Citation
Counseling with rejected nursing school applicants : some correlates of selected outcome measures

Material Information

Title:
Counseling with rejected nursing school applicants : some correlates of selected outcome measures
Creator:
Johnson, Douglas Gilmore, 1938-
Publication Date:
Language:
English
Physical Description:
xii, 174 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Anxiety ( jstor )
Basic skills ( jstor )
Counselor training ( jstor )
Dyadic relations ( jstor )
Health care outcome assessment ( jstor )
Nursing ( jstor )
Psychological counseling ( jstor )
Psychotherapy ( jstor )
School counseling ( jstor )
School counselors ( jstor )
Counseling -- Vocational guidance ( lcsh )
Counselor Education thesis Ph. D
Dissertations, Academic -- Counselor Education -- UF
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 167-173.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Douglas G. Johnson.

Record Information

Source Institution:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator(ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
025839710 ( ALEPH )
03384144 ( OCLC )

Downloads

This item has the following downloads:


Full Text

















COUNSELING WITH REJECTED NURSING SCHOOL APPLICANTS:
SOME CORRELATES OF SELECTED OUTCOME MEASURES








. Jy



DOUGLAS G. JOHNSO.N


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIRE-!ENTS FOR THE
DEGREE OF DOCTOR OF EDUCATION





UNIVERSITv n7 FLORIDA


1977


Pr


A
































TO GRACE












ACKNOWLEDGMENTS

This dissertation has resulted from the cooperative

efforts of a large number of people. For their cooperation. and assistance, I would like to recognize the following:

-Dr. .Ted Landsman, committee chairman and pacemaker.

-Dr. Richard Anderson, committee member.

-Dr. A. Garr Cranney, committee member.

-Dr. Robert Soar, design and statistical consultant.

-Dr. David Lane and Dr. Bert Sharp, former committee chairmen who were unable to continue service -due to retirement and position ch-ange.

-Santa Fe Community College.

-The subjects who participated in the study.

-Mr. George Huber, Mr. Tom McCullough, Mr. Don Mott,

Mr. Bernie Murphy, and Mr. Gene Wigington, the participating counselors, who made special schedule arrangements to be available for this study.

-Miss Lillie Covert who made, all of the subject-counselor appointments in addition to her normal duties.

-Mr. Mike Conlon, computer analysis consultant.

-Judy Johnson for editorial assistance, manuscript

preparation, patience and encouragement which were central to the completion of the study.

-Alison Johnson and Meredith Johnson for understandingcooperation, and independence necessary to allow their


iii









father to devote the necessary time to the project.

-Mr. John Dumbauld for continuing to be my friend.


iv








TABLE OF CONTENTS
Page

ACKNOWLEDGMENTS iii

LIST- OF TABLES vii

LIST OF FIGURES ix

ABSTRACT x

CHAPTER I INTRODUCTION 1

Purpose of the Study 3
Setting for the Study 4
Need for the Study . 5

CHAPTER II REVIEW OF THE LITERATURE 7

Facilitative Therapeutic Conditions 7
Facially Mixed Counseling Dyads 9
State Anxiety 11
Client and Counselor Relationships 13
Initial Interview or Short Term Counseling 13
Client Counselor Compatibility 14
The Concept of Compatibility 16
Research Using Schutz's Concept and Indices
of Compatibility 18
Summary 19

CHAPTER III PURPOSE AND DESIGN 21

Definitions 21
Hypotheses 21
Hypothesis 1 22
Hypothesis 2 22
Hypothesis 3 22
General Design 22
Subjects 25
Counselors 25
Instrumentation 26
Fundamental Interpersonal Relationship
Orientation - Behavior *26
Importance of Becoming Admitted to the
Nursing Program 28
Subject's Feeling about Examination Performance 29
The Affect Adjective Check List 32
Interview Rating Scale 35
Preference for Counselor Indicator 35


v







Page


Counselor's Level of Facilitative Conditions 35
Barrett-Lennard Relationship Inventory 35
Specific Design Consideration Precautions 38
Procedures 40
Counselor Training 40
Initial Data Collection 42
The Study Interviews 44
.Counseling Interview 45

CHAPTER IV RESULTS AND DISCUSSION 49

Confidence Level Parameters 49
Control Study for Facilitative Conditions
Generally Offered by Participating Counselors 49
Subject Pool 54
Independent Variables 59
Subjects 59
Subject-Counselor Compatibility 65
Length of the Interview 69
Dependent Variables 70
Anxiety'Change 70
Rating of the Interview 74
Choice of Counselor for Future Occasion 77
Testing of the Hypotheses 77
Preliminary Analysis of the Data 79
Hypothesis 1 83
Hypothesis 2 91
Hypothesis'3 100

CHAPTER V SUMMARY AND CONCLUSIONS 106

Summary 106
Conclusions 113

APPENDICES 116

REFERENCES 167

BIOGRAPHICAL SKETCH 174


vi










LIST OF TABLES


Table Page

1 Facilitative Conditions Generally Offered 50
to Female Client in Separate Control
Study

2 Control Study Client Ages 52

3 Length of Control Study Interviews 53

4 Control Study Racial Group Membership 55

5 Comparison of Participating Subjects and Non-Participating Subject Pool Members 57


6 Comparison -of Participating Subjects and 58
Non-Participating Subject Pool Members

7 Age of Subjects 60

S Subject Racial Group Membership .61

9 -Number of Years of Schooling Completed by 62
Subjects

10 Importance Placed on Becoming Admitted 64
to Program

11 Subject Feeling About Entrance Test 66
Performance

12 Discrepancy from Passing Score on Entrance 67
Examination

13 Subject-Counselor Compatibility 68

14 Length of Interview 71

15 First Anxiety Measure 72

16 Second Anxiety Measure 73

17 Anxiety Change 75

18 Interview Rating Scale 76

19 Preference for Particular Counselor for- 78
Future Occasion


vii












Table Page

20 Summary - Multiple Correlation with Anxiety
Reduction 85

21 Summary - Multiple Correlation with Rating
of Interview 93

22 Categorical Weight Coefficients 103


viii











LIST OF FIGURES


Figure Page

1 Level of Importance Scale 30

2 Subject Feeling about Entrance Examination
Performance 31

3 Opinion of Future Counselor Choice 36


ix




I


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


COUNSELING WITH REJECTED NURSING SCHOOL APPLICANTS:
SOME CORRELATES OF SELECTED OUTCOME MEASURES


.By

Douglas G. Johnson

June, 1977

Chairman: Theodore Landsman Major Department: Counselor Education

Five white male counselors engaged in a total of 95 interviews. during which a female applicant to an Associate Degree Nursing Program was informed she had not passed an entrance examination to qualify for admission to that program.

The framework for the study was a theory of interpersonal needs (Schutz, 1960) with the primary focus being the effect of reciprocal.compatibility in the area of affection on the outcome of these stressful interviews. Subject anxiety change over the course of the interview, a rating of the 'interview by the subject, and an indication of the subject's feeling about seeing the counselor at some future occasion were selected as measures of outcome.

The study was done as field research with other measurable factors used as additional independent variables. A system for random assignment of subjects to counselors was used. A basic interview format was used with each counselor allowed to operate according to his usual manner or style.

x










Using a multiple regression analysis procedure, subjectcounselor compatibility was found to be related to anxiety reduction at a low level. The compatibility accounted for only about two percent of the anxiety reduction.variance and was less related than the importance the subject placed on being admitted to the program. Only about 13 percent of the anxiety reduction variance was accounted for at a significant level by a combination of variables.

A multiple regression analysis was done with respect to the subjects' ratings of the interviews. All the variables including. a variety of combination variables indicative of specific effects based on categorical subgroup were included in the analysis with the multiple correlation coefficient remaining significant. The major finding about the -ratings of the interviews was that over 50 percent of the variance was associated with variables which were related to lower ratings of the interviews. Subject-counselor compatibility was not supported as related to better or worse ratings of the interviews.

A discriminant analysis- procedure was done to determine if subject-counselor compatibility was related to subjects indicating a preference to return to the same counselor on a future occasion. The hypothesis that compatibility would be an.indicator of an expression to that effect by subjects was rejected. The distribution of actual responses was so heavily weighted in favor of subjects indicating a preference to return to the same counselor that any conclusions would be


Xi










highlyy suspect, The overwhelming tendency of subjects to indicate a preference to see the same counselor on some future counseling occasion was not explained by the results.

A subject's perception of failure as either a "personal" problem or an "educational-vocational" problem in relation t.o a particular counselor's style or manner was discussed as appropriate for further investigation.

The hypotheses that higher levels of subject-counselor

compatibility would be positively related- to higher levels of. anxiety reduction, better ratings of the interviews, and a showing of preference to return to the same counselor in the future were upheld at a low level -in the first case and not upheld in the other two instances. The hope that the reciprocal subject-counselor compatibility in the area of affection proposed by Schutz (1960), which could be easily used as a possible basis for client-to-counsel-or assignment might be utilized, was not realized. The concept of compatibility as used in this study seemed to shed little light on the complex relationship between subject and counselor.


xii













CHAPTER I

INTRODUCTION


The premise that' the nature of the interaction of client and counselor personal characteristicsis of importance and is in need of further investigation has received considerable support (Ferreira, 1964; Garfield, 1973; Kiesler, 1966; Lorion, 1974; Sheiner, 1967; and Strupp, 1962, 1973a, 1973b). Several writers have attempted to define some elements of the interaction between counselor and client. :Patterson (1967) stated, "It would appear that the relationship [:of counselor with client jis an important factor in any interpersonal interaction, and therefore is basic to counseling or psychotherapy" (p. 85). Friedman (1960) drawing on the ideas of Martin Buber, made

the point that even though psychotherapy is one-sided, it

is still an 'I-Thou' relation founded on mutuality,

trust and a partnership in a. common situation .." (p. 31). Bordin (1955) encouraged counselors to be friendly, accepting, and willing to work with clients where they are (emphasis added). Hansen (1963) used the; term rapport to describe the appropriate interpersonal relationship between client and counselor. Shoben (1954) suggests that conditions which . . mediate affective security . . ." (p. 47) are the first order of business in counseling.

It is widely accepted that counselors should possess a genuine,- positive feeling toward their clients as specific


1







2


human beings. This concern is, in fact, a part of the definition of Unconditional Positive Regard (later called Non-Possessive Warmth) as originally posited by Rogers in his statement about the necessary and sufficient conditions of therapeutic personality change (1957). Non-Poss.essive Warmth has more recently emerged as one condition which seems to be highly dependent on both the client and the. therapist in the initial phase of therapy (Truax, Wargo, Frank, Imber; Battle, Hoehn-Savic, Nash, & Stone, 1966). The inference can be made that the level of warmth is a function of both the client and the counselor,. at least in initial phases of counseling or therapy.

A situation in which the client's and counselor's

personal characteristics are at -some optimal level, or are in balance, or are compatible, with respect to one another could be an important factor with regard to the nature and outcome of the counseling relationship. Lorion (1974) urged delineation of characteristics which relate to ". . positive outcome with any social or ethnic group" (p. 352). Beutler (1973) suggests that the patient-therapist dyad be considered ". . as a separate treatment whose effectiveness depends not only on the individual characteristics of patients and therapists but upon characteristics of their mutual compatibility" (p. 305).

It would seem then that compatibility and some

mutuality of warmth, at least.in the initial stages of





3


counseling, might be beneficial to the re-lationship and thereby to the outcome of the counseling which occurs within that relationship.

Schutz (1960) gave.the term compatibility a very specific explanation, stating that "Compatibility is a property of a relation between two or more persons . . . that leads to mutual satisfaction of interpersonal needs and harmonious coexistence" (p. 105). Compatibility-canbe measured for any pair of persons based on a standardized instrument called the Fundamental Interpersonal Relationship Orientation - Behavior, devised by Schutz (1960), and the use of a specific formula for comparing scores obtained on that instrument by the two people. As the warmth of the relationship between client and counselor is of importance, it was Schutz's formula for compatibility which was used in this study.


Purpose of the Study

It was the purpose of this study to examine some of the effects of client-counselor compatibility on clients in an initial one-to-one, educational-vocationa.1 counseling interview during which a female applicant to a registerednursing program is told she is not admissible to that program, based on her earned nursing program entrance test scores. .Specifically the study focused on the client's state anxiety reduction over the course of the intervew. her evaluation of the counseling interview, and her





4


reported selection of a counselor should she decide to seek out a counselor again.


Setting for the Study

Female applicants to the.Santa Fe Community College Associate of Science Registered Nursing program not achieving a grade level score of 12.0 on both the math and reading sections of the Comprehensive Test of Basic Skills were requested to meet with a professional counselor. The purposes of the counseling interview were as follows:

-To inform the applicant that she had not scored

high enough on the Comprehensive Test of Basic Skills to enter the Registered Nursing program for which she had applied.

-To discuss with the applicant the implications of her scores for possible future admission to the program.

-To discuss with the applicant her educationalvocational goals in light of the current situation.

-To discuss with the applicant any other matters

which seemed important within this general context of not meeting admission requirements fo' the Registered Nursing program.

It was intended that through this process the applicant would be assisted to deal with her confrontation with failure in a constructive and positive way.

Each counselor involved in the study conducted the

interview in his own manner and style. Some elements were





5


added to the normal overall testing and counseling process in order to gather data pertinent to the completion of this study. A complete description of the procedures can be found beginning on page 40.


Need for the Study

Brough's study in 1965 suggests that the single most

important source of perceptions about counselors among high school students is actual experience in counseling. Assuming the sane to be true for potential college students, being able to assign prospective college students to 'counselors with whom they are interpersonally compatible would seem to merit consideration, especially if the prospective students must be informed that they are not admissible to a particular program due to entrance scores.

Blocher (1967) supported the idea that matching client

and counselors in a variety of ways may be an important issue relating to the outcome. of counseling.

Graff, Rogue, and Danish (1974) suggested that there

is currently a specific need to do research which can help to understand the outcomes of educational-vocational counseling, which both high school and community college counselors deal with frequently.

The issue of client-counselor compatibility was the subject-of this study. The concept of compatibility as presented by Schutz (1960) addresses itself to the specific relationship between two persons. Schutz's index of reciprocal





6


compatibility was designed to be an indicator of the likelihood of two persons expressed and wanted needs being complementary.

The research undertaken in this study examined the effect of the compatibility of a client-counselor dyad on specific counseling outcomes, The outcomes were client anxiety reduction, client rating of the counseling interview, and the client's expressed preference to see the same counselor on another occasion. It was- posited that greater compatibility between the client and the counselor would be related to greater anxiety reduction in the client, a better rating of the interview by the client, and greater likelihood that the client would prefer to return to the same counselor on a future- occasion. Positive findings would suggest that client-counselor compatibility would be useful as a method for assignment of clients and counselors in better than a random fashion.














CHAPTER II

REVIEW OF THE LITERATURE


That the relationship between the client and the counselor is of importance was presented as the major focus of this study in Chapter I. The review of the literature deals specifically with issues of importance to this study which are of an a priori nature, with a review of the research on client-counselor relationships, and with a specific review of the research utilizing Schutz's (1960) postulate of compatibility.


Facilitative Therapeutic Cond-itions

The research'dealing with facilitative therapeutic

conditions based on Rogers' (1957) idea of the conditions

necessary for personal growth in therapy is massive. The presence of high levels of facilitative conditions related to positive outcomes in therapy and low levels of facilitative conditions related to negative outcomes in therapy was adeptly shown by Truax (1966). These conditions were also shown to be largely the contribution of the therapist (Truax et al., 1966).

Though originally intended as indices related to

positive outcome in therapy with a client centered orientation, facilitative conditions have been shown to have an


7






8


effect on the progress of verbal conditioning even using the inexperienced -counselors with a behavioristic orientation (Mickelson &. Stevic, 1971), which seems to show their relevance in a greater range of situations than Feidler's (1950) finding that experience may be more crucial than theoretical orientati.on.or technique.

Rogers (1957) stressed the importance of the patient's awareness of facilitative conditions. Truax (1966) pointed out, however, that interpersonal perceptions are not likely to be accurate in most patient populations. and may not, therefore, be as reflective of some "actual" level of presence of the conditions.

Barrett-Lennard (1962) presented an instrument to measure the.level of facilitative conditions from the perspectivTe of the client. He considered.this to be the intent of Rogers' thinking and built a "Pelationship Inventory" which would give five scores: Empathic Understanding, Level of Regard, Unconditionality of Regard, Congrience, and Willingness to be Known.

