were not.

The result of consensus of opinion is illustrated in a study of 485 high school students who attributed credibility to particular counselors. Littrell, Caffrey, and Hopper (1987) reported that counselors' reputations influenced the subjects' precounseling preferences for counselors and the subjects' perceptions of credibility of the counselors.

Does consistent evidence of counselor expertise promote counselor credibility? Strong (1968) stated that in addition to diplomas, certificates, and professional libraries, certain behaviors by counselors are indicators of their expertise. The initial interview with clients serves not only diagnostic purposes, but also affords counselors the opportunity to effectively







43

communicate their expertise by discussing the roles and responsibilities of clients and counselors as well as the process of counseling and problem solution. In 1981 Sue wrote that if clients perceive counselors as lacking credibility, clients may be reluctant to trust them. Counselor education students may learn from their supervisors how to convey a perception of credibility.

Clinicians' view of clients

Research has linked attribution theory to clinicians' judgments of clients' mental states. Since the cognitive-behavioral approach to counseling gained importance in the 1960s, attribution theory sometimes has been a foundation for study of clinicians' judgments (i.e., diagnoses, predictions, and etiology of clients' conditions) (Lopez & Wolkenstein, 1990) and for the study of clients' perceptions of their mental states. Success in using attribution theory to explain clients' mental states led to the development of attribution therapies in which clients' perceptions of causal events for their problems are used in diagnosis (Graham & Folkes, 1990). The effectiveness of attribution therapy has not been tested in the field, but attribution theory has been a foundation for understanding and alleviating maladaptive conditions such as depression and anger (Forsterling, 1990).







44


Clinical Training of Counselor Education Students

Counselor education students prepare to work in settings where

diagnoses generally are required and where knowledge of mental disorders enables counselors to treat clients competently. These sites include schools, businesses, and other places where most counselors are now employed and/or are projected to be employed (i.e., community mental health agencies, inpatient and outpatient treatment facilities, and individual private practices) (Ritchie, Piazza, & Lewton, 199 1; Vacc & Loesch, 1994).

It may be assumed that a growing impact of mental health counselors on the mental health profession would be observed if descriptive data gathered in 1982 1983 were made current. In 1982 1983, the majority of people who sought help for relationship problems involving marriage, family, job, peers, school, and community saw psychologists and social workers. The same survey showed that 38% of psychologists' weekly hours and 32% of social workers' weekly hours were spent working with clients diagnosed as having neuroses, anxiety, and personality disorders (Knesper & Pagnucco, 1987).

The DSM-IV (APA, 1994) serves as a basis for common lexicon

when mental health providers discuss assessment, diagnosis, and treatment. DSM-IV language also is a basis for referring clients who counselors prefer not to assist and a basis for third party payments for the rendering of







45


mental health services. Government and private health insurance companies require diagnostic codes on health insurance claim forms. Clinicians are called upon to testify in court cases where expert opinions are sought. An expert may be asked to testify about specifics such as diagnoses of individuals, and are expected to be familiar with prevailing norms (Faust & Ziskin, 1988). Finally, DSM-IV criteria may be used for testing the effectiveness of various treatments.

Many counselors believe that the process of diagnosis provides a focal point for consideration of clients' (a) patterns of behavior, (b) structural components such as self-image and temperament, and

(c) functional components such as cognitive style and psychodynamics. Contextual factors also are considered. These include situations in which problems arise for clients (Kottler, 1991). Counselor education students are expected to learn to incorporate all of these types of data into meaningful and useful information in the clinical reasoning process. Clinical reasoning

Elstein and Bordage (1988) described the ideal clinical reasoning process as composed of four stages: (a) cue acquisition, wherein clients' histories are taken, psychological testing is done, and the influence of possible medical problems is considered; (b) hypothesis generation, wherein alternative explanations for clients' problems are considered and either accepted or ruled out; (c) cue interpretation, wherein data are







46

analyzed; and (d) hypothesis evaluation, wherein hypotheses are confirmed or not confirmed, and if necessary, new hypotheses formed until verification is accomplished and diagnoses are determined.

The development of clinical reasoning skills is ongoing during the

education of counselor education students and is vital to their being able to offer effective counseling. The American Psychiatric Association warns that the diagnostic criteria found in their diagnostic manual are only guidelines and that the proper use of the manual requires "familiarity with the description of the diagnostic categories and with definitions of technical terms" (APA, 1987, p. v), as well as "specialized clinical training that provides both a body of knowledge and clinical skills" (p. xxix). Educational preparation for diagnosis

Counselor education students probably are receiving inadequate

instruction in diagnosis and treatment planning. Of the 146 respondents to Ritchie, Piazza, and Lewton's (1991) survey of counselor education programs, 31 taught nothing about the DSM-III-R (APA, 1987), and only 34 of the 146 programs taught the use of the DSM-III-R as an independent course.

Professionals in counseling are calling for counselor educators to be more aware of the need to adequately prepare counselor education students to meet the clinical demands of the workplace. Theodore R. Remley, Jr., recent Executive Director of the American Counseling Association,







47


advocated that the ACA Governing Council recommend that "CACREP [require] core work in diagnosis and treatment planning of emotional and mental disorders, regardless of specialty" (1993, p. 4). His opinion was echoed by Gail Robinson, past president of the American Mental Health Counselors Association:

I support a core [curriculum] including courses in diagnosing
and treating mental and emotional disorders. All practicing
counselors should understand the continuum of mental health care and all must have the knowledge and skills to diagnose, treat, and
refer clients with mental and emotional problems. (1993, p. 16)

Waldo, Brotherton, and Horswill (1993) described a counselor education program in which School Counseling, Marriage and Family Therapy, and Mental Health Counseling tracks incorporated instruction in the use of the DSM-1II-R (APA, 1987) into their curriculum on the premise that it is complementary, rather than antithetical, to all three modalities of counseling. They explained that school counselors' humanistic perspective is found in the DSM-III-R focus on clients' experiences and self-reported symptoms. Students can be perceived as individuals worthy of respect who are having problems that can be alleviated.

Hohenshil (1993) has noted the trend toward training school counselors to diagnose using the DSM-III-R (APA, 1987). School counselors need to be able to converse knowledgeably with school psychologists, special education teachers, and referral agencies. School counselors should be aware of the circumstances in which students need to







48


be referred to mental health providers who can offer clinical help that is not available in schools.

Therapy based on a systemic philosophy is consistent with use of the DSM (APA, 1994) to identify clusters of symptoms and psychosocial stressors. Rather than focusing solely on etiology of disturbances, viewing individual complaints as symptomatic of a dysfunctional family system is congruent with the philosophy of Marriage and Family therapists (Waldo, Brotherton, & Horswill, 1993).

Mental health counselors are encouraged to use diagnoses based on DSM-IV (APA, 1994) criteria to design effective treatment plans. Attention to all five axes in a DSM-IV based diagnosis enables mental health counselors to treat clients holistically.

The importance of teaching proper use of the DSM-IV (APA, 1994) should not be ignored. Fong (1993) hypothesized that counselor education students have difficulty diagnosing accurately and the reason is because they are frequently taught the content of the diagnostic manual (i.e., signs, symptoms, diagnostic codes) but not the process of diagnosis. The process of diagnosis involves assessing clients by observing their behavior, interviewing them (and perhaps their family members), examining their mental status, and, finally, generating hypotheses about the clients' conditions by noting patterns such as clients' psychological and physiological abnormalities and consulting the DSM (APA, 1994) decision







49

trees. Further, Fong (1993) provided the analogy that teaching diagnostic manual content, but not diagnostic process, is akin to expecting counselor education students to be competent counselors after teaching them counseling theories, but not techniques.

Furlong and Hayden (1993) suggested using available computer

programs to enhance the teaching of diagnostic skills. They proposed that counselor educators make available to their students programs such as DTREE (First, Williams, & Spitzer, 1988), wherein students could become familiar with using the diagnostic manual decision trees, SCID SCORE (First, Gibbon, Williams, & Spitzer, 1991), in which students would use data from the Structured Clinical Interview for DSM-III-R (APA, 1987) to frame diagnoses, and the DICA-R (Reich, Weiner, & Herjanic, 1990) to arrive at diagnoses for children and adolescents. Clinical Practice of Supervisors and Supervisees

Knowledge of beginning counselors' attitudes toward their

supervisors' diagnostic credibility could affect client care and management procedures in staffing facilities. Carey, Williams, and Wells' (1988) study of 3 1 counseling students in their first practica showed that trustworthiness (a dimension of credibility) in supervisors was one of the most esteemed characteristics of supervisors. Their findings were compatible with the earlier study of Heppner and Handley (198 1) who found that trustworthiness and expertness (i.e., credibility) correlated with







50


supervisees' satisfaction with supervision. Because novice counselors perceive the need to trust their supervisors, and because counselor education students are expected to consult with supervisors while providing counseling services, client care could be imperiled if the people who assess, diagnose, and suggest treatment modalities were viewed as incompetent diagnosticians by counselor education students. Counselors are expected to have logical grounds for the diagnoses they make and the treatments they offer. If counselor education students were to believe that the judgments of their supervisors were untrustworthy, counselor education students could experience ethical dilemmas and emotional conflicts.

Counselor education students who have received inadequate

instruction in diagnostic procedures could possibly jeopardize client care. They could fail to have a reasoned approach to treatment or referral because they might lack reliable and relevant information, and could fail to understand the need to be informed. They also could compromise the integrity of the mental health care system by failing to act knowledgeably and responsibly, or by unquestioningly following the lead of supervisors or other diagnosticians, none of whom would be infallible.

In addition, assigning students to practicum and internship sites and to supervisors could be influenced by results of this study. The feasibility and desirability of using doctoral students or mental health professionals not associated with the counselor education department as clinical







51

supervisors also could be of interest to faculty in departments of counselor education.

Perceptions of Credibility of Clinical Supervisors

Counselor education students' supervisors are often unseen

influences in the counseling relationship. Supervisors are sources of information for students to use in assessing, diagnosing, and treating clients. Worthington (1984) discovered a relationship between supervisors providing acceptance and support for supervisees as well as teaching them case conceptualization, intervention, and other skills to perceived competence of supervisors. The sensitivity to evaluation felt by counselor education students, particularly during their first practica and their internships (Worthington, 1984), warrant providing supervisory assistance and encouragement to those students. Worthington (1984) proposed that it is during the first practica that counselor education students first view themselves as counselors, and it is during their internships that they first perceive themselves as professionals. It is beneficial to counselor education students during these times of professional identity formation for supervisors to be aware of the progressive nature by which students become counselors.

On what basis can it be assumed that if counselor education students find their clinical supervisors to be credible diagnosticians the students will conform to the paradigms preferred by their supervisors? Social









psychologists have presented an answer to this question. Tedeschi and Linkskold (1976) wrote that social influence is the impact one person has on another to the extent that a change in behavior or attitude consequently is produced in the target person. They further explained that people perceived as experts are capable of encouraging conformity in others. This opinion is supported by the theory of reciprocal determinism in which Bandura (1978) implied that counselor education students, perceiving their supervisors as credible and themselves as needing guidance in clinical matters, would seek direction from their supervisors. If the students were satisfied with the outcome, they would incorporate the knowledge into their repertoire of skills and information.

Another position was offered by Heppner and Handley (1981). In a 15-week study, they tested students in their first practica and found that students' perceptions of supervisor expertness and trustworthiness correlated with satisfaction with the supervisory experience; however, these perceptions were not highly correlated with the students' ratings of supervisor impact. Heppner and Handley (1981) argued that expertness is necessary, but not sufficient to influence students' counseling behavior. They go on to suggest that other variables such as supervisors' legitimate power, their ability to effectively produce change, and supervisees' characteristics also may be correlated with supervisor impact.









Doctoral students

Heppner and Handley's (1981) study of 33 first practicum students' perceptions of their 20 supervisors whom the subjects knew to be advanced doctoral students revealed that supervisors' trustworthiness was rated as more meaningful than expertness. It is important to recall from the discussion above of Heppner and Handley's (1981) study that students' perceptions of supervisor expertness and trustworthiness correlated with satisfaction with the supervisory experience. However, these perceptions were not highly correlated with the students' ratings of supervisor impact on personal and professional behavior. Therefore, it can be concluded that credibility of doctoral students is necessary for satisfactory supervisory experiences for entry-level students, but it does not necessarily follow that supervisory experiences for these supervisees will be substantially productive.

Poidevant, Loesch, and Wittmer (1991) surveyed 105 doctoral

students in 13 CACREP-accredited programs. The object of the study was to determine the professional work settings and work activities preferred by counselor education students and how their perceived levels of efficacy related to their preference for work activities. Of thirteen possible choices of work settings, the respondents showed the greatest preference for college or university teaching and private practice. The respondents also showed a preference for four work-related activities: (a) counseling,







54


(b) teaching and training, (c) consultation, and (d) supervision and training. Students chose supervision and training as the area in which they had the greatest efficacy, followed by the fields of teaching and training, consultation, and counseling.

"Doctoral students' perceptions of self-efficacy are likely to have an impact on their professional functioning relative to setting, activities, and competence" (Poidevant, Loesch, & Wittmer, 1991, p. 290). All of the preferred jobs and all of the preferred work activities indicated by the respondents in Poidevant, Loesch, and Wittmer's (1991) study potentially require a thorough knowledge of assessment, diagnosis, and treatment. Apparently counselor education doctoral students think they have acquired the skills necessary to properly meet the needs of clients, students, consultees, and supervisees. However, it was unknown to what extent supervisees find doctoral students to be credible diagnosticians. Doctorate-holding psychologists

Worthington (1984) learned from a nationwide survey of 237 counseling supervisees who were supervised by doctoral students in psychology and by Ph.D. holding psychologists that doctoral students and psychologists were not perceived differently in terms of being competent, providing satisfactory supervisory experiences, or having equal impact on supervisees' counseling ability. In addition, of 48 supervisor behaviors (e.g., established goals and provided feedback ), only one, the use of









humor, appeared as a significant difference between the groups. Humor was used more often by more experienced supervisors than by less experienced supervisors.

In a study that determined the differences in how 60 undergraduate female students in psychology classes attributed credibility and attractiveness based on fee charged and title to counseling psychologists and professional counselors with master's degrees, Conoley and Bonner (1991) found there was no significant main effect for variations in professional title. Apparently, perceptions of credibility of psychologists and counselors based on title among undergraduate female psychology students do not differ. Some students who enter counselor education programs come from this population. Doctorate-holdiny- faculty members

In a 1985 study, counselor education and counseling psychology

doctoral students were asked to complete anonymous questionnaires rating their current clinical supervisors. Supervisors who were rated in this study by Worthington and Stern were faculty members or doctoral candidates associated with that department or with affiliated practicum sites. Results showed that supervisees believed that, regardless of supervisor status, the supervision relationship improved as supervision progressed, and that faculty member status was not a factor in supervisees' ratings of







56


supervisors' effect, competence, and the quality of the supervisory relationships.

Psychiatrists

Although subjects in Truatt and Bloom's (1982) study were not

counselor education students, the results can be extrapolated. Truatt and Bloom (1982) studied university students' ratings of credibility for psychiatrists, clinical psychologists, and counselors. Although aspects of the methodology used in Truatt and Bloom's study are suspect and therefore not useful for general comparison among psychiatrists, psychologists, and counselors, it is interesting to note that subjects attributed the greatest credibility to psychiatrists because of their title of "doctor." The clinical psychologist and the counselor were not so denoted in the study.

Greenberg and Wursten (1988) questioned if psychiatrists are

perceived as more credible and more persuasive than psychologists when they provide mental health testimony in court. Students in an introduction to psychology course rated the psychiatrists as more credible than the psychologists. The authors admitted that credibility was the only variable studied, which enhanced the internal validity of their study, but because they did not test attractiveness, likability, and dynamism, all characteristics that could influence a judge or jury, the study's generalizability was limited.







57

Although the subjects in Greenberg and Wursten's (1988) study were not counselor education students, the results may be indicative of the thinking of the general public and therefore of counselor education students as well. The limitations of their study apply to the present study in that the generalizability of the results are constrained by lack of attention to other variables such as Greenberg and Wursten (1988) noted and which could affect supervisees' perceptions of credibility. Genders of supervisors and supervisees

Because recent research on the influence of gender on the

perceptions of the supervisory process by supervisees has been scant, it was unknown if perceptions of the degree of credibility as diagnosticians of female clinical supervisors is different from perceptions of credibility as diagnosticians of male clinical supervisors. Therefore, results of this study may be significant to departments of counselor education faculty and to mental health facility administrators when assigning supervisors to supervisees.

Worthington and Stern (1985) addressed the issue of gender in the supervisory relationship. They found that male supervisees believed they had better relationships with supervisors, regardless of gender, than did female supervisees. Supervisor supervisee dyads matched by gender received more favorable ratings by supervisees than did unmatched dyads.







58


Schiavone and Jessell (1988) studied 86 master's level counselor education students. Their study revealed that during initial interactions, counselor education students' perceptions of supervisors' expertness and competence does not differ as a function of the gender of supervisors or supervisees.

Proportion of program hours completed by supervisees

Apparently no research related to the number of program hours completed by counselor education students and their attitudes toward supervisors' diagnostic credibility has existed. Results from this study however, reinforced Stoltenberg and Delworth's (1987) theory of development in counselors and Usher and Border's (1993) study of counselors' preferences for supervisory style. It is known that supervisees' needs differ according to their development as counselors and that developmental level and amount of training are not interchangeable concepts (Wiley & Ray, 1986).

Stoltenberg and Delworth (1987) outlined the changes in counselor characteristics as counselors progress through three levels of development. If Stoltenberg and Delworth (1987) are correct, and if the supervisory conditions present for counselors during this progression meet students' needs, then it would be expected that beginning counselors, (i.e., counselors in Stoltenberg and Delworth's Level I) would rely to a large extent on supervisors' diagnoses. Stoltenberg and Delworth's (1987) belief







59


was that by the time counselors move beyond Level 1, through Level 2, which is characterized by "fluctuating motivation, striving for independence, becoming more self-assertive and less imitative" (p. 70) and reach Level 3, they can function fairly independently.

Usher and Borders' (1993) study supports Stoltenberg's (1981) thesis that different supervisory experiences are more likely to be perceived as more beneficial by practicing counselors than by counselors who have developed less professionally. Using the Supervisor Emphasis Rating Form-Revised (SERF-R), Usher and Borders (1993) learned that professional counselors who are not school counselors are more interested in having assistance in conceptualizing cases and choosing interventions than in learning counseling skills and behaviors. Therefore, it would be expected that counselor education students who were completing their final clinical field experiences would seek less direction from supervisors, and would be more likely to question debatable diagnoses and feel comfortable in confirming those diagnoses with which they have no argument.

Summary

It is currently believed that mental disorders are caused solely, or in combination, by biological, psychological, and sociological factors. Therefore, it is prudent to codify mental disorders and, using diagnoses and theoretical approaches to etiology and counseling as guides, design appropriate treatment plans for clients. Since 1952, The Diagnostic and







60


Statistical Manual of Mental Disorders (APA) has been the authoritative source for diagnostic criteria. Ongoing controversy has surrounded the philosophical approach to diagnosis, and the medical and psychosocial models (both at times favored by psychiatrists) have been incorporated into a descriptive, generally atheoretical, biopsychosocial model currently used in the DSM-IV (APA, 1994).

Codifying diagnoses of mental disorders affords clinicians and researchers a common means to communicate among themselves, and provides a universal and predictable framework from which to diagnose, group, and treat clients. A thorough understanding of clients' psychological, emotional, cultural, and developmental stages and needs are necessary to effectively diagnose and counsel clients.

Counselor education students engage in coursework and field experiences to gain the expertise needed to be effective counselors. During their practica and internships, counselor education students receive assistance and direction from qualified clinical supervisors. Research has shown that expertness and trustworthiness, the components of credibility, are characteristics counselor education students value highly in their supervisors. However, it was unknown to what extent counselor education students consider their clinical supervisors to be credible diagnosticians, people whom they can trust to be reliable role models and tutors. In the present study, based on Kelley's (1973) ANOVA model of attribution







61


theory, the extent to which counselor education students grant credibility to clinical supervisors based on supervisors' educational credentials and gender, and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation was investigated.

Male and female counselor education students who had completed

varying proportions of their entry-level counselor education program hours were subjects in this study. Counselor education students' perceptions of the diagnostic credibility of male and female counselor education doctoral students, male and female psychologists with doctoral degrees, male and female counselor education department faculty members, and male and female psychiatrists acting as clinical supervisors were subjected to analysis of variance and post-hoc follow-up analyses of significant interaction. Qualitative analyses of subjects' replies were done to achieve a broader understanding of the subjects' views on clinical supervisors' diagnostic credibility.

Researchers have examined perceptions among parties in the

counseling process. However, no known previous study had examined the extent to which counselor education students attribute credibility to clinical supervisors. Therefore, the results of this study may be used in clinical training of counselor education students, staffing practices of







62

mental health facilities, and assignment of clinical supervisors in counselor education departments.














CHAPTER 3
METHODOLOGY

The extent to which counselor education students consider their

clinical supervisors to be credible diagnosticians was unknown. Therefore, in this study, Kelley's (1973) ANOVA model of attribution theory was used as a basis to determine the degree of diagnostic credibility counselor education students grant their clinical supervisors.

Delineation of the Variables

Supervisors' types of credentials (i.e., M.D., doctorate-holding

counselor education faculty, doctorate-holding psychologist, or doctoral student holding master's degree in counselor education) and gender and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation (i.e., <1/3, 1/3 to 2/3, or >2/3) were the independent variables examined in this study. The dependent variables were the scores obtained from a Likerttype index of subjects' approval of clinical supervisors' diagnoses. Additionally, through analyses of subjects' written narratives, a qualitative description of subjects' reasons for arriving at approval scores was framed.


63





64


Population

Responses that met criteria for use in this study came from subjects in nine of the 27 counselor education departments in the United States that offer both CACREP-accredited doctoral-level programs (i.e., Ph.D. and Ed.D.) in Counselor Education and Supervision and CACREPaccredited entry-level (i.e., master's degree) programs in Community Counseling or Mental Health Counseling. Representatives of three other programs agreed to participate in this study, but failed to return any of the 100 instruments sent to them, even though follow-up post cards (Appendix H) were sent to them and telephone calls were made reminding them of their agreement and requesting their cooperation. In addition, responses from 60 completed instruments were not included in the data analyses because the respondents did not meet the criteria for inclusion in the study.

Because doctoral students holding master's degrees in counselor

education were one of the types of clinical supervisors whose educational credentials were being examined in this study, it was preferable to use subjects from counselor education programs who potentially would have experience as supervisees of doctoral students. Therefore, only male and female entry-level counselor education students who had completed varying proportions of their counselor education program hours required for graduation from mental health counseling or community counseling tracks were subjects in this study.






65


Sampling Procedure

Attempts were made to contact 27 CACREP liaison persons

associated with the programs identified as suitable for study. Those persons who were available to be spoken with by telephone were asked if their Community Counseling or Mental Health Counseling programs and their doctoral-level programs in Counselor Education and Supervision were still CACREP-accredited. If their programs currently were accredited, the liaison persons were asked to suggest the names of professors in their departments who could be asked to cooperate for this study. Nine CACREP liaisons offered to cooperate in helping to conduct the study. They were sent letters of introduction (Appendix A) and the number of packets of testing materials they suggested would be appropriate. Six other CACREP liaisons offered the names of 12 counselor education department faculty as contacts. Seven of these people were able to be reached by telephone. Five of the contacts were not available when calls were made to them, or they did not return calls when messages were left. Those who agreed to participate were sent letters of introduction and appropriate testing materials. The target size for the sample was 300. Unfortunately, almost nothing is known currently about the demographic characteristics of students currently enrolled in counselor education curricula. Therefore, professors who administered the instruments were asked to do so in a manner as random as possible, and it is assumed that






66


the resultant sample was an effective representation of entry-level counselor education students.