Though the matter of facilitative conditions is

clearly of importance, the scales.as developed by Truax (1961, 1962a,*1962b) were designed for use by independent raters working from typescripts or tape recordings of psychotherapy interviews with disturbed patients-, The conditions under which the interviews used as a context for the present study occurred were expected to be stressful





9


and/or ego threatening. Recording the interview would have introduced additional design problems (Campbell & Stanley, 1963) and recording of interviews has been shown to have at least some effect on client functioning (Gelso, 1973). The Barrett-Lennard Relationship I-nventory and measurements of facilitative conditions obtained from tape recordings have been shown to be related to outcome in therapy (Barrett-Lennard, 1962; Truax, 1966).

It was decided, therefore, that a measurement of the ability to provide facilitative conditions from the point

- of view of-the client, but independent of the research interviews, be obtained for each of the counselors who participated in this study. As research with the instrument has shown the Relationship Inventory to be most parsimonious when used as a total measure (Mills & Zytowski, 1967; Lanning & Lemons, 1974), total scores were used in this control phase. Characteristics of the Barrett-Lennard Relationship Inventory and specific treatment of these data as a control measure are discussed in the section on preliminary procedures in Chapter III.


Racially Mixed Counseling Dyads

A-preliminary inspection of the situation in which

this study was conducted suggested a subject pool with at least a 50:50 black-white ratio, with a possibly much higher black-white ratio, potentially as high as 85:15. When given a choice of a counselor similar to or different





10


from the client on racial and cultural bases, Stranges & Riccio (1970) found that preferences were strongly in favor of similarity of counselor chosen to the choosing client.

Studying conditions of race, sex, and age mix of

clients and counselors in a survey followup of vocational counseling with lower socio-economic group clients, Beckner (1970) got mixed results with the'most poin-ted finding being that while 25 percent of the client respondents said that racial and ethnic backgrounds were of importance, 68 percent said those issues were not of importance.

Russell (1970) stated that blacks were no different

from whites in terms of feelings, wishes, hopes, fears -and frustrations, yet several writers asserted that a white counselor .just cannot hope to help a black person as they are so culturally different. (Thomas, 1969; Williams & Kirkland, 1971). Walkin, Moriwaki, & Williams (1973) noted that, "The conditions which foster trust at the beginning of a relationship must be explored and then exploited to improve the.effectiveness of therapy" (p. 316). They noted that blacks seem to be more reserved and recommended more -black therapists.

The whole problem in this area seems to have been touched upon experimentally by Ewing (1974) in which neither favorability of rating nor effectiveness was shown to be significantly related to the race of the co.unselor, except that white clients did tend to rate black counselors





11


. as less helpful in initial interviews in which precollege

*students.were being-seen by experienced counselors.

The review by Dreger & Miller (1968) suggests that lumping blacks into one category is similar to lumping any -arge group of people into one category, with most. such categories standing a very good chance of not really being homogeneous along a number of dimensions.


State Anxiety

Being told that her entrance test scores are not high enough to allow her to be admitted to a Registered Nursing program would be expected to raise the level of anxiety in a subject. This is a situation analogous to a task performance (digit -symbol) feedback experiment in which subjects who were informed they were doing poorly showed a significant rise.in state anxiety (Fremont, Means, & Means, 1970). The general theory of interpersonal relationship needs posed by Schutz (1960) states:

A discrepancy between the satisfaction of
an interpersonal need and the present state or the organism engenders a feeling in the
organism that shall be called anxiety (p. 16). The effects of failure stress on anxiety (Hodges & Spielberger, 196.9) and poor interpersonal compatibility based on not having interpersonal needs well met would certainly seem of importance i.n this sttidy.

The Affect Adjective Check List (Zuckerman, 1960) has been successfully used to demonstrate changes in state anxiety associated with information that subjects were





12


doing poorly at digit span tests, noted above as failure stress. It has been shown to be sensitive when used over short time periods (Hodges & Spielberger, 1969; .Zuckerman, Lubin, Vogel, & Valerius, 1964), and can be experimen tally calibrated when used as a dependent variable (Zuckerman et al., 1964)..

Of particular interest with respect to this study

were the results obtained with sophomore nursing students in a course crit-ical for continuance in that program (Hayes, 1966). The nonexamination day mean scores on the Affect Adjective Check List are not much higher than those present'ed by Zuckerman (1960), but the examihation day mean scores were higher. Hayes. suggested that this'may have been due to the necessity of passing the course in order for the students to continue in the program, which is similar to the conditions under which. the present study was conducted. This finding seems consistent with and supportive of the results obtained with students'- reporting differing levels of examination worry and the sensitivity of the Affect Adjective Check List to such personal states (Zuckerman & Biase, 1962). Ten subjects rating themselves "low" in examination worry showed a differential from the class mean of -1.42 on examination days; 15 students rating

themselves "moderate" in examination worry showed a mean change of +1.81 on examination days; and seven subjects rating themselves "high" in examination worry showed a mean change of +4.23 on examination days '(F = 5.71, p( .01).





13


Spielberger (1972) noted, ". . . the usefulness of the Affect Adjective Check List as a measure of day-to-day fluctuation in state anxiety has been clearly demons.trated" (p. 35).


Client and Counselor Relationships. Initial Interview or Short Term Counseling

A variety of research on first or initial interviews is available in the literature. Greenberg (1969), in an analogue study, noted that just telling a subject that the therapist was warm or cold combined with information about his being experienced or inexperienced had the effect of having a "warm" therapist seen as more attractive, more effective, and influential over the subject than.a "cold" therapist.

Inexperienced counselors were better 1.iked than exper-ienced counselors, controlling for facilitative conditions, in another analogue study of initial interviews (Pope, Nudler, Vonkorff, & McGhee, 1974). The explanation was perceived similarity of subjects to analogue therapists, both subjects and analogue counselors being of similar age and class (college students).

Pope & Siegman (1962, 1965, & 1968) and-Pope, Siegman & Blass (.1970) discovered that initial interview verbal behavior varies with changes in the specificity of therapist remarks, the warmth or coldness of the interviewer,

and induced ego threat anxiety (in this case inferred poor family adjustment).





14


Grater (1964) discovered that clients presenting themselves to- a college counseling center showed more willingness to discuss personal-social concerns with counselors who were considered to be more "affective" in

approach and manner while fewer clients who met with more "cognitive" counselors did so, preferring instead to discuss educational-vocationa.l matters, The "compatibility"or preference for style was not overwhelming but present nonetheless.

In a study of neurotic versus normal students' levels of self-disclosure during a first short telephone conversation with either-a high or low self-disclosing experimenter

-as the other party, normals, presumably more similar to and

compatible with the experimenter, reciprocated levels of self-dis-closure while neurotic students' levels of self-disclosure remained'in the middle range and did not fluctuate much (Chaikin, Derlaga, Bayma, & Shaw, 1975).

A study of three groups of students grouped for high, medium, and low similarity to the counselor on the Allport Vernon- Lindzey Study of Values showed that the middle similarity group changed more in the meaning of two concepts than either the high similarity or low similarity groups (Cook, 1966),


Client-Counselor Compatibility

Psychoanalysts, who have long considered the "relationship" between.patient and therapist to be crucial, are





15


reporting that therapists with particular personality dynamics prefer patients with. certain diagnoses and have differentially successful results depending upon the "compatibility" of the therapist's and patient's personality dynamics (Reimann, 1968).

Similarity with respect to personality "type" as

measured by the Myers Briggs-Type Indicator (.MBTI) has been shown to be related to longevity in counseling (Mendelsohn & Geller, 1963; Mendelsohn, 1966). Middle ranges of similarity of personality type (MBTI) have been discovered to be -related to a-.better evaluation of the client-therapist contact than were measures of very high or very low similarity (Mendelsohn & Geller, 1965). High client-counselor similarity levels have also been related to initial interpersonal attraction (Izard, 1960, 1961).

The classification of therapists as either "A" or "B" type therapists (Whitehorn & Betz., 1960) with respect to differential outcome with schizophrenic patients marked the beginning of an .avalanche of studies using that classification system as a theoretical base. The controversial nature of that research is well demonstrated in a short article by Eysenck and a reply by James & Foreman (Eysenck, 1975). Recent research has begun to include classifications of patients along the same dimensions (Berzins, Friedman, & Seidman, 1969) in therapy analogue.

A fitting summary statement with respect to the entire matter of inferred compatibility or matching was made by





16


Beutler (1973) while writing about both the learning theory issues and the "A" or "B" type therapist issues as they pertain to learning theory:

The alternative . . . is'to consider each
patient-therapist pair as a separate treatment.whose effectiveness depends not only
upon the individual characteristics of patients
and therapists but upon characteristics of
their mutual compatibility (p. 305).

The concept of reciprocal compatibility proposed by Schutz (1960) meets Beutler's criterion both in theory and as a usable numeric index.


The Concept of Compatibility

Schutz (1960) presented a specific definition of the nature of interpersonal compatibility as a part of a

-broader theory of interpersonal interaction based-on a postulated set of interpersonal needs. The basic postulates of the theory were:

1. Every person has three interpersonal needs: inclusion, control, and affection.

2. Inclusion, control, and affection constitute a

sufficient set of areas of interpersonal behavior for the' prediction and explanation of interpersonal phenomena (p. 13).

The postulate of compatibility is stated, "If the. compatibility of one group, h, is greater than that of another group, m, then the goal achievement of h will

exceed that of m" (p. 105). A theorem of dyadic preference is also posed:





17


If the compatibility of one dyad, yl, is
greater than the compatibility.of another dyad, Y2, theh the members-of yi are more likely to prefer each other for continued
personal contact (p. -120). He further stated;

For the dyad, this theory means that a
person wants to act a certain way toward others and wants to be acted toward in a
certain way . . . . By comparing A's
description of how he likes to be acted
toward with B's description of how he
likes to act toward people, and vice versa,
a measure of mutual need satisfaction
emerges (p. 107).

Reciprocal compatibility is the specific term used to describe this status. The measure (index) of reciprocal compatibi-lity of persons i and j is expressed by the formula

.rKij = ei - wj +Iej - wi

in which ei and.ej are the measures of how i and j typically behave toward others, and wi and wj are the measures of how i and j typically want others to behave toward them. The value of rKij. when computations are used based on the instrument created by Schutz to scale the postulated needs (pp. 61-65), becomes an index of the reciprocal compatibility for the dyad, ij. The value of the index can only be positive due to the use of absolute values in the final summation and the value can range from 0 to 18. The degree of compatibility is highest when the index number is 0 and lowest when the index number is 18. A conversion procedure using .18 minus the actual index level computed for a dyad





18


can be used to produce an ascending scale of compatibility indices (Schutz., 1960).


Research Using Schutz's~Concept and Indices of Compatibility

Two person compatibility based on.the use of the

Fundamental Interpersonal Relationship Orientation - Behavior formulations has been demonstrated to be related to interpersonal relationships in several ways. Schutz (1960) showed that while chare would have any two members of a fraternity select each other for roommates and for.reciprocal compatibility to occur in about 13 percent of the cases, 31.3 percent actually made such choices (p< .001 for difference) (p. 124).

Gassner (1970) showed that personal reactions toward

therapists were positively related to reciprocal compatibility in a clinical setting in which juvenile delinquents and pastors formed the dyads.

Verbal conditioning under either the "set" of interpersonal compatibility or measured total compatibility, using a broader range of formulae with rk as one of them, showed that the effect of the conditioning was present during the presence of the experimenter when.compatibility was high or was withheld until after the experimenter left the room when the compatibility was low (Sapolsky. 1960).

In other research the Pearson correlation between

resident therapist/patient total compatibility and super.visor's rating of patient improvement was shown to be .45 (p( .05) (Sapolsky, 1965).





19


. Mendelsohn & Rankin (1969) in studying clients' perceptions of the relationship and evaluations of both the counselor and the usefulness -of counseling.where N = 115 over an eight session therapy or counseling regimen, discovered that compatibility was differentially related to various outcome measures for male and female clients. The Pearson product moment correlation between general evaluation of counseling and compatibility was .34 (p .05) thus showing reciprocal compatibility in the affection area to be negatively correlated with general evaluation of counseling by clients.- Their discussion points to the possibility ". . that factors which foster strong emotional attachments in typical social relations can, unless carefully handled, lead to an excessive personalization of the special relationship which characterizes counseling and psychotherapy" (p. 163)..


Summary

Various compatibility indices seem to produce differing correlations with a variety of outcome criteria for counseling or therapeutic dyadic interpersonal interaction. The emphasis placed on the importance of warmth in the beginning phase of therapy (Shoben, 1954; Truax, 1966) and the conflicting results of Sapolsky (1965) and Mendelsohn & Rankin (1969) are adequately summarized by Sapolsky (1965) when writing of Schutz's theory and instrumentation, he states, '. . . an important underlying personality variable





20


contributing to the establishment of 'good' therapeutic relationships . . . may have been identified" (p. 75).

Schutz's unique theory provides just such a framework

within which both client and counselor personality variables in identifiable areas can be viewed. His postulates, formulae, and indices provide the tools, based on and integrated with that theory, which prompted the writer to propose that the-Fundamental Interpersonal Relationship. Orientation - Behavior theoretical framework should be utilized to examine a "typical" counseling situation using reciprocal compatibility in the area of affection as the primary treatment variable of interest.












CHAPTER III

PURPOSE AND DESIGN


Within the stressful setting in which a counselor .informs a client, on the basis of entrance examination scores, she has not -qualified for admission into an Associate of Science Registered Nursing program for which she- has applied, this study addressed three major questions:

l.. During the interview does the level of mutual

compatibility of the client-counselor dyad have an effect

-on the anxiety level of the client?

2. Does the level of mutual compatibility of the

client-counselor dyad have an effect on the client's evaluation of the interview?

3. Does the level of mutual compatibility of the client-counselor dyad have an effect on the client's indication of preference for selection of a counselor on another occasion?


Definitions

Dyad: a meaningful dialogue, encounter, or relationship. between two people. In this study the client-counselor dyad is defined as the client and counselor when they are meeting in a counseling interview.

Compatibility: the ability.to exist together;

the ability to get on well together. The level of mutual


21




22



compatibility of the client-counselor dyad in this study is represented by an index number. The lowest index number, 0, indicates a minimum level of compatibility while an index number of 18 indicates a maximum level of compatibility.


Hypotheses

.Three major hypotheses were tested by the research

project.

Hypothesis 1

Clients in higher compatibility client-counselor dyads will show greater reduction in anxiety over

the course of the interview than will-clients in

lower compatibility. client-counselor dyads. Hypothesis .2

Clients in higher compatibility client-counselor

dyads will rate the counseling session more favorably

than will clients in lower compatibility clientcounselor dyads.

Hypothesis 3

Clients in higher compatibility client-counselor

dyads will indicate a preference for seeing the same

counselor again significantly more often than will

clients in lower client-counselor compatibility

dyads.


General Design

The research was designed to study the effect of client-counselor compatibility on selected outcomes of





23


a stressful counseling interview. The outcomes selected for examination were the client's reduction in anxiety over the course of the interview, the rating of the interview by the client, and the client's expressed preference with respect to seeing the same counselor should counseling be considered appropriate in the future.

Various procedures required for data collection were introduced into an ongoing series of -occurrences which constitute the application .sequence for the Santa Fe Community College Registered Nursing program. The data collection procedures were introduced into the achievement testing and the counseling with rejected-female applicants segments of the sequence.

Hypothesis 1 was tested through the use of a "One

Group Pretest-Posttest Design" (Campbell & Stanley, 1963). Hypotheses 2 and 3 were tested through the use of a "Static Group Comparison" design, except that rather than comparing a group which received no treatment with a group which did. receive' treatment as the design provides, this study compared differing levels of the treatment (compatibility) (Campbell & Stanley, 1963). Due to the relatively large number of independent variables necessary to provide adequate control for unknown sources of variance in the outcome measures, a statistical procedure which couldshow the relationships of a large number of independent variables with the outcome measures was necessary.

Walberg (1971) encouraged the use of general




24


regression models when analyzing outcome measures to test hypotheses. He particularly emphasized their efficacy when large numbers of independent variables were used.and when the data wereprimarily of a continuous nature. Cohen (1968) suggested the appropriateness of multiple regression analysis as an appropriate tool in such- cases. He also, confirmed the robust nature of the regression procedure to analyze data .which reflect membership in a group or category, e.g., race, sex, etc., when it has been properly coded and thereby can be treated as continuous. The fact that the outcome measures, anxiety reduction and rating of the interview were continuous and reasonably normally distributed permitted the selection of multiple regression analysis as an analytic tool to test Hypotheses 1 and 2.