Research Procedure

Assessment materials were sent to the counselor education

department faculty who agreed to assist in conducting this study. Fourpage assessment packets were included for each subject and contained (a) a Demographic Inventory (Appendix C), (b) an assessment instrument (i.e., a two-page profile of a fictitious client) (Appendix D), and (c) an Approval Rating Scale and Qualitative Data Form (Appendix F).

Administrators were provided with instructions about how to

distribute, retrieve, and return the assessments (Appendix B) and were provided self-addressed, stamped envelopes in which to return the completed materials.

Demographic Inventory

Personal information pertaining to the subjects was necessary to conduct this study. Therefore, relevant questions were asked of the subjects about the types of programs in which they were enrolled, number of program hours they had completed, and their gender (Appendix C). Assessment Instrument

Each subject was presented with intake interview data about a

fictitious client who was purportedly seeking help from the subject for a mental health problem (Appendix D). The case presented in this study was





67


a profile of a woman who was diagnosed with Borderline Personality Disorder, one of more than 200 real case studies compiled by Spitzer et al., (1994, p. 395). Spitzer et al., (1994) drew from their own and others' clinical and teaching experiences to present edited cases that contain information needed for differential diagnosis.

Each assessment instrument contained one of eight descriptions of fictitious clinical supervisors who had diagnosed the fictitious client as having Borderline Personality Disorder. The client's clinical profile and diagnosis was constant across conditions presented to subjects, however, the credentials and gender of the diagnosticians were not constant. They reflected potentially different credibility based on diagnosticians' educational credentials and gender. The fictitious clinical supervisors who were presented as having arrived at the diagnosis of Borderline Personality Disorder for the fictitious client were described as either male or female and as holding one of four educational degrees (Appendix E).

Approval Rating Scales (Appendix F) on which subjects recorded the extent to which they agreed with their clinical supervisors' diagnoses were coded with ciphers that corresponded to the categories of variables in which the fictitious clinical supervisors were positioned in the assessment instrument. Genders and educational credentials were presented in combination to allow for eight variants (see Appendix G).






68


Agreement Rating Scale and Qualitative Data Form

Each subject was asked to indicate on a Likert-type scale the extent to which they agreed with the supervisor's diagnosis. In addition, subjects were asked to explain the bases of their responses by describing the factors or considerations that influenced their responses.

A pilot study with a small number of students indicated that minor changes needed to be made in the instructions to the subjects. A few respondents claimed they could not indicate the extent to which they agreed with their clinical supervisors' diagnosis because they believed they were untrained to arrive at a conclusion as to the accuracy of their clinical supervisors' diagnosis. Therefore, rather than simply asking the respondents the extent to which they agreed or disagreed with their clinical supervisors' diagnosis, respondents were advised that it was recognized that there may be limitations to their ability to diagnose the client at the time they were asked to do so, but they were asked to indicate, nonetheless, the extent to which they agreed with the diagnosis provided.

Data Analyses

Quantitative Analyses

Data were analyzed using a four-way factorial analysis of variance. The independent variables were the supervisors' gender (i.e., male or female) and educational credentials (i.e., M.D., Ph.D. holding counselor education faculty, Ph.D. holding psychologist, and doctoral student with






69


master's degree in counselor education) as well as subjects' proportion of program hours completed that were required for graduation (i.e., <1/3, 1/3 to 2/3, and >2/3) and subjects' gender (i.e., male or female). Post hoc comparisons of group means were done using Tukey's Test of Honestly Significant Differences, with a minimum significant difference of 2.53 using a significance level of .02.

Qualitative Analyses

To answer questions such as "How did the subjects in this study

think about the diagnosis of the client?" and "What factors did the subjects consider when arriving at conclusions about the diagnostic credibility of their supervisors?," the context in which the hypothetical supervisors and supervisees were interacting when diagnostic decisions were made was investigated through an examination of counselor education students' thinking when they responded to the supervisors' diagnoses. Subjects had the opportunity to express their convictions, viewpoints, and values in order that a deeper and broader understanding of their responses could be attained and a sharper awareness of their perspectives achieved.

Subjects were presented an open-ended question that solicited discussion of why the subjects chose to agree to the extent they did (Appendix F). The qualitative analyses of the data were accomplished in three phases. In Phase I, central ideas or prominent opinions expressed by






70


the subjects as the reasons for agreeing or disagreeing with the supervisors' diagnoses were identified.

In Phase II, responses showing similar rationales were grouped. For example, those responses wherein various supervisors' perceived diagnostic expertise based on their educational credentials stimulated supervisees' attribution of credibility were grouped together. Likewise, those responses wherein some supervisors' diagnostic credibility was questioned because of perceived negligent practices in their process of diagnosis were combined.

In Phase III, the themes that were distinguished, and the "recurring ideas and patterns of beliefs" that were identified (Marshall & Rossman, 1989, p. 116) were arranged further, with the result being three categories of responses that indicated the extent to which counselor education students attributed credibility to their clinical supervisors: (a) supervisors' diagnosis rejected, (b) supervisors' diagnosis provisionally accepted, and

(c) supervisors' diagnosis accepted without qualification.














CHAPTER 4
RESULTS

Quantitative Analyses

Analyses of variance and a post hoc comparison of group means were used to examine the following null hypotheses:

1. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials.

2. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' gender.

3. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students.

4. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation.

5. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' gender.


71





72


Supervisors' types of credentials (i.e., M.D., doctorate-holding

counselor education faculty, doctorate-holding psychologist, or doctoral student holding master's degree in counselor education) and gender and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation (i.e., <1/3, 1/3 to 2/3, or >2/3) were the independent variables examined in this study. The dependent variables were the scores obtained from a Likerttype index of subjects' approval of clinical supervisors' diagnoses.

A four-way factorial analysis of variance was performed on the data resulting from the responses of 191 subjects. Tukey's Test of Honestly Significant Difference was performed for variables having more than two levels.

In Table 4 1 are the summaries of the ratings given by counselor

education students of the clinical supervisors by their levels of educational credentials and proportion of program hours completed by the subjects. The ANOVA source table is found in Table 4 2.

The significant interaction between supervisors' educational

credentials and the proportion of program hours completed by students for graduation can be seen in Figure 1. There is a difference greater than the Tukey minimum HSD of 2.53 in two data sets: credentials and program hours. One in regard to the latter is the difference of 3.95 between






73)


Table 4 1

Summaries of Ratings by Levels of Educational Credentials and Proportion of Program Hours Completed

Variable Value Label

Psychiatrist 1.00 Doctorate-Holding Faculty 2.00 Doctorate-Holding Psychologist 3.00 Masters-Holding Doctoral Student 4.00

Less Than One-Third of Program Completed 1.00 One-Third to Two-Thirds of Program Completed 2.00 Greater than Two-Thirds of Program Completed 3.00

Variable Value Label Mean Std. Dev Cases

For Entire
Sample 5.2356 2.2295 191 CRED 1.00 5.0000 2.5199 41 HRS 1.00 5.1579 2.3157 19 HRS 2.00 7.3750 2.3867 8 HRS 3.00 3.4286 1.6968 14

CRED 2.00 5.5600 2.0619 50 HRS 1.00 5.3600 1.9553 25 HRS 2.00 6.2857 2.2887 7 HRS 3.00 5.5556 2.1753 18

CRED 3.00 4.9149 2.3390 47 HRS 1.00 5.0435 2.0993 23 HRS 2.00 4.2857 2.2887 7 HRS 3.00 5.0000 2.7386 17

CRED 4.00 5.3962 2.0413 53 HRS 1.00 5.1613 2.2226 31 HRS 2.00 5.2857 1.2536 7 HRS 3.00 5.9333 1.9445 15






74


Table 4 2

Analysis of Variance of Ratings of Diagnostic Credibility by Educational Credentials, Gender of Supervisors, Gender of Subjects, and Proportion of Program Hours Completed

Rating
By Educational Credentials (CRED)
Gender Of Supervisor (GENSUP)
Gender Of Subject (GENSUB)
Proportion Of Program Hours Completed (HRS)

Sum of DF Mean F Signif Source of Variation Squares Square of F

Main Effects 32.849 7 4.693 .983 .446 CRED 14.728 3 4.909 1.029 .382 GENSUP 1.205 1 1.205 .252 .616 GENSUB 2.669 1 2.669 .559 .456 HRS 15.312 2 7.656 1.604 .205

2-way
Interactions 135.255 17 7.956 1.667. .055 CRED x GENSUP 13.247 3 4.416 .925 .430 CRED x GENSUB 13.665 3 4.555 .954. .416 CRED x HRS 74.552 6 12.425 2.603 .020* GENSUP x GENSUB 7.373 1 7.373 1.545 .216 GENSUP x HRS 16.577 2 8.288 1.736 .180 GENSUB x HRS .466 2 .233 .049 .952

3-way
Interactions 65.062 17 3.827 .802 .689 CRED x GENSUP x 4.660 3 1,553 .325 .807 GENSUB
CRED x GENSUP x HRS 24.270 6 4.045 .847 .535 CRED x GENSUB x HRS 23.376 6 3.896 .816 .559 GENSUP x GENSUB x 4.929 2 2.465 .516 .224 HRS

* Significant at .02 level

attribution of credibility to psychiatrists by students having completed

more than one-third and less than two-thirds of their programs (mean =

7.38) and the students who had completed more than two-thirds of their

programs (mean = 3.43). A second significant difference of 3.09 is in






75


regard to the former between the attribution of credibility to psychiatrists (mean = 7.38) and to doctorate-holding psychologists (mean = 4.29) by students who had completed more than one-third and less than two-thirds of their programs.

It also can be seen in Figure 1 that as the proportion of program

hours required for graduation increases from less than one-third to greater than two-thirds, attribution of diagnostic credibility to doctoral students rises from 5.16 to 5.29 to 5.93. In addition, attribution of diagnostic credibility to psychiatrists rises from a mean of 5.16 among students who had completed less than one-third of their programs to 7.38 among students who had completed between one-third and two-thirds of their programs, but then falls to 3.43 among students who had completed more than two-thirds of their programs.


8


C 6
e

4

e 3
a
n s2



Psvchiatrists Faculty <1/3 his 1/3


> <2/3


Psychologists
>2/3 his


.... ... .... ..


Students


Figure 1. Cell means of educational credentials by proportion of program hours completed.






76


Qualitative Analyses

An analysis of counselor education students' thinking when they

indicated the extent to which they agreed with their supervisors' diagnosis was done to determine how the subjects thought about the diagnostic credibility of their supervisors and which factors the subjects considered when arriving at their conclusions. Subjects were presented an open-ended question that solicited discussion of why the subjects chose to agree to the extent they did with the various types of supervisors (Appendix F). The qualitative analyses were accomplished in three phases. Central ideas or prominent opinions expressed by the subjects as the reasons for agreeing or disagreeing with the supervisors' diagnoses were identified in Phase I.

In Phase II, those responses from subjects that indicated reasons for agreeing to the extent they did with their supervisors were arranged into eleven groups based on the rationales of the responses. For example, various supervisors' perceived diagnostic expertise based on their educational credentials stimulated some supervisees' unquestioning attribution of credibility. Likewise, some supervisors' diagnostic credibility was questioned because of perceived negligent practices in their process of diagnosis.

The themes that were distinguished, and the "recurring ideas and patterns of beliefs" that were identified (Marshall & Rossman, 1989, p. 116) were arranged further in Phase III so that three major categories of






77


responses were identified which indicated the extent to which counselor education students attributed credibility to their clinical supervisors: (a) supervisors' diagnosis rejected--groups #1 #4 (n=70), (b) supervisors' diagnosis provisionally accepted--groups #5 8 (n=54), and (c) supervisors' diagnosis accepted without qualification--groups #9 10 (n=50). A fourth category was composed of nine subjects who would not accept or reject the diagnosis until the issue of abuse was addressed with the client. Those responses in which subjects did not provide an explanation for their ratings were excluded.

A representation of the four groups who rejected the supervisors' diagnosis and their reasons for doing so are shown in Figure 2. Group #1 was composed of subjects (n=21) who thought the diagnosis was premature. They believed the supervisor had insufficient data to justify the diagnosis of Borderline Personality Disorder (BPD). The subjects in Group #2 (n=9) believed the diagnosis was inaccurate, but acknowledged having limited knowledge of how to diagnose, so their rejection of the diagnosis was based solely on conjecture. Thirty three subjects (Group #3) rejected the diagnosis because they thought the symptoms presented were not those of BPD. The fourth group who rejected the supervisors' diagnosis (a=7) did so saying they were unaware of the symptoms of BPD, but they suggested other diagnoses as being more suitable than BPD.





78


Alternative Offered (7)

Preimture (21)






Inaccurate Based on SynMptoms (33)
Inaccurate Based on Conjecture (9)




Figure 2. Reasons for rejecting supervisors' diagnosis and number of subjects who offered the reasons cited.

Figures 3 6 show the response frequencies of male and female subjects in groups #1 #4 to the diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students. Subjects had completed varying proportions of their program hours required for graduation [i.e., less than one-third (<), between onethird and two-thirds (< >), or more than two-thirds (>)]. For reasons cited above, subjects in these groups rejected the diagnosis given by their clinical supervisors. Figure 7 is a summary of the response frequencies of the subjects in groups #1 #4.

Figure 8 depicts the reasons some subjects gave for accepting with qualifications the supervisors' diagnosis. The diagnosis was accepted as possibly correct by the 34 subjects in Group #5; however, they offered other diagnoses for consideration as well as BPD. Although the diagnostic






79


3.0 2.5

2.0

1.5.


0.5


F PsychiatM Psychiat F Fac


M Fac F Psychol M Psychol


FSiuden M Studen


M Female <
0 Male <


] Female <> [ Male <>


Mj, Female >
Male >


Figure 3. Group #1 response frequencies. Subjects diagnosis as reason for rejection.


(n2 1) cited premature


.P
F PsychiatM Psychia


F Fac


MFac


F Psychol M Psychol F Stiden M Suden


Female < P Male <


,-- Female <> [ Male <>


f.11H1l, Female >
Male>


Figure 4. Group #2 response frequencies. Subjects (1=9) acknowledged limited knowledge of diagnosis, but rejected the diagnosis based on conjecture.


2.0.



1.5.


1.0



0.5






80


4







2







F PsychiaM Psych F Fac MFac F PsycholM Psycho F Sluden M Siuden Female < Female <> M T Female > E Male < Male <> Male >


Figure 5. Group #3 response frequencies. Subjects (n=33) cited symptoms presented as being incompatible with diagnosis as reason for rejection.






2.0







1.0 0.5



0.04
FPsychiatMPsychiat FFac MFac FPsycholMPsychol FESluden MSiuden Female < Female <> 3M Female > E Male < Male <> jj Male>


Figure 6. Group #4 response frequencies. Although they were unaware of the symptoms of BPD, subjects (n=7) offered alternative diagnoses to BPD.









7

6

51


3





F PsychiairMPsychiat F Fac N Female <
W Male <


MFac F Psychol [1] Female <> [ ] Male <>


MPsychol FSiuden MSiuden

E Female >
ON Male >


Figure 7. Summary of response frequencies for groups #1 #4. Subjects (n=70) rejected diagnosis.


criteria for BPD were unfamiliar to the five subjects in Group #6, they did not rule out BPD as a possible diagnosis.

Subjects in Group #7 (n=13) believed BPD was a possible diagnosis, but that other diagnoses should be ruled out before designing a treatment plan. Finally, the two subjects in Group #8 believed "professionals can be wrong," so they would not fully commit to the diagnosis offered by the supervisors.

Figures 9 12 show the response frequencies of male and female subjects in groups #5 #8 to the diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students. Subjects had completed varying proportions of their program


81






82


Cautiously Accepted (2)


Tentatively Accepted (13)






Unfamiliar Citeria (5) possible (34)





Figure 8. Reasons for provisionally accepting supervisors' diagnosis and number of subjects who offered the reasons cited. hours required for graduation [i.e., less than one-third (<), between onethird and two-thirds (< >), or more than two-thirds (>)]. For reasons cited above, subjects in these groups provisionally accepted the diagnosis given by their clinical supervisors. Figure 13 is a summary of the responses of subjects in groups #5 #8.

In Figure 14 the numbers of subjects and the reasons for accepting the supervisors' diagnosis without question are shown. Group #9 (n=35) wrote that they thought the diagnosis was accurate because the symptoms presented met the criteria for BPD. The subjects in Group #10 (n= 15) were less certain about the validity of the diagnosis based on their knowledge, but they said that they accepted the diagnostic judgment of their supervisors because they were willing to defer to their supervisors' experience and training.






83


4







21



1



0 L F PsychialrM Psychiat F Fac M Fac F Psychol M Psychol F Siuden M Siuden


Female < F Male <


Female <> Q Male <>


BE1 Female >
Male>


Figure 9. Group #5 response frequencies. Subjects (acould be one diagnosis among several possibilities they


2.01 1.51


1.G



0.5



0.0F PytPs FFa
F PsychiatlMPsychiat F Fac


Female < r Male <


MFac


:34) wrote that BPD offered.


FPsycholMPsychol FStuden MStIuden


Female <> Male <>


Ma Female >
Male >


Figure 10. Group #6 response frequencies. Subjects (n=5) were not familiar with the criteria for BPD diagnosis, but indicated that it could be one diagnosis among others.






84


2.0F 1.5 1.0 0.5



F Psychia M Psychi F Fac M Fac F PsychoiM Psycho F Sluden M Stiden


M Female < E Male <


Female <> L--I Male <>


ZI Female > 01 Male >


Figure 11. Group #7 response frequencies. BPD was considered a possible diagnosis, but subjects (n=13) thought other diagnoses should be ruled out.






1.01 0.94 0.8i 0.7 0.6
0.5
0.4 o.3~ 0.2, 0.1 0.0
FPsychiatMPsychiat FFac MFac F Psychol M Psychol F Siuden M Sluden


m Female < Sin Male <


Female <> Male <>


ill Female > EM |Male>


Figure 12. Group #8 response frequencies. Subjects (n=2) had reservations and would not commit fully to the diagnosis offered by the supervisors.






85


6,


4

3

21




F PsychiatM Psychia F Fac Female <
P Male <


M Fac F Psychol

Female <> [__] Male <>


MPsychol FSiuden MSiuden

Ii Female >
Male >


Figure 13. Summary of response frequencies for groups #5 #8. Subjects (n=54) provisionally accepted diagnosis.

Figure 15 and Figure 16 show the response frequencies to the

diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students by male and female subjects in Groups #9 and #10 Subjects had completed varying proportions of their program hours required for graduation [i.e., less than one-third (<), between one-third and two-thirds (< >), or more than two-thirds (>)]. Subjects in these groups accepted the diagnosis given by their clinical supervisors. A summary of the response frequencies for these two groups is shown in Figure 17.






86


Inability to diagnose (15)


Based on Symptoms (35)


Figure 14. Reasons for accepting supervisors' diagnosis without question and number of subjects who offered the reasons cited.


4


2,







F PsychiatrMPsychiat F Fac MFac F Psychol MPsychol


FStiuden MSiuden


M Female < M Male <


- Female <> [ ] Male <>


Ki Female >
Male>


Figure 15. Group #9 response frequencies. Subjects (a=35) reported diagnosis was accurate because symptoms presented met criteria for BPD.










3.0O 2.5


1.5

1F0P

05111


FPsychiatMPsychiat FFac MFac FPsycholM~Psychol F'Siden M Studen


Female < Male <


Female <> ] Male <>


lIE Female >
Male>


Figure 16. Group #10 response frequencies. Subjects (n= 15) uncertain about validity of diagnosis based on their knowledge, so they deferred to supervisors.


4



3



2







F PsychialrM Psychiat F Fac MFac F Psych M Psych


FSiuden MStuden


Female < OM Male <


Female <> LII Male <>


ElI Female >
Male>


Figure 17. Summary of response frequencies for groups #9 and #10 subjects (a=50) accepted diagnosis.


87














CHAPTER V
DISCUSSION


None of the null hypotheses addressed in this research were rejected. No significant main effects were found. It was discovered that there is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials or gender. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation or on students' gender.

However, a significant interaction was found between the level of

attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials and counselor education students' proportion of program hours completed that are required for graduation.

Limitations of the Study

The results of this study are limited in regard to generalizability for several reasons. Strict controls for random selection of the sample were


88






89


impractical. However, because the instrument was administered to counselor education students in CACREP-accredited programs at various institutions throughout the United States by counselor education faculty at those institutions, the instruments were administered in a manner as random as possible. Therefore, it can be assumed that the resultant sample was a satisfactory representation of entry-level counselor education students.

The educational credentials of supervisors chosen for study here were not inclusive of all credentials supervisors of counselor education students in field experiences might hold. For example, licensed clinical social workers, licensed marriage and family therapists, and licensed and unlicensed mental health counselors with master's degrees were not represented as supervisors in this study.

Only students from CACREP-accredited counselor education programs were selected for inclusion in this study. Students from programs with less rigorous academic standards or from programs that, for whatever reason, have not become CACREP-accredited were not part of the sample.

The hypothetical race, age, theoretical orientation, and working environment of the clinical supervisors were not addressed. Also, the actual race, age, theoretical orientation, and extent and types of field experiences of the subjects were not investigated.






90


As in Greenberg and Wursten's (1988) study, this investigation did not test attractiveness, likability, or dynamism, all characteristics that could affect supervisees' perceptions of credibility of their supervisors. Subjects' and supervisors' perceived self-efficacy, expectations, or personalities also were not studied.

Three levels of proportion of program hours completed that are

required for graduation for counselor education students were used without a theoretical basis for cutoff points. It was assumed that in the beginning of their programs students would be less experienced in clinical matters than they would be in the middle or last thirds of their programs.

Developmental levels of subjects as counselors were not ascertained. Stoltenberg (1981) and Stoltenberg and Delworth (1987) wrote, and Wiley and Ray (1986) showed, that counselor education students exhibit qualitatively different needs, behavioral motivations, and resistance at various developmental levels during training; however, developmental level and amount of training (proportion of program hours completed that are required for graduation) are not interchangeable concepts.

The limitations of this study render its results under-representative of all counselor education students. Therefore, generalizing conclusions to groups that are not similar would be inappropriate. However, the results should adequately represent counselor education students in CACREPaccredited community or mental health counseling programs.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EACZRZE4L_WOECX1 INGEST_TIME 2017-10-31T17:44:55Z PACKAGE UF00100972_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ERDVJ1OMN_1RYQQ6 INGEST_TIME 2017-07-11T21:32:13Z PACKAGE UF00100972_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

ATTRIBUTION OF DIAGNOSTIC CREDIBILITY TO CLINICAL SUPERVISORS AMONG COUNSELOR EDUCATION STUDENTS By JANE J CARROLL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1995

PAGE 2

To my clients and students, those I have known and those I have yet to meet. May my work always reflect the exceptional education and supervision I received as I prepared to be a counselor and counselor educator.

PAGE 3

ACKNOWLEDGMENTS Without the influence of people who encouraged and supported me in this undertaking, I'd have had to sacrifice the pleasure. I learned this is a journey I could not have traveled alone. The spirit of my father, James Judy, lives on. His powerful example and his belief in me were essential to this effort. My mother, Mildred Judy, lovingly gave me invaluable guidance and assistance. James Carroll and Timothy Carroll, my children, endorsed my plan to take a different path. Their advocacy made the striving that much easier. My grandchildren Kaleigh, Jackson, and those to be born, make the striving that much more meaningful. I cannot overstate the importance of James Bazan's many contributions. I am particularly grateful for his belief in my competence, pride in my work, and steadfast presence. Dr. Eve Hershberger's friendship and unrivaled clinical supervision empowered me to develop personally and professionally. Rick Sayers and Dr. David Miller brought light to a dark place--my brain on statistics.

PAGE 4

Candy Spires, Donna Johnson, and Sharon Lake in 1215 Norman were unselfish with their time and encouraging with their words. Their friendship relieved some of the most stressful moments. My committee: I appreciate Dr. Jin-Wen Hsu's eagerness to help me understand concepts in quantitative analysis. Dr. Rob Sherman introduced me to the world of the qualitative researcher--a professional gift. He nurtured my homesick soul with his familiar and charming Maine-speak. Dr. Joe Wittmer is truly a gentle person. His confidence in me helped me learn how it feels to be a counselor educator. His influence will never end. Dr. Larry Loesch has been the single greatest influence in my professional life. I hold this man in high esteem not only for the example he sets, but for his superb management and supervision of my academic experiences, reassuring words and presence, and confidence in my ability. He allowed me to learn, but never to suffer. I am proud and grateful that he is my friend and mentor.

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS iii LIST OF TABLES vii LIST OF FIGURES viii AB STRACT ix CHAPTERS 1 INTRODUCTION 1 Overview 3 Person Perception and Attribution 5 Statement of the Problem 7 Need for the Study 8 Purpose of the Study 14 Rationale for the Approach 15 Hypotheses 16 Definitions 16 Overview of Remainder of Study 18 2 REVIEW OF LITERATURE 19 Application of Counseling Theories and Diagnostic Skills 22 Relevance of Theoretical Framework 35 Elaboration on Need for the Study 41 Research on Perceptions and Attributions of Parties in Counseling Process 41 Summary 59 3 METHODOLOGY 63 Delineation of the Variables 63 Population 64

PAGE 6

Sampling Procedure 65 Research Procedure 66 Data Analyses 68 4 RESULTS 71 Quantitative Analyses 71 Qualitative Analyses 76 5 DISCUSSION 88 Limitations of the Study 88 Relevance of Attribution Theory 91 Conclusions and Interpretations 92 Implications 103 Recommendations for Future Research 107 Summary 110 APPENDICES A LETTER OF INTRODUCTION 112 B INSTRUCTIONS FOR ADMINISTERING INSTRUMENT 113 C DEMOGRAPHIC INVENTORY 114 D PROFILE OF A FICTIOUS CLIENT 115 E DESCRIPTIONS OF FICTITIOUS CLINICAL SUPERVISORS 117 F AGREEMENT RATING SCALE AND QUALITATIVE DATA 119 G GENDER-CREDENTIALS CODE 120 H POST CARD REMINDER 121 REFERENCES 122 BIOGRAPHICAL SKETCH 131

PAGE 7

LIST OF TABLES Table page 4-1 Summaries of Rating by Levels of Educational Credentials and Proportion of Program Hours Completed 73 4-2 Analysis of Variance of Ratings of Diagnostic Credibility by Educational Credentials, Gender, and Proportion of Program Hours Completed .... 74

PAGE 8

LIST OF FIGURES Figure page 1 Cell means of educational credentials by proportion of program hours completed 75 2 Reasons for rejecting supervisors' diagnosis 78 3 Group#l response frequencies 79 4 Group #2 response frequencies 79 5 Group #3 response frequencies 80 6 Group #4 response frequencies 80 7 Summary of response frequencies for groups #1 #4 81 8 Reasons for provisionally accepting supervisors' diagnosis 82 9 Group #5 response frequencies 83 10 Group #6 response frequencies 83 1 1 Group #7 response frequencies 84 12 Group #8 response frequencies 84 13 Summary of response frequencies for groups #5 #8 85 14 Reasons for accepting supervisors' diagnosis without question 86 15 Group #9 response frequencies 86 16 Group #10 response frequencies 87 17 Summary of response frequencies for groups #9 and #10 87

PAGE 9

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ATTRIBUTION OF DIAGNOSTIC CREDIBILITY TO CLINICAL SUPERVISORS AMONG COUNSELOR EDUCATION STUDENTS By Jane J. Carroll August, 1995 Chairman. Larry C Loesch, Ph.D. Major Department: Counselor Education This study investigated the extent to which counselor education students differentially attribute diagnostic credibility to clinical supervisors based on clinical supervisors' educational credentials and gender, and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation. Subjects were 191 male and female counselor education students who had completed varying proportions of their entry-level program hours from CACREP-accredited counselor education programs. They rated their perceptions of the diagnostic credibility of male and female: (a) counselor education doctoral student, (b) doctoral-level psychologist, (c) counselor

PAGE 10

education department faculty member, and (d) psychiatrist acting as supervisors. Quantitative and qualitative analyses were conducted. An analysis of variance and post hoc analyses of significant interactions between supervisors' educational credentials and students' proportion of program hours completed required for graduation were done. Qualitative analyses of subjects' narrative replies were done to achieve broader understanding of the subjects' views on clinical supervisors' diagnostic credibility. Counselor education students who had completed varying proportions of program hours required for graduation did not attribute diagnostic credibility differentially to their clinical supervisors. Also, gender of supervisors and supervisees was not a significant factor in attribution of diagnostic credibility. The results have implications for clinical training of counselor education students, counselor education curricula, staffing practices in mental health facilities and managed care programs, and student supervising practices.

PAGE 11

CHAPTER 1 INTRODUCTION In his novel The Prince of Tides Conroy described a psychiatrist as a "sojourner among wounded, disabused souls" (1986, p. 386). Conroy's description is not peculiar to the fictional doctor. It applies as well to many mental health counselors whose primary professional ambition is to help people who have problems that interfere with normal functioning. The State of Florida defines the practice of mental health counseling as the use of scientific and applied behavioral science theories, methods, and techniques for the purpose of describing, preventing, and treating undesired behavior and enhancing mental health and human development. Such practice includes the use of methods of a psychological nature to evaluate, assess, diagnose, and treat emotional and mental dysfunctions or disorders, whether cognitive, affective, or behavioral; behavioral disorders; interpersonal relationships; sexual dysfunction; alcoholism; and substance abuse. (Department of Business, 1994, p. 2) When counselors and clients work together effectively they can identify clients' problems and needs. When the right facilitative action is taken, clients' lives often become more fulfilled and happier than they had been previously. At the time a client seeks help, the counselor's first task is to determine accurately the nature and range of the client's problem or disorder (Seligman, 1990). This undertaking requires specialized skills,

PAGE 12

the development of which begins early in counselors' professional preparation and continues throughout training. Need for assessment and diagnostic skills typically is not evident until counselor education students start supervised field placement experiences. At the beginning of supervised field experiences, counselor education students are relatively inexperienced as clinicians and, consequently, are unprepared to recognize and conceptualize easily the meaning of signs and symptoms of mental distress. Therefore, counselor education students are expected to gain proficiency in assessment and diagnosis of clients, primarily through instruction and clinical supervision in field placement experiences. Cognitive change may lead to behavioral change (Forsterling, 1988). Social learning theory, commonly used to explain cognitive change, suggests that the confidence counselor education students have in their respective supervisors' diagnostic competence would be instrumental in determining the degree to which counselor education students develop effective diagnostic skills (Bandura, 1977). However, do counselor education students conclude that confidence in their supervisors as instructors or models is deserved? Attribution theory suggests that belief strength varies as a function of source credibility (Secord & Backman, 1974). Thus, the degree of confidence counselor education students have in their supervisors'

PAGE 13

3 diagnostic abilities could vary as a function of their supervisors' educational credentials and gender and of counselor education students' gender and proportion of counselor education program hours completed that are required for graduation. It was unknown, however, the extent to which counselor education students' confidence in their supervisors' diagnostic credibility is a function of these variables. Overview Counselor education students in the United States have the opportunity to become competent professional mental health counselors through more than 90 departments of counselor education that offer programs accredited by the Council for the Accreditation of Counseling and Related Educational Programs (CACREP). In particular, counselor education students enrolled in CACREP-accredited mental health counseling preparation programs must complete supervised practica and internships as curricular requirements. Counselor education students, having been taught counseling theories and techniques in the classroom, are expected to transfer the knowledge gained in classes to field experiences. There, presumably capable professionals supervise student progress and practice as they deal with "client welfare, counseling relationship, assessment, diagnosis, clinical intervention, prognosis, and appropriate referral techniques" (Bradley, 1989, p. 4).

PAGE 14

4 Students most likely to be interested in diagnostic credibility are those in CACREP-accredited community or mental health counseling programs. These are people who practice as supervised mental health counselors at sites such as community mental health agencies, inpatient and outpatient treatment facilities, human resource centers, businesses and industries, criminal justice settings, or individual private practices. These students shall be referred to hereafter as counselor education students, with the understanding that they are students in community counseling or mental health counseling programs. Mental health service providers generally require nationally recognized and accepted descriptive diagnoses of clients. When based primarily on criteria found in the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (APA. 1994), these diagnoses identify clients' relevant biological, psychological, and sociological conditions. Thus, diagnosis is generally accepted as a prelude to designing appropriate treatment plans. Supervisors guide counselor education students in practica or internships as they determine the nature of their clients' problems. This guidance frequently takes the form of advising counselor education students to "accept" (i.e., use and act upon) diagnoses developed by someone other than the students, usually their immediate supervisors. In essence, supervisors typically expect counselor education students to

PAGE 15

assume the correctness of others' diagnoses and to employ established treatment plans or to develop treatment plans based on others' diagnoses. Several questions arise in this regard. Do counselor education students assess the diagnostic credibility clinical supervisors differently as a function of the supervisors' credentials'^ Does the gender of supervisors influence counselor education students' attribution of credibility? How does the proportion of counselor education program hours completed that are required for graduation and the gender of counselor education students relate to attribution of credibility to supervisors'' Person Perception and Attribution An individual's reality is formed subjectively by his or her perceptions (Heider, 1958). However, there are a limited number of details to which people can attend simultaneously, and several factors determine which details will be perceived. For example. Postman, Bruner, and McGinnies (1976) wrote that perceivers' personal values, interests, and survival needs play major roles in determining the stimuli individuals select to perceive and those they defend against or reject. Individuals choose information they find personally significant. According to Secord and Backman (1974), one of the main sources of information used in evaluating other people are their actions and characteristics (e.g., credentials attained and gender).

PAGE 16

6 Attribution theory addresses how people use information they have collected to answer questions that ask "why"^" (Hewstone, 1983; Kelley, 1973). Kelley (1973) admitted that an individual's process of attribution is not free from bias, is based on incomplete evidence, and is scientifically inconclusive. In his ANOVA model of attribution in social interaction, however, Kelley (1973) proposed that causes (independent variables) covary with effects (dependent variables). Kelley (1973) said that effects are caused by (a) entities that are present in the situation, (b) the situation or circumstances in which the action occurs, or (c) the perceptions or characteristics of observers who are present, or combinations of these three factors. The measure of validity of people's perceptions, judgments, and evaluations can be determined by any of three criteria. Kelley (1973) proposed that the first criterion for validity is consensus Is a response to a stimulus like that of other people's? May the respondent therefore have confidence in her/his belief? The second criterion is consistency Is a response consistent over time'^ This criterion includes the presumption that like responses have taken place prior to the time in question. The third criterion of validity of a perception is distinctiveness Is a response associated with a particular stimulus? Attribution of credibility to an entity and the validity of the attribution can be illustrated by examining a hypothetical example of a

PAGE 17

counselor education student in a clinical internship (the circumstances) attributing credibility to a supervisor (an entity) by responding (based on student's personal perceptions) to the question, "why do you believe your supervisor, Dr. Doe, arrived at a correct diagnosis for that client?" The counselor education student, using information gained from observations and experience, could answer, "because my colleagues in the office agree (consensus) that s/he is correct, the criteria are present for this diagnosis (distinctive stimulus), and I think Dr. Doe's diagnoses are usually correct (consistency)." A perceiver attributes credibility to a source of information based on the nature and the strength of her or his perceptions (Kelley, 1973). In addition, as Postman, Bruner, and McGinnies (1976) and Secord and Backman (1974) suggested, the characteristics of the perceiver (e.g., gender and proportion of counselor education program hours completed that are required for graduation) influence the extent to which an impression of credibility is formed. Statement of the Problem Although fundamental principles of attribution theory have been examined in many contexts, they have been examined infrequently in the counseling profession Rarer still is investigation of these principles within the context of clinical supervision of counselor education students. Therefore, the problem addressed in this study was that differences among

PAGE 18

counselor education students' attribution of diagnostic credibility to psychiatrists, doctorate-holding counselor education program faculty, doctorate-holding psychologists, and doctoral students holding master's degrees in counselor education were unknown. It was also unknown if attribution of credibility varied as a function of diagnosticians' gender or of counselor education students' gender and proportion of counselor education program hours completed that are required for graduation. Need for the Study Do clinical supervisors' educational credentials or gender or supervisees' gender or proportion of counselor education program hours completed that are required for graduation affect the extent to which supervisees attribute diagnostic credibility to supervisors? Among those who most frequently supervise counselor education students in practica and internships are psychiatrists, counselor education department faculty members holding doctoral degrees, doctorate-holding psychologists, and doctoral students who have earned at least a master's degree in counseling. Unfortunately, little research has existed about supervisees' beliefs in the credibility of supervisors' assessments of clients' problems. Results of this study included implications for the mental health counseling profession and for counselor education departments. Differences among counselor education students' attribution of diagnostic credibility to supervisors with varying educational credentials by students

PAGE 19

who have completed different proportions of their counselor education programs could affect students' education and treatment of clients. Educational Credentials Heppner and Handley (1981), Worthington (1984), and Worthington and Stern (1985) found that supervisees' perceptions of counselors' competence are not related to educational credentials. College students perceive psychiatrists, however, to be more credible than other mental health providers at assessing mental health problems (Truatt & Bloom, 1982; Greenberg & Wursten, 1988). Gender There has been meager research on how gender relates to perceived credibility of clinical supervisors. Supervisors' gender, at least in initial supervisory encounters, is not related to supervisees' perceptions of supervisors' competence (Schiavone & Jessell, 1988). Apparently it has been unknown if gender of supervisors or supervisees affects counselor education students' perceptions of supervisors' diagnostic credibility. Proportion of Program Hours Completed Evidently there has been no published research on the relationship between attribution of diagnostic credibility by counselor education students to clinical supervisors and proportion of counselor education program hours completed that are required for graduation. However, Stoltenberg and Delworth's (1987) proposed levels of counselor

PAGE 20

10 development implies that counselor education students would experience gradually changing perceptions of expertness and trustworthiness of clinical supervisors. Usher and Borders (1993), who examined professional counselors' preferred areas for supervisory concentration, discovered that they favor supervisors who emphasized conceptual skills to those who emphasized professional behaviors, professional practice, and process skills. The information learned in this study also can be meaningfully related to (a) research on perceptions and attributions of immediate parties in the counseling process, (b) clinical training of counselor education students in counselor education programs, and (c) relationships of clinical supervisors and counselor education students in practica and internships. Additionally, the information could be applied to investigations related to other groups of people such as (a) counselors who have completed their basic education and have had some professional counseling experience, (b) counselors in different work settings who work with various types of clientele, and (c) counselors who are disinclined to diagnose. The information also could be applied to studies of educational programs that place differing emphases on teaching the art and science of diagnosis, as well as in situations where supervision varies in the extent of cooperation or cohesiveness between supervisors and counselor education students. These results also could be used in the study of relevant personality or

PAGE 21

attitude variables in the context of supervision. For example, characteristics of supervisors, such as the extent to which they project confidence, might be significant to their being considered credible, as might be the (a) role that counselor education students' expectations play in attribution of credibility, (b) influence of counselor education students' self-efficacy on their perceptions of supervisors' credibility, and (c) personality types of counselor education students and supervisors. Research on Perceptions of Parties in Counseling Process The early application of attribution theory, wherein it was used to explain how people perceive and interpret others' actions (Heider, 1958; Rotter, 1966), has explained clinicians' inferences about clients' psychological states. "The clinician's conceptual framework--whether behavioral, dynamic, or humanistic--is a set of systematic attributions for the causes of human behavior" (Jordan, Harvey, & Weary, 1988, p. 91). Further, "virtually all clinical endeavors, diagnostic and therapeutic, have attributional processes at their core" (Jordan, Harvey, & Weary, 1988, p. 91). Clients' and counselors' perceptions of each other result from the fundamental interactive nature of counseling. Counselors make clinical judgments of clients based on knowledge gained through assessment (Smith, 1988), and clients value certain counselor attributes as significant to the success of the counseling experience. Researchers have shown that

PAGE 22

counselor expertness is a variable clients consider favorable to the counseling process (Holland, Atkinson, & Johnson, 1987; Littrell, Caffrey, & Hopper, 1987; Ponce & Atkinson, 1989). Apparently no research exists, however, that examines counselor education students' perceptions of their supervisors' diagnostic expertise. Therefore, it was unknown whether counselor education students considered their supervisors as credible diagnosticians and how the attribution of credibility, or lack of it, influences the supervisory experience and ultimately client care. Clinical Training of Counselor Education Students Clinical judgment develops in an ongoing process. The struggle to understand and make sense of others' feelings, thoughts, and behavior originates early in counselor education students' classroom education. Counselor education students refine their ability to diagnose, however, when they are supervised in practical field experiences. Counselor education students are exposed to a number of theories from which to choose viable explanations for clients' problems, as well as suitable approaches to treatment. The ultimate decisions, however, regarding diagnosis and treatment planning must be, for moral, ethical, and legal reasons, a result of considered judgment. Counselor education students may not receive the level of training in diagnosis that some counselor educators and students think they need (Fong, 1993; Furlong & Hayden, 1993; Hohenshil, 1993; Seligman, 1990;

PAGE 23

Waldo, Brotherton, & Horswill, 1993). By discovering the factors that counselor education students consider when attributing credibility to diagnosticians, this study may help determine if counselor education programs adequately prepare counselor education students to diagnose clients thoughtfully and knowledgeably. Relationships Among Clinical Supervisors and Supervisees The practice of professional mental health counseling may benefit from this study. Knowledge of beginning counselors' attitudes toward their supervisors' diagnostic credibility affects two primary areas of concern, client care and facility staffing. Novice counselors could jeopardize the quality of client care if they (a) view the staff who assess, diagnose, and suggest treatment modalities as either incompetent or infallible; (b) are unable to accept the judgments of certain diagnosticians because of personal biases; (c) base their clients' treatments on diagnoses other than those on record, or, likewise, compromise the integrity of the system by unquestioningly following the lead of others; (d) philosophically or theoretically differ with particular diagnosticians; and (e) have no philosophical or theoretical basis for agreeing with or disagreeing with particular diagnosticians. Those who staff facilities that offer mental health counseling could benefit from knowing (a) the likelihood of power differentials among staff based on educational attainment or gender; (b) the likelihood of counselors

PAGE 24

14 to accept or reject diagnoses, regardless of source; and (c) which staff member(s) would most likely be respected as having diagnostic credibility. Mahoney (1988, p. 156) proposed that a counselor will "influence those whom he assesses, and counsels. One would hope that the selection and training of such professionals would assure that their assessment [a prelude to diagnosis] and advice are qualitatively superior to that of their untrained peers." Clearly, counselor educators, clinical supervisors, and staffers of facilities could use the results of this study when planning curricula and supervising students' performances in practica and internships. As Herman (1993) noted, therapist training may be the decisive factor in therapy outcome. Purpose of the Study The purpose of this study was to investigate an application of the nature of principles of attribution theory to clinical supervision of counselor education students. The study was conducted to determine the extent to which counselor education students think clinical supervisors' diagnoses of clients are credible, to ascertain whether supervisors' educational credentials and gender influence counselor education students' views of supervisors' credibility, and whether counselor education students' gender and proportion of counselor education program hours completed that are required for graduation influence their beliefs about

PAGE 25

15 supervisors' credibility. Data were subjected to quantitative and qualitative analyses and interpretations. Although other, potentially pertinent variables (e.g., race or theoretical orientation of the supervisors or students) could have been investigated, their inclusion would have made the design prohibitively complex and beyond the resources available for this study. Therefore, this study focused upon those variables deemed most significant in this initial, exploratory research. Rationale for the Approach This experimental research was conducted to determine the extent to which counselor education students differentially attribute credibility to clinical supervisors. There were a variety of potential ways to assess subjects' attitudes However, one of the most commonly accepted techniques is to use a bogus profile to which subjects can respond. The essential content of the profile of a fictitious client, as well as the diagnosis, were constant across conditions presented to subjects. The credentials of the diagnostician were not constant, but rather reflected potentially different levels of credibility This approach allowed for variation in responses, anonymity for subjects, time for thinking before responding, administration to many people at one time, and responding to otherwise identical content (Bailey, 1987; Henerson, Morris, & FitzGibbon, 1987).

PAGE 26

16 Hypotheses The following null hypotheses were tested in this study: 1. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials. 2. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' gender. 3. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students. 4. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation. 5. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' gender. Definitions The following operational definitions were used throughout this dissertation: Age is a period of human life. In this study, a subject self-reported years of age.

PAGE 27

17 Attribution theory is an explanation of how people explain causes of events and the actions that follow Credibility means to be believable, trustworthy, and worthy of confidence. The believability of diagnosticians was addressed in this study. Diagnosis is the judgment reached about a person's mental health status based on an assessment of the client, including the presence or absence of signs and symptoms as described in the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (APA, 1994). Gender is defined in this study as self-reported as male or female. Educational credentials are defined operationally as the highest educational degree acquired (M.D., doctorate, master's). Signs are objective indications of a mental disorder (e.g., weight loss). Supervisees are counselor education students who work with clients as part of their practica and internship (i.e., supervised field placement) experiences. Supervisors are counselors who are designated to "directly oversee the professional clinical work of counselors" (Ethical guidelines, 1993). Symptoms are phenomena that accompany a mental disorder (e.g., insomnia).

PAGE 28

18 Overview of Remainder of the Study The introduction to the study has been presented in Chapter 1. The review of related literature is presented in Chapter 2, the methodology in Chapter 3, and the results of the study in Chapter 4. Chapter 5 includes a summary of the study, discussion of the results, implications, and recommendations.