The fact that the outcome measure of preference for seeing the same counselor on some future occasion should

counseling seem appropriate allowed only one of three choices made the use of some other data analytic tool necessary. Discriminant analysis, as available in the Statistical Analytic Systems, Inc. package (Barr., Goodnight, Sall, & Helwig, 1976) provides weight coefficients for each independent variable with respect to each of the possible indications on an outcome measure. The relationship .of a particular independent variable to a given response on an outcome measure .can thereby be determined. As the weight coefficients could be interpreted to test .Hypothesis 3, discriminant analysis was selected as the





25


data -analytic tool to test Hypothesis 3.


Subjects

The pool of subjects for this study was the female applicants to the Associate of Science-Registered Nursing Program at Santa Fe Community College who did not reach the required admissions scores of the 12.0 grade levelon both the math and reading sections of the Comprehensive Test of Basic Skills and who were at least 18 years old or had parental permission to participate. The sample consisted of all such cases beginning with those taking the Comprehensive Test of Basic Skills Test on or after August 19, 1975 and ending when 95 subjects had completed all phases of the experimental procedures.

As males had historically represented less than ten

percent of the applicants.to the program not meeting minimum admissions scores, it was decided that they would not be considered within the scope of this study. Counselors

Counselors were selected to present characteristics which would replicate the normally occurring conditions of the study interviews as created by the staffing- patterns of the college. The five counselors who agreed to participate were male, white, trained at the master's degree level, had five or more years of experience with community college students, were between 35 and 55 years old, and espoused a client-centered orientation.





26



Instrumentation

The instruments used in the study are described

below. Specific utilization of each instrument is presented in the section of this chapter entitled "Procedures." Fundamental Interpersonal Relationship Orientation - Behavior

The Fundamental interpersonal Relationship OrientationBehavior instrument was used to determine the compatibility of the client-counselor dyad in this study. Schutz's (1960) concept of compatibility addresses itself to the specific relationship between two persons. His index of reciprocal compatibility was designed as an indicator of the likelihood of two persons expressed and wanted needs being complementary. The complementary nature of the affection needs of the client and the counselor was considered to be related to the existence of mutual warmth in a beginning counseling relationship which has been shown to be dependent on both client and counselor in the initial phases of their relationship (Truax et al., 1966).


The need scales

The Fundamental Interpersonal Relationship OrientationBehavior (Schutz, 1960) is a 54 item self-report questionnaire composed of six Guttmann type scales (Inclusion-Expressed, Inclusion-Wanted, Control-Expressed, Control-Wanted, AffectionExpressed, Affection-Wanted). Schutz reports internal . consistency in terms of reproductibility for five of-the six scales at .94 with the Control-Expressed scale at .93


.. I




.27


based on primarily college student samples averaging 1543 subjects.

Test-retest reliability reported for the individual

scales includes Inclusion-Expressed, .82 (p( .01); InclusionWanted, .75 (p (.01); Control-Expressed, .74 (p( .01); Control-Wanted, .71 (p ( .01) based on a time interval of one month with a mean number of- 140- (Range 125-183). A one week interval with 57 college students yielded test-.retest reliabilities of .73 (p < .01) for Affection-Expressed and .80 (p< .01) for Affection-Wanted.

Intercorrelation of the Expressed and Wanted dimensions of each of the areas is Inclusion = .62 (p( .01); Control = .25 (p< .05), and Affection = .70 (p <.01). The compatibility scales

The reciprocal compatibility of any dyad (ij) is computed by the formula

rKij =ei - wj + ej - wi

in which ei and ej are the measures of how i and j typically behave toward others and.wi and wj are the measures of how i and j typically want others to behave toward them. In the formula e stands for the scale score on the expressed scale in any of the areas of inclusion, control, or affection, and w stands for the scale score on the wanted scale in any of the areas. Prior to the final summation, signs are dropped using only the absolute values resulting in compatibility scores ranging from 0 to 18 with 0 meaning complete compatibility and 18 meaning complete incompatibility.




28



The conversion formula 18 minus the computed index value was used (Schutz, 1960) to produce an ascending scale of indices.

Based on the computation of reciprocal compatibility in the affection area for all possible combinations of pairs of two from the first to the second testing, the group of college student's upon which test-retest reliability was determined (N = 57), a full scale Pearson product moment correlation of .67 (p( .01) was achieved. See Appendix A for a sample of the Fundamental Interpersonal Relationship Inventory - Behavior as used in this study. Importance of Becoming Admitted to the Nui-sing Program

Failure stress was shown by Hodges & Spielberger (1969) to have the effect of significantly raising reported state anxiety on the.~Affect Adjective Check List (Zuckerman, 1960). They showed that information that a subject was doing poorly on digit span test was related to increased state anxiety. Increasing levels of reported examination worry have been shown to be significantly related to increasing levels of state anxiety prior to an exami-nation (Zuckerman & Biase, 1962). Extremely high levels of state anxiety were shown to be present in sophomore nursing students on examination days in a course which determined continuance or noncontinuance in that program (Hayes, 1966). It seemed reasonable that the level of importance a subject might place on becoming admitted to the nursing program might play a part in her state anxiety when told she had not passed the entrance examination.





29


It also seemed reasonable that asking a direct question of the subjects about how important admission to a given program was to them personally would be the most direct way to obtain the information. A question in scale form was developed within the context of asking students how important becoming admitted.to the nursing program was to them compared to other important events in their lives. This scale is shown in Figure 1. Although the level of importance indicator does not have known validity or reliability it was decided that some attempt to assess the perceived importance of becoming admitted to the program might add relevant information which could shed light on the hypotheses -being tested in the study (see Appendix A for specific format). Subject's Feeling about Examination Performance

Zuckerman (1960).demonstrated that state anxiety

increased in subjects who were told they were doing poorly at digit span tests. Based on this finding, it was considered to -be important to discover each subject's feeling about her performance -on the entrance examination, as it would possibly affect her level of anxiety.during the interview. A simple question (see Figure 2) was devised and asked of each subject at the conclusion of the examination session (see Appendix A for specific format).





30


Many events in our lives are of importance (graduatingfrom school, marrying, having children, getting a new car, finding a genuine friend, learning a new skill, et-c.). Each of us considers some of these events to be of greater importance than others.

DIRECTIONS: Taking into account the events in your
life which are now or have been of
IMPORTANCE to you, please place an '."
at the point on the scale shown below
which BEST indicates the importance YOU
place on becoming admitted to the program
for which you are applying.



IMPORTANCE

(some) (most)


2- 4 67




(NOTE: Marking "1" indicates some importance
while marking "7" indicates that becoming
admitted to the program is as important
as anything in your life.)






FIGURE 1


Level of Importance Scale




31


DIRECTIONS:


Please place an "X" on the line to the left of one statement below which BEST -reflects your feelings about'how you did on the Comprehensive Test of Basic Skills you took a little while ago.


I probably didn't pass. I HONESTLY have no feelings about it either way.

I probably passed.


FIGURE 2


Subject Feeling about Entrance Examination Performance





32



The Affect Adjective Check List

The measurement of state anxiety was central to this study. The Affect Adjective Check List (Zuckerman, 1960) was selected as an appropriate instrument to.measure state anxiety. The Affect Adjective Check List is a list of 61 words, 11 of which, if checked, are scored as indicative of state anxiety and 10 of. which, if not checked, are considered indicative of state anxiety (see Appendix E). The other 39 words are neutral and all words are listed alphabetically. The range of scores.is 0 to 21. The internal reliability using the Kuder-Richardson Formula 20 is .85 and is significant beyond the .001 level. The test-retest reliability is .31 (p< .05). Validation was demonstrated by students (34) scoring a differentially higher score mean of 2.41 .53 (p( .005) for three preexamination occasions compared to ten nonexamination days. These results were replicated with sophomore nursing students in a course which meant continuance or noncontinuance in the program (Hayes, 1966). Specific data for these nursing students were:

Mean SD

All- class periods 5.86 3.19
Test days 13.99 3.15

Difference 8.13 (p< .05)

Two forms of the checklist were devised for-use in the present -study. Each was called the Feeling Word Check List. Form #1 (printed on green paper) was derived by using the middle word (#31) as the first word and then


. I




33



working-alternately backward-then-forward, one word at a time, to arrive at the new ordering of words. Form #2 (printed on yellow paper) was derived in the same way but by alternately working forward-then-backward, one word at a time to arrive at the new ordering of words. The word "gay," a neutral word, was then dropped from both lists based on the loaded connotation which that particular word has acquired over the past several years (see Appendix E).


Interview Rating Scale

A scale which could be used to give a numeric rating

of the counseling session was required to test Hypothesis 2. The Counseling Evaluation Inventory was developed in order to measure client ratings of counselors (Linden, St.one, & Shertzer, 1965). A 68 item Counseling Evaluation Inventory was developed including the .50 items of the Interview Rating Scale described by Anderson & Anderson (1962)'. Following this initial development phase by Linden et al. which included 446 counselors and 289 students, all 68 items were retained. These 68 items were then submitted to 336 university stu.dents to evaluate their counseling experience, and the items were keyed based on weightings derived from the initial screening of the 68 items.

Factor analysis produced four factors, Counseling Climate (X), Counselor Comfort (Y), and two parts of a Client Satisfaction factor (Z-1 and Z-2). Only items that loaded .40. on one but- less than .40 on all other




, .4 ~ -


34



factors were retained in the final 21- item short form. One item proved to be unreliable at the .05 level and -was omitted from the shorter form though it- was reliable on the longer form. Two additional items-were deleted from use in this study. One item requested an evaluation of the counselor's effectiveness in showing the subject the value

of testing. The other asked.the subject to evaluate the counselor's effectiveness in commenting on how the subj6.ct could better reach her goals.. Both items were considered potential sources of bias by suggesting to the subject what the outcome of the session should have been in terms of tests and future plans, two topics which might well carry significant weight in the subject's thinking at the close of the session.

The test-retest reliability of the Counseling Evaluation Inventory (Linden et al., 1965) over a 14 day period with 16.3 secondary school students is .83 for the total score while the coefficients.for the subscales are X = .78, Y = .63, and Z = .74, all significant at beyond the .05 level. The total score was used in this study. The name "Interview Rating Scale." as originally used by Anderson & Anderson was selected on the premise that it would be less- likely to foster a spurious rating of the session in a subject's attempt to give the -counselor a "good" or "bad" evaluation.




35



Preference for Counselor Indicator

In order to test the applicability of Schutz's (1960)

theorem of dyadic compatibility as it relates to preferences 'for continued contact with a particular person. (Hypothesis 3) some measure was needed to permit each- subject to express

her opinion with respect to continued contact with a particular counselor, should that be necessary. As what was. wanted from each subject was her opinion about how-she might behave in the future, it was decided to ask each subject her opinion-directly. Figure 3 illustrates the format in which the .question was asked (see. Appendix F for placement of the question as part of the Interview Rating Scale).



Counselor's Level of Facilitative Conditions

A measure to determine the comparability of the facilitative conditions offered by the counselors participating in the study was considered essential. The Barrett-Lennard Relationship Inventory (Barrett-Lennard, 1962) was adopted for this purpose.


Barrett-Lennard Relationship Inventory

- The Relationship Inventory (Barrett-Lennard, 1962) is a 64 item questionnaire (Form OS-M-64) with a Likert type response pattern. It was developed to assess the necessary conditions for establishment of a helping relationship as described by Rogers (1957) and yields separate scores for each of four scales (level of regard, empathic understanding, congruence, and unconditionality of regard)




36


A..


If you should decide to seek out a counselor on another occasion, would you seek out (check only one):


This same counselor No preference


A different counselor


FIGURE 3


Opinion of Future Counselor Choice





37



and a total score. The scores are composed of eight positively-scored items and eight negatively-scored items per scale.

The items of the -Relationship .Inventory were judged

by a panel of five client centered counselors (Barrett-Lennard, 1962). All but one item was placed in either the positive or negative category by all the judges. Odd-even internal consistency coefficients based on a sample of 40 patient judges ranged from .82 (p< .01) to .93 (p < .01) with a mean of .86 (p( .01), compared to therapist coefficients ranging from .88 (p< .01) to .96 (p< .01) with a mean of .92 (p< .01). Test-retest correlations over a four week period based on a sample of 36 college students rating a close, long-standing personal relationship they felt was stable and unchanging were .89 (p< .01)-on an empathic understanding scale, .84 (p< .01) on a level of regard scale, .90 (p< .01) on an unconditionality of regard scale, .86 (p< .01) on a congruence scale, and .95 (p< .01) total.

Mills & Zytowski (1967) found that test-retest correlations under the -directions of estimating their mother's feelings about them (N 79) were .90 (p< .01) for accurate empathy (Empathic Understanding), .74 (p< .01) for level of regard, .80 (p< .0l)-for unconditionality of regard, and .88 (p(..01) for congruence, all'significant (p( .01). These four scales were reported to account for approximately 70 percent of the variance loading on a single factor. They




38


subsequently recommended that the.instrument be used to assess the total relationship rather than components of it. This point is further supported by the recommendation that the use of the subscales of the relationship seemed unjustified, again on a factor. analytic basis (Lanning & Lemons, 1974). See Appendix D for a copy of .the BarrettLennard Relationship Inventory and a tabulation scoring sheet for the Inventory.


Specific Design Consideration Precautions

Cautions were exercised, in accordance with recognized research design techniques, to minimize any intruding factors which could jeopardize the validity of the research (Campbell & Stanley, 1963).. Hypothesis 1 is of the "One. Group Pretest-Posttest 'Design" (Campbell & Stanley, 1963). The problem of factors other than the treatment (compatibility) accounting for differences in outcome among subjects (anxiety change) is not present in this study to any large degree. The short time span between the first and second testing of anxiety used to determine the anxiety change protected against the effects of other intervening variables accounting for observed differences in outcome. The use of two different forms of the instrument used to measure anxiety helped protect against the effect of the first test influencing the second. The nature of the instrument was such that "learning" from one testing to the next was not likely to occur. The fact that the





39


meaning of the instrument was not likely to change based

on an observer's interpretation and that its use over a

short time span had been previously demonstrated as appropriate (Hodges & Spielburger, 1969; Zuckerman, Lubin, Vogel, & Valerius, 1964) protected against 'problems inherent to the

instrument. All levels of anxiety change and compatibility were used in the study, thereby eliminating error from the.

use of only extreme scores. Clients were assigned to counselors on a rotation basis and interruptions in counseling

- were minimized by hanging a "do not disturb" sign on the

counselor's door and having all telephone calls held.

Whatever effect the measurement of anxiety from the first testing.to the second testing might have on the change in

anxiety is not known. Introduction of the anxiety measuring

instrument as an aid to help the subjects focus on their

feelings was considered to minimize this possible source of

extraneous variation. No reports of adverse reactions were observed by participating counselors. In addition to the precautions noted above,' the possible-effects of

the'subject's age, race, discrepancy from the required

passing score, feeling about her entrance examination

performance, the importance she placed on being admitted

to the program, the length of the interview, the number of years of schooling the subject had completed, and the

particular counselor she saw in the interview were treated as additional variables in the analysis of the data so that possible effects of those factors could be determined.





40



Hypotheses 2 and 3 were of the "Static-Oroup Comparison" design except that rather than comparing a group which received no treatment with a group which did as the design provides, this study compared differing levels of the treatment (compatibility)... To this extent the design of the study with respect to.Hypotheses 2 and 3 was also considered analogous to the "Recurrent Institutional- Cycle Design" (-Campbell & Stanley, 1963) as the experimental group was exposed to the various treatments (differing levels of compatibility) over a period of time.' The major difficulty with this type of design is that the nature of the situation might influence the completion of the measurement device. A quiet place to complete the instrument and assurance of the subject's anonymity were used as precautions against social or personal pressures to rate the interview positively or.negatively or to indicate a particular preference for a counselor on another occasion.


Procedures

Counselor Training

Prior to the beginning of the study, the-five

counselors participated in.a three hour training session with the write. During the session the counselors and the researcher discussed the procedures to be followed and the use of the instruments they were to administer. A standard format for the procedural parts of the interview was agreed upon by all of the counselors. The counselors





41



practiced the use of the instruments with each other and were able to conduct two simulated interviews with each other until a standard method for administration of those instruments was achieved.

The "Counselor Guidelines" statement (see Appendix

C) based on the format developed during the counselor training session was distributed to each counselor. Each counselor was also requested to complete the Fundamental Interpersonal Relationship Orientation - Behavior instrument. Each was asked.to complete it according to the directions with an indication that it was needed as a general descriptive measure. The counselors were not apprised of the specific nature of the research. They were told that the project would best be served if they behaved in. their usual manner and that it was not their counseling style which was being investigated. All counselors agreed to participate under these conditions.and*.to follow the Counselor Guidelines as distributed.