PAGE 29

CHAPTER 2 REVIEW OF RELATED LITERATURE It is the same thing which makes us mad or delirious, inspires us with dread and fear brings sleeplessness, inopportune mistakes, aimless anxieties, absentmindedness, and acts that are contrary to habit. These things that we suffer all come from the brain, when it is not healthy, but becomes abnormally hot, cold, moist, or dry. (Hippocrates, 460 377, B.C.) (Kellerman & Burry, 1988) This archaic, biologically based explanation for mental illness eventually replaced the ancient belief that divine or demonic interventions caused psychopathology. Historically, the etiology of mental disorders and how to codify and treat them has been controversial. "What is behind the symptom?" is a question the developmentalist Menninger posed in 1963 (p. 325), but one which mental health professionals had been debating for centuries. Currently, researchers and clinicians propose that there are physiological, psychological and sociological reasons for what is perceived as abnormal behavior and that identifying commonalties in mental disorders is the critical first step toward diagnosis and effective client treatment. Only a little more than a half century ago, however, "lunatics" were still being housed in grim "insane asylums" where, over the years, patients were regarded with varying degrees of insight, empathy, and concern. 19

PAGE 30

20 In the early 1900s the organic psychiatrist Kraepelin's conviction that patients' mental problems were physiological (i.e., the consequences of diseased brains) influenced the treatment of people with mental disorders. Abnormal psychological phenomena were thought to be symptoms of those afflicted brains (Weckowicz, 1984). Although much of what Kraepelin proposed is no longer accepted, his conception is alive in the current belief that biochemical imbalance is a contributing cause of some mental disorders. Several other explanations have been offered for why people think or behave in ways that are not customary in contemporary Western cultures. Alternative theories to Kraepelin's conjecture of the causes of mental disorders have been proposed and are bases upon which clinicians diagnose and treat clients. They include the following theoretical perspectives, briefly presented. Psychodynamic Psychodynamics has its basis in Freudian psychoanalysis. Clinicians who practice psychodynamic therapy view mental disorders as being caused by individuals' desires, motives, and conflicts (Weckowicz, 1984). Behavioral Behaviorism was a response to psychodynamics and is rooted in the research related to learning. Theorists, such as Bandura (1977), advocate encouraging positive behavioral change as a means of eliminating maladaptive behaviors. They believe clients' problems are learned behaviors.

PAGE 31

21 Cognitive Therapies based on the concept that people act according to their perceptions are termed cognitive therapies. Cognitive therapists, such as Ellis (1979), theorize that clients' constructs, ideas, and expectations determine their behavior. Developmental Developmentalists, such as Erikson (1963) and Menninger (1963), hold that psychopathology results from regression to a lesser stage of psychosexual, cognitive, or emotional development. They believe that human psychological development is directional and that difficulties develop if individuals are confronted with the demands of stages with which they are not yet equipped to cope (Crain, 1985; Weckowicz, 1984). Sociocultural Sociocultural theorists contend that the causes of mental disorders are either socially or biologically influenced and that the form mental disorder takes is determined by social factors. The various sociocultural constructions of mental disorders are based on the premise that individuals are links in society and therefore are part of a problem related to the whole (Weckowicz, 1984). This concept is found in family systems theory, such as that espoused by Minuchin (1974), which stresses the interaction and interdependency of family members. Humanistic Humanists assert that peoples' needs and aspirations do not vary across cultures and that all people have a need to move toward self-actualization. Rogers (1961) developed his response to psychoanalytic

PAGE 32

22 theory by testing the philosophical hypothesis that clients' potential for positive change is inherent and that with proper facilitation, normal growth and development occur. He proposed that psychological disturbances result from three influences: (a) the clients' need for self-respect, (b) important others' estimations of clients, and (c) the dichotomy between what clients perceive themselves as being and what they need to be in order to have self-respect (Lynn & Garske, 1985; Weckowicz, 1984). Application of Counseling Theories and Diagnostic Skills I used to be astonished ... to see humane physicians going daily round the wards of asylums, mere spectators of every form of distressing coercion, without a word of sympathy. The first principle in the treatment of lunatics was fear, and the best means of producing fear was said to be punishment. Thus torture became more and more ingenious. (Conolly, 1856, p. 13) Regardless of differences in current mental health providers' theoretical approaches to etiology and treatment of mental disorders, all mental health counselors, when interacting with clients, make inferences about their clients' relative states of well-being. Mental health counselors are expected to base their therapeutic actions on verifiable theoretical grounds. "[The counselor] will have some underlying rationale for what he does, some hunches about what different clients will do in different situations, and some general ideas about the counseling procedures that are apt to be effective" (Pepinsky & Pepinsky, 1954, p. iii).

PAGE 33

23 Although researchers and practitioners are regularly testing theoretical bases of mental health counseling, the fundamental objectives of counseling are constant. That is, mental health counselors routinely gather data about clients, diagnose the nature of problems, and provide appropriate professional interventions (Gambrill, 1990). Historical Perspective on Diagnosis Prognostic diagnosis, based on a medical model, was favored by Kraepelin, and is still practiced when clinicians use clients' signs, symptoms, and history of their development to predict the course of events (Weckowicz, 1984). However, from the end of World War U until the mid1970s, a psychosocial (rather than a medical model) of mental disorders prevailed. During that time, the first Diagnostic and Statistical Manual of Mental Disorders (APA. 1 952) was published. Because psychopathology resulting from wartime experiences had to be explained, descriptions of mental disorders, largely based on psychodynamic theory and following that language, reflected the influence of the psychosocial model. Former "functional" syndromes became "reactions" (Wilson, 1993). Biologically oriented psychiatrists who advocated more research and a return to the medical model in the practice of mental health criticized the use of the psychosocial model to describe psychopathology. Academicians and other critics argued that if mental disorders were psychosocial in nature, they lay outside the province of psychiatrists and were more

PAGE 34

24 properly in the social, political, and legal domains. Therefore, mental illness was a lie and psychiatric labeling a dangerous practice (Goffman, 1961; Scheff, 1966; Szasz, 1961). However, little empirical support for the argument that labeling alters self-concept has been produced. Morey, Skinner, and Blashfield (1986) pointed out that the behaviors that cause people to seek counseling were there before the diagnostic label, so it is obvious the label did not cause the clients' conditions. By the 1970s, research dollars, and consequently the amount of research on mental disorders, had dwindled. President Jimmy Carter's commission on mental health bemoaned the lack of research data, while insurance companies reduced allowed services and payments (Wilson, 1993). In short, a confluence of pressures on the profession [psychiatry] from outside influences--the major providers of resources for psychiatric research and treatment and the gathering momentum of the antipsychiatry critique--threw the issues of diagnostic reliability, treatment outcomes, and accountability for those outcomes into bold relief. Under these unfavorable professional conditions, the psychosocial model, as the dominant organizing model of psychiatric knowledge and the source of many of these problems, would have to be significantly altered, if not jettisoned altogether. (Wilson, 1993, p. 403) The DSM-III. published in 1980 (APA), was considered by some to be a return to the medical model of diagnosing mental and personality disorders, although the majority of the disorders catalogued have no known organic cause (Wilson, 1993). The DSM-III (APA. 1980) was authored mainly by medical researchers, but because it was atheoretical and

PAGE 35

25 descriptive, no explicit etiologic premise was represented. The American Psychiatric Association's Task Force on Nomenclature and Statistics decided the manual would be a "descriptive manual which would emphasize the assessment of easily observable symptoms" (Wilson, 1993, p. 405). The DSM-IV (1994) amends and enhances the DSM-III-R (1987) Changes in the DSM were made after work groups reviewed published literature, reanalyzed data sets, and conducted field trials. The DSM-IV is not a philosophical departure, however, from the DSM-III-R While the DSM-IV is coming into widespread use, the DSM-III-R will continue to be a reference for some clinicians. Also, for a while, clinicians will be working with clients diagnosed from criteria from the DSM-III-R There are three indications that diagnosis using the DSM-III (APA, 1980), (and likewise DSM-III-R 1987, and DSM-IV 1994), requires a more narrow focus than does a psychosocial model. First, the DSM-III its revised version ( DSM-III-R 1987), and DSM-IV do not attend to the unconscious mind; focus is on the visible workings of the mind. Second, symptoms are not placed within the context of individual development; rather, symptoms are viewed at a particular time in the client's life. Finally, clinical concerns are narrowed so that personality development, family dynamics, and social factors are deemphasized (Wilson, 1993). The American Psychiatric Association acknowledged that the DSMIII (APA. 1980) diagnostic criteria were hypotheses to be tested; however.

PAGE 36

26 "reification of the discourse of descriptive psychiatry through daily use of DSM-III and DSM-III-R [and, by extension, the DSM-IV (1994)] seems unavoidable" (Wilson, 1993, p. 408). Although the DSM-III was scrutinized and subjected to field trials by psychiatrists prior to its publication, the scientific quest to arrive at total diagnostic reliability in using the DSM-III was not achieved. Research has validated, however, the overall accuracy of the Diagnostic and Statistical Manual of Mental Disorders (Third Edition Revised ) (APA, 1987) when clinicians are welltrained in interviewing processes and application of the DSM-III-R (APA. 1987) (Hohenshil, 1993; Morey, Skinner, & Blashfield, 1986). Current Diagnostic Standards "Mental disorder" is an imprecise term that has been defined rather loosely as a clinically significant behavioral or psychological syndrome or pattern that occurs in a person and that is associated with present distress (a painful symptom) or disability (impairment in one or more important areas of functioning) or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom. In addition, this syndrome or pattern must not be merely an expectable response to a particular event. Whatever its original cause, it must currently be considered a manifestation of a behavioral, psychological, or biological dysfunction in the person. Neither deviant behavior nor conflicts that are primarily between the individual and society are mental disorders unless the deviance or conflict is a symptom of a dysfunction in the person, as described above. ( DSM-IV 1994, p. xxi) The approach of the DSM-IV (APA. 1994) is generally atheoretical regarding the etiology or pathology of mental disorders, except where the

PAGE 37

27 etiology is known and is presented in the definition of the disorder (e.g., Alcohol Withdrawal, Vascular Dementia). It is expected that, over time, there will be an increase in the certainty of the etiology of disorders. Some disorders will be found to be caused by biological factors, and others by psychological, sociological, or interaction of those three elements ( DSMIV, APA, 1994). The DSM-III-R (APA, 1987) and DSM-IV (APA, 1994) present descriptive criteria which many clinicians and researchers agree are clinical manifestations of the disorders named in the manual. A multiaxial system is used to integrate clients' biological, psychological, and social data and allows for a comprehensive first step in assessing clients' problems. The information on which diagnoses are based in the DSM-III-R however, are incomplete in important correlate areas such as the outcome of the disorder, clients' family histories, and treatment responses (APA, 1987). A recognition that ethnic and cultural beliefs and practices influence thinking and behavior prompted the writers of DSM-IV to attend to ethnic and cultural considerations when discussing criteria for psychopathology. Counselor Preparation for Diagnosis of Clients The objective of diagnosis depends on the clinician's purpose and theoretical commitment to diagnosis. Diagnostic reliability is more likely, however, when clinicians are trained in, and use, universal diagnostic criteria such as those found in the DSM (APA, 1994) (Weckowicz, 1984).

PAGE 38

28 Diagnosis provides (a) standard nomenclature that facilitates communication among clinicians and researchers; (b) descriptions of disorders or syndromes so that individuals can be compared, and therefore, allows categorization of disorders; (c) prediction of etiology, prognosis, and responses to treatment; (d) justification for treatment; and (e) a system for retrieving information when descriptive, etiologic, and treatment information is needed (Morey, Skinner, & Blashfield, 1986; Weckowicz, 1984). Background for development of diagnostic skills To diagnose clients and to provide them with effective treatment, counselors must see clients' symptoms in meaningful contexts. Mental health counselors are aided by knowledge of social, political, legal, and educational influences on clients. Quantitative and qualitative results of clinical assessments, conclusions reached through subjective observations of behavior, and knowledge of how clients are affected by family and other social dynamics all contribute relevant data about clients' conditions. In addition, familiarity with personality theory, human development theory, and awareness of recent scientific findings related to mental disorders are essential for broadly focused assessment and diagnosis. CACREP-accredited mental health counselor education programs offer curricula that prepare counselor education students to diagnose clients. These courses are designed to enhance awareness of the forces

PAGE 39

29 that may affect diagnosis. For example, graduating counselor education students are required to have successfully completed courses in (a) Human Growth and Development--the study of individuals at their different maturation levels; (b) Social and Cultural Foundations--the study of individuals' concerns in a multicultural society; (c) The Helping Relationship--the study of counseling and consulting processes; (d) Group Dynamics, Processes, and Counseling--the study of the practices and processes in observing and counseling individuals in groups; (e) Career and Lifestyle Development--the study of individuals' career development concerns; (f) Appraisal--the study of assessment and evaluation; (g) Research and Program Evaluation--the study of scientific inquiry and statistical methods; and (h) Professional Orientation--the study of how counselors function in professional roles (Vacc & Loesch, 1994; Wittmer, 1993). Counselor education students in CACREP-accredited programs who are preparing to be mental health counselors also learn the procedures for differential diagnosis using criteria in the latest Diagnostic and Statistical Manual of Mental Disorders (APA. 1994). Counselor education students are instructed in the use of the DSM (APA. 1994) because it is the "most widely accepted system for describing and diagnosing mental disorders" (Seligman, 1990, p. 3).

PAGE 40

30 Supervised Field experiences Each entry level student in a CACREP-accredited counselor education program must complete two supervised practica that each total 100 clock hours and a supervised internship of 600 hours (Wittmer, 1993). Experiences in field placements are intended to provide students with the opportunities to develop and refine their diagnostic skills. Supervision in field experiences Counselor education students engaged in mental health counseling practica and internships are under the charge of clinical supervisors who help supervisees develop assessment, diagnostic, and counseling skills. Counselor education students are relatively inexperienced clinicians, and therefore, in order to function effectively, they work to develop the aforementioned skills as well as the confidence that they can adequately fulfill their other responsibilities as counselors. It is also helpful for supervisors to provide feedback to counselor education students about how personal emotional reactions may affect their ability to diagnose. Borders and Leddick (1987) wrote that the responsibilities of supervisors, after having established supportive relationships with counselor education students, are to facilitate growth through teaching, identifying students' needs, and giving valuable feedback to students.

PAGE 41

31 Conditions of supervision Bradley (1989) reported that "warmth, acceptance, understanding, respect, and trust" (p. 37) are qualities supervisees think are important in their supervisors. It was unknown, however, to what extent counselor education students consider their supervisors credible or what factors are associated with variations in attributed credibility. Attribution theory can be used not only to evaluate action (e.g., diagnosis) based on the observer's perceptions, but also to predict (trust) that the action will be repeated. Hovland, Janis, and Kelley (1953) wrote that trustworthiness and expertness are the two components of credibility. Trustworthiness was defined by Strong (1968) as being a function of the communicator's (a) reputation for honesty; (b) social role, such as professional status; (c) sincerity and openness; and (d) perceived lack of reason for personal gain. Strong (1968, p. 216)) wrote that a communicator is perceived as expert (i.e., "a source of valid assertions") if there exists (a) objective evidence of specialized training [e.g., advanced degrees], (b) behavioral indicators of rational and knowledgeable arguments and confident presentations of views, and (c) consensus among others that the individual is an expert. Studies have shown that trustworthiness and expertness are characteristics that counselor education students value in supervisors. Beginning counselor education students, in particular, feeling less efficacious than more experienced counselors, indicate they want

PAGE 42

32 supervisors whom they can trust to provide reliable (i.e., expert) direction (Carey, Williams, & Wells, 1988; Hillerbrand & Claiborn, 1990; Tracey, Ellickson, & Sherry, 1989; Wiley & Ray, 1986). Counselor education students' need for structure in supervision is described by Stoltenberg (1981) and Stoltenberg and Delworth (1987) as being related not only to their level of skill and knowledge of theories, but to their level of development (i.e., qualitative differences among their needs, behavioral motivation, and resistance at different stages). Stoltenberg's (1981) theory was tested by Wiley and Ray's (1986) study of 107 dyads of supervisors and supervisees from nine university counseling centers. Wiley and Ray (1986) determined that supervisees' needs differ according to developmental level during training, and that developmental level and amount of training are not interchangeable concepts. Hillerbrand and Claiborn (1990) studied the needs of supervisees at different training levels. They discovered that, often, novice counselors with limited clinical and educational backgrounds cannot identify the features inherent in clients' problems, so the nature of clients' problems are not easily recognized. Conversely, more experienced and better educated counselors, who felt more confident about their skills, perceived cases more clearly than did novice counselors. Tracey, Ellickson, and Sherry (1989) studied counseling students' preferences for supervision style. They found that 40 novice counseling

PAGE 43

33 Students who had completed up to one semester of practicum training preferred a more authoritative and prescriptive supervisory experience than did 38 more advanced counselor education students who had completed two or more semesters of practicum but had not yet enrolled in internships. Beginning counselor education students not only desire expert direction in supervision, but appreciate supervisors whom they can trust to help them manage anxiety in clinical situations (Carey, Williams, & Wells, 1988). Uncertainty about case conceptualization and client diagnosis could be a source of trepidation to novice counselors. Counselor education students could benefit from instructional supervision in problem-structuring skills. Knowing how to clarify clients' problems enables counselors to redefine ill-structured problems and therefore be better able to arrive at deliberated diagnoses. Counselor education students sometimes doubt their ability to be effective counselors. Sipps, Sugden, and Faiver (1988) found that selfefficacy in counseling among counselor education students rose directly as the level of graduate school training, with the exception of the second year. The first year students' efficacy levels were significantly lower than the efficacy levels of third and fourth year students. The second year students' efficacy levels were significantly lower than the efficacy levels of the first, third, and fourth year students. Sipps, Sugden, and Faiver (1988) attributed the second year efficacy score as reaction to counseling

PAGE 44

34 Students' relatively naive first year attempts at counseling. Johnson, Baker, Kopala, Kiselica, and Thompson (1989) found that counseling students' efficacy increased with training over an eight-week period. In conducting their study, Johnson, Baker, Kopala, Kiselica, and Thompson's (1989) subjects engaged in ten classroom contact hours of case conceptualization that included exploration and clarification of a client's problem. Meeting counselor education students' need for expert and trustworthy supervision and instruction in diagnosis appears to be justified. Students' counseling efficacy and effectiveness presumably would improve with attention to diagnostic skill-building. Confidence in supervisors Principles of person perception and attribution theory suggest that clinicians' reasoning is subjective and imprecise. Additionally, although the DSM-IV (APA. 1994) criteria are useful because they are predicated on observation and research, there is no question that diagnoses based on the DSM-IV descriptive criteria are limited. Also, Faust (1986) warned of the folly of failing to acknowledge that clinical decisions are influenced not only by diagnosticians' perceptual selectivity and the frequency of occurrence of certain mental disorders, but by diagnosticians falsely associating frequently occurring signs with certain conditions with which the signs do not covary. Diagnosticians may also seek confirming evidence

PAGE 45

35 for their hypotheses and ignore disconfirming evidence. Faust (1986) also offered the reminder that diagnosticians have cognitive limitations. Again, to what extent do counselor education students have confidence that their supervisors' opinions are credible? To what extent are counselor education students certain that their supervisors will be worthy role models from whom diagnostic skills can be learned? Kelley's (1973) attribution theory, which was built on Heider's (1958) hypotheses, was used to infer that counselor education students perceive degrees of credibility in their supervisors based on levels of educational attainment and genders of the supervisors and proportion of counselor education program hours completed that are required for graduation and genders of counselor education students. Relevance of Theoretical Framework This study was an attempt to use the conceptual framework of attribution theory to determine the extent to which counselor education students attribute credibility to clinical supervisors. Attribution theory is a rather loosely structured set of hypotheses, rather than a tightly organized theory. In 1958, Heider, one of the pioneers in attribution theory, set forth his ideas about how people perceive each other, how they feel about each other, how they react to each other, and what they expect from each other. Heider (1958) did not believe that explaining the workings of the

PAGE 46

36 unconscious mind was essential when describing the tasks of perception and attribution. Rather, he believed that people attribute characteristics to other people by relying on common sense, a key component of which is person perception. Person Perception Individuals construct their cognitive reality by starting with what Dewey (1984, p. 249) called the "unanalyzed whole." People allow their interests to fuel new thoughts and new understandings. They then associate those ideas with previous thoughts, interests, or values (i.e., "dispositional properties" ) (Heider, 1958). When original thoughts, interests, and values ally with new thoughts and understandings, new inferences and judgments are formed (Kelly, 1963; Sherman, 1985). It is in the act of person perception that individuals make inferences or judgments about other people. Observations of others' characteristics are incorporated into one's reality in the form of social cognitive responses to the people observed (Tagiuri,1958). How are individuals motivated to select those stimuli they perceive from the constant inundation of stimuli to which they are exposed"^ Tagiuri (1958) contended that three factors, which can be viewed singularly or in interaction, influence person perception: (a) the attributes of the person being perceived, (b) the characteristics of the perceiver, and (c) the nature of the situation in which the association occurs. Tagiuri's (1958) position

PAGE 47

37 is consistent with that of Postman, Bruner, and McGinnies (1976) who argued that individuals' values, interests, and survival needs determine which stimuli will be perceived. Tagiuri's (1958) reasoning also is consistent with Heider's (1958) proposal of how individuals form perceptions. Heider (1958) wrote that a perception is the result of mediation between an observer and an object. In the act of person perception, the other person, an individual who cannot be experienced in totality by the observer, is the object (distal stimulus) toward which the observer's perception (proximal stimulus) is directed. Mediation occurs between the distal stimulus and the formation of a percept. Mediation involves psychological functioning in the observer that leads to a response in the form of a perception of the object. "The proximal stimulus, this message from the object, is formed and determined not only by the object, but also by facts that are really independent of the object and that therefore falsify or cloud the message" (Heider, 1980, p. 5). This perception, albeit constructed in part by the observer, becomes a part of the observer's reality. Common Sense, Common-Sense Psychology, and Attribution Heider (1958, p. 2) described common sense as an "unformulated" and "vaguely conceived" awareness of a situation that, because individuals understand themselves and others, allows those individuals to interact with

PAGE 48

38 Others. Common sense also could be defined as a system of beliefs that are customarily accepted to be true in everyday life (Hewstone, 1983). Common-sense psychology makes use of the "unformulated or halfformulated knowledge of interpersonal relations as it is expressed in our everyday language and experience" (Heider, 1958, p. 4). Common-sense psychology explains attributory behavior by asserting that individuals' perceptions of themselves and their environment, and the individuals' resultant reactions concern "perceptions about the causes of behavior and not the actual causes of social behavior" (Snyder, 1976, p. 54). Heider (1958) wrote that common-sense psychology has scientific value for two reasons. First, it provides the principles used in social perceptions and reactions (i.e., individuals interpret other people's actions, and from there predict what those people will do). Second, common-sense psychology contains subjective truth based on data from individuals' environments. That is, it relies on the obvious. Scientific Thinking and Attribution Kelley (1973) acknowledged that Heider's (1958) writings were the principle source of the hypotheses in attribution theory; however, Kelley (1973) apparently thought effective understanding of attribution theory involves gathering clarifying information from the attributer and, therefore, that Heider's (1958) theses insufficiently explained many phenomena. Kelley (1973) offered an addition to Heider's (1958) work, which he called

PAGE 49

39 his ANOVA model of the attribution process. Kelley's (1973) covariation theory of attribution proposes that people act as social scientists when constructing their reality. Kelley (1973, p. 108) stated that it is possible to infer cause of an event from observation of covariance. "An effect [dependent variable; e.g., attribution of credibility] is attributed to the possible cause(s) [independent variables; e.g., characteristics of entities observed, nature of circumstances, characteristics of observers] with which, over time, it covaries." Kelley (1973) claimed that there are three classes of causes for attribution: (a) the entities that are present (e.g., clinical supervisors), (b) the time or circumstances in which the action occurs (e.g., diagnostic action by clinical supervisors in practica/internships settings), and (c) the perceptions and characteristics of the observers who are present (e.g., counselor education students). How does an individual determine if her/his perceptions are valid foundations upon which to base attributions'^ Kelley (1973) maintained that the test for validity of a perception is (a) consensus of perceptual response among observers of the event, (b) consistency of observers' responses over time or modality, (c) the presence of a distinctive stimulus that initiated the perception, or (d) a combination of these factors. Various combinations of high and low distinctiveness, high and low consensus, and high and low consistency lead to attributions being granted

PAGE 50

40 to a single cause of an event or to a combination of causes. For example, high distinctiveness, high consensus, and high consistency lead to attribution to an entity; whereas, low distinctiveness, high consensus, and high consistency lead to attribution to the observer and the entity (Jones, 1990). The purpose of this study was to determine, through responses to a bogus profile of a troubled client, the extent to which credibility is granted to clinical supervisors (entities) by counselor education students (observers). The covariation of the causes (independent variables) with the effect (dependent variable) were examined. The independent variables were supervisors' educational credentials, supervisors' gender, students' gender, and proportion of counselor education program hours completed that are required for graduation by students. The dependent variable in this study was the degree of attribution of credibility granted to clinical supervisors by counselor education students. The tests for validity of the attribution were consensus (agreement among subjects regarding level of attributed credibility), reaction to distinctive stimuli (supervisors' diagnosis portrayed in bogus profile), and consistency (agreement among students in various counselor education departments with the assumption that these attributions have been made previously). Kelley's (1973) ANOVA model of attribution was the basis for explaining the extent to which counselor education students grant

PAGE 51

41 credibility to clinical supervisors based on their educational credentials and gender and counselor education students' proportion of counselor education program hours completed that are required for graduation and gender. However, quantitative analyses were not sufficient to explain subjects' relevant thoughts and experiences, the "complex whole" (Dewey, 1984, p. 246) being examined in this study. For that reason, the data were subjected to qualitative analyses and interpretation as well. Fischoff (1976) offered the intriguing suggestion that people are not intuitive scientists as Kelley (1973) wrote, but rather are intuitive historians. He wondered if it might be valuable to produce "an integrated psychology of predictive and explanatory behavior that accommodates the historians' observations, the philosopher's formalizations, and the psychologists' and the sociologists' theories and empirical findings" (Fischoff, 1976, p. 445). Elaboration on Need for the Study Research on Perceptions and Attributions of Parties in Counseling Process The act of counseling is not done in a vacuum with interaction only between clients and counselors who are unaffected by extraneous elements. The proceedings are mediated by biological, psychological, and social forces emanating from and acting on clients and counselors who are frequently supervised. For this reason, parties immediately involved in a counseling experience (i.e., clients and counselors) form perceptual inferences about those with whom they are interacting.

PAGE 52

42 Clients' view of clinicians Fundamentals of attribution theory, although not always identified as such, have been used to describe the circumstances in which clients attribute credibility to counselors. The impact of distinctive stimuli is apparent in a study in which Holland, Atkinson, and Johnson (1987) tested 209 undergraduates and found that counselors who expressed opinions about sexual behavior that were similar to those of students were perceived as credible. There were no findings that likenesses in gender between counselors and respondents affected the results. Ponce and Atkinson (1989) found in a study of 169 Mexican-Americans that counselors of the same ethnicity as the subjects were perceived as more credible than those who were not. The result of consensus of opinion is illustrated in a study of 485 high school students who attributed credibility to particular counselors. Littrell, Caffrey, and Hopper (1987) reported that counselors' reputations influenced the subjects' precounseling preferences for counselors and the subjects' perceptions of credibility of the counselors. Does consistent evidence of counselor expertise promote counselor credibility'!' Strong (1968) stated that in addition to diplomas, certificates, and professional libraries, certain behaviors by counselors are indicators of their expertise. The initial interview with clients serves not only diagnostic purposes, but also affords counselors the opportunity to effectively

PAGE 53

43 communicate their expertise by discussing the roles and responsibilities of clients and counselors as well as the process of counseling and problem solution. In 1981 Sue wrote that if clients perceive counselors as lacking credibility, clients may be reluctant to trust them. Counselor education students may learn from their supervisors how to convey a perception of credibility. Clinicians' view of clients Research has linked attribution theory to clinicians' judgments of clients' mental states. Since the cognitive-behavioral approach to counseling gained importance in the 1960s, attribution theory sometimes has been a foundation for study of clinicians' judgments (i.e., diagnoses, predictions, and etiology of clients' conditions) (Lopez & Wolkenstein, 1990) and for the study of clients' perceptions of their mental states. Success in using attribution theory to explain clients' mental states led to the development of attribution therapies in which clients' perceptions of causal events for their problems are used in diagnosis (Graham & Folkes, 1990). The effectiveness of attribution therapy has not been tested in the field, but attribution theory has been a foundation for understanding and alleviating maladaptive conditions such as depression and anger (Forsterling, 1990).

PAGE 54

44 Clinical Training of Counselor Education Students Counselor education students prepare to work in settings where diagnoses generally are required and where knowledge of mental disorders enables counselors to treat clients competently. These sites include schools, businesses, and other places where most counselors are now employed and/or are projected to be employed (i.e., community mental health agencies, inpatient and outpatient treatment facilities, and individual private practices) (Ritchie, Piazza, & Lewton, 1991, Vacc & Loesch, 1994). It may be assumed that a growing impact of mental health counselors on the mental health profession would be observed if descriptive data gathered in 1982 1983 were made current. In 1982 1983, the majority of people who sought help for relationship problems involving marriage, family, job, peers, school, and community saw psychologists and social workers. The same survey showed that 38% of psychologists' weekly hours and 32% of social workers' weekly hours were spent working with clients diagnosed as having neuroses, anxiety, and personality disorders (Knesper & Pagnucco, 1987). The DSM-IV (APA, 1994) serves as a basis for common lexicon when mental health providers discuss assessment, diagnosis, and treatment. DSM-IV language also is a basis for referring clients who counselors prefer not to assist and a basis for third party payments for the rendering of

PAGE 55

45 mental health services. Government and private health insurance companies require diagnostic codes on health insurance claim forms. Clinicians are called upon to testify in court cases where expert opinions are sought. An expert may be asked to testify about specifics such as diagnoses of individuals, and are expected to be familiar with prevailing norms (Faust & Ziskin, 1988). Finally, DSM-IV criteria may be used for testing the effectiveness of various treatments. Many counselors believe that the process of diagnosis provides a focal point for consideration of clients' (a) patterns of behavior, (b) structural components such as self-image and temperament, and (c) functional components such as cognitive style and psychodynamics. Contextual factors also are considered. These include situations in which problems arise for clients (Kottler, 1991). Counselor education students are expected to learn to incorporate all of these types of data into meaningful and useful information in the clinical reasoning process. Clinical reasoning Elstein and Bordage (1988) described the ideal clinical reasoning process as composed of four stages: (a) cue acquisition, wherein clients' histories are taken, psychological testing is done, and the influence of possible medical problems is considered; (b) hypothesis generation, wherein alternative explanations for clients' problems are considered and either accepted or ruled out; (c) cue interpretation, wherein data are

PAGE 56

46 analyzed; and (d) hypothesis evaluation, wherein hypotheses are confirmed or not confirmed, and if necessary, new hypotheses formed until verification is accomplished and diagnoses are determined. The development of clinical reasoning skills is ongoing during the education of counselor education students and is vital to their being able to offer effective counseling. The American Psychiatric Association warns that the diagnostic criteria found in their diagnostic manual are only guidelines and that the proper use of the manual requires "familiarity with the description of the diagnostic categories and with definitions of technical terms" (APA, 1987, p. v), as well as "specialized clinical training that provides both a body of knowledge and clinical skills" (p. xxix). Educational preparation for diagnosis Counselor education students probably are receiving inadequate instruction in diagnosis and treatment planning. Of the 146 respondents to Ritchie, Piazza, and Lewton's (1991) survey of counselor education programs, 3 1 taught nothing about the DSM-III-R (APA, 1987), and only 34 of the 146 programs taught the use of the DSM-III-R as an independent course. Professionals in counseling are calling for counselor educators to be more aware of the need to adequately prepare counselor education students to meet the clinical demands of the workplace. Theodore R. Remley, Jr., recent Executive Director of the American Counseling Association,

PAGE 57

47 advocated that the ACA Governing Council recommend that "CACREP [require] core work in diagnosis and treatment planning of emotional and mental disorders, regardless of specialty" (1993, p. 4). His opinion was echoed by Gail Robinson, past president of the American Mental Health Counselors Association: I support a core [curriculum] including courses in diagnosing and treating mental and emotional disorders. All practicing counselors should understand the continuum of mental health care and all must have the knowledge and skills to diagnose, treat, and refer clients with mental and emotional problems. (1993, p. 16) Waldo, Brotherton, and Horswill (1993) described a counselor education program in which School Counseling, Marriage and Family Therapy, and Mental Health Counseling tracks incorporated instruction in the use of the DSM-HI-R (APA, 1987) into their curriculum on the premise that it is complementary, rather than antithetical, to all three modalities of counseling. They explained that school counselors' humanistic perspective is found in the DSM-HI-R focus on clients' experiences and self-reported symptoms. Students can be perceived as individuals worthy of respect who are having problems that can be alleviated. Hohenshil (1993) has noted the trend toward training school counselors to diagnose using the DSM-HI-R (APA, 1987). School counselors need to be able to converse knowledgeably with school psychologists, special education teachers, and referral agencies. School counselors should be aware of the circumstances in which students need to

PAGE 58

48 be referred to mental health providers who can offer clinical help that is not available in schools. Therapy based on a systemic philosophy is consistent with use of the DSM (APA, 1994) to identify clusters of symptoms and psychosocial stressors. Rather than focusing solely on etiology of disturbances, viewing individual complaints as symptomatic of a dysfunctional family system is congruent with the philosophy of Marriage and Family therapists (Waldo, Brotherton, & Horswill, 1993). Mental health counselors are encouraged to use diagnoses based on DSM-IV (APA, 1994) criteria to design effective treatment plans. Attention to all five axes in a DSM-IV based diagnosis enables mental health counselors to treat clients holistically. The importance of teaching proper use of the DSM-IV (APA, 1994) should not be ignored. Fong (1993) hypothesized that counselor education students have difficulty diagnosing accurately and the reason is because they are frequently taught the content of the diagnostic manual (i.e., signs, symptoms, diagnostic codes) but not the process of diagnosis. The process of diagnosis involves assessing clients by observing their behavior, interviewing them (and perhaps their family members), examining their mental status, and, finally, generating hypotheses about the clients' conditions by noting patterns such as clients' psychological and physiological abnormalities and consulting the DSM (APA, 1994) decision

PAGE 59

49 trees. Further, Fong (1993) provided the analogy that teaching diagnostic manual content, but not diagnostic process, is akin to expecting counselor education students to be competent counselors after teaching them counseling theories, but not techniques. Furlong and Hayden (1993) suggested using available computer programs to enhance the teaching of diagnostic skills. They proposed that counselor educators make available to their students programs such as DTREE (First, Williams, & Spitzer, 1988), wherein students could become familiar with using the diagnostic manual decision trees, SCID SCORE (First, Gibbon, Williams, & Spitzer, 1991), in which students would use data from the Structured Clinical Interview for DSM-III-R (APA, 1987) to frame diagnoses, and the DICA-R (Reich, Welner, & Herjanic, 1990) to arrive at diagnoses for children and adolescents. Clinical Practice of Supervisors and Supervisees Knowledge of beginning counselors' attitudes toward their supervisors' diagnostic credibility could affect client care and management procedures in staffing facilities. Carey, Williams, and Wells' (1988) study of 3 1 counseling students in their first practica showed that trustworthiness (a dimension of credibility) in supervisors was one of the most esteemed characteristics of supervisors. Their findings were compatible with the earlier study of Heppner and Handley (1981) who found that trustworthiness and expertness (i.e., credibility) correlated with

PAGE 60

50 supervisees' satisfaction with supervision. Because novice counselors perceive the need to trust their supervisors, and because counselor education students are expected to consult with supervisors while providing counseling services, client care could be imperiled if the people who assess, diagnose, and suggest treatment modalities were viewed as incompetent diagnosticians by counselor education students. Counselors are expected to have logical grounds for the diagnoses they make and the treatments they offer. If counselor education students were to believe that the judgments of their supervisors were untrustworthy, counselor education students could experience ethical dilemmas and emotional conflicts. Counselor education students who have received inadequate instruction in diagnostic procedures could possibly jeopardize client care. They could fail to have a reasoned approach to treatment or referral because they might lack reliable and relevant information, and could fail to understand the need to be informed. They also could compromise the integrity of the mental health care system by failing to act knowledgeably and responsibly, or by unquestioningly following the lead of supervisors or other diagnosticians, none of whom would be infallible. In addition, assigning students to practicum and internship sites and to supervisors could be influenced by results of this study. The feasibility and desirability of using doctoral students or mental health professionals not associated with the counselor education department as clinical

PAGE 61

supervisors also could be of interest to faculty in departments of counselor education. Perceptions of Credibility of Clinical Supervisors Counselor education students' supervisors are often unseen influences in the counseling relationship. Supervisors are sources of information for students to use in assessing, diagnosing, and treating clients. Worthington (1984) discovered a relationship between supervisors providing acceptance and support for supervisees as well as teaching them case conceptualization, intervention, and other skills to perceived competence of supervisors. The sensitivity to evaluation felt by counselor education students, particularly during their first practica and their internships (Worthington, 1984), warrant providing supervisory assistance and encouragement to those students. Worthington (1984) proposed that it is during the first practica that counselor education students first view themselves as counselors, and it is during their internships that they first perceive themselves as professionals. It is beneficial to counselor education students during these times of professional identity formation for supervisors to be aware of the progressive nature by which students become counselors. On what basis can it be assumed that if counselor education students find their clinical supervisors to be credible diagnosticians the students will conform to the paradigms preferred by their supervisors? Social

PAGE 62

52 psychologists have presented an answer to this question. Tedeschi and Linkskold (1976) wrote that social influence is the impact one person has on another to the extent that a change in behavior or attitude consequently is produced in the target person. They further explained that people perceived as experts are capable of encouraging conformity in others. This opinion is supported by the theory of reciprocal determinism in which Bandura (1978) implied that counselor education students, perceiving their supervisors as credible and themselves as needing guidance in clinical matters, would seek direction from their supervisors. If the students were satisfied with the outcome, they would incorporate the knowledge into their repertoire of skills and information. Another position was offered by Heppner and Handley (1981). In a 15-week study, they tested students in their first practica and found that students' perceptions of supervisor expertness and trustworthiness correlated with satisfaction with the supervisory experience; however, these perceptions were not highly correlated with the students' ratings of supervisor impact. Heppner and Handley (1981) argued that expertness is necessary, but not sufficient to influence students' counseling behavior. They go on to suggest that other variables such as supervisors' legitimate power, their ability to effectively produce change, and supervisees' characteristics also may be correlated with supervisor impact.

PAGE 63

53 Doctoral students Heppner and Handley's (1981) study of 33 first practicum students' perceptions of their 20 supervisors whom the subjects knew to be advanced doctoral students revealed that supervisors' trustworthiness was rated as more meaningful than expertness. It is important to recall from the discussion above of Heppner and Handley's (1981) study that students' perceptions of supervisor expertness and trustworthiness correlated with satisfaction with the supervisory experience. However, these perceptions were not highly correlated with the students' ratings of supervisor impact on personal and professional behavior. Therefore, it can be concluded that credibility of doctoral students is necessary for satisfactory supervisory experiences for entry-level students, but it does not necessarily follow that supervisory experiences for these supervisees will be substantially productive. Poidevant, Loesch, and Wittmer (1991) surveyed 105 doctoral students in 13 CACREP-accredited programs. The object of the study was to determine the professional work settings and work activities preferred by counselor education students and how their perceived levels of efficacy related to their preference for work activities. Of thirteen possible choices of work settings, the respondents showed the greatest preference for college or university teaching and private practice. The respondents also showed a preference for four work-related activities: (a) counseling,

PAGE 64

54 (b) teaching and training, (c) consultation, and (d) supervision and training. Students chose supervision and training as the area in which they had the greatest efficacy, followed by the fields of teaching and training, consultation, and counseling. "Doctoral students' perceptions of self-efficacy are likely to have an impact on their professional functioning relative to setting, activities, and competence" (Poidevant, Loesch, & Wittmer, 1991, p. 290). All of the preferred jobs and all of the preferred work activities indicated by the respondents in Poidevant, Loesch, and Wittmer's (1991) study potentially require a thorough knowledge of assessment, diagnosis, and treatment. Apparently counselor education doctoral students think they have acquired the skills necessary to properly meet the needs of clients, students, consultees, and supervisees. However, it was unknown to what extent supervisees find doctoral students to be credible diagnosticians. Doctorate-holding psychologists Worthington (1984) learned from a nationwide survey of 237 counseling supervisees who were supervised by doctoral students in psychology and by Ph.D. holding psychologists that doctoral students and psychologists were not perceived differently in terms of being competent, providing satisfactory supervisory experiences, or having equal impact on supervisees' counseling ability. In addition, of 48 supervisor behaviors (e.g., established goals and provided feedback ), only one, the use of

PAGE 65

55 humor, appeared as a significant difference between the groups. Humor was used more often by more experienced supervisors than by less experienced supervisors. In a study that determined the differences in how 60 undergraduate female students in psychology classes attributed credibility and attractiveness based on fee charged and title to counseling psychologists and professional counselors with master's degrees, Conoley and Bonner (1991) found there was no significant main effect for variations in professional title. Apparently, perceptions of credibility of psychologists and counselors based on title among undergraduate female psychology students do not differ. Some students who enter counselor education programs come from this population. Doctorate-holding faculty members In a 1985 study, counselor education and counseling psychology doctoral students were asked to complete anonymous questionnaires rating their current clinical supervisors. Supervisors who were rated in this study by Worthington and Stern were faculty members or doctoral candidates associated with that department or with affiliated practicum sites. Results showed that supervisees believed that, regardless of supervisor status, the supervision relationship improved as supervision progressed, and that faculty member status was not a factor in supervisees' ratings of

PAGE 66

56 supervisors' effect, competence, and the quality of the supervisory relationships. Psychiatrists Although subjects in Truatt and Bloom's (1982) study were not counselor education students, the results can be extrapolated. Truatt and Bloom (1982) studied university students' ratings of credibility for psychiatrists, clinical psychologists, and counselors. Although aspects of the methodology used in Truatt and Bloom's study are suspect and therefore not useful for general comparison among psychiatrists, psychologists, and counselors, it is interesting to note that subjects attributed the greatest credibility to psychiatrists because of their title of "doctor." The clinical psychologist and the counselor were not so denoted in the study. Greenberg and Wursten (1988) questioned if psychiatrists are perceived as more credible and more persuasive than psychologists when they provide mental health testimony in court. Students in an introduction to psychology course rated the psychiatrists as more credible than the psychologists. The authors admitted that credibility was the only variable studied, which enhanced the internal validity of their study, but because they did not test attractiveness, likability, and dynamism, all characteristics that could influence a judge or jury, the study's generalizability was limited.

PAGE 67

57 Although the subjects in Greenberg and Wursten's (1988) study were not counselor education students, the results may be indicative of the thinking of the general public and therefore of counselor education students as well. The limitations of their study apply to the present study in that the generalizability of the results are constrained by lack of attention to other variables such as Greenberg and Wursten (1988) noted and which could affect supervisees' perceptions of credibility. Genders of supervisors and supervisees Because recent research on the influence of gender on the perceptions of the supervisory process by supervisees has been scant, it was unknown if perceptions of the degree of credibility as diagnosticians of female clinical supervisors is different from perceptions of credibility as diagnosticians of male clinical supervisors. Therefore, results of this study may be significant to departments of counselor education faculty and to mental health facility administrators when assigning supervisors to supervisees. Worthington and Stern (1985) addressed the issue of gender in the supervisory relationship. They found that male supervisees believed they had better relationships with supervisors, regardless of gender, than did female supervisees. Supervisor supervisee dyads matched by gender received more favorable ratings by supervisees than did unmatched dyads.

PAGE 68

58 Schiavone and Jessell (1988) studied 86 master's level counselor education students. Their study revealed that during initial interactions, counselor education students' perceptions of supervisors' expertness and competence does not differ as a function of the gender of supervisors or supervisees. Proportion of program hours completed by supervisees Apparently no research related to the number of program hours completed by counselor education students and their attitudes toward supervisors' diagnostic credibility has existed. Results from this study however, reinforced Stoltenberg and Delworth's (1987) theory of development in counselors and Usher and Border's (1993) study of counselors' preferences for supervisory style. It is known that supervisees' needs differ according to their development as counselors and that developmental level and amount of training are not interchangeable concepts (Wiley & Ray, 1986). Stoltenberg and Delworth (1987) outlined the changes in counselor characteristics as counselors progress through three levels of development. If Stoltenberg and Delworth (1987) are correct, and if the supervisory conditions present for counselors during this progression meet students' needs, then it would be expected that beginning counselors, (i.e., counselors in Stoltenberg and Delworth's Level 1) would rely to a large extent on supervisors' diagnoses. Stoltenberg and Delworth's (1987) belief

PAGE 69

59 was that by the time counselors move beyond Level 1, through Level 2, which is characterized by "fluctuating motivation, striving for independence, becoming more self-assertive and less imitative" (p. 70) and reach Level 3, they can function fairly independently. Usher and Borders' (1993) study supports Stoltenberg's (1981) thesis that different supervisory experiences are more likely to be perceived as more beneficial by practicing counselors than by counselors who have developed less professionally. Using the Supervisor Emphasis Rating Form-Revised (SERF-R), Usher and Borders (1993) learned that professional counselors who are not school counselors are more interested in having assistance in conceptualizing cases and choosing interventions than in learning counseling skills and behaviors. Therefore, it would be expected that counselor education students who were completing their final clinical field experiences would seek less direction from supervisors, and would be more likely to question debatable diagnoses and feel comfortable in confirming those diagnoses with which they have no argument. Summary It is currently believed that mental disorders are caused solely, or in combination, by biological, psychological, and sociological factors. Therefore, it is prudent to codify mental disorders and, using diagnoses and theoretical approaches to etiology and counseling as guides, design appropriate treatment plans for clients. Since 1952, The Diagnostic and

PAGE 70

60 Statistical Manual of Mental Disorders (APA) has been the authoritative source for diagnostic criteria. Ongoing controversy has surrounded the philosophical approach to diagnosis, and the medical and psychosocial models (both at times favored by psychiatrists) have been incorporated into a descriptive, generally atheoretical, biopsychosocial model currently used in the DSM-IV (APA. 1994). Codifying diagnoses of mental disorders affords clinicians and researchers a common means to communicate among themselves, and provides a universal and predictable framework from which to diagnose, group, and treat clients. A thorough understanding of clients' psychological, emotional, cultural, and developmental stages and needs are necessary to effectively diagnose and counsel clients. Counselor education students engage in coursework and field experiences to gain the expertise needed to be effective counselors. During their practica and internships, counselor education students receive assistance and direction from qualified clinical supervisors. Research has shown that expertness and trustworthiness, the components of credibility, are characteristics counselor education students value highly in their supervisors. However, it was unknown to what extent counselor education students consider their clinical supervisors to be credible diagnosticians, people whom they can trust to be reliable role models and tutors. In the present study, based on Kelley's (1973) ANOVA model of attribution

PAGE 71

61 theory, the extent to which counselor education students grant credibility to clinical supervisors based on supervisors' educational credentials and gender, and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation was investigated. Male and female counselor education students who had completed varying proportions of their entry-level counselor education program hours were subjects in this study. Counselor education students' perceptions of the diagnostic credibility of male and female counselor education doctoral students, male and female psychologists with doctoral degrees, male and female counselor education department faculty members, and male and female psychiatrists acting as clinical supervisors were subjected to analysis of variance and post-hoc follow-up analyses of significant interaction. Qualitative analyses of subjects' replies were done to achieve a broader understanding of the subjects' views on clinical supervisors' diagnostic credibility. Researchers have examined perceptions among parties in the counseling process. However, no known previous study had examined the extent to which counselor education students attribute credibility to clinical supervisors. Therefore, the results of this study may be used in clinical training of counselor education students, staffing practices of

PAGE 72

62 mental health facilities, and assignment of clinical supervisors in counselor education departments.