To protect against differing levels of facilitative conditions generally offered by the participating counselors being introduced as an uncontrolled source -of variance in -the outcome of the study, each counselor agreed to participate in a control study separate from the study of the rejected applicants. The counselors each agreed to present several female clients he -had seen for the first time for 30 minutes or more to his secretary.' He told the client that the secretary had a questionnaire for her to





42


complete to assist the college in improving its services. The nature of the counseling session was not specified except that it could not-be one in which the counselor was required to tell the client that she did not meet entrance requirements for a particular program or curriculum.

The writer provided each counselor's secretary

with Barrett-Lennard Relationship Inventory questionnaires and instructed him or her to ask female clients to be brought in by the counselor to complete one questionnaire. The client was encouraged to be honest in her responses on the questionnaire items and was- told that neither she nor the counselor would be identified on the questionnaire.


Initial Data Collection

Each female applicant (potential subject) to the

Associate of Science Registered Nursing program was required to schedule the reading and mathematics sections of the Comprehensive Test of Basic Skills. Each applicant was allowed to choose her own test date from a list of prescheduled testing dates. She was told how long the testingsession would take. Twenty minutes beyond the time normally used to administer the Comprehensive Test of Basic Skills was included in that statement to allow time 'to gather data required for this study. When administration of the entrance examination had been completed, the initial data gathering instrument was distributed (see Appendix A).





43


Also distributed was a written statement about the research entitled "Consent Form''which had been .approved by the University of Florida's Committee for the Protection of Human Subjects (see Appendix B).

As applicants to other health related programs were also taking the Comprehensive Test of Basic Skills during these- testing sessions, all persons taking the test were asked to complete the data gathering instruments and to read and sign .the consent form. Emphasis was placed on the point that the information gathered-in this research would have absolutely no effect on admission to the programs for which students were applying.

The data gathering instrument included spaces for the- applicants to indicate name, the program applied for, age, sex, race, telephone number, highest level of schooling completed, the Level of Importance Indicator, the item indicating perception of performance on the Comprehensive Test of Basic Skills, and the Fundamental Interpersonal Relationship Orientation - Behavior. It was possible that age, race, highest level of schooling completed, the level. of importance the person placed on being admitted to the program, and the feeling about examination performance items would have a potential bearing on the client's response to the main treatment and they were gathered as away of providing caution against confounding from unknown sources. A copy of the data gathering instrument can be found in Appendix A.





44


All applicants were thanked for their cooperation in this research effort and assured that their specific responses to the questionnaire would be confidential. Name and telephone number information requests were explained as necessary if follow-up were to become desirable. Upon departure all applicants were handed a sheet thanking them again for their cooperation in the research and instructing them about making an appointment to get the results of the Comprehensive Test of Basic Skills and reassuring them of the confidentiality of the research information (see Appendix B). The Study Interviews

The instruction sheet given applicants upon completion of the Comprehensive Test of Basic Skills and the first phase of the data collection for this study instructed them to call the researcher's secretary to make an appointment to see a counselor about their Comprehensive Test of. Basic -Skills results. The procedures in which the secretary would participate were discussed thoroughly with her prior to the beginning of the interviews. These procedures were considered practical by her and were practiced until they were done consistently with five consecutive surrogate subjects.

The secretary was furnished with a.roster of applicants taking the Comprehensive Test of Basic Skills after each testing session. Those applicants to the Registered








Nursing program who had not passed the Comprehensive Test of Basic Skills at the required 12.0 grade level and.had completed the initial data gathering instruments were noted on each roster as appropriate for an appointment with one of the five participating counselors. These applicants are hereinafter referred to as subjects. Those applicants not so identified, including applicants to other health related programs, were scheduled for an appointment with the writer or another nonparticipating counselor.

The secretary maintained a master roster of counselor availability and made an appointment for each subject to see one .of the participating counselors on the basis of the first appointment time available to the subject after she called for an appointment. Subjects were not told by

the secretary whether or not they had passed the Comprehensive Test of Basic Skills but were informed that the counselor would have that information for them at the scheduled appoitnment. Subjects were informed that the counseling session would take approximately an hour. Subjects' files were available to the counselor for reference during the interviews.


Counseling Interview

Seating arrangement for the interview placed the counselor at his desk with the subject at the side of the desk. This provided space for review of the subject's folder and space for the subject to complete necessary




46


data collection forms. This arrangement should have had neither a positive nor negative effect on subject anxiety as reported by Myers (1969). The secretary allowed no interruptions and a "do not disturb" sign was placed on the door to further insure privacy and continuity in the interview. While getting settled for the counseling interview, the counselor informed the subject that the interview would last approximately 45 minutes and that the secretary would want to see her for about five minutes when the interview was over. An easily read clock was within clear view of the client during the- interview. No attempt- was made to terminate the interview prematurely simply to meet the time frame stated unless other obligations of the counselor required it. In that case the counselor would, at the end of the interview, invite the client to make another appointment.

By looking at but without opening the subject's

folder, the counselor informed the subject that she had not passed the Comprehensive Te-st of Basic Skills (based on a coded notation on the tab of the folder .just below the subject's name). At this point, the first function of the interview had been completed. In accordance with the standard procedure stated in the Counselor Guidelines and training received prior to the beginning of the study, the counselor immediately asked the subject to complete the "Feeling Word Check List #1" while he 'studied her test protocols. This administration of the Feeling Word Check




47



List was considered the initial (pretreatment) measure

of state anxiety used in the computation of data to test

Hypothesis 1.

The counselor recorded the time the Feeling Word

Check List #1 was completed on the Process Notes form (see

Appendix C) provided in the subject's -folder. This notation

is used in the computation of data included to test the hypotheses. Once form #1 of the Feeling Word Check List

was completed the counselor placed it aside and continued

with the interview in a manner of his own choosing. He was

required to include each of the purposes of the interview,

- which were the following:

- -To discuss with the subject the implications of

her scores for possible future admission to the program.

-To discuss with the subject her' educationalvocational goals in light of the current situation.

-To discuss with the subject any other matters

which seemed important within this general context of not meeting admission requirements for the Registered

Nursing program.

When both subject and counselor agreed that the

discussions had reached a natural termination point, the subject was asked to complete Feeling Word Check List #2.

The counselor reco-rded the time Feeling Word Check List #2

was begun on the Process Notes form. This notation was

used in computation of data to test the hypotheses. The

score on Form'#2 was the posttreatment measure of state




48


anxiety and was used in the computation of data to test Hypothesis 1.

The counselor then escorted the subject back to the secretary and returned to his office. He made notes of events, impressions and feelings on the Process Notes form which he believedto be of importance as he had counseled with the subject.

The secretary asked the subject to complete a short opinionaire, ostensibly as the final par.t of the college's attempt to gain-insight into how to improve its services to students and prospective students. The subject-was seated in.a quiet area to complete the Interview Rating Scale which was used to test Hypothesis 2. One additional item was added to the Interview Rating Scale (#19) and was used to test Hypothesis 3. A copy of the Interview Rating Scale including the added item can be found in Appendix F.












CHAPTER IVRESULTS AND DISCUSSION

Confidence Level Parameters

The .05 level of -confidence was established as the necessary.indication that the results obtained in statistical procedure did not occur by chance.


Control Study for Facilitative Conditions
Generally Offered by Participating Counselors

The scores obtained on the Barrett-Lennard Relationship Inventory completed by the control study clients represented the level of facilitative conditions offered by participating counselors. The scores ranged from 46 points to 117 points. The mean was 113-.30 points with a standard deviation of 32.72. Table 1 presents the -means and standard deviations computed for the facilitative conditions variable for the entire group of control study clients and the subgroup of clients seen by each counselor. The analysis of variance for the facilitative conditions variable indicates that the level of facilitative conditions generally offered to female clients by the participating counselors is not significantly -different from counselor to counselor as measured by the Barrett-Lennard.Relationship Inventory (Barrett-Lennard, 1962).

A comparison of the age of the clients seen by the participating counselors as a part of the control study 49




30


TABLE 1

Facilitative Conditions Generally Offered to
Female Client .by Counselors in Separate Control Study



Total Score on the Barrett-Lennard Relationship Inventory


Group


Mean


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


117.38 111.88

114.88 114.25 108.13


Entire Group


113.30


SD N


26.28 28.18

30.81 32.15

46.19



32.72


8

8 8*

8

8



40


Analysis of Variance


Sum/Squares


df - Mean Square


Between groups Within groups


616.938


39217.850


4


90


154.234 1120.512


Total


39217.850


*Counselor #3 did 10 control study interviews while the other counselors did 8. The highest and lowest scores attained .by Counselor #3 were dropped in this computation to balance cell size.


Source


F


.1376

NS


SD


N





51


was made with age of the clients seen by the counselors in the actual study interviews. The ages of the clients in the control study ranged from 17 years to'51 years. The mean age was 23.92 years with a standard deviation of 8.31 years. Table 2 presents the means and standard deviations computed for the entire control study group and the subgroup seen by each counselor. The analysis of variance of the ages of the control study clients shows that the mean.ages of the control group clients and the subgroup of control group clients seen by each counselor were not significantly different. Table 2 also shows the results of a t test for group means which shows the control study client -mean age was not significantly different from the subject

sample mean age.

A comparison of the length of the control study interviews was made-with the length of the actual study interviews. The length of the control study interviews ranged from 15 minutes to 45 minutes. The mean length of the control study interviews was 28.95 minutes with a standard deviation of 7.45.minutes. Table 3 presents the means and standard deviations for the length of the control study group and the subgroup seen by each counselor. The analysis of variance of the length of the control study interviews shows that the mean interview length -for the group and the subgroup seen by each counselor is not significantly different. Table 3 also shows the results of a t test for group means which shows the mean length





TABLE 2

Control Study Client Ages


Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 All Clients


27.50 21.87 20.25

24.87 25..12


23.92


13.72 6.79

3.45 7.45 6.62 8.31


8

8

8

8

8


40


Source

Between groups Within groups


Ana-lysi.s of Variance

Sum/Squares df

310.475 4

2420.131 35


Mean Square

77.61

69.14


2720.131



t Test


Control Study Clients

23.92 8.13

.98


Subject Sample

23.03

6.19

.64


.609 p( .01


F


1.1253

NS


39


Mean SD

SEM


52


Total


Value of t











Length of


TABLE 3

Control Study Interviews.


Group Mean. SD N


Counselor -1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group


Analysis


Source

Between groups .Within groups.


Sum/Squares

226.271 1105.502


of Variance

df Mean Square

4 56.56

35 31.58


1331.773


t Test

Control Group Clients

28.95

7.43 1.18


Subject - Sample

27.87 18-.89

1.94


Value of t


53


25.75 27.50 30.00 31.50 30.00 27.95


3.99 3.99 0.00 9.95

5.34


7.43


8

8

8

8

8


40


Total


F


1.79 NS


39


Mean SD

SEM


.404 p 4. 01





54


- of the control study interviews is not significantly different from the mean length of the study interviews.

The control-study group was composed of five black and 35 white clients. Tabl-e 4 presents the.absolute frequency of black and white control study clients for the entire group and for the subgroup seen by each counselor. It also shows the percentage of the entire group or subgroup falling into each category. A Chi-Square test was done to compare the control study racial group balance with the subject sample racial group balance. Table 4 shows that the Chi-Square is significant beyond the .01 level indicating that the control study group is not of the same racial balance as the study sample. Although it can-be said that the facilitative conditions generally offered to female clients arenot significantly different from counselor to counselor, the small number of black clients seen as part of the control study does not allow that assurance to be made with respect to black clients seen by the counselors.


Subject Pool

- Applicants to the Registered Nursing pr-ogram who

did not pass the Comprehensive Test of Basic Skills at the required 12.0 grade level on or after August 19, 1975, and continuing through November 22, 1976, became members of. the subject pool. A total of 277 took and did not pass the Comprehensive Test of Basic Skills at the necessary level during this period. There were 70 of that number








TABLE 4

Control Study Racial Group Membership


Black


White


N

1

1

2

1

0


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group


cv. c~.


%4 14 14 33

14

0


N

7

7

6

7

8


5 13


%0 86 87 67 86

100


N

8

8



8

8


35 87


100

100 100 100 100


40 100


Chi Square Test


Subject Sample Control Study* Group


Chi Square value (2df)


X2 = 34.95 (p4 .05 = 3.841)'


Group


Total


Black

N

31


%3

33


White

N

.64


5 13


67 87


35




56



who did not complete the initial data collecting instruments at the time of the administration .of the Comprehenisve Test of Basic Skills, leaving a pool of 207. Ninety-five subjects made appointments with a counselor -to get their test results, were seen by the counselors participating in this study, and completed all data gathering instruments, thereby becoming the sample of subjects for this study.

A comparison of the subjects who participated in

this study and of the 112 subject pool- members who did not participate was made to determine whether the subjects of this study represented the broader group. Table 5presents the comparative information. The t tests shown.in the table demonstrate that the subject group and the nonparticipating subject pool group are not able to be distinguished from one another with respect to their discrepancies from the required passing score.on the entrance examination, their ages, the level of importance- they placed on being admitted to the program,, and the number of years of.schooling they had completed. Chi-Square tests shown in Table 6 suggest the groups were not significantly different in racial group membership and their feeling about their entrance examination performances. The subjects who participated in this study, therefore, cannot be differentiated from the total population of applicants who failed to meet the required passing score on the entrance examination.




57


TABLE 5

Comparison of Participating Subjects
and Non-Participating Subject Pool Members


Participants Non-Participants

Value of t
Mean SD N Me an SD N


Discrepancy from passing score


24.04 21.91 95


.64
p = .52
(two-tailed)


22.15 20.34 112


23.03 6.19 94


23.73 7.48 112


-.72
p = .47 (two-tailed)


Importance of. Being Admitted


.85 95


-.42
p = .67
(two-tailed)


Years of Schooling Completed


12.61


.93 95


12.79 1.24 112


-1.13
p = .26
(two-tailed)


Age


6.13


6.18


.82 108









TABLE 6


Comparison of Participating Subjects
and Non-Participating Subject Pool Members


Participants Non-Participants

N Row % N Row%

Black 31 57.4 23 42.6

White 64 41.8 89 58.2


2
Chi Square Value (1 df) X = .145
p @.O.5 = 3.841




Felt passed
examination 56 47.5 62 52.5

"No feeling"
about examination
performance 19 45.2 23 54.8

Felt not
passed
examination 20 44.4 25 55.6

Chi Square Value (2 df) X2= 3.298
-p @ .05 = 5.991.




59


Independent Variables

Subjects

Age. The subjects' ages ranged from 17 years to

50 years. The mean was 23.032 years with a standard deviation of 6.193 years. Table 7 presents the means and standard: deviations computed for -the age variable for the entire saniple .of subjects and for the subsample of subjects seen by each counselor. The results of the analysis of variance for the age variable show that the mean ages of the subjects in the entire sample and the subgroup of subjects seen by each counselor were not significantly different.

Race. Sixty-four of the subjects identified themselves as white while 31 identified themselves as black.. Table 8 presents the absolute frequencies of black and white subjects for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category and the results of a Chi-Square test showing that the subgroups of subjects seen by each counselor are not sufficiently out of.balance to indicate that the subgroups are likely to be from a different population of subjects.

Years of Schooling. The number of years of schooling completed by the sample subjects ranged from 10 years to 16 years. The mean was 12.611 years with a standard deviation of .926 years. Table 9 presents the means and standard deviations computed for the years of schooling





60


TABLE 7

Age of Subjects


Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group


24.300

*23.105

22.222 24.400 20.706 23.032


7.658

4.421 5.589

8.127

2.974 6.193


20 19 18

20

17*


94*


Analysis of Variance


Source Between groups Within -groups

Total*


Sum/Squares

173.4648 3393.4609 3566.9258


df

4

89. 92


Mean Square

43.3662 38.1288


.*The age of one subject was not obtained. Computations basedon N shown.


F

1.1374 NS




61


TABLE 8

Subject.Racial Group Membersh-ip


Group Black White Total


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


N

8

3

4

9

7


Entire Sample 31


%O 30 16

22 45 39


N

12 16

14 11 11


Cr /0

70

84 78 55 61


N

20

-19 18

20 18


33 64 67 95


Chi Square Test


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


Sum: Dev2/Exp Sum: Dev2/Exp Value of X2


Black Exp Obs

6.5 8 6.2 3 5.9 4 6.5 9 5.9 7


Dev2 2.25

10.24 3.61 6.35

-1.21


White Exp Obs .13.5 12 12.8 16

12.1 14 13.5 11 12.1 11


Dev2

2.25

10.24 3.61 6.35

1.21


3.792


(black plus white) (p( .05, df = 4)


5.627 (p< .05)

.9.488


/0

21 20 19

21 19


100


Total



20 19 18

2.0 18




62


TABLE 9

Number of .Years of Schooling Completed by Subjects



Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire group


12.75

12.95 12.33 12.60 12.39


12.61


1.21

0.97 0.68

0.82 0.78


0.93


20 19 18

20 18 95


Source

Between groups Within groups


Total


Analysis of Variance

Sum/Squares df

4.8135 4

75.7764 90


80.5898 94


Mean Square

1.2034 .0.8420


F

1.4292

NS




63 .


variable for the entire sample and the subsample seen by each counselor. The results of the analysis of variance for the yedrs of schooling variable show that the mean number of years of schooling completed by subjects in the entire sample and the subsample of subjects seen by each counselor were not significantly different.

Importance Placed on Becoming Admitted to Program. The point at which the level of importance indicator was marked was scaled graphically to the nearest half point and was labeled as the subject's level of importance score. The marked level of importance scores ranged numerically from a low of three to a high of seven. The possible scale scores ranged from one to seven. The mean was 6.13 with a standard deviation of 0.85. Table 10 presents the means and standard deviations computed for the level of importance variable for the entire sample of subjects and for the subsample of subjects seen by each counselor. The results of the analysis of variance for the level of importance variable.show that the mean level of importance checked by subject in the entire sample and the subsample of subjects seen by each counselor are not significantly different.

Feeling about Entrance Examination Performance. Subjects were asked to place an "X" next to one of three statements which best reflected how they felt about their performance on the Comprehensive Test of Basic Skills. Fifty.six subjects indicated they felt they had passed, 19 indicated they hon-estly had no feelings either way, and 20 indicated




64


TABLE 10

Importance Placed on Becoming Admitted to Program


Group Mean SD N


Counselor 1 6.20 1.11 20

Counselor 2 6.03 0.98 19

Counselor 3 6.03 0.79 18

Counselor 4 6.29 0.64 20

Counselor 5 6.11 0.72 18


Entire group 6.13 0.85 95



Analysis of Variance

Source Sum/Squares df Mean Square F

Between groups 0.9167 - 4 0.2292 0.3047

Within groups 67.6840 90 0.7521 NS


Total 68.6057 94




L)3


they felt they had not passed. Table 11 presents the absolute frequencies by response category for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category and the results of a Chi-Square test showing that the subgroups of subjects seen by each counselor are not sufficiently out of balance to show.the subgroups lilcely to be from a different population of subjects.

Discrepancy from Necessary Passing Score. The number of points by which each subject failed to meet the required 12.0 grade level on the Comprehensive Test of Basic Skills was labeled the subject's discrepancy score. The discrepancies ranged from- one point to 89 points. The mean was 24.04 points with a standard deviation of 21.91 points. Table 12 presents the means and standard deviations computed for the discrepancy score variable for the entire-sample and the subsample seen by each counselor. The analysis of variance for the discrepancy score variable shows that the mean discrepancy scores for the entire sample of subjects and the 'subsample of subjects by each counselor were not significantly different.


Subject-Counselor Compatibility

The subject-counselor dyad compatibility index

values as measured by the Fundamental Interpersonal Relationship Orientation - Behavior ranged from a minimum of two to the maximum possible of 18. The mean value was 12.505 with a standard deviation of 3.727. Table 13 presents the means




66


TABLE 11

Subject Feeling About Entrance Test Performance


Group . Pass Don't Know Didn't Pass


Counselor -I Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Sample


N

11

14

-11

14

6


10

55

74 61 70 33


. 56 59.


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


Chi Square

Pass

Exp Obs Dev2 11.8 11 .64 11.2 14 7.84 10.6 11 .16 11.8 14 4.84 10..6 6 21.16


Test

Don't Know

Exp Obs Dev2

4 5 1.

.3.8 4 .04

3.6 1 6.76

4 2 4

3.6 7 11.56


Didn't Pass Exp Obs Dev2 4.2 4 .04

4 1 9

3.8 6 4.84 4.2 4 .04 3.8 5 1.44


Sum: Dev2./Exp. 3.176 Sum: Dev2/ExD 'All) Value of X2 (p( .05, df = 12).


6.539 13.636 (p< .05) 21.026


N

5

4

1

2

7


-19


10

25'

21

6

10 39


20


N

4

1

6

4

5


20


20

5

33

20 28


21


3.921




Ii


TABLE 12

Discrepancy from Passing Score on Entrance Examination



Group Mean SD N




Counselor 1 26.80 23.04 20

Counselor 2 18.05 1.9.09 19

Counselor 3 26.22 24.68 18'

Counselor 4 20.70 16.87 20

Counselor 5 28.83 25.62 18


Entire group 24.04 21.91 95




Analysis of Variance

Source Sum/Squares df Mean Square F

Between groups 155.8672 4 388.9668 0.8033

Within groups 43579.9805 90 484.2219 NS


Total 45135.8477












TABLE 13

.Subject-Counselor Compatibility


Group Me an SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


All dyads


14.000 13.526 9.611* 10.850*

14.500 12.505


2.000 1.867

4.461 4.660 2.093


3.727


Analysis of Variance


Source. Between groups Within groups


Total


Sum/Squares

341.6846 964.0654 1305.7500


df

4

90


94


Mean Square

85.4211 10.7118


F

7.9745 (p( .01)


*Significantly different.from each other and the group mean through the application of Tukey's test for significant gap (Edwards, 1954).


68


20 19 is

20 18 95




69



and standard deviations. computed for the compatibility index values for all of the subjedt-counselor dyads and for.the subsample of dyads involving each of the counselors. The results of the analysis of variance of the compatibility index values indicate that the mean compatibility index for the subsample of subject-counselor dyads involving Counselors #3 and #4 was lower than for the subject-counselor-dyads of the other three counselors. This lowered

- mean compatibility index for Counselors #3 and #4 did not preclude their inclusion in any further analysis as the specific compatibility of the subject-counselor dyad was treated on a case-by-case basis with each of the outcome measures.. It did mean that there would be. a lowered possibility for the compatibility of those dyads to show as strong a relationship to the outcome measures.. It also suggested that each counselor should be considered as a special subgroup in preliminary data analysis procedures used to-discover interaction effects.


. Length of the.Interview

The length of each study interview in minutes was determined by computing the difference between the time the subject completed the first measurement of anxiety taken immediately after the subject was told she had not passed the Comprehensive Test of Basic Skills at the required level and the time she began the second measurement of anxiety at the conclusion of the interview.





70


The length of the interviews ranged from six minutes to 96 minutes. The mean length was 27.874 minutes with a standard deviation of 18.893 minutes. Table 14 presents the means and standard deviations computed for the length of interview variable for all of the subject-counselor dyads and for the subsample of dyads involving each of the counselors. The analysis of variance for the length of interview variable demonstrated that Counselors #1 and #4 engaged in interviews of significantly greater mean length .than did the other three counselors. The possibility that longer or shorter interviews might have differential effects on- outcome measures suggested that each counselor should be considered a separate subgroup when examining preliminary data for interaction effects.


Dependent Variables

Anxiety Change

Change in client anxiety over the course of the

interview was determined by subtracting the measured anxiety

-level of each subject at the end of.the interview from her measured anxiety level at the beginning of the interview for each subject. As the anxiety change value is derived, an examination was made of the scores on the first and second anxiety measures and a t test for paired measures was done to determine the validity of the anxiety change. values. Tables 15 and ]6 show the mean and standard deviation values for the first and second anxiety measures













TABLE 14

Length of Interview


Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5


28. 000* 23.105 18.333 52.300** 15.167 27.874


All Groups


Source Between groups Within groups


Total


Analysis of

Sum/Squares 16910.0430 16644.5195


33554.5625


Variance

df Mean Square 4 .4227.5078

90 184.9391


F

22.8589 (p( .01)


*Significantly different at the .05 level through the application of Tukey's test for stragglers (Edwards, 1954).

**Significantly different at the .05 level through the application of Tukey's test for significant gap (Edwards, 1954).


71


-8.820 8.110. 4.863

26.274 5.-227 18.893


20 19 18

20

18. 95














First


TABLE i5 Anxiety Measure


Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 All groups'


Source

Between groups Within groups


Analysis of

Sum/Squares

0.8423

1203.3062


Variance

df Mean Square

4 0.2106

90 13.3701


1204.1484


72


11.000 11.000 11.111 11.250

11. 111


11.095


4.554 3.559 2.720 3.508

3.612 3.579


20 19 18

20 18 95


F

0.0157 NS


Total











TABLE 16

Second Anxiety Measure


Group Mean . SD N



Counselor 1 7.800 4.099 20

Counselor 2 7.263 3.297 19

Counselor 3 9.222 3.209 18

Counselor 4 8.200 2.285 20

Counselor 5 8.444 4.090 18


Entire group 8.'168 3.441 95




Analysis of Variance

Source Sum/Squares df Mean Square F

Between. groups 39..6680 4 9.9170 0.8313

Within groups 1073.6406 90 11L9293 NS


Total 1113.3086 94





74


for the entire sample of subjects and the subsamples of subjects seen by each counselor. The analysis of variance for those variables shows that the mean anxiety score for the entire sample of subjects and the subsample of subjects seen by each counselor did not differ significantly.

The anxiety change values for the sample ranged

from.minus ten to 12.- The mean anxiety change was found to be.2.905 with a standard deviation of 3.573. Table 17 8hows the mean and standard deviation values for the entire sample of subjects and subsample of subjects seen by each counselor.. The analysis of variance of the anxiety change variable shows that the mean anxiety change value for the entire sample of subjects and the subsample of subjects seen by each counselor did not significantly differ. The results of a t.test for mean differences between paired measures shows that the mean difference in measured anxiety over the course of the interview was significant across the pool of subjects.


Rating of the Interview

A rating of each interview was taken using the

Interview Rating Scale. The scores on the Interview Rating Scale ranged from 0 to 38. The mean score was found to be 28.579 with a standard deviation of 9.313. Table 18 shows the means and standard deviations computed for the Interview Rating Scale scores for the entire sample of subjects and the subsamples seen. by each counselor. The analysis





;; 5


TABLE *17

Anxiety Change


Group Mean SD N SNK Range
p = .05


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire group


3.200 3.737 1.778 3.050 2.667 2.905


3.750

2.864 3.766 3.790

3.964 3.573


20 19 18

20 18 95


3*.37 3.70

3.94



3.94 2.82


Source

Between groups Within groups. Total


Analysis

Sum/square

39.202

1160.945 1200.147


of Variance s df Mean Square

1 4 9.8005

3 90 12.8994

5 94


t Test - First Anxiety Measure with Second Anxiety Measure

First Measure Second Measure

Mean 11.095 8,168

SD 3.579 3.441


Dif f

SD/Diff Value of t


2.905 3.573 8.02


p< .01


F

0.7598

NS





76


TABLE 18

Interview Rating Scale


Group Mean SD N


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire group


31.250

26.474 28.333 29.750

- 26.778


28.579


Source

Between groups Within groups


Analysis of

Sum/Squares

313.8867

7839.3.633


Variance

df Mean Square

4 78.4717

90 87.1040


8153.2500


7.629 12.607 8.139 6.382

10.773 9.313


20 19 18

20 18


95


F

0.9009

.NS


jF9F3!FVC1F-


Total





77


of variance of the results on the Interview Rating Scale shows that the mean rating of the interview scores for the entire sample of subjects and for the subsample of subjects seen by each counselor did not significantly differ. Choice of Counselor for Future Occasion

- An indication of each subject's choice of counselor,

should seeing a counselor at some unspecified future time seem appropriate, was obtained. Subjects were asked to check one of three response categories--to see the.same counselor, no preference, or to see a different counselor. Table 19 presents the absolute frequencies by response category for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category.


Testing of the Hypotheses

The study was undertaken to discover if a relationship exists between certain outcomes of stressful counseling interviews and several .variables associated with those interviews. Ninety-five subjects completed all phases of the study. They were seen by one of five counselors.

Three measures of outcome were chosen for examination. They ware as follows:

1. Change in subject anxiety over the course of

the interview.

2. A rating of the interview by the subject.





7 3


TABLE 19

Preference for Particular Counselor for Future Occasion Group Same No Preference different


C.
IC


Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group


is 13 11 16

14


90.0

68.4 61.1 80.0 77.8


72 75.8


2

6

6

4


22


10.0 31.6 33.3

20.0

22.2


23 . 2


0

0

1

0

0


0

0



0

0


1 1.1


Chi Square Value (4 df )


0 = 2.90




1117 Tyl ,_


3. An indication of the subject's choice of

counselor if seeing a counselor. again should

seem appropriate.

The primary focus of the study was to determine if the subject-counselor reciprocal compatibility in the area of affection (Schutz, 1960) was correlated with the measures of outcomes. Other variables which were thought to be related

-to the measures of outcome were also considered. They included the following:

1. The age of the subject.

-2. The racial group to which the subject belonged.

3. The number of years of schooling the subject.

had completed.

-4. The importance of becoming admitted to the

.program as reported by the subject.

5. The subject's feeling about her performance on

the entrance examination.

6. The discrepancy from the required passing score

on the entrance examination.

7. The length of the interview. Preliminary Analysis of the Dat'a.

A series of Pearson product moment correlations were computed between each of the outcome measures and each of the independent variables for each of several subgroups of subjects. The subgroups were black subjects,

subjects who reported "no feeling" about entrance examination performance, and subjects who reported they felt they


79




"0


had failed the entrance examination. Thi-s was done to inspect for the possibility that these subgroups of subjects might show a difference in their relationship-to an outcome measure on one of t.he independent variables which might not appear when the entire sample of subjects was considered as a group. Subjects who saw-Counselors #2, #3, #4, or #5 were also considered subgroups for this same reason. This use of "Counselor" as a subgroup was necessitated by the fact that the client-counselor compatibility scores and the length of the interviews conducted by each counselor were not comparable as demonstrated in the analyses of variance of those variables.

The subgroups of white subjects, subjects who

expressed the feeling they had passed- the entrance examination, and subjects seen by Counselor #1 were not considered subgroups for this purpose as their scores on independent variables were automatically considered in the general data analytic procedures used -to test the hypotheses.

The intercorrelations for each subgroup (see Appendix G) were inspected for correlations between independent variables and outcome measure. An interaction effect was considered to.exist when a significant-correlation between an independent variable and an outcome measure was found for a particular.subgroup. An interaction term was considered .not to exist, however, when an even number of subgroups of similar nature, e.g., Coun.selor #2 and Counselor #4 both showed a correlation between an independent variable and





81


an outcome measure and the signs of the correlation coefficient were in opposite directions, thereby having offsetting effects. Neither was an interaction effect considered to exist when the.number of subjects in a subgroup fell to seven or below, as statistical analyses based on such a small number were considered unsound. Interaction terms were computed for the interaction effects noted by application of these criteria.

Eleven such interactions were found to exist among the independent variables. The five with respect to the subject's anxiety change over the course of the interview were as follows:

-Reporting "no feeling" about performance on the

entrance examination combined with the discrepancy from the required passing score (r = .39, p .05,

N = 19).

-Reporting "no feeling" about performance on the

entrance examination combined with the importance

of becoming admitted to the program (r = .41,

Sp( .05, N = 19).

-Having seen Counselor #3 combined with the length

of the interview (r = .46, p<.05, N = 18).

-Having seen Counselor #5 combined with the clientcounselor compatibility score (4 = .43, p<.05-,

N = 18).

-Having seen -Counselor #5 combined with the

importance of becoming admitted to the program





82


(r = .46, p(.05, N = 18).

The eight interactions with respect to the rating of the interview by the client were. as follows:

-Being black combined with the length of the interview (r = .40, p <.05, N = 31).

-Being black combined with the discrepancy from the required passing score on the entrance examination

(r = -.47, p.01, N = 31).

-Being black combined with reporting "no feeling"

about- performance on the entrance examination (passed

or not passed) (r = -.32, p<.05, N = 31).

-Reporting "no feeling" about performance on the

entrance examination combined with.the discrepancy from the required passing score (r = -.45, p<.05,

N = 19).

-Reporting "no feeling" about performance on' the

entrance examination combined with the importance of being admitted to the program (r = .71, p<.Ol,

N 19).

-Reporting -f'eeling the entrance examination not passed combined with the. discrepancy from the

required passing score (r = -.60, p<.Ol, N = 20).