PAGE 73

CHAPTER 3 METHODOLOGY The extent to which counselor education students consider their clinical supervisors to be credible diagnosticians was unknown. Therefore, in this study, Kelley's (1973) ANOVA model of attribution theory was used as a basis to determine the degree of diagnostic credibility counselor education students grant their clinical supervisors. Delineation of the Variables Supervisors' types of credentials (i.e., M.D., doctorate-holding counselor education faculty, doctorate-holding psychologist, or doctoral student holding master's degree in counselor education) and gender and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation (i.e., 2/3) were the independent variables examined in this study. The dependent variables were the scores obtained from a Likerttype index of subjects' approval of clinical supervisors' diagnoses. Additionally, through analyses of subjects' written narratives, a qualitative description of subjects' reasons for arriving at approval scores was framed. 63

PAGE 74

64 Population Responses that met criteria for use in this study came from subjects in nine of the 27 counselor education departments in the United States that offer both CACREP-accredited doctoral-level programs (i.e., Ph.D. and Ed.D.) in Counselor Education and Supervision and CACREPaccredited entry-level (i.e., master's degree) programs in Community Counseling or Mental Health Counseling. Representatives of three other programs agreed to participate in this study, but failed to return any of the 100 instruments sent to them, even though follow-up post cards (Appendix H) were sent to them and telephone calls were made reminding them of their agreement and requesting their cooperation. In addition, responses from 60 completed instruments were not included in the data analyses because the respondents did not meet the criteria for inclusion in the study. Because doctoral students holding master's degrees in counselor education were one of the types of clinical supervisors whose educational credentials were being examined in this study, it was preferable to use subjects from counselor education programs who potentially would have experience as supervisees of doctoral students. Therefore, only male and female entry-level counselor education students who had completed varying proportions of their counselor education program hours required for graduation from mental health counseling or community counseling tracks were subjects in this study.

PAGE 75

65 Sampling Procedure Attempts were made to contact 27 CACREP liaison persons associated with the programs identified as suitable for study. Those persons who were available to be spoken with by telephone were asked if their Community Counseling or Mental Health Counseling programs and their doctoral-level programs in Counselor Education and Supervision were still CACREP-accredited, If their programs currently were accredited, the liaison persons were asked to suggest the names of professors in their departments who could be asked to cooperate for this study. Nine CACREP liaisons offered to cooperate in helping to conduct the study. They were sent letters of introduction (Appendix A) and the number of packets of testing materials they suggested would be appropriate. Six other CACREP liaisons offered the names of 12 counselor education department faculty as contacts. Seven of these people were able to be reached by telephone. Five of the contacts were not available when calls were made to them, or they did not return calls when messages were left. Those who agreed to participate were sent letters of introduction and appropriate testing materials. The target size for the sample was 300 Unfortunately, almost nothing is known currently about the demographic characteristics of students currently enrolled in counselor education curricula. Therefore, professors who administered the instruments were asked to do so in a manner as random as possible, and it is assumed that

PAGE 76

66 the resultant sample was an effective representation of entry-level counselor education students. Research Procedure Assessment materials were sent to the counselor education department faculty who agreed to assist in conducting this study. Fourpage assessment packets were included for each subject and contained (a) a Demographic Inventory (Appendix C), (b) an assessment instrument (i.e., a two-page profile of a fictitious client) (Appendix D), and (c) an Approval Rating Scale and Qualitative Data Form (Appendix F). Administrators were provided with instructions about how to distribute, retrieve, and return the assessments (Appendix B) and were provided self-addressed, stamped envelopes in which to return the completed materials. Demographic Inventory Personal information pertaining to the subjects was necessary to conduct this study. Therefore, relevant questions were asked of the subjects about the types of programs in which they were enrolled, number of program hours they had completed, and their gender (Appendix C). Assessment Instrument Each subject was presented with intake interview data about a fictitious client who was purportedly seeking help from the subject for a mental health problem (Appendix D). The case presented in this study was

PAGE 77

67 a profile of a woman who was diagnosed with Borderline Personality Disorder, one of more than 200 real case studies compiled by Spitzer et al,, (1994, p. 395). Spitzer et al., (1994) drew from their own and others' clinical and teaching experiences to present edited cases that contain information needed for differential diagnosis. Each assessment instrument contained one of eight descriptions of fictitious clinical supervisors who had diagnosed the fictitious client as having Borderline Personality Disorder. The client's clinical profile and diagnosis was constant across conditions presented to subjects; however, the credentials and gender of the diagnosticians were not constant They reflected potentially different credibility based on diagnosticians' educational credentials and gender. The fictitious clinical supervisors who were presented as having arrived at the diagnosis of Borderline Personality Disorder for the fictitious client were described as either male or female and as holding one of four educational degrees (Appendix E). Approval Rating Scales (Appendix F) on which subjects recorded the extent to which they agreed with their clinical supervisors' diagnoses were coded with ciphers that corresponded to the categories of variables in which the fictitious clinical supervisors were positioned in the assessment instrument. Genders and educational credentials were presented in combination to allow for eight variants (see Appendix G).

PAGE 78

68 Agreement Rating Scale and Qualitative Data Form Each subject was asked to indicate on a Likert-type scale the extent to which they agreed with the supervisor's diagnosis. In addition, subjects were asked to explain the bases of their responses by describing the factors or considerations that influenced their responses. A pilot study with a small number of students indicated that minor changes needed to be made in the instructions to the subjects. A few respondents claimed they could not indicate the extent to which they agreed with their clinical supervisors' diagnosis because they believed they were untrained to arrive at a conclusion as to the accuracy of their clinical supervisors' diagnosis. Therefore, rather than simply asking the respondents the extent to which they agreed or disagreed with their clinical supervisors' diagnosis, respondents were advised that it was recognized that there may be limitations to their ability to diagnose the client at the time they were asked to do so, but they were asked to indicate, nonetheless, the extent to which they agreed with the diagnosis provided. Data Analyses Quantitative Analyses Data were analyzed using a four-way factorial analysis of variance. The independent variables were the supervisors' gender (i.e., male or female) and educational credentials (i.e., M.D., Ph.D. holding counselor education faculty, Ph.D. holding psychologist, and doctoral student with

PAGE 79

69 master's degree in counselor education) as well as subjects' proportion of program hours completed that were required for graduation (i.e., 2/3) and subjects' gender (i.e., male or female). Post hoc comparisons of group means were done using Tukey's Test of Honestly Significant Differences, with a minimum significant difference of 2.53 using a significance level of .02. Qualitative Analyses To answer questions such as "How did the subjects in this study think about the diagnosis of the client'^" and "What factors did the subjects consider when arriving at conclusions about the diagnostic credibility of their supervisors?," the context in which the hypothetical supervisors and supervisees were interacting when diagnostic decisions were made was investigated through an examination of counselor education students' thinking when they responded to the supervisors' diagnoses. Subjects had the opportunity to express their convictions, viewpoints, and values in order that a deeper and broader understanding of their responses could be attained and a sharper awareness of their perspectives achieved. Subjects were presented an open-ended question that solicited discussion of why the subjects chose to agree to the extent they did (Appendix F). The qualitative analyses of the data were accomplished in three phases. In Phase I, central ideas or prominent opinions expressed by

PAGE 80

70 the subjects as the reasons for agreeing or disagreeing with the supervisors' diagnoses were identified. In Phase II, responses showing similar rationales were grouped. For example, those responses wherein various supervisors' perceived diagnostic expertise based on their educational credentials stimulated supervisees' attribution of credibility were grouped together. Likewise, those responses wherein some supervisors' diagnostic credibility was questioned because of perceived negligent practices in their process of diagnosis were combined. In Phase III, the themes that were distinguished, and the "recurring ideas and patterns of beliefs" that were identified (Marshall & Rossman, 1989, p. 116) were arranged further, with the result being three categories of responses that indicated the extent to which counselor education students attributed credibility to their clinical supervisors: (a) supervisors' diagnosis rejected, (b) supervisors' diagnosis provisionally accepted, and (c) supervisors' diagnosis accepted without qualification.

PAGE 81

CHAPTER 4 RESULTS Quantitative Analyses Analyses of variance and a post hoc comparison of group means were used to examine the following null hypotheses: 1. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials. 2. There is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' gender. 3. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students. 4. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation. 5. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' gender. 71

PAGE 82

72 Supervisors' types of credentials (i.e., M.D., doctorate-holding counselor education faculty, doctorate-holding psychologist, or doctoral student holding master's degree in counselor education) and gender and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation (i.e., 2/3) were the independent variables examined in this study. The dependent variables were the scores obtained from a Likerttype index of subjects' approval of clinical supervisors' diagnoses. A four-way factorial analysis of variance was performed on the data resulting from the responses of 191 subjects. Tukey's Test of Honestly Significant Difference was performed for variables having more than two levels. In Table 4 1 are the summaries of the ratings given by counselor education students of the clinical supervisors by their levels of educational credentials and proportion of program hours completed by the subjects. The ANOVA source table is found in Table 4-2. The significant interaction between supervisors' educational credentials and the proportion of program hours completed by students for graduation can be seen in Figure 1. There is a difference greater than the Tukey minimum HSD of 2.53 in two data sets: credentials and program hours. One in regard to the latter is the difference of 3.95 between

PAGE 83

Table 4 1 Summaries of Ratings by Levels of Educational Credentials and Proportion of Program Hours Completed 73 Variable Value Label Psychiatrist Doctorate-Holding Faculty Doctorate-Holding Psychologist Masters-Holding Doctoral Student 1.00 2.00 3.00 4.00 Less Than One-Third of Program Completed 1.00 One-Third to Two-Thirds of Program Completed 2.00 Greater than Two-Thirds of Program Completed 3.00 Variable

PAGE 84

74 Table 4 2 Analysis of Variance of Ratings of Diagnostic Credibility by Educational Credentials, Gender of Supervisors, Gender of Subjects, and Proportion of Program Hours Completed Rating By Educational Credentials (CRED) Gender Of Supervisor (GENSUP) Gender Of Subject (GENSUB) Proportion Of Program Hours Completed (HRS) Sum of

PAGE 85

75 regard to the former between the attribution of credibility to psychiatrists (mean = 7.38) and to doctorate-holding psychologists (mean = 4.29) by students who had completed more than one-third and less than two-thirds of their programs. It also can be seen in Figure 1 that as the proportion of program hours required for graduation increases from less than one-third to greater than two-thirds, attribution of diagnostic credibility to doctoral students rises from 5.16 to 5.29 to 5.93. In addition, attribution of diagnostic credibility to psychiatrists rises from a mean of 5. 1 6 among students who had completed less than one-third of their programs to 7.38 among students who had completed between one-third and two-thirds of their programs, but then falls to 3.43 among students who had completed more than two-thirds of their programs. Psychiatrists Faculty Psychologists Students -^ 2/3 hrs Figure 1 Cell means of educational credentials by proportion of program hours completed.

PAGE 86

76 Qualitative Analyses An analysis of counselor education students' thinking when they indicated the extent to which they agreed with their supervisors' diagnosis was done to determine how the subjects thought about the diagnostic credibility of their supervisors and which factors the subjects considered when arriving at their conclusions. Subjects were presented an open-ended question that solicited discussion of why the subjects chose to agree to the extent they did with the various types of supervisors (Appendix F). The qualitative analyses were accomplished in three phases. Central ideas or prominent opinions expressed by the subjects as the reasons for agreeing or disagreeing with the supervisors' diagnoses were identified in Phase I. In Phase II, those responses from subjects that indicated reasons for agreeing to the extent they did with their supervisors were arranged into eleven groups based on the rationales of the responses. For example, various supervisors' perceived diagnostic expertise based on their educational credentials stimulated some supervisees' unquestioning attribution of credibility. Likewise, some supervisors' diagnostic credibility was questioned because of perceived negligent practices in their process of diagnosis. The themes that were distinguished, and the "recurring ideas and patterns of beliefs" that were identified (Marshall & Rossman, 1989, p. 1 16) were arranged further in Phase III so that three major categories of

PAGE 87

77 responses were identified which indicated the extent to which counselor education students attributed credibility to their clinical supervisors: (a) supervisors' diagnosis rejected--groups #1 #4 (n=70), (b) supervisors' diagnosis provisionally accepted—groups #5-8 (n=54), and (c) supervisors' diagnosis accepted without qualification--groups #9 10 (n=50). A fourth category was composed of nine subjects who would not accept or reject the diagnosis until the issue of abuse was addressed with the client. Those responses in which subjects did not provide an explanation for their ratings were excluded. A representation of the four groups who rejected the supervisors' diagnosis and their reasons for doing so are shown in Figure 2. Group #1 was composed of subjects (n=21) who thought the diagnosis was premature. They believed the supervisor had insufficient data to justify the diagnosis of Borderline Personality Disorder (BPD). The subjects in Group #2 (n=9) believed the diagnosis was inaccurate, but acknowledged having limited knowledge of how to diagnose, so their rejection of the diagnosis was based solely on conjecture. Thirty three subjects (Group #3) rejected the diagnosis because they thought the symptoms presented were not those of BPD. The fourth group who rejected the supervisors' diagnosis (n=7) did so saying they were unaware of the symptoms of BPD, but they suggested other diagnoses as being more suitable than BPD.

PAGE 88

78 Alternative OfFercd (7) Inaccurate Based on Symptoms (33) Premature (21) Inaccurate Based on Conjecture (9) Figure 2 Reasons for rejecting supervisors' diagnosis and number of subjects who offered the reasons cited. Figures 3-6 show the response frequencies of male and female subjects in groups #1 #4 to the diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students. Subjects had completed varying proportions of their program hours required for graduation [i.e., less than one-third (<), between onethird and two-thirds (< >), or more than two-thirds (>)]. For reasons cited above, subjects in these groups rejected the diagnosis given by their clinical supervisors. Figure 7 is a summary of the response frequencies of the subjects in groups #1 #4. Figure 8 depicts the reasons some subjects gave for accepting with qualifications the supervisors' diagnosis. The diagnosis was accepted as possibly correct by the 34 subjects in Group #5; however, they offered other diagnoses for consideration as well as BPD. Although the diagnostic

PAGE 89

79 3.02.5 2.01 1.51.0 0.5 0.0 F PsychiatM Psychiat F Fee MFac F Psychol M Psychol F Sluden MSluden Female < Male < I I Female <> f~~] Male <> Female > Male > Figure 3 Group #1 response frequencies. Subjects (n=21) cited premature diagnosis as reason for rejection. 2.a 1.5 1.0 0.5 0.0 F PsychiatM Psychiat F Fac MFac F Psychol M Psychol F Sluden MSluden Female < Male < Female <> Male <> Female > Male > Figure 4 Group #2 response frequencies. Subjects (n=9) acknowledged limited knowledge of diagnosis, but rejected the diagnosis based on conjecture.

PAGE 90

80 F PsychiaM Psychi F Fac MFac F PsycholM Psycho F Sluden MSluden Female < Male < Female <> Male <> Female > Male > Figure 5 Group #3 response frequencies. Subjects (n=33) cited symptoms presented as being incompatible with diagnosis as reason for rejection. 2.0 1.& 1.0 0.5 0.0 F PsychiatMPsychiat F Fac MFac F Psychol M Psychol F Sluden M Sluden Female < Male < [ I Female <> I I Male <> Female > Male > Figure 6 Group #4 response frequencies. Although they were unaware of the symptoms of BPD, subjects (n=7) offered alternative diagnoses to BPD.

PAGE 91

81 F PsychiatrMPsychiat F Fac MFac F Psychol M Psychol F Sluden MStJden Female < Male < I I Female <> I I Male <> Female > Male > Figure 7 Summary of response frequencies for groups #1 #4. Subjects (n=70) rejected diagnosis. criteria for BPD were unfamiliar to the five subjects in Group #6, they did not rule out BPD as a possible diagnosis. Subjects in Group #7 (n=13) believed BPD was a possible diagnosis, but that other diagnoses should be ruled out before designing a treatment plan. Finally, the two subjects in Group #8 believed "professionals can be wrong," so they would not fully commit to the diagnosis offered by the supervisors. Figures 9-12 show the response frequencies of male and female subjects in groups #5 #8 to the diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students. Subjects had completed varying proportions of their program

PAGE 92

82 Cautiously Accepted (2) Tentatively Accepted (13) Unfamiliar Criteria (5) Possible (34) Figure 8 Reasons for provisionally accepting supervisors' diagnosis and number of subjects who offered the reasons cited. hours required for graduation [i.e., less than one-third (<), between onethird and two-thirds (< >), or more than two-thirds (>)]. For reasons cited above, subjects in these groups provisionally accepted the diagnosis given by their clinical supervisors. Figure 13 is a summary of the responses of subjects in groups #5 #8. In Figure 14 the numbers of subjects and the reasons for accepting the supervisors' diagnosis without question are shown. Group #9 (n=35) wrote that they thought the diagnosis was accurate because the symptoms presented met the criteria for BPD. The subjects in Group #10 (n=15) were less certain about the validity of the diagnosis based on their knowledge, but they said that they accepted the diagnostic judgment of their supervisors because they were willing to defer to their supervisors' experience and training.

PAGE 93

83 Ol F PsychiatrMPsychiat F Fac MFac F Psychol M Psychol F Sluden MStiden Female < Male < n Female <> Male <> Female > Male > Figure 9 Group #5 response frequencies. Subjects (n=34) wrote that BPD could be one diagnosis among several possibilities they offered. 2.(1 1.51.0 0.5 0.0 F PsychiatMPsychiat F Fac MFac F Psychol M Psychol F Sluden M Sluden Female < Male< I I Female <> I I Male <> Female > Male > Figure 10 Group #6 response frequencies. Subjects (n=5) were not familiar with the criteria for BPD diagnosis, but indicated that it could be one diagnosis among others.

PAGE 94

2.0, 84 1.& 1.0 0.5. O.OU F Psychia M Psychi F Fac MFac F Psychol M Psycho FStuden MSluden Female < Male < I I Female <> Male <> Female > Male > Figure 1 1 Group #7 response frequencies. BPD was considered a possible diagnosis, but subjects (n=13) thought other diagnoses should be ruled out. 1.0, 0.9 0.81 0.7 0.6 0.5 0.4 o.a 0.2 0.1 0.0 F PsychiatMPsychiat F Fac MFac F Psychol M Psychol F Sluden MSluden Female < Male < Female <> Male o Female > Male > Figure 12 Group #8 response frequencies. Subjects (n=2) had reservations and would not commit fully to the diagnosis offered by the supervisors.

PAGE 95

u J F PsychiatMPsychia F Fac MFac F Psychol M Psychol F Stud en MStjden Female < Male < I I Female <> Male <> Female > Male> Figure 13 Summary of response frequencies for groups #5 #8. Subjects (n=54) provisionally accepted diagnosis. Figure 15 and Figure 16 show the response frequencies to the diagnosis offered by male and female psychiatrists, counselor education faculty, psychologists, and doctoral students by male and female subjects in Groups #9 and #10 Subjects had completed varying proportions of their program hours required for graduation [i.e., less than one-third (<), between one-third and two-thirds (< >), or more than two-thirds (>)]. Subjects in these groups accepted the diagnosis given by their clinical supervisors. A summary of the response frequencies for these two groups is shown in Figure 17.

PAGE 96

86 Inability to diagnose (1 Based on Symptoms (35) Figure 14 Reasons for accepting supervisors' diagnosis without question and number of subjects who offered the reasons cited. F PsychiatrMPsychiat F Fac MFac F Psychol M Psychol F Sluden MSluden Female < Male < I I Female <> S Male <> Female > Male > Figure 1 5 Group #9 response frequencies. Subjects (n=35) reported diagnosis was accurate because symptoms presented met criteria for BPD.

PAGE 97

87 3.0 2.52.0 1.& 1.0 0.5 O.Ol F PsychiatMPsychiat F Fac MFac F Psychol M Psychol F Sluden MSuden Female < Male < I I Female <> Male <> Female > Male > Figure 16 Group #10 response frequencies. Subjects (n=l 5) uncertain about validity of diagnosis based on their knowledge, so they deferred to supervisors. F PsychiaIrM Psychiat F Fac MFac F Psych M Psych F Sluden M Sluden Female < Male < I I Female <> I I Male <> Female > Male > Figure 17 Summary of response frequencies for groups #9 and #10. subjects (n=50) accepted diagnosis.

PAGE 98

CHAPTER V DISCUSSION None of the null hypotheses addressed in this research were rejected. No significant main effects were found. It was discovered that there is no difference in level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials or gender. There is no significant supervisors' educational credentials by gender interaction for level of attribution of diagnostic credibility among counselor education students. There is no difference in level of attribution of diagnostic credibility to clinical supervisors based on counselor education students' proportion of program hours completed that are required for graduation or on students' gender. However, a significant interaction was found between the level of attribution of diagnostic credibility to clinical supervisors among counselor education students based on supervisors' educational credentials and counselor education students' proportion of program hours completed that are required for graduation. Limitations of the Study The results of this study are limited in regard to generalizability for several reasons. Strict controls for random selection of the sample were

PAGE 99

89 impractical. However, because the instrument was administered to counselor education students in CACREP-accredited programs at various institutions throughout the United States by counselor education faculty at those institutions, the instruments were administered in a manner as random as possible. Therefore, it can be assumed that the resultant sample was a satisfactory representation of entry-level counselor education students. The educational credentials of supervisors chosen for study here were not inclusive of all credentials supervisors of counselor education students in field experiences might hold. For example, licensed clinical social workers, licensed marriage and family therapists, and licensed and unlicensed mental health counselors with master's degrees were not represented as supervisors in this study. Only students from CACREP-accredited counselor education programs were selected for inclusion in this study. Students from programs with less rigorous academic standards or from programs that, for whatever reason, have not become CACREP-accredited were not part of the sample. The hypothetical race, age, theoretical orientation, and working environment of the clinical supervisors were not addressed. Also, the actual race, age, theoretical orientation, and extent and types of field experiences of the subjects were not investigated.