-Having seen Counselor #2 combined with the discrepancy from the requited passing score (r = -.73,

p(.01, N = 19).





83



-Having seen Counselor #4 combined with the importance of becoming admitted to the program (r = .48, p (.05,

N = 20).


The four interactions with respect to the subject's

choice of counselor for some future occasion were as follows:

-Reporting "no feeling" about performance on the

entrance examination combined with the importance

of becoming admitted to the program (r = -.75,

p <.01, N = 19).

-Having seen Counselor #2 combined with the subject's

age (r = -.57, p( .01, N = 19).

-Having seen Counselor #4 combined with the subjectcounselor compatibility (r = -.53, p( .01, N = 20).

-Having seen Counselor #2 combined with the discrepaicy from the required passing score (r = .43,

p< .05, N = 19).

Interaction terms consistent with these interactions of independent variables were computed and were used as additional independent variables in the analyses of data with respect to the dependent variables. Hypothesis 1

Clients in higher compatibility client-counselor dyads will show greater reduction in anxiety over

the course of a stressful interview than will

clients in lower client-counselor compatibility

dyads.





84


Procedure. A forward stepwise multiple regression procedure was performed with the subject's anxiety change as the dependent variable. The independent variables were as follows:

1. Subject-counselor compatibility

2. Age of subject

3. Being black

4. Years of schooling

5. Importance of being admitted

6. Reporting "no feeling" about examination per. formance

7.. Reporting feeling entrance examination not passed

8. Discrepancy from passing score

9. Length of interview.

10. Reporting "no feeling" about examination performance

- combined with discrepancy from required score

11. Reporting "no feeling" about examination performance

combined with importance of being admitted

12. Saw Counselor #3 combined with length of interview

13. Saw Counselor #5 combined with subject-counselbr

compatibility

14. Saw counselor #5 combined with importance of

being admitted.


Findings. Table 20 presents a summary of the results of the regression- analysis with respect to anxiety reduction scores as'the dependent variable. The table shows only the first six steps of the analysis as.the significance of the






TABLE 20


Summary - Multiple Correlation. with Anxiety Reduction


Entry Independent Multiple Add to Total % Add % Partial df F
on Variable.. . R R Variance Variance r before and
Step entry sig


1 Importance of
being admitted .216


2 Subject-counselor
compatibility .264


3 Lentgth of
interview .307


4 "No feeling" abbut
test performance with
discrepancy from
passing score ,331


5 "No feeling" about
test performance .347


6 Being black .368-


.216



.048 .039





.024 .016


.021


4.68



6.96



9.40





10.99



12.06


13.54


4.68 2.28 2.08 1.59 1.07


1.48


- 1/92 4.520
(p( .05)


+.15 +.16 +.13



-.11


2/91 3.403
(p ( .05)


3/90 3.114
(p< .05)




4/89 2.747
(p ( .05)


5/88 2.414
(p ( .05


-.13 6/87 2.272
(p( .05)





86


multiple correlation coefficient did not reach the .05 level when the seventh independent variable was entered into the analysis. Inspection of Table 20 reveals that the single best predictor of anxiety reduction was the importance the subject placed on being admitted to the program. It accounted for approximately 4.7 percent of the variance and created a multiple correlation coefficient of ..216. Once the importance of being admitted was considered, the next most important factor realted to subject anxiety reduction was the subjectcounselor compatibility which contributed '.048 to the multiple correlation and accounted for approximately 2.3 percent more of the variance in the anxiety reduction scores of the subjects. Subjects in higher compatibility dyads did show greater anxiety reduction over the course of the interview than did subjects -in lower compatibility dyads. The subject-counselor compatibility was not, however, the most important variable'with respect to anxiety reduction and contributed only a small amount to subjects' anxiety reduction. The Pearson product moment correlation between subject-counselor compatibility and anxiety reduction for all dyads was found to be .18 (p( .05). It should be noted that subject-counselor compatibility combined with having seen Counselor #5 met a criterion for becoming an interaction term with respect to anxiety reduction. It was possible that the strength of the relationship between compatibility and anxiety reduction in the case of one counselor (r = .43, p 4.05) was strong enough to produce the appearance of a








general effect. That interadtion term did not enter the multiple regression analysis while the multiple correlation coefficient was significant or prior to the entry of subjectcounselor compatibility as a general effect. The general effect of subject-counselor compatibility would appear, therefore, to be stronger with respect -to anxiety reduction over the course of the interview.

The hypothesis that subjects in higher compatibility dyads did show greater anxiety reduction over the course of the interview than subjects in lower compatibility dyads was upheld.

Discussion. The other four variables which entered the regression analysis before the multiple correlation ceased to be significant are also shown-in order of entry in Table 20. As can be seen, all six variables accounted for only about 13.5 percent of the-variance of the anxiety reduction scores and the last two to enter the regression model (reporting "no feeling" about test performance and being black) were negatively related to- anxiety reduction as they entered the multiple regression correlation analysis. This should not be interpreted as suggestive of. withholding counseling from.black female applicants who do not meet entrance examination requirements or from female applicants not meeting entrance examination requirements who do not feel that they either passed or did not pass the examination. It does suggest that with those subjects there was an increase in anxiety over the course of the interview and that other





88


methodologies might be more appropriate when anxiety reduction is the goal.

Two of the six variables related to anxiety reduction involved the subject's indicating "no feeling" about her performance on the entrance examination. When combined

with greater discrepancy from the required passing score on the entrance examination there was a slight positive relationship with anxiety reduction, but when considered alone the relationship was negative though still small. This suggests that where the subjects reported that they did not have an internal feeling about having passed or not passed the examination and had not passed the examination by greater amounts they experienced some anxiety reduction during the interview. 'When subjects reporting "no feeling" about entrance examination performance are considered on that

basis alone, however, there was a slight increase in anxiety over the course of the interview. It appears that anxiety increased under the stress of discovering that the entrance examination had not been passed for subjects as a whole but decreased slightly -for subjects who had not passed by wider margins.

That the length of the interview should have a positive effect on anxiety reduction was not surprising. The more time spent in a supportive.relationship where subjects discovered they had not met entrance examination requirements did promote a reduction in anxiety over the course of that interview.




Full Text

PAGE 1

COUNSELING WITH REJECTED NURSING SCHOOL APPLICANT SOME CORRELATES OF SELECTED OUTCOME MEASURES DOUGLAS G. JOHNSON A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL 0] THE UNIVERSITV Or FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIRE^'lENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY' FLORIDA 1977

PAGE 2

TO GRACE

PAGE 3

ACKNOWLEDGMENTS This dissert?>.tion has resulted from the cooperative efforts of a lar^^e number of people. For their cooperation and assistance, I would like to recognize the following: -Dr. Ted Landsman, committee chairman and pacemaker. -Dr. Richard Anderson, committee member. -Dr. A. Garr Cranney , committee member. -Dr. Robert Soar, design and statistical consultant. -Dr. David Lane and Dr. Bert Sharp, former comimittee chairmen who were unable to continue service due to retiremiont and position change. -Santa Fe Community College. -The subjects who participated in the study. -Mr. George Huber, Mr. Tom McCul lough, Mr. Don Mott , Mr. Bernie Murphy, and Mr. Gene Wigington , the participatin counselors, who made special scheduL^ arrangements to be available for this study. -Miss Lillie Covert who made all of the sub j ect -counselor appointments in addition to her normal duties. -Mr. Mike Conlon, comiputer analysis consultant. -Judy Johnson for editorial assistance, manuscript preparation, patience and encouragement which were central to the completion of the study. -Alison Johnson and Meredith Johnson for understanding cooperation, and independence necessary to allow their iii

PAGE 4

father to devote the necessary time to the proje -Mr. John Dumbauld for continuing to be my iv

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMEN TS i i i LISTOF TABLES ^ vii LIST OF FIGURES ABSTRACT ^ CHAPTER I INTRODUCTION 1 Purpose of the Study 3 Setting for the Study 4 Need for zhe Study • ^ CHAPTER II • REVIEW OF THE LITERATURE 7 Facilitative Therapeutic Conditions 7 Facially Mixed Counseling Dyads 9 State Anxiety H Client and Counselor Relationships 13 Initial Interview or Short Term Counseling 13 Client Counselor Compatibility 14 The Concept of Compatibility _ 16 Research Using Schutz's Concept and indices of Compatibility 18 Summary 19 CHAPTER III PURPOSE AKD DESIGN 21 Definitions 21 Hypotheses 21 Hypothesis 1 22 Hypothesis 2 . 22 Hypothesis 3 22 General Design 22 Subjects 2o Counselors 25 Instrumentation 26 Fundamental Interpersonal Relationship Orientation Behavior 26 Importance of Becoming Admitted to the Nursing Program 28 Subject's Feeling about Examination Performance 29 The Affect Adjective Check List 32 Interview Rating Scale 35 Preference for Counselor Indicator 35 V

PAGE 6

Counselor's Level of Facilitative Conditions Barrott-Lennard Relationship Inventory Specific Design Consideration Precautions Procedures Counselor Training Initial Data Collection The Study Interviews Counseling Interview CHAPTER IV RESULTS AND DISCUSSION Confidence Level Parameters Control Study for Facilitative Conditions Generally Offered by Participating Counselors Subject Pool Independent Variables SulD.jects Sub.j ect -Counselor Conipat ibility Length of the InterviewDependent Variables Anxiety Change Rating of the Interview Choice of Counselor for Future Occasion Testing of the Hypotheses Preliminary Analysis of the Data Hypothesis 1 Hypothesis 2 Hypothesis ' 3 CHAPTER V SUMMARY AND CONCLUSIONS Summary Conclusions APPENDICES REFERENCES BIOGRAPHICAL SKETCH vi

PAGE 7

LIST OF TABLES Table 1 Facilitative Conditions Generally Offered to Female Client in Separate Control Study 2 Control Study Client Ages 3 Length of Control Study Interviews 4 Control Study Racial Group Membership 5 Comparison of Participating Subjects and Non-Participating Subject Pool Mem.bers 6 Comparison of Participating Subjects and Non-Participating Subject Pool Members 7 Age of Subjects 8 Subject Racial Group Membership 9 Number of Years of Schooling Completed by Subjects 10 Importance Placed on Becoming Admitted to Program 11 Subject Feeling About Entrance Test Performance 12 Discrepancy from Passing Score on Entrance Examinat ion 13 Subject-Counselor Com.pat ibi lity 14 Length of Interview 15 First Anxiety Measure 16 Second Anxiety Measure 17 Anxiety Change 18 Interview Rating Scale 19 Preference for Particular Counselor for Future Occasion vi i

PAGE 8

Table Page 20 Summary Multiple Correlation with Anxiety Reduction 85 21 Summary Multiple Coi'relation with Rating of Interview 93 22 Categorical Weight Coefficients 103 vi ii

PAGE 9

LIST OF FIGURES Figure Pa-g 1 Level of Importance Scale 30 2 Subject Feeling about Entrance Examination Performance 31 3 Opinion of Future Counselor Choice 36 ix

PAGE 10

Abstract of Dissertation Tresented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education COUNSELING WITH REJECTED NURSING SCHOOL APPLICANTS: SOME CORRELATES OF SELECTED OUTCOME MEASURES By Douglas G. Johnson June . 1977 Chairman: Theodore Landsman Major Department: Counselor Education Five white male counselors engaged in a total of 95 interviews during which a female applicant to an Associate Degree Nursing Program was informed she had not passed an entrance examination to f4ualiiy for admission to that program. The framework for the study was a theory of interpersonal needs (Schutz, 1960) with the primary focus being the effect of reciprocal com.patibility in the area of affection on the outcome of these stressful interviews. Subject anxiety change over the course of the interview, a rating of the interview by the subject, and an indication of the subject's feeling about seeing the counselor at some future occasion were selected as measures of outcome. The study was done as field research with other measurable factors used as additional independent variables. A system for random assignment of subjects to counselors was used. A basic interview format was used with each counselor allowed to operate according to his usual manner or style.

PAGE 11

Using a multiple regression analysis procedure, subjectcounselor compatibility was found to be related to anxiety reduction at a lev/ level. The compatibility accounted for only about two percent of the anxiety reduction variance and was less related than the importance the subject placed on " being admitted to the progra.m. Only about 13 percent of the anxiety reduction variance was accounted for at a significant level by a combination of variables. A multiple regression analysis was done v.'ith respect to the subjects' ratings of the interviews. All the variables includinga variety of combination variables indicative of specific effects based on categorical subgroup were included in the analysis with the multiple correlation coefficient remaining significant. The major finding about the ratings of the interviews was that over 50 percent of the \'a.riance was associated with \-ariables which were related to lower ratings of the interviews. Subj ect -counselor compatibility was not supported as related to better or worse ratings of the interviews. A discriminant analysis procedure was done to determine if subject-counselor com.patibility was related to subjects indicating a preference to return to the same counselor on a future occasion. The hypothesis that compatibility would be an indicator of an expression to that effect by subjects was rejected. The distribution of actual responses was so heavily weighted in favor of subjects indicating a preference to return to the same counselor that any conclusions would be

PAGE 12

aighly suspect. The overwhelming tendency of subjects to indicate a preference to see the same counselor on some future counseling occasion was not explained by the results. A subject's perception of failure as either a "personal" problem or an •educat iona 1 -vocat lonal" problem in relation to a particular counselor's style or manner was discussed as appropriate for further investigation. The hypotheses that higher levels of subject-counselor compatibility would be positively related to higher levels of anxiety reduction, better ratings of the interviews, and a showing of preference to return to the same counselor in the future were upheld at a low level in the first case and not upheld in the other two instances. The hope that the reciprocal subject-counselor compatibility in the area of affection proposed by Schutz (1960), which could be easily used as a possible basis for client-to-counselor assignment might be utiii2:ed , was not realized. The concept of compatibility as used in this study seemed to shed little light on the complex relationship between subject and counselor. xii

PAGE 13

CHAPTER I INTRODUCTION The premise that the nature of the interaction of client and counselor personal characteristics is of importance and is in need of further investigation has received considerable support (Ferreira, 1964; Garfield. 1973; Xiesler, 1966; Lorion , 1974; Sheiner. 1967; and Strupp , 1962, 1973a, 1973b). Several writers have attempted to define some elements of the interaction between counselor and client. Patterson (1967) stated, "It would appear that the relationship [of counselor with client ] is an important factor in any interpersonal interaction and therefore is basic to counseling or psychotherapy" (p. 85). Friedman (1960 ) drawing on the ideas of .Martin Eubor , made the point that even though psychotherapy is one-sided, it ". . .is still an 'I-Thou' relation founded on mutuality, trust and a partnership in a common situation ..." (p. 31). Bordin (1955) encouraged counselors to be friendly, accepting, and willing to work with clients where they are (emphasis added) Hansen (1963) used the terra raoport to describe the appropriate interpersonal relationship between client and counselor. Shoben (1954) suggests that conditions which " . . . mediate affective security . . ." (p. 47) are the first order of business in counseling. It is widely accepted that counselors should possess a genuine, positive feeling toward their clients as specific 1

PAGE 14

9 human beings. This concern is, in fact, a part of the definition of Unconditional Positive Regard (later called Non-Possessive Warmth) as originally posited by Rogers in his statement about the necessary and sufficient conditions of therapeutic personality change (1957). Non-Possessive Warmth has more recently emerged as one condition which seems to be highly dependent on both the client and the therapist in the initial phase of therapy (Truax, Wargo , Frank, Imber, Battle, Hoehn-Savic. Nash, & Stone, 1966). The inference can be made that the level of warmth is a function of both the client and the counselor, at least in initial phases of counseling or therapy. A situation in which the client's and counselor's personal characteristics are at some optimal level, or are in balance, or are compatible, with respect to one another could be an important factor with regard to the nature and outcome of the counseling relationship. Lorion (1974) urged delineation of characteristics which relate to " . . . positive outcome with any social or ethnic group" (p. 352). Beutler (1973) suggests that the patient-therapist dyaa be considered ". . . as a separate treatment whose effectiv( uess depends not only on the individual ch arac cerist ics of patients and therapists bur upon characteristics of their mutual com.patibility" (p. 305). It would seem then that compatibility and some mutuality of warmth, at least. in the initial stages of

PAGE 15

3 counseling, mig-ht be beneficial to the relationship and thereby t.o the outcome of the counseling which occurs within that relationship. Schutz (1960) gave the term compatibility a very specific explanation, stating that "Compatibility is a property of a relation between two or more persons . . . that leads to mutual satisfaction of interpersonal needs and harmonious coexistence" (p. 105). Compatibility can be measured for any pair of persons based on a star.dardized instrument called the Fundamental Interpersonal Relationship Orientation Behavior, devised by Schutz (1980), and the use of a specific formula for comparing scores obtained on that instrument by the two people. As the warmth of the relationship between client and counselor is of importance, it was Schutz 's formula fo)compatibility whicli was used in this study. Purpose of the Study It was the purpose of this study to examine some of the effects of c 1 lent -counse lor com.patibility on clients in an initial one-to-one, educational-vocational counseling interview during which a female applicant to a registered' nursing program is told she is not admissible to that program, based on her earned nursing program entrance test scores. Specifically the study focused on the client's state anxiety reduction over the course of the intervew. her evaluation of the counseling interview, and her

PAGE 16

4 reported seleclion of a counselor should she decide to seek out a counselor again. Setting Tor the Study Female ai^plicants to the Santa Fe Community College Associate of Science Registered Nursing program not achieving a grade level score of 12.0 on both the math and reading sections of the Comprehensive Test of Basic Skills were requested to meet with a professional counselor. The purposes of the counseling interview were as follows: -To inform the applicant that she had not scored high enough on the Comprehensive Test of Basic Skills to enter the Registered Nursing program for which she had applied. -To discuss with the applicant the implications of her scores for possible future a.dmission to the program. -To discuss with the applicant her educationalvocational goals in light of the current situation. -To discuss with the applicant any other matters which seemed important within this general context of not meeting admission requirements for the Registered Nursing program. It was intended that through this process the applicant would be assisted to deal with her confrontation with failure in a constructive and positive v.-ay . Each counselor involved in the study conducted the interview in his own manner and style. Some elements were

PAGE 17

5 added to the normal overall testing and counseling process in order to gather data pertinent to the completion of this study. A complete description of the procedures can be found beginning on page 40. Need for the Study Brough ' s study in 1965 suggests that the single most important source of perceptions about counselors among high school students is actual experience in counseling. Assuming the same to be true for potential college students, being able to assign prospectiv^e college students to counselors with whom they are interpersonal ly compatible would seem to merit consideration, especially if the prospective students must be informed that they are not admissible to a particular program due to entrance scores. Blocher (1967) supported the idea that matching client and counselors in a variety of ways may be an important issue relating to the outccm.e of counseling. Graff, Rogi;e, and Danish (1974) suggested that there is currently a specific need to do research which can help to understand the outcomes of educat ional -vocational counseling, which both high school and community college counselors deal with frequently. The issue of c 1 lent -counse lor compatibility was the subject of this study. The concept of compatibility as presented by Schutz (1960) addresses itself to the specific relationship between two persons. Schutz 's index of reciprocal

PAGE 18

6 compatibility was designed to be an indicator of the likelihood of two persons expressed and wanted needs being co:Tip leraent ary . The research undertaken in this study examined the effect of the cor.ipat ibil ity of a client-counselor dyad on specific counseling outcomes, The outcomes were client anxiety reduction, client rating of the counseling interview, and the client's expressed preference to see the same counselor on another occasion. It was posited that greater compatibility between the client and the counselor would be related to greater anxiety reduction in the client, a better rating of the interview by the client, and grea.ter likelihood that the client would prefer to return to the same counselor on a future' occasion . Positive findings would suggest that client-counselor compatibility would be useful as a method for assignment of clients and counselors in better than a random fashion.

PAGE 19

I CHAPTER II REVIEV; OF THE LITERATURE That the relationship between the client and the counselor is of importance was presented as the major focus of thi.s study in Chapter I. The review of the literature deals specifically with issues of importance to this study which are of an a priori nature, with a review of the research on client-counselor relationships, and with a specific review of the research utilizirxg Schutz's (1960) postulate of compatibility. Facilitative Therapeutic Condi tions The research dealing with facilitative therapeutic conditions based on Rogers' (1957) idea of the conditions necessary for personal grov/th in therapy is massive. Tlie presence of high levels of facilitative conditions related to positive outcomes in therapy and low levels of facilitative conditions related to negative outcomes in therapy was adeptly shown by Truax (1966). These conditions were also shov n to be largely the contribution of the therapist (Truax et al . , 1966) . Though originally intended as indices related to positive outcom.e in therapy with a client centered orientation, facilitative conditions have been shown to have an 7

PAGE 20

s effect on the progress of verbal conditioning; even using the inexperienced counselors v.'ith a behavioristic orientation (Mickelson S: Stevic, 1971), which seems to show their relevance in a greater range of situations than Feidler's (1950) finding that experience may be more crucial than theoretical orientation or technique. Rogers (1957) stressed the importance of the patient's awareness of facilitative conditions. Truax (1966) pointed out, however, that interpersonal perceptions are not likely to be accurate in most patient populations, and may not, therefore, be as reflective of some "actual" level of presence of the conditions. Barrett -Lennard (1962) presented an instrument to measure the level of facilitative conditions from the perspective of the client. He considered this to be the intent of Rogers' thinking and built a "Relationship Inventory" which would give five scores; Empathic Understanding. Level of Regard, Unconditionality of Regard, Congruence, and -Vi 1 lingness to be Known. Though the matter of facilitative conditions is clearly of importance, the scales as developed by Truax (1961, 1962a, i962b) were designed for use by independent raters working from typescripts or tape recordings of psychotherapy interviews with disturbed patients. The conditions under which the interviews used as a context for the present study occurred were expected to be stressful

PAGE 21

9 and/or ego threatening. Recording the interview would have introduced additional design problems (Campbell Sc Stanley, 1963) and recording of interviews has been shown to have at least some effect on client functioning (Gelso, 1973). The Barrett-Lennard Relationship Inventory and measurements of facilitative conditions obtained from tape recordings have been shown zo be related to outcome in therapy (Barrett-Lennard, 1962; Truax, 19G6). It was decided, therefore, that a measurement of the ability to provide facilitative conditions from the i^oint of view of the client, but independent of the research interviews, be obtained for each of the counselors who participated in this study. As research with the instrument has shown the Relationship Inventory to be most parsimonious when used as a total measure (Mills & Zytowski , 1967; Lanning &: Lemons, 1974), total scores were used in this control phase. Characteristics of the Barrett-Lennard Relationship Inventory and specific treatment of these data as a control measure are discussed in the section on preliminary procedures in Chapter III. Racially Mixed Counseling D yads A preliminary inspection of the situation in which this study was conducted suggested a subject pool with at least a 50:50 black-white ratio, with a possibly much higher black-white ratio, potentially as hign as 85:15. V/hen given a choice of a counselor similar to or different

PAGE 22

10 from the client on racial and cultural bases, Stranges & Riccio (1970) found that preferences were strongly in favor of similarity of counselor chosen to the choosing cl lent . uStudying conditions of race, sex, and age mix of clients and counselors in a survey followup of vocational counseling with lower socio-economic group clients. Beckner (1970) got mixed results with the most pointed finding being that while 25 percent of the client respondents said that racial and ethnic backgrounds were of importance, 68 percent said those issues were not of importance. Russell (1970) stated that blacks were no different from whites in terms of feelings, wishes, hopes, fears and frustrations, yet several writers asserted that a white counselor just cannot hope to help a black person as they are so culturally different (Thomas, 1969; Williams & Kirkland, 1971). V.'alkin , ?,:oriwaki , & Williams (1973) noted that, "The conditions which foster trust at the beginning of a relationship must be explored and then exploited to improve the effectiveness of therapy" (p. 316). They noted that blacks seem to be more reserved and recommended more black therapists. The whole problem in this area seems to have been touched upon experimentally by Ewing (1974) in which neither favorability of rating nor effectiveness was shown to be significantly related to the race of the counselor, except that white clients did tend to rate black counselors

PAGE 23

11 as less helpful in initial interviews in which precollege students were being seen by experienced counselors. The review by Dreger Miller (1963) suggests that lumping blacks into one category is similar to lumping any large group of people into one category, with most^ such categories standing a very good chance of not really being homogeneous along a number of dimensions. State Anxiety Being told that her entrance test scores are not high enough to allow her to be admitted to a Registered Mursing program would be expected to raise the level of anxiety in a subject. This is a situation analogous to a task performance (digit sj-mbol) feedback experiment in which subjects who were informed they were doing poorly showed a significant rise in state anxiety (Fremont, Means, & .Means, 1970). The general theory of interpersonal relationship needs posed by Schutz (1960) states: A discrepancy between the satisfaction of an interpersonal need and the present state or the organism engenders a feeling in the organism that sha.ll be called anxiety (p. 16). The effects of failure stress on anxiety (Hodges & Spielberger, 1969) and poor interpersonal compatibility based on not having interpersonal needs well met would certainly seem of importance in this ..ludy. The Affect Adjective Check List (Zuckerman, 1960) has been successfully used to demonstrate changes in state anxiety associated with information that subjects were

PAGE 24

12 » doing poorly at digit span tests, noted above as failure stress. It has been shown zo be sensitive when used over short time periods (Hodges & Spielberger, 1969; Zuckerman, Lubin, Vogel, & Valerius. 1964), and can be experimentally calibrated when used as a dependent variable (Zuckerman et al., 1964).Of particular interest with respect to this study were the results obtained with sophomore nursing students in a course critical for continuance in that program (Hayes, 1966). The nonexaminat ion day mean scores on the Affect Adjective Check List are not much higher than those presented by Zuckerman (1960), but the examination day mean scores were higher. Mayes suggested that this may have been due to the necessity of passing the course in order for the students to continue in the program, which is similar to the conditions under which the present study was conducted. This finding seems consistent with and supportive of the results obtained with students' reporting differing levels of examination worry and the sensitivity of the Affect Adjective Check List to such personal states (Zuckerman & Biase, 1962). Ten subjects rating themselves "low" in examination worry showed a differential from the class m.ean of -1.42 on examination days; 15 students rating themselves "moderate" in examination worry showed a mean change of +1.81 on examination days; and seven subjects rating themselves "high" in examination worry showed a mean change of +4.23 on examination days (F =^ 5.71, p< .01).

PAGE 25

13 Spielberger (1972) noted, "... the usefulness of the Affect Adjective Check List as a measure of day-to-day fluctuation in state anxiety has been clearly demonstrated" (p. 35). Client and Counselor Relations hips Initial Interview or Short Term Counseling A variety of research on first or initial interviews is available in the literature. Greenberg (1969), in an analogue study, noted that just telling a subject that the therapist was warm or cold combined with information about his being experienced or inexperienced had the effect of having a "warm" therapist seen as more attractive, more effective, and influential over the subject than a "cold" therapist . Inexperienced counselors were better liked than experienced counselors, controlling for facilitative conditions, in another analogue study of initial interviews (Pope, Nudler, Vonkorff, & McGhee , 1974). The explanation was perceived similarity of subjects to analogue therapists, both subjects and analogue counselors being of similar age and class (college students). Pope'S: Siegraan (1962, 1965, & 1968) and Pope, Siegman & Blass (1970) discovered that initial interview verbal behavior varies with changes in the specificity of therapist remarks, the warmth or coldness of the interviewer, and induced ego threat anxiety (in this case inferred poor family adjystment ) .

PAGE 26

14 Grater (1964) discovered that, clients presenting themselves to a college counseling center showed more willingness to discuss personal-social concerns with counselors who were considered to be more "affective" in approach and manner while fewer clients who met with more "cognitive" counselors did so, preferring instead to discuss educational-vocational matters, The "compatibility" or preference for style was not overwhelming but present nonetheless. In a study of neurotic versus normal students' levels of self-disclosure during a first short telephone conversation with either a high or low self-disclosing experimenter as the other party, normals, presumably m.oi-e similar to and compatible with the experimenter, reciprocated levels of self-disclosure while neurotic students' levels of self-disclosure remained in the middle range and did not fluctuate much (Chaikin, Derlaga, Bayma , k Shaw, 1975). A study of three groups of students grouped for high, medium, and low similarity to the counselor on the Allport Vernon Lindzey Study of Values showed that the middle similarity group changed more in the mieaning of two concepts than either the high similarity or low similarity groups (Cook, 1966). CI i en t-Cuun se 1 or Compa t ib i 1 i ty Psychoanalysts, who have long considered the "relationship" between patient and therapist to be crucial, are

PAGE 27

15 reporting that therapists with particular personality dynamics prefer patients with certain diagnoses and have ' differentially successful results depending upon the "compatibility" of the therapist's and patient's personality dynamics (Reimann, 1968). Similarity with respect to personality "type" as measured by the Myers Briggs Type Indicator (MBTI) has been shown to be related to longevity in counseling (Mendelsohn S: Geller, 1963; Mendelsohn, 1966). Middle ranges of similarity of personality type (.MBTI) have been discovered to be related to a better evaluation of the client-therapist contact than were measures of very high or very low similarity (Mendelsohn & Geller, 1965). High client-counselor similarity levels have also been related to initial interpersonal attraction ' (Izard, 1960, 1961). The classification of therapists as either "A"' or "B" type therapists (Whitehorn & Betz-, 1960) with respect to differential outcome with schizophrenic patients marked the beginning of an avalanche of studies using that classification system as a theoretical base. The controversial nature of that research is well demonstrated in a short article by Eysenck and a reply by James & Foreman (Eysenck, 1975). Recent research has begun to include classifications of patients along the same dimensions (Berzins. Friedman, & Seidinan, 1969) in therapy analogue, A fitting summary statement with respect to the entire matter of inferred compatibility or matching was made by

PAGE 28

16 Beutler (1973) while writing about both the learning theory issues and the "A" or "B" type therapist issues as they pertain to learning theory: The alternative ... is to consider each patient-therapist pair as a separate treatment whose effectiveness, depends not only upon the individual characteristics of patients and therapists but upon characteristics of their mutual compatibility (p. 305). The concept of reciprocal compatibility proposed by Schutz (1960) meets Beutler 's criterion both in theory and as a usable numeric index. The Concept of Compatibility • Schutz (1960) presented a s'pecific definition of the nature of interpersonal compatibility as a part of a broader theory of interpersonal interaction based on a postulated set of interpersonal needs. The basic postulates of the theory were; 1. Every person has three interpersonal needs: inclusion, control, and affection. 2. Inclusion, control, and affection constitute a sufficient set of areas of interpersonal behavior for the prediction and explanation of interpersonal phenomena (p. 13). The postulate of compatibility is stated. "If the compatibility of one group, h, is greater than that of another group, .m , then the goal achievement of h will exceed that of m" (p. 105). A theorem of dyadic preference is also posed:

PAGE 29

17 If the compatibility of one dyad, y-j^ , is greater than the compatibility of another dyad. y2 , then the members of yi are more likely to prefer each other for continued personal contact (p. 120). Ke further stated: For the dyad, this theory means that a person wants to act a certain way toward others and wants to be acted toward in a certain way .... By comparing A's description of how he likes to be acted toward with B's description of how he likes to act toward people, and vice versa, a measure of mutual need satisfaction emerges (p . 107) . Reciprocal compatibility is the specific term used to describe this status. The measure (index) of reciprocal compatibility of persons i and j is expressed by the formula . rKij ei wj + | e j wi in which ei and ej are the measures of how i and j typically behave toward others, and wi and wj are the measures of how i and j typically want others to behave toward them. The value of rKij . when computations are used based on the instrument created by Schutz to scale the postulated needs (pp. 61-65), becomes an index of the reciprocal compatibility for the dyad, ij . The value of the index can only be positive due to the use of absolute values in the final summation and the value can range from 0 to IS. The degree of compatibility is highest when the index number is 0 and lowest when the index number is IS. A conversion procedure using 18 minus the actual index level computed for a dyad

PAGE 30

18 can be used to produce an ascending scale of compatibility indices (Schutz, 19G0). Research Using Schutz's Concept and Indices of Compatibility Two person compatibility based on the use of the Fundamental Interpersonal Relationship Orientation Behavior formulations has been demonstrated to be related to interpersonal relationships in several ways. Schutz (1960) showed that while charce would have any two members of a fraternity select each other for roommates and for reciprocal compatibility to occur in about 13 percent of the cases, 31.3 percent actually m.ade such choices (p <( .001 for difference) (p. 124). Gassner (1970) showed that personal reactions toward therapists were positively related to reciprocal compatibility in a clinical setting in which juvenile delinquents and pastors formed the dyads. Verbal conditioning under either the ''set" of interpersonal compatibility or measured total compatibility, using a broader range of formulae with rk as one of them, showed that the effect of the conditioning was present during the presence of the experimenter when compatibility was high or was withheld until after the experimenter left the room when the compatibility was low (Sapolsky. 1960). In other research the Pearson correlation between resident therapist/patient total compaLibility and supervisor's rating of patient improvement was shown to be .45 (p< .05) (Sapolsky, 1965).