PAGE 100

90 As in Greenberg and Wursten's (1988) study, this investigation did not test attractiveness, likability, or dynamism, ail ciiaracteristics that could affect supervisees' perceptions of credibility of their supervisors. Subjects' and supervisors' perceived self-efficacy, expectations, or personalities also were not studied. Three levels of proportion of program hours completed that are required for graduation for counselor education students were used without a theoretical basis for cutoff points. It was assumed that in the beginning of their programs students would be less experienced in clinical matters than they would be in the middle or last thirds of their programs. Developmental levels of subjects as counselors were not ascertained. Stoltenberg (1981) and Stoltenberg and Deiworth (1987) wrote, and Wiley and Ray (1986) showed, that counselor education students exhibit qualitatively different needs, behavioral motivations, and resistance at various developmental levels during training; however, developmental level and amount of training (proportion of program hours completed that are required for graduation) are not interchangeable concepts. The limitations of this study render its results under-representative of all counselor education students. Therefore, generalizing conclusions to groups that are not similar would be inappropriate. However, the results should adequately represent counselor education students in CACREPaccredited community or mental health counseling programs.

PAGE 101

91 Relevance of Attribution Theory It appears that within the parameters of this study Kelley's ANOVA model of attribution theory aids in explaining the results. Kelley (1973) asserts that it is possible to infer cause of an event from observation of covariance. "An effect [e.g., rating of diagnostic credibility] is attributed to the possible cause(s) [e.g., characteristics of entities observed, nature of circumstances, characteristics of observers] with which, over time, it covaries." Kelley (1973) claimed that there are three classes of causes for attribution: (a) the entities that are present (e.g., clinical supervisors with various educational credentials), (b) the time or circumstances in which the action occurs (e.g., diagnosis by clinical supervisors in practica/internships settings), and (c) the perceptions and characteristics of the observers who are present (e.g., counselor education students with varying proportion of program hours completed that are required for graduation). Furthermore, Kelley's (1973) test for validity of the conclusions reached here is applicable. He wrote that the test for validity of a perception (diagnostic credibility or relative lack of it) is (a) consensus of perceptual response among observers of the event (seen in the significant interaction between supervisors' educational credentials and proportion of program hours completed that are required for graduation); (b) consistency of observers' responses over time or modality (data were collected from counselor education students in CACREP-accredited programs in various

PAGE 102

92 parts of the United States); (c) the presence of a distinctive stimulus which initiated the perception (identical diagnosis and supervisor for four different treatment groups); or (d) a combination of these factors. Conclusions and Interpretations What do we know about counselor education students' attribution of diagnostic credibility to clinical supervisors? If the analyses were restricted to information learned from the quantitative data we would know that there were no significant main effects, but that there was a significant ratings interaction between clinical supervisors' educational credentials and proportion of program hours completed that are required for graduation by counselor education students. We would also know that gender of supervisors and counselor education students was not a significant factor in the ratings of diagnostic credibility by counselor education students. Analyses of qualitative data garnered from a request that subjects provide an explanation why they agreed to the extent they did with the clinical supervisors' diagnosis illuminates and gives meaning to the quantitative data. The following discussion blends data from the qualitative analyses with that of the quantitative analyses in order to present a more holistic picture than is possible if only quantitative data are presented.

PAGE 103

93 Attribution of Credibility by Students With Less Than One-Third of Program Hours Completed Students who had completed less than one-third of the program hours required for graduation (n=98) did not attribute diagnostic credibility differentially to supervising psychiatrists, doctorate-holding counselor education faculty, doctorate-holding psychologists, and master's-holding doctoral students. These results do not conflict with the results of Heppner and Handley's (1981) study of first practicum students' perceptions of their supervisors in which it was learned that supervisors' trustworthiness was rated as more meaningful than expertness. In the present study, subjects who had completed less than one-third of their program hours did not consider educational credentials or gender to be critical factors when evaluating supervisors' diagnostic expertise. They considered all supervisors to be equally credible diagnosticians. Subjects who had completed less than one-third of their program hours required for graduation indicated their reliance on clinical supervisors by remarks such as "I don't have the knowledge to diagnose, so I let my supervisor do it" and "I have faith in my supervisor." Deference to clinical supervisors is characteristic of students in Level One of counselor development (Stoltenberg & Delworth, 1987). The subjects in this study who were relatively inexperienced wanted the supervisors to assume responsibility for case conceptualization. The

PAGE 104

94 supervisor was viewed as a "role model and resident expert" (Stoltenberg & Delworth, 1987, p. 62). Attribution of Credibility by Students With Greater Than One-Third of Program Hours Completed Attribution of credibility to psychiatrists Differences in attribution of credibility to psychiatrists among subjects was based on the proportion of program hours completed. Students with one-third to two-thirds of their program hours completed, and who were supervised by psychiatrists (n=8), attributed a significantly greater level of credibility to psychiatrists (mean = 7.38) than did students who had completed more than two-thirds of the program hours required for graduation and who were supervised by psychiatrists (n=14) (mean = 3.43). Dia.gnosis accepted. Subjects with one-third to two-thirds of their program hours completed who agreed strongly with psychiatrists cited client symptoms as the reason for agreeing with the supervisor. Subjects in this group generally accepted the supervisors' diagnosis, but were reluctant to offer unconditional support for their supervisors' expertness and trustworthiness. Diagnosis accepted with reservations Notwithstanding gender being insignificant in the quantitative results of this study, the subjects with onethird to two-thirds of their programs completed, and who gave reasons for having reservations about the psychiatrists' diagnosis, saying that it was

PAGE 105

95 not one to which they could commit without hesitation, were responding to the female psychiatrists' diagnosis. An example of responses from this group is: "Since I would be working as a counselor-in-training and I would most likely have to agree with my supervisor anyway, I would be so-so with my answer." Only one of the respondents with more than two-thirds of her program completed accepted the diagnosis given by psychiatrists either provisionally or fully. The remainder of the subjects with more than twothirds of their programs completed offered several reasons for questioning the diagnosis. Some subjects with more than two-thirds of their program completed wrote that the diagnosis was possibly correct, but other diagnoses needed to be considered. They also mentioned that they would be more comfortable "consulting DSM IV about the diagnosis." These students probably had more highly developed clinical reasoning skills than those who were less advanced in their programs. Therefore, after having completed more than two-thirds of their programs, they may have been less likely to accept others' diagnoses and to be more autonomous. Stoltenberg and Delworth (1987) wrote that this behavior is to be expected at the Level Three stage of development of counselors. Students with more than two-thirds of their program hours completed were less inclined than subjects who had completed one-third to

PAGE 106

96 two-thirds of their program hours to accept the diagnosis offered by psychiatrists. Students with more than two-thirds of their programs completed appear to have been inclined to use clinical reasoning skills, but they also may lack the information and educational background necessary to do so effectively. It is unclear from this study if more advanced students are inclined to be more autonomous when making diagnostic decisions. The conclusion that students with more than two-thirds of their program hours completed are more inclined than students with one-third to two-thirds of their programs completed to attempt to engage in a standard clinical reasoning process is speculative. Diagnosis rejected The group of subjects who rejected the psychiatrists' diagnosis because they thought the diagnosis was premature wrote that they did not have enough information to allow them to agree or disagree with the supervisor's diagnosis. They claimed they lacked evidence to make a diagnosis. Several subjects in this group rejected the supervisors' diagnosis because they thought it inaccurate based on the symptoms presented. These subjects strongly disagreed with the psychiatrists, citing symptoms of Axis I disorders such as depression and offering diagnoses such as "loneliness," "isolation," "low self-esteem," and "sadness, all symptoms mentioned in the client profile.

PAGE 107

97 Perhaps the perceived philosophy of the psychiatrists contributed to some counselors' distrust of psychiatrists' diagnosis. It may be that a response to the medical model of determining client problems and needs affected the attribution of diagnostic credibility to psychiatrists. Some mental health practitioners consider diagnosis to be an imposition of the medical model on the practice of mental health counseling. Specifically, this opinion is held particularly by relationship-oriented counselors who warn of self-fulfilling prophecies, labeling, and the negative existential implications of diagnosis to clients (Corey, 1986). Attribution of credibility to psychologists Psychologists were not considered to be more credible diagnosticians than any other type of supervisor by any group of subjects, but they were considered less credible than psychiatrists by students who had completed one-third to two-thirds of their program hours. As noted above, students who had completed between one-third and two-thirds of the program hours required for graduation attributed a significantly different level of diagnostic credibility to psychiatrists than to psychologists. Diagnosis rejected The respondents who rejected the diagnosis offered by the psychologists apparently thought that the diagnostic process was incomplete or otherwise flawed. Subjects offered alternative diagnoses such as major depression (an Axis I disorder, not an Axis II disorder as was the client's diagnosis). Some respondents went so far as to suggest

PAGE 108

98 medications, including one highly-addictive medication frequently prescribed for anxiety, but which is not an antidepressant, and potentially could further disinhibit a client with Borderline Personality Disorder. Some of these subjects also proposed conditions such as a lack of self esteem as diagnoses. This tendency to overaccommodate clients is representative of students in Level Two as described by Stoltenberg and Delworth (1987). Students in Level Two also are less imitative and more assertive than students in Level One. Stoltenberg and Delworth (1987) asserted that students at this level have their own ideas, but want to rely to some extent on supervisors. These subjects demonstrated a lack of knowledge about the fundamentals of multiaxial diagnosis as used in the DSM-IV (APA, 1994). In addition, an incomplete understanding of pharmaceuticals available for use with clients with depression was evidenced. Had these respondents been practicing in clinical settings, their rejection of the supervisors' diagnosis and the subjects' comfort with offering alternative diagnoses without consideration that Borderline Personality Disorder might be an additional diagnosis to another primary diagnosis supports Fong's (1993) position that students are not effective diagnosticians.

PAGE 109

99 Fong stated that because students use a diagnostic manual to look for signs and symptoms, but do not consider other indices used in proper diagnosis, they are often inept diagnosticians. That is, they often fail to assess clients by observing their behavior, interviewing them (and perhaps their family members), examining their mental status, and ultimately generating hypotheses about the clients' conditions by noting patterns such as clients' psychological and physiological abnormalities. Therefore, it can be concluded that students who are mid-way in their programs may incompletely diagnose and often do not have solid grounds for recognizing the accuracy or inaccuracy of supervisors' diagnoses. Attribution of credibility to counselor education faculty No quantitative results indicated that counselor education students at any level of their programs considered counselor education faculty to be more or less credible diagnosticians than psychiatrists, psychologists, or doctoral students. Counselor education faculty were not rated as more or less expert and trustworthy than were other supervisors. Attribution of credibility to faculty also was not influenced by gender of subjects or supervisors. Female counselor education faculty, however, received higher ratings overall than did male faculty. Reasons for ratings Mixed sentiments were found in the responses offered by subjects who rated counselor education faculty as clinical supervisors. Those who rejected the diagnosis offered explanations such as "This is a

PAGE 110

100 hasty diagnosis. More time is needed with the client" and "The supervisor has not seen the client and is relying on the counselor's word." Some subjects admitted they were not knowledgeable about diagnosis and for that reason strongly agreed with their supervisors. Subjects who agreed with counselor education facultys' diagnosis listed "confirming factors" that led them to agree with their faculty supervisors. Faust (1986) warned of diagnosticians seeking confirming evidence for their hypotheses and ignoring disconfirming evidence. There is no doubt that the client presented in this study could have been diagnosed with major depression on Axis I; however, other signs and symptoms pointed to the efficacy of considering additional diagnoses. This condition was met by supervisors diagnosing the client as having Borderline Personality Disorder on Axis IT Faust (1986) noted that diagnosticians have cognitive limitations. Clinical decisions are influenced by diagnosticians' perceptual selectivity. Some respondents in this group recognized the signs and symptoms of depression and suggested it as an alternative diagnosis, thereby indicating a lack of knowledge about the multiaxial system of diagnosis. Proportion of program hours completed There was no notable distinction among the groups of subjects as to their rating of credibility to counselor education faculty based on the proportion of program hours the subjects

PAGE 111

101 had completed. No group attributed more or less diagnostic credibility to counselor education faculty. Attribution of credibility to doctoral students Reasons for ratings Subjects who rated the diagnostic credibility of doctoral students were equivocal in their responses. The majority of this group of subjects was uncertain about the diagnosis, admitted they were unfamiliar with the criteria for meeting the diagnosis, and were not knowledgeable about how to diagnose; therefore, they were unwilling to accept their supervisors' diagnosis without question. However, some subjects couched their responses by writing that although they agreed with the supervisor, they would have preferred having more information about the client. One person responded that a medical workup was in order. However, generally, they were willing to risk trusting their supervisors' expertise: "I would take my supervisor's 'word' and of course ask for an explanation." Subjects were not inclined to strongly dispute doctoral student supervisors' diagnosis. They wrote such explanations as, "I agree with the supervisor because she has more experience than I. I do not feel like I know enough to challenge someone who is more educated." Also, "I cannot challenge the diagnosis of the supervisor, and not having a DSM-IV to refer to, I have no choice but to agree to the accuracy of the statement."

PAGE 112

102 Eleven subjects accepted the doctoral students' diagnosis to some extent. Only three respondents rejected the doctoral students' diagnosis, citing (a) the need to know why the supervisor diagnosed as she did, (b) that the diagnosis was based on too little information, and (c) no information was given about the client's use, if any, of addictive substances. Tagiuri's (1958) contention that (a) the attributes of the person being perceived, (b) the characteristics of the perceiver, and (c) the nature of the situation in which the association occurs influences people's perceptions of others should be considered here. Entry-level students may perceive doctoral counselor education students as credible diagnosticians not only because of perceived expertness ("better educated"), but also as having a distinct understanding of the clinical problems faced by counselor education students in field experiences. Program hours completed Subjects, regardless of number of program hours completed, yielded no significant difference in the diagnostic credibility between doctoral students (mean = 5.39) and psychiatrists (mean = 5.00), doctorate-holding counselor education faculty (5.56), and psychologists (4.92).

PAGE 113

103 Implications Provision of mental health services It is important for staffers of mental health treatment facilities, including managed mental health care systems, counselor education faculty, and clinical supervisors to know how counselors-in-training rate the credibility of their clinical supervisors. The extent to which counselors-in training are confident of their supervisors' diagnostic credibility could affect the student counselors' quality and quantity of learning, the nature of their experiences as professional counselors, and their clients' wellbeing. The structure of mental health care systems and the functions of mental health care providers in the future are uncertain. Winegar (1992) asked if mental health counselors will more often find themselves part of managed mental health care systems? In those circumstances, counselors would be asked to help clients who have been diagnosed by the Primary Care Physician (PCP), most likely a psychiatrist, or they will be part of a multidisciplinary team that determines the diagnostic status of clients and designs treatment plans. In managed mental health care systems the "gatekeeping function" (Winegar, 1992, p. 95) of diagnosis and treatment planning is usually conducted by nonphysicians in consultation with a psychiatrist in a cooperative, non-competitive spirit. The consequences of diagnosing

PAGE 114

104 clients accurately and providing treatment relevant to the diagnoses of clients become relevant not only to clients and counselors but also to insurance companies or other payors. Mental health providers incurring preventable, unnecessary, or additional costs to insurance providers could mean lost contracts and elimination of professional positions for mental health counselors. Clinical reasoning If mental health counselors have weak clinical reasoning skills or inferior diagnostic skills, or if they are not philosophically compatible with those with whom they would provide services, difficulties for both providers and clients would develop. It is essential that counselors be aware of theoretical differences and similarities between themselves and colleagues. It is imperative that counselors be aware of limitations in their clinical skills and not be disposed toward compensating for the lack of those skills by taking unprofessional, unethical, or irresponsible action. For example, being aware of recent developments in psychopharmacology and being aware of when to refer clients could prevent counselors from acting on impulses or basing clinical decisions on less-than-adequate knowledge or judgment. Excluding the responses from subjects who did not offer reasons for their ratings, the results of this research showed that among the subjects with more than two-thirds of their program hours completed (n=64), there

PAGE 115

105 was a propensity not only to accept or reject the diagnosis, but also to offer alternative diagnoses or courses of action (n=49). However, they often admitted they did not know how to diagnose or were unfamiliar with the criteria for the diagnosis offered by their supervisors. This situation calls into question the practice of counselors arriving at premature and uninformed decisions about client problems and counseling needs. It also suggests that more training in clinical reasoning skills and in the process and procedure of diagnosis is called for in counselor education programs. Counselor education programs It is important to novice counselors that they be able to trust their supervisors (Heppner & Handley, 1981). As revealed in this study, counselor education students do not become more selective as they move toward completion of their programs. At no point did any of the subjects in this study attribute significantly less credibility to counselor education faculty or to counselor education doctoral students than they did to psychiatrists or to psychologists. This information could be meaningful to counselor education faculty when planning supervision assignments for their students and when planning curricula. Counselors are expected to have sound reasons for the diagnoses they give and the treatments they offer. This study revealed that counselor education students who reject their supervisors' diagnosis (n=79) are less inclined to offer alternative suggestions for diagnosis than they are to want

PAGE 116

106 to accumulate more information before making diagnoses. This cautious approach to diagnosis may mean that counselor education students are open-minded to information that can be offered by clinical supervisors. Conversely, it is important to note that the subjects who did offer alternative diagnoses (n=55) seemed to attach a great deal of significance to the necessity to have an explanation for the clients' problems, regardless of the extent to which they agreed with the supervisors' diagnosis. Thirty-one of 146 counselor education programs that responded to Ritchie, Piazza, and Lewton's (1991) survey taught nothing about use of the DSM-III-R (APA. 1987). This information and the results of this study demonstrate the need and importance of counselor education students more frequently being offered substantive courses in assessment and diagnosis so that counselors can make well-reasoned judgments about the credibility of their supervisors' diagnoses. It is also important that clinical supervisors thoroughly explain the reasons for their positions on the nature of clients' problems. The development of clinical skills and acquisition of clinical knowledge should be on-going processes not only for counselors but for supervisors as well. "Supervisors should pursue professional and personal continuing education activities such as advanced courses, seminars, and professional conferences on a regular and on-going basis. These activities

PAGE 117

107 should include both counseling and supervision topics and skills" (ACA Ethical guidelines, p. 6). Recommendations for Future Research Perhaps this research has raised more questions than it has answered. Primary attention in future studies should be given to some of the issues raised. For example, this study could be modified with closer controls on educational and experiential levels of subjects. However, Stoltenberg (1987) made it clear that developmental level is not equivalent to educational or experiential level, so researchers should investigate attribution of diagnostic credibility to clinical supervisors among counselor education students at their various developmental levels as counselors. The present study has shown that attribution of credibility is not differentially attributed to clinical supervisors, but there is some question as to the bases from which their ratings were derived. Do counselor education students place ultimate responsibility for diagnosis on themselves or on their clinical supervisors? The answer to this question may lie in an investigation of attribution of diagnostic credibility based on developmental levels as described by Stoltenberg (1987). He noted that in Level One students defer to supervisors, while in Level Two students either develop a better understanding of clients than they were able to do in Level One or they overaccommodate or overassimilate with clients in an attempt to explain all discrepancies. Level

PAGE 118

108 Three students neither overaccommodate or overassimilate, but rather are able to defend theoretical views and enjoy dialogue with supervisors. Would conducting this research as a qualitative study further support the findings here, or would it provide different results? A grounded theory approach in which the researcher gathers data "about the 'lived' experience of participants" (Sherman & Webb, 1990, p. 125) would allow discovery of how counselor education students make sense of their supervised clinical field experiences. Research also could focus on whether the counselor education profession is addressing the practical implications of disagreement about diagnosis within the profession. Counselor education students are introduced both to the perspectives of the proponents of the medical model and to the perspectives of the existential or relationship-oriented advocates of counseling theory. The cognitive dissonance some students might experience when confronted with different theoretical or philosophical perspectives regarding diagnosis could be addressed. Researchers also could investigate subjects enrolled in educational programs that place differing emphases on teaching the art and science of diagnosis. Situations where clinical supervision varies in the extent of cooperation or cohesiveness between supervisors and counselor education students also could be studied.

PAGE 119

109 Different attributes or attitudes than were used in this study of counselor education students and supervisors could be used as independent variables in future research. For example, counselor education students' degree of confidence about their ability to be effective counselors could be addressed in an examination of counselor education students' self-efficacy as diagnosticians. Other characteristics of students and supervisors that could be investigated in studies similar to the present one include the hypothetical race, age, theoretical orientation, self-efficacy, expectations, and working environment of clinical supervisors. Also, the actual race, age, expectations of the clinical experience and of supervisors, as well as the theoretical orientation of the subjects, could be investigated as independent variables. In addition, the educational credentials of supervisors could be changed to include licensed clinical social workers, licensed marriage and family therapists, and licensed and unlicensed mental health counselors with master's degrees. Other characteristics that could be investigated for their effect on perceived credibility are perceived attractiveness, likability, and dynamism of supervisors. Students from programs with less rigorous academic standards than CACREP-accredited programs or who are from programs that, for whatever reason, have not become CACREP-accredited could be the

PAGE 120

no subjects of a future study. In addition, the numbers and types of supervised field experiences could be included as independent variables. Future research efforts also could be expanded beyond the use of counselor education students as subjects and include practicing professional counselors who have completed their basic preparation and have had some professional counseling experience. This also might include counselors in different work settings with various types of clientele, and counselors who are disinclined to diagnose. Summarv Counselor education students engage in field experiences to gain the expertise needed to be effective mental health counselors. During their practica and internships, counselor education students receive assistance and direction from qualified clinical supervisors. However, it was unknown to what extent counselor education students consider their clinical supervisors to be credible diagnosticians. In the present study, based on Kelley's (1973) ANOVA model of attribution theory, the extent to which counselor education students grant credibility to clinical supervisors based on supervisors' educational credentials and gender and counselor education students' gender and proportion of counselor education program hours completed that are required for graduation was investigated.

PAGE 121

Ill Male and female counselor education students who had completed varying proportions of their entry-level counselor education program hours were subjects in this study. Counselor education students' ratings of the diagnostic credibility of male and female counselor education doctoral students, male and female psychologists with doctoral degrees, male and female counselor education department faculty members, and male and female psychiatrists acting as clinical supervisors were subjected to a factorial analysis of variance and a post hoc analysis of the interaction between ratings of credibility and supervisors' educational credentials and the proportion of program hours required of students for graduation. Qualitative analysis of subjects' comments was done to achieve broader understanding of the subjects' views on clinical supervisors' diagnostic credibility. These narrative responses provided understanding of the reasons why subjects at different stages in their education as counselors rated clinical supervisors as they did. From the results of this study it was learned that counselor education students who have completed varying proportions of program hours that are required for graduation do not attribute diagnostic credibility differentially to their clinical supervisors. However, it was also learned that students are not unified in their reasons for awarding or denying credibility to supervisors.