PAGE 31

19 Mendelsohn & Rankin (1969) in studying clients' percept jLons of the relationship and evaluations of both the counselor and the usefulness of counseling where N = 115 over an eight session therapy or counseling regimen, discovered that compatibility was differentially related to various outcome measures for male and female clients. The Pearson product moment correlation between general evaluation of counseling and compatibility was .34 (pC.05) thus showing reciprocal compatibility in the affection area to be negatively correlated with general evaluation of counseling by clients. Their discussion points to the possibility ". . . that factors which foster strong emotional attachments in typical social relations can, unless carefully handled, lead to an excessive personalization of the special relationship which characterizes counseling and psychotherapy" (p. 163).^ Summary Various compatibility indices seem to produce differing correlations with a variety of outcome criteria for counseling or therapeutic dyadic interpersonal interaction. The emphasis placed on the importance of warmth in the beginning phase of therapy (Shoben, 1954; Truax , 1965) and the conflicting results of Sapolsky (1965) and Mendelsohn &; Rankin (1969) are adequately smnmarized by Sapolsky ( 1965) when writing of Schutz's theory and instrumentation, he states, . .an important underlying personality variable

PAGE 32

20 conti-ibuting to the establishment of 'good' therapeutic relationships . . . may have been identified" (p. 75). Schutz's unique theory provides just such a framework within which both client and counselor personality variables in identifiable areas can be viewed. His postulates, formulae, and indices provide the tools, based on and integrated with that theory, which prompted the writer to propose that the Fundamental Interpersonal Relationship Orientation Behavior theoretical framework should be utilized to examine a ''typical" counseling situation using reciprocal compatibility in the area of affection as the primary treatment varia'ble of interest .

PAGE 33

CHAPTER III PURPOSE AND DESIGN Within the stressful setting in which a counselor informs a client, on the basis of entrance examination scores, she has not qualified for admission into an Associate of Science Registered Nursing program for which she has applied, this study addressed three major questions 1. During the interview does the level of mutual compatibility of the client-counselor dyad have an effect on the anxiety level of the client? 2. Does the level of mutual compatibility of the client-counselor dyad have an effect on the client's evalua tion of the interview? 3. Does the level of mutual compatibility of the client-counselor dyad have an effect on the client's indication of preference for selection of a counselor on another occasion? Definitions Dyad: a meaningful dialogue, encounter, or relationship between two people. In this study the client-counselor dyad is defined as the client and counselor when they are meeting in a counseling interview. Compatibility : the ability to exist together; the ability to get on well together. The level of mutual 21

PAGE 34

compatibility of the client-counselor dyad in this study is represented by an index number. The lowest index number, 0, indicates a minimum level of compatibility while an index number of IS indicates a maximum level of compatibility. Hypotheses Three m.ajor hypotheses were tested by the research proj ect . Hypothesis 1 Clients in higher corripat ibil ity client-counselor dyads will show greater reduction in anxiety over the course of the interview than will clients in lower compatibility client-counselor dyads. Hypothesis 2 Clients in higher compatibility client-counselor dyads will rate the counseling session m.ore favorably than will clients in lower compatibility clientcounselor dyads. Hypothesis 3 Clients in higher com.pat ibility client-counselor dyads will indicate a preference for seeing the same counselor again significantly more often than will clients in lower client-counselor compatibility dyads . General Design The research was designed to study the effect of client-counselor compatibility on selected outcomes of

PAGE 35

23 a stressful counseling interview. The outcomes selected for examination were the client's reduction in anxiety over the course of the interview, the rating of the interview by the client, and the client's expressed preference with respect to seeing the same counselor should counseling be considered appropriate in the future. Various procedures required for data collection were introduced into an ongoing series of occurrences vdiich ' constitute the application sequence for the Santa Fe Community College Registered Nursing program. The data collection procedures were introduced into the achievement testing and the counseling with rejected female applicants segments of the sequence. Hypothesis 1 was tested through the use of a "One Group Pretest-Posttest Design" (Campbell Si Stanley, 1963). Hypotheses 2 and 3 were tested through the use of a "Static Group Comparison" design, except that rather than comparing a group which received no treatment with a group which did receive treatment as the design provides, this study compared differing levels of the treatment (compatibility) (Campbell & Stanley, 1963). Due to the relatively large number of independent variables necessary to provide adequate control for unknown sources of variance in the outcome measures, a statistical procedure which could show the relationships of a large number of independent variables with the outcome measures was necessary. Walberg (1971) encouraged the use of general

PAGE 36

24 regression models when analyzing outcome measures to test hypotheses. He particularly emphasized their efficacy when large numbers of independent variables were used and when the data were primarily of a continuous nature. Cohen (1968) suggested the appropriateness of multiple regression analysis as an appropriate tool in such cases. He also . confirmed the robust nature of the regression procedure to analyze data which reflect membership in a group or category, e.g., race, sex, etc. , when it has been properly coded and thereby can be trea.ted as continuous. The fact that the outcome measures, anxiety reduction and rating of the interview were continuous and reasonably normally distributed permitted the selection of multiple regression analysis as an analytic tool to test Hypotheses 1 and 2. The fact that the outcome measure of preference for seeing the same counselor on some future occasion should counseling seem appropriate allowed only one of three choices made the use of some other data analytic tool necessary. Discriminant analysis, as available in the Statistical Analytic Systems, Inc. package (Barr, Goodnight, Sail, & Helwig, 1976) provides weight coefficients for each independent variable with respect to each of the possible indications on an outcome measure. The relationship of a particular independent variable to a given response on an outcome measure can thereby be determined. As the weight coefficients could be interpreted to test Hypothesis 3, discriminant analysis was selected as the

PAGE 37

25 data analytic, tool to test Hypothesis 3. Sub j acts The pool of subjects for this study was the female applicants to the Associate of Science Registered Nursing Program at Santa Fe Community College who did not reach the required admissions scores of the 12.0 grade level on both the math and reading sections of the Comprehensive Test of Basic Skills and who were at least 18 years old or had parental permission to participate. The sample consisted of all such cases beginning with those taking the Comprehensive Test of Basic Skills Test on or after August 19, 1975 and ending when 95 subjects had completed all phases of the experimental procedures. As males had historically represented less than ten percent of the applicants . to the program not meeting minimum admissions scores, it was decided that they would not be considered within the scope of this study. Counselors Counselors were selected to present characteristics which would replicate the normally occurring conditions of the study interviews as created by the staffingpatterns of the college. The five counselors who agreed to participate were male, white, trained at the master's degree level , had five or more years of experience with community college students, were between 35 and 55 years old, and espoused a client-centered orientation.

PAGE 38

26 Instrument at ion The instruments used in the study are described below. Specific utilization of each instrum.ent is presented in the section of this chapter entitled "Procedures." Fundamental Interpersonal Relationship Orientation Behavior The Fundamental Interpersonal Relationship OrientationBehavior instrum.ent was used to determine the com.pat ibi lity of the client-counselor dyad in this study. Schutz's (1960) concept of compatibility addresses itself to the specific relationship between two persons. His index of reciprocal compatibility was designed as an indicator of the likelihood of two persons expressed and wanted needs being complementary. The complementary nature of the affection needs of the client and the counselor was considered to be related to the existence of m.utual warmth in a beginning counseling relationship which has been shown to be dependent on both client and counselor in the initial phases of their relationship (Truax et al . , 1966) . The need scales The Fundamental Interpersonal Relationship OrientationBehavior (Schutz, 1960) is a 54 item, self-report questionnaire composed of six Guttmann type scales (Inclusion-Expressed, Inclusion-Wanted, Control -Expressed , Control-Wanted. AffectionExpressed, Affection-Wanted). Schutz reports internal consistency in terms of reproduct ibi 1 ity for five of " the six scales at .94 with the Control-Expressed scale at .93

PAGE 39

based on primarily college student samples averaging 1543 subjects . Test-retest reliability reported for the individual scales includes Inclusion-Expressed, .82 (p<.01); InclusionWanted, .75 (p<.01); Control-Expressed, .74 (p<.Cl); Control-Vi'anted , .71 (p<.01) based on a time interval of one month with a mean number of 140 (Range 125-183). A one week interval with 57 college students yielded testretest reliabilities of .73 (p<.01) for Affection-Expressed and .80 (p< .01) for Affection-Wanted. Intercorrelation of the Expressed and Wanted dimensions of each of the areas is Inclusion = .52 (p < .01); Control = .25 (p<; .05), and Affection = .70 (p< .01). The compatibility scales The reciprocal compatibility of any dyad (ij) is computed by rhe formula rKij = I ei wj + ej wi in which ei and ej are the measures of how i and j typically behave toward others and wi and wj are the measures of hov; i and j typically want others to behave toward thera. In the formula e stands for the scale score on the expressed scale in any of the areas of inclusion, control, or affection, and w stands for the scale score on the wanted scale in any of the areas. Prior to the final summation, signs are dropped using only the absolute values resulting in compatibility scores ranging from 0 to 18 with 0 meaning complete compatibility and 18 meaning complete incompatibility

PAGE 40

28 The conversion formula IS minus the computed index value was used (Schutz, 1960) to produce an ascending scale of indices. Based on the computation of reciprocal compatibility in the affection area for all possible combinations of pairs of two from the first to the second testing, the group of college students upon which test-retest reliability v/as determined (N = 57), a full scale Pearson product moment correlation of .67 (pc;; .01) was achieved. See Appendix A for a sample of the Fundamental Interpersonal Relationship Inventory Behavior as used in this study. Importance of Becoming Admitted to the Nursing Program Failure stress was shown by Hodges & Spielberger (1969) to have the effect of significantly raising reported state anxiety on the Affect Adjective Check List (Zuckerman, 1960). They showed that information that a subject was doing poorly on digit span test was related to increased state anxiety. Increasing levels of reported examination worry have been shown to be significantly related to increasing levels of state anxiety prior to an examination (Zuckerman & Biase, 1962) Extremely high levels of state anxiety were shown to be present in sophomore nursing students on examination days in a course which determined continuance or noncont inuance in that program (Hayes, 1966). It seemed reasonable that the level of im.portance a subject might place on becoming admitted to the nursing program might play a part in her state anxiety when told she had not passed the entrance examination.

PAGE 41

29 It also seemed reasonable that asking a direct question of the subject.s about how important admission to a given program was to them personally would be the most direct way to obtain the information. A question in scale form was developed within the context of asking students how important becoming admitted to the nursing program was to them compared to other important events in their lives. This scale is shown in Figure 1. Although the level of importance indicator does not have known validity or reliability it was decided that some attempt to assess the perceived importance of becoming admitted to the program might add relevant information which could shed light on the hypotheses being tested in the study (see Appendix A for specific format). Subject's Feeling about Examination Performanc e Zuckerman (1960) demonstrated that state anxiety increased in subjects who were told they wore doing poorly at digit span tests. Based on this finding, it was considered to be important to discover each subject's feeling about her performance on the entrance exiunination , as it would possibly affect her level of anxiety during the interview. A simple question (see Figure 2) was devised and asked of each subject at the conclusion of the examination session (see Appendix A for specific format ) .

PAGE 42

Many events in our lives are of importance (graduating from school, marrying, having children, getting a new car, finding a genuine friend, learning a new skill, etc.). Each of us considers some of these events to be of greater importance than others. DIRECTIONS: Taking into account the events in your life which are now or have been of IMPORTANCE to you, please place an "X" at the point on the scale shown below which BEST indicates the importance YOU place on becoming admitted to the program for which you are applying. IMPORTANCE (some) (most) (NOTE: Marking "1" indicates some importance while marking ''7" indicates that becoming admitted to the program is as important as anything in your life.) FIGURE 1 Level of Importance Scale

PAGE 43

31 DIRECTIONS : Please place an "X" on the line to the left of one statement below which BEST reflects your feelings about how you did on the Comprehensive Test of Basic Skills you took a little while ago I probably didn't pass. I HONESTLY have no feelings about it either way I probably passed. FIGURE 2 Subject Feeling about Entrance Examination Performance

PAGE 44

32 The Affect Adjective Check List The measurement of state anxiety was central to this study. The Affect Adjective Check List (Zuckerman, 1960) v/as selected as an appropriate instrument to measure state anxiety. The Affect Adjective Check List is a list of 61 words, 11 of which, if checked, are scored as indicative of state anxiety and 10 of which, if not checked, are considered indicative of state anxiety (see Appendix E). The other 39 words are neutral and all words are listed alphabetically. The range of scores . is 0 to 21. The internal reliability using the Kuder-Richardson Formula 20 is .85 and is significant beyond the .001 level. The test-retest reliability is .31 (p<(.05). Validation was demonstrated by students (34) scoring a differentially higher score mean of 2.41 ± .53 (p<;; .005) for 'three preexamination occasions compared to ten nonexamination days. These results were replicated with sophomore nursing students in a course which meant continuance or noncontinuance in the program (Hayes, 1966). Specific data for these nursing students were : Mean ' SD All class periods 5.86 3.19 Test days 13.99 3.15 Difference 8.13 (p< .05) Two forms of the checklist were devised for use in the present study. Each was called the Feeling Word Check List. Form ^^1 (printed on green paper) was derived by using the middle word (^^31) as the first word and then

PAGE 45

working alternately backward-then-f orwarcl , one word at a time, to arrive at the new ordering of words. Form rr2 (printed on yellow paper) was derived in the same way but by alternately working forward-then-backward, one word at a time to arrive at the new ordering of w'ords . The word "gay," a neutral word , was then dropped from both lists based on the loaded connotation which that particular word has acquired over the past several years (see Appendix E). Interview Rating Scale A scale which could be used to give a numeric rating of the counseling session was required to test Hypothesis 2. The Counseling Evaluation Inventory was developed in order to measure client ratings of counselors (Linden, Stone, & Shertzer, 1965). A 68 item Counseling Evaluation Inventory was developed including the 50 items of the Interview Rating Scale described by Anderson Cc Anderson (1962) . Following this initial development phase by Linden et al . which included 446 counselors and 289 students, all 68 item.s were retained. These 68 items were then submitted to 336 university students to evaluate their counseling experience, and the items were keyed based on weightings derived from the initial screening of the 68 items. Factor analysis produced four factors, Counseling Climate (X), Counselor Comfort (Y), and two parts of a Client Satisfaction factor (Z-1 and Z-2). Only items that loaded .40 on one butless than .40 on all other

PAGE 46

34 i > factors were retained in the final 21 item short form. One item proved to be unreliable at the .05 level and was omitted from the shorter form though itwas reliable on the longer form. Two additional items were deleted from use in this study. One item requested an evaluation of the counselor's effectiveness in showing the subject the value of testing. The other asked the subject to evaluate the counselor's effectiveness in commenting on how the subject could better reach her goals. Both items were considered potential sources of bias by suggesting to the subject what the outcome of the session should have been in terms of tests and future plans, two topics which m.ight well carry significant weight in the subject's thinking at the close of the session. The test-retest reliability of the Counseling Evaluation Inventory (Linden et al . , 1965) over a 14 day period with 16.3 secondary school students is .83 for the total score while the coefficients for the subscales are X = .78, Y = .63, and Z = .74, all significant at beyond the .05 level. The total score was used in this study. The name "Interview Rating Scale" as originally used by Anderson & Anderson was selected on the premise that it would be lesslikely to foster a spurious rating of the session in a subject's attempt to give the counselor a "good" or "bad" evaluation.

PAGE 47

35 Preference for Counseloi^ Indicator In order to test the applicability of Schutz's (1960) theorem of dyadic compatibility as it relates to preferences •for continued contact with a particular person (Hypothesis 3) some measure was needed to permit each subject to express her opinion with respect to continued contact with a particular counselor, should that be necessary. As what was wanted from each subject was her opinion about how she might behave in the future, it was decided to ask each subject her opinion directly . Figure 3 illustrates the format in which the question was asked (see Appendix F for placement of the question as part of the Interview Rating Scale). Counselor's Level of Facilitative Conditions A measure to determine the comparability of the facilitative conditions offered by the counselors participating in the study was considered essential. The Barrett-Lennard Relationship Inventory (Barrett-Lennard, 1962) was adopted for this purpose. Barrett-Lennard Relationship Inventory The Relationship Inventory (Barrett-Lennard, 1962) is a 64 item questionnaire (Form OS-M-64) with a Likert type response pattern. It was developed to assess the necessary conditions for establishment of a helping relationship as described by Rogers (1957) and yields separate scores for each of four scales (level of regard, empathic understanding, congruence, and unconditionality of regard)

PAGE 48

If you should decide to seek out a counselor