PAGE 122

APPENDIX A LETTER OF INTRODUCTION (Mailing Address) (Date) Dear Dr : As a doctoral candidate in Counselor Education and Supervision in the Department of Counselor Education at the University of Florida, I have chosen to conduct research on counselor education students' perceptions of their clinical supervisors' credibility. My study focuses on how students enrolled in community counseling or mental health counseling programs attribute diagnostic credibility to clinical supervisors. Among the independent variables I have chosen to examine are students' gender and proportion of program hours completed that are required for graduation. I am requesting your cooperation in conducting this study. My sample will be enrollees in master's degree level programs who engage in practica and internships at sites such as community mental health agencies, inpatient and outpatient treatment facilities, criminal justice settings, or individual private practices. You would need approximately 10-15 minutes to administer the assessment instrument to groups of students. I am requesting the completed materials be returned to me in postage-paid mailers I will provide. I appreciate your cooperation. I also hope the results of this study will be useful to you in your work as a counselor educator and supervisor. Sincerely, 112

PAGE 123

APPENDIX B INSTRUCTIONS FOR ADMINISTERING INSTRUMENT Instructions to the administrator : (1) Prior to distributing the test packets, please ask those who may have already participated in this study to excuse themselves from participating at this time. (2) Prior to distributing the test packets, please announce that this is a study of opinions among counselor education students. Subjects should not use outside resources (such as DSM-III-R ) or converse with other people while participating in this study. The work is to be done in the presence of the administrator, no test packets may be taken from the room, and upon completion, test packets are to be returned to the administrator. (3) Distribute the test packets. Ask the subjects to verify that there are four pages in each packet: (a) a demographic inventory, (b) a two-page clinical profile, and (c) a response sheet with instructions. Ask that the subjects NOT unstaple the packets. (4) Ask the subjects who choose to participate to continue by (a) completing the demographic inventory (b) reading the profile of the fictitious client (c) indicating their answers on the agreement rating scale and providing the narrative information requested under the agreement rating scale (5) Collect the test packets and return them in the enclosed stamped mailers to: Jane Carroll (Mailing Address) Thank You. 113

PAGE 124

APPENDIX C DEMOGRAPHIC INVENTORY Some personal information is needed about the subjects in this study of counselor education students' opinions. Please respond to the following: (1) Age: (2) Gender: (3) Race: (4) Are you enrolled in a community counseling or mental health counseling program in a counselor education department? Circle: (a)Yes (b) No (5) What degree are you currently seeking? (6) How many academic credit hours are you required to complete in order to graduate from your program? Circle: (a) Semester hours (b) Quarter hours (7) How many academic credit hours have you completed? Do NOT include course hours for courses in which you are currently enrolled. Circle: (a) Semester hours (b) Quarter hours (8) Of the following supervised field experiences, indicate those you have completed. Do NOT include those in which you are currently enrolled. (a) Pre-Practicum (b) First practicum (c) Second practicum (d) First half of a split internship (e) Internship 114

PAGE 125

APPENDIX D PROFILE OF A FICTITIOUS CLIENT Client's Name : Ellen Farber Summary of intake interview : Ellen Farber, a 35-year-old, single, former insurance company executive, came to a clinic in which you work as a counselor-in-training. She was assigned to you as a client. Ms. Farber complained of depression and said she had thought of driving her car off a cliff. An articulate, moderately overweight, sophisticated woman, Ms. Farber appeared to be in considerable distress. She reported a 6-month period of increasingly persistent dysphoria and lack of energy and pleasure. Feeling as if she were "made of lead," Ms. Farber had recently been spending 15 20 hours a day in her bed. She also reported daily episodes of binge eating, when she would consume "anything I can find" including entire chocolate cakes or boxes of cookies. She reported problems of intermittent binge eating since adolescence, but these had recently increased in frequency, resulting in a 20-pound weight gain over the last few months. In the past, her weight had often varied greatly as she had gone on and off a variety of diets. She denied a preoccupation with thinness or a history of episodes of vomiting or other weight-reduction procedures to compensate for the binge eating. She attributed her increasing symptoms to financial difficulties. Ms. Farber had been fired from her job two weeks before coming to the clinic. She claimed it was because she "owed a small amount of money." When asked to be more specific, she reported owing $150,000 to her former employers and another $100,000 to various local banks. Further questioning revealed she had always had difficulty managing her money and had been forced to declare bankruptcy at age 27. From age 30 to age 33 she had used her employer's credit cards to finance weekly "buying binges," accumulating the $150,000 debt. She denied past or present symptoms of mania, obsessive thoughts, or a compulsion to buy, but instead reported that spending alleviated her chronic feelings of loneliness, isolation, and sadness. Experiencing only temporary relief, every few days 115

PAGE 126

116 she would impulsively buy expensive jewelry, watches, or multiple pairs of the same shoes. Two years ago, when her employers noticed the massive credit card bills, Ms. Farber did not have enough assets she could sell that could significantly reduce the debt. Her employers allowed her to pay off the debts by continuing to work for them and giving them part of her salary. However, she could not stop her spending. She financed further purchases by a process she called "check kiting." She would open a checking account at one bank, overdraw from that account to open a second account at another bank, and then overdraw from the second account to open an account at a third bank. Over two years this escalating process led to her additional $100,000 debt. When the banks discovered the fraudulent practice two weeks ago, they contacted Ms. Farber's employers, who promptly fired her. This course of action led to her current desperate state. In addition to lifelong feelings of emptiness, Ms. Farber described chronic uncertainty about what she wanted to do in life and with whom she wanted to be friends. She had many brief, intense relationships with both men and women, but her quick temper led to frequent arguments and even physical fights. Although she had always though of her childhood as happy and carefree, when she became depressed, she began to recall episodes of abuse by her mother. Initially, she said she had dreamed that her mother had pushed her down the stairs when she was only six, but she then began to report previously unrecognized memories of beatings or verbal assaults by her mother.

PAGE 127

APPENDIX E DESCRIPTIONS OF FICTITIOUS CLINICAL SUPERVISORS Instructions to subjects : You have completed your initial intake interview with your new client named Ellen Farber. At the subsequent meeting with your clinical supervisor. [0101] Mr. James Doe, a doctoral student in a counselor education program, he says he believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Mr. Doe's diagnosis? [0102] Dr. James Doe, a psychologist, he says he believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis? [0103] Dr. James Doe, a professor in a counselor education department, he says he believes the client's Axis II diagnosis should Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis? [0104] Dr. James Doe, a psychiatrist, he says he believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis? 117

PAGE 128

118 [0105] Ms. Jane Doe, a doctoral student in a counselor education program, she says she believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Ms. Doe's diagnosis? [0106] Dr. Jane Doe, a psychologist, she says she believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis? [0107] Dr. Jane Doe, a professor in a counselor education department, she says she believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis? [0108] Dr. Jane Doe, a psychiatrist, she says she believes the client's Axis II diagnosis should be Borderline Personality Disorder. To what extent do you agree with Dr. Doe's diagnosis?

PAGE 129

APPENDIX F AGREEMENT RATING SCALE AND QUALITATIVE DATA Instructions to Subjects: Realizing there may be limitations to your ability to reach a diagnosis of this client at this time, please indicate, nonetheless, on the scale below, the extent to which you agree with the diagnosis of Ellen Farber provided by your clinical supervisor. Strongly Disagree Strongly Agree 123456789 10 In the space below, and on the back of the sheet if necessary, please explain what factors you considered when deciding the extent to which you agreed with your supervisor. That is, why did you agree or disagree as you did? 119

PAGE 130

APPENDIX G GENDER-CREDENTIALS CODE 0101 male master's degree level counselor education doctoral student 0102 male psychologist holding doctorate 0103 male counselor education professor holding doctorate 0104 male psychiatrist 0105 female master's degree level counselor education doctoral student 0106 female psychologist holding doctorate 0107 female counselor education professor holding doctorate 0108 female psychiatrist 120

PAGE 131

APPENDIX H POST CARD REMINDER Dear On (date) I sent you (number) copies of an assessment instrument you agreed to administer to a group of students. I appreciate your willingness to help me in my doctoral research, and I hope you will be able to return the completed instruments to me soon in the prepaid mailers I provided. Thank you. Jane Carroll (Address) 121

PAGE 132

REFERENCES American Psychiatric Association (1952). Diagnostic and statistical manual of mental disorders Washington, D. C: Author. American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, D. C: Author. American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders (3rd ed. rev). Washington, D. C: Author. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders Washington, D. C: Author. Bailey, K. D. (1987). Methods of social research (3rd ed.). New York: Free Press. Bandura, A. (1977). Social learning theory Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1978). The self system in reciprocal determinism. American Psychologist. 33_(4), 344 358. Borders, L.D., & Leddick, G.R. (1987). Handbook of counseling supervision Alexandria, VA: Association for Counselor Education and Supervision. Bradley, L. J. (1989). Counselor supervision (2nd ed). Muncie, IN: Accelerated Development. Carey, J. C, Williams, K. S., & Wells, M. (1988). Relationships between dimensions of supervisors' influence and counselor trainees' performance. Counselor Education and Supervision, 28 130 139. Conolly, J. (1856). The treatment of the insane without mechanical restraints. London: Smith Elder. In G.N. Grob (Ed), Mental illness and social policy: The American experience (pp.1 380). New York: Arno Press. Conroy, P. (1986). The prince of tides Boston: Houghton Mifflin. 122

PAGE 133

123 Corey, G. C. (1986). Theory and practice of counseling and psychotherapy Pacific Grove, CA: Brooks/Cole Grain, W.C. (1985). Theories of development: Concepts and applications (2nd ed.) Englewood Cliffs, NJ: Prentice Hall. Department of Business and Professional Regulation, Division of Medical Quality Assurance, Board of Clinical Social Workers, Marriage and Family Therapists, and Mental Health Counselors (1994). Laws and rules study book Tallahassee: Author. Dewey, J. (1984). Qualitative thought. In J. A. Boydston & K. E. Poulos (Eds.), John Dewey: The later works. 1925 1953 (Vol. 5, 1929 1930, pp. 243 262). Carbondale, IL: Southern Illinois University Press. (Original work published 1930) Ellis, A. (1979). The theory of rational-emotive therapy. In A. Ellis & J. Whiteley (Eds), Theoretical and empirical foundations of rationalemotive therapy Monterey, CA: Brooks/Cole. Elstein, A. S., & Bordage, G. (1988). Psychology of clinical reasoning. In J. Dowie & A. Elstein (Eds), Professional judgment: A reader in clinical decision making (pp. 109 129). New York: Cambridge University Press. Erikson, E. (1963). Childhood and society New York: Norton. Ethical guidelines for counseling supervisors. (1993, Summer). ACES Spectrum p. 5 6. Faust, D. (1986). Research on human judgment and its application to clinical practice. Professional Psychology: Research and Practice. 17.(5), 420 430. Faust, D. & Ziskin, J. (1988). The expert witness in psychology and psychiatry. Science, 241 (4861), pp. 31 35. First, M. B., Gibbon, M., Williams, J. B. W., & Spitzer, R. L. (1991). AutoSCID II : Structured Clinical Interview for DSM-III-R North Tonawanda, NY: Multi-Health Systems. First, M. B., Williams, J. B. W., & Spitzer, R. L. (1988). DTREE: The electronic DSM III-R (Axis 1). North Tonawanda, NY: Multi-Health Systems.

PAGE 134

124 Fischoff, B. (1976). Attribution theory and judgment under uncertainty. In J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New directions in attribution research (vol.1). Hillsdale, NJ: Erlbaum. Fong, M. L. (1993). Teaching assessment and diagnosis within a DSM-IIIR framework. Counselor Education and Supervision, 32(4), 276 286. Forsterling, F. (1988). Attribution theory in clinical psychology New York: John Wiley & Sons. Forsterling, F. (1990). Attributional therapies. In S. Graham & V. S. Folkes (Eds), Attribution theory: Applications to achievement, mental health, and interpersonal conflict (pp. 123 139). Hillsdale, NJ: Lawrence Erlbaum. Furlong, M. J., & Hayden, D. C. (1993). Computer-assisted training of the DSM-III-R in counselor education. Counselor Education and Supervision. 11(4), 298 3 10. Gambrill, E. (1990). Critical thinking in clinical practice San Francisco: Jossey Bass. Goffman, E. (1961). Asylums Garden City, NY: Anchor Books. Graham, S., & Folkes, V. S. (Eds). (1990) Attribution theory: Applications to achievement, mental health, and interpersonal conflict Hillsdale, NJ: Lawrence Erlbaum. Greenberg, J., & Wursten, A. (1988). The psychologist and the psychiatrist as expert witnesses: Perceived credibility and influence. Professional Psychology: Research and Practice. 19(4), 373 378. Heider, F. (1958). The psychology of interpersonal relations New York: John Wiley & Sons. Heider, F. (1980). Perception and attribution. In Gorlitz, D. (Ed), Perspectives on attribution research and theory (pp. 3 8). The Bielefeld Symposium. Cambridge, MA: Ballinger. Henerson, M. E., Morris, L. L., & Fitz-Gibbon, C. T. (1987). How to measure attitudes Newbury Park, CA.: Sage.

PAGE 135

125 Heppner, P. O., & Handley, P. (1981). A study of the interpersonal influence process in supervision. Journal of Counseling Psychology. 28, 437 444. Herman, K. C. (1993). Reassessing predictors of therapist competence. Journal of Counseling and Development, 72(1), 29 3 2. Hewstone, M. (Ed). (1983). Attribution theory Oxford, England: Basil Blackwell. Hillerbrand, E., & Claiborn, C. D. (1990). Examining reasoning skill differences between expert and novice counselors. Journal of Counseling and Development, 68 684 691. Hippocrates (460 377, B.C.) (1988). On the sacred disease. In H. Kellerman & A. Burry. Psychopathoiogy and differential diagnosis: A primer Volume One: History of Psychopathoiogy (p. 14). New York: Columbia University. Hohenshil, T. H. (1993). Teaching the DSM-HI-R in counselor education. Counselor Education and Supervision, 32(4), 267 275. Holland, A. L., Atkinson, D. R., & Johnson, M. E. (1987). Effects of sexual attitude and sex similarity on perceptions of the counselor. Journal of Counseling Psychology, 34(3), 322 325. Hovland, C. I., Janis, I. L., & Kelley, H H. (1953). Communication and persuasion New Haven: Yale University Press. Johnson, E., Baker, S. B., Kopala, M., Kiselica, M. S., & Thompson, E. C. HI (1989). Counseling self-efficacy and counseling competence in prepracticum training. Counselor Education and Supervision 28 205 218. Jones, E. E. (1990). Interpersonal perception New York: W. H. Freeman. Jordan, J. S., Harvey, J. H., & Weary, G. (1988). Attributional biases in clinical decision making. In DC. Turk & P. Salovey (Eds), Reasoning, inference, and judgment in clinical psychology (pp. 90 106). New York: The Free Press. Kellerman, H., & Burry, A. (1988). Psychopathoiogy and differential diagnosis: A primer Vol. 1: History of Psychopathoiogy New York: Columbia University Press.

PAGE 136

126 Kelley, H. H. (1973). The processes of causal attribution. American Psychologist. 28, 107 128. Kelly, G. A. (1963). A theory of personality: The psychology of personal constructs. New York: Norton. Knesper, D. J., & Pagnucco, D. J. (1987). Estimated distribution of effort by proyiders of mental health services to U.S. adults in 1982 and 1983. The American Journal of Psychiatry. 144(7). 883 888. Kottler, J. A. (1991). The compleat therapist San Francisco: Jossey Bass Littrell, J. M., Caffrey, P., & Hopper, G. C. (1987). Counselor's reputation: An important precounseling variable for adolescents. Journal of Counseling Psychology, 34(2), 228 23 1. Lopez, S. R., & Wolkenstein, B. H. (1990). Attributions, person perception, and clinical issues. In S. Graham & V. S. Folkes (Eds), Attribution theory: Applications to achievement, mental health, and interpersonal conflict (pp. 103 121). Hillsdale, NJ: Lawrence Erlbaum. Lynn, S. J., & Garske, J. P. (1985). Contemporary psychotherapies Columbus, OH: Merrill. Mahoney, M. J. (1988). Rationalism and constructivism in clinical judgment. In DC. Turk & P. Salovey (Eds), Reasoning, inference, and judgment in clinical psychology (pp. 155 181). New York: Free Press. Marshall, C, & Rossman, G. B. (1989). Designing qualitative research Newbury Park, CA: Sage. Menninger, K. (1963). The vital balance New York: Viking. Minuchin, S. (1974). Families and family therapy Cambridge, MA: Harvard University Press. Morey, L. C, Skinner, H. A., & Blashfield, R. K. (1986). Trends in the classification of abnormal behavior. In A.R. Ciminero, K.S. Calhoun, & H.E. Adams (Eds), Handbook of behavior assessment (2nd Ed.) (pp. 47 75). New York: John Wiley.

PAGE 137

127 Pepinsky, H.B., & Pepinsky, P.N. (1954). Counseling theory and practice New York: Ronald Press. Poidevant, J. M., Loesch, L. C, & Wittmer, J. (1991). Vocational aspirations and perceived self-efficacy of doctoral students in the counseling professions. Counselor Education and Supervision. 30, 289 300. Ponce, F. Q., & Atkinson, D. R. (1989). Mexican-American acculturation, counselor ethnicity, counseling style, and perceived counselor credibility. Journal of Counseling Psychology. 36(2), 203 208. Postman, L., Bruner, J. S., & McGinnies, E. (1976). Personal values as selective factors in perception. In G. M. Murch (Ed), Studies in perception (pp. 279 291). Indianapolis, IN: Bobbs Merrill. Reich, W., Welner, Z., & Herjanic, B. (1990). Diagnostic interview for children and adolescents revised -Computer program: Child/Adolescent Version and Parent Version North Tonawanda, NY: Multi-Health Systems. Remley, T. R. (1993, November). We are all mental health counselors. Guidepost p. 4. Ritchie, M. H., Piazza, N. J., & Lewton, J. C. (1991). Current use of the DSM-III-R in counselor education. Counselor Education and Supervision, 30. 205 211. Robinson, G. (1993, November). ACA divisions split counseling profession. Guidepost p. 16. Rogers, C. (1961). On becoming a person Boston: Houghton Miffiin. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs. 80 (1, Whole No. 609). Scheff, T. J. ((1966). Being mentally ill: A sociological theory Chicago: Aldine. Schiavone, C. D., & Jessell, J. C. (1988). Influence of attributed expertness and gender in counselor supervision. Counselor Education and Supervision, 28. 29 42.

PAGE 138

128 Secord, P. F., & Backman, C, W, (1974). Social psychology (2nd ed.). New York: McGraw-Hill. Seligman, L. (1990). Selecting effective treatments San Francisco: JosseyBass. Sherman, R. R. (1985). Philosophy with guts. Journal of Thought, 20(2), 3 11. Sherman R. R. & Webb, R. B. (1990). Qualitative research in education: Focus and methods New York: Falmer. Sipps, G. J., Sugden, G. J., & Faiver, C. M. (1988). Counselor training level and verbal response type: Their relationship to efficacy and outcome expectations. Journal of Counseling Psychology. 3 5(4). 397 401. Smith, A. F. (1988). Perceiving the client. In DC. Turk & P. Salovey (Eds.), Reasoning, inference, and judgment in clinical psychology (pp.73 89). New York: The Free Press. Snyder, M. (1976). Attribution and behavior: Social perception and social causation. In J.H. Harvey, W.J. Ickes, and R.F. Kidd (Eds), New directions in attribution research. Vol 1 (pp. 53 72). Hillsdale, NJ: Lawrence Erlbaum. Spitzer, R. L., Gibbon, M., Skodol, A E., Williams, J. B. W., & First, M B. (1994). DSM-IV Case Book Washington, D. C: American Psychiatric Press. Stoltenberg, C. (1981). Approaching supervision from a developmental perspective: The Counselor Complexity Model. Journal of Counseling Psychology. 28 ( 1 ) 59 65. Stoltenberg, C. D., & Delworth, U. (1987). Supervising counselors and therapists San Francisco: Jossey Bass. Strong, S. R. (1968). Counseling: An interpersonal influence process. Journal of Counseling Psychology 1 5(3), 215 224. Sue, D. W. (1981). Counseling the culturally different New York: John Wiley. Szasz, T. ((1961). The myth of mental illness: Foundations of a theory of personal conduct New York: Harper & Row.

PAGE 139

129 Tagiuri, R., & Petrullo, L. (Eds.) (1958). Person perception and interpersonal behavior Stanford, CA: Stanford University Press. Tedeschi, J. T., & Linkskold, S. (1976). Social psychology: Interdependence, interaction, and influence New York: Wiley. Tracey, T. J., Ellickson, J. L., & Sherry, P. (1989). Reactance in relation to different supervisory environments and counselor development. Journal of Counseling Psychology. 36(3), 336 344. Truatt, G. M., & Bloom, L. J. (1982). Therapeugenic factors in psychotherapy: The effects of fee and title on credibility and attraction. Journal of Clinical Psychology. 38(2), 274 279. Usher, C. H., & Borders, L. D. (1993). Practicing counselors' preferences for supervisory style and supervisory emphasis. Counselor Education and Supervision. 33(2), 66 79. Vacc, N. A., & Loesch, L. C. (1994), A professional orientation to counseling (3rd ed.). Muncie, IN: Accelerated Development. Waldo, M., Brotherton, W. D., & Horswill, R. (1993). Integrating DSMIII-R training into school, marriage and family, and mental health counselor preparation. Counselor Education and Supervision, 32(4), 332 342. Weckowicz, T. E. (1984). Models of mental illness Springfield, IL: Charles C. Thomas. Wiley, M. O. & Ray, P. B. (1986). Counseling supervision by developmental level. Journal of Counseling Psychology. 33(4), 439 445. Wilson, M. (1993). DSM-III and the transformation of American psychiatry: A history. The American Journal of Psychiatry. 1 50 (3), 399 410. Winegar, N. W. The clinician's guide to managed mental health care New York: Haworth. Wittmer, J. (1993). Managing your school counseling program: K 12 developmental strategies Minneapolis: Educational Media.

PAGE 140

130 Worthington, E. L., Jr. (1984). Empirical investigation of supervision of counselors as they gain experience. Journal of Counseling Psychology. 31(1), 63 75. Worthington, E. L., Jr., & Stern, A. (1985). Effects of supervisor and supervisee degree level and gender on the supervisory relationship. Journal of Counseling Psychology. 32.(2), 252 262.

PAGE 141

BIOGRAPHICAL SKETCH Jane J. Carroll graduated with a Ph.D. degree in mental health counseling from the University of Florida in 1995 after having received the master and educational specialist degrees in agency, correctional, and developmental counseling from UF in 1991. Jane's supervised field experiences and the focus of her studies were substance abuse counseling and counselor education and supervision. While at UF Jane received several professional awards for outstanding achievement and was active in extracurricular activities within the Department of Counselor Education. Jane taught biological sciences at Vero Beach High School, Vero Beach, Florida, for almost 20 years before she enrolled at UF in 1989 to study counseling. In 1981 she earned a master's degree in science education from Florida Institute of Technology. Jane was born and reared in Maine where she received her early education. She received a B.S. in education with a major in biology in 1961 from the University of Maine. Currently Jane works with a psychiatrist in private practice in Gainesville, Florida, where she provides counseling for adults with a wide range of mental health problems. Jane will join the faculty at the University of North Carolina at Charlotte in the fall of 1995. Development 131

PAGE 142

of a Substance Abuse Counseling track in the Department of Human Services at UNCC will be among her responsibilities as an assistant professor. She also plans to continue in private practice as a licensed mental health counselor. Jane is the mother of two sons. James, married to Pamela Wiggins, and Timothy, married to Krissy Valente, have growing families. Jane hopes her work will contribute to the quality of life for her children and grandchildren. 132

PAGE 143

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Larpy C Loesch, Chair Professor of Counselor Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ^^rJi^y^:!L^fr:^
PAGE 144

This dissertation was submitted to the Graduate Facuitv of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy August. 1995 Dean7 Colleaa'^f Education*^ Dean. Graduate School

PAGE 145

1780 1995 UNIVERSITY OF FLORIDA 3 1262 08285 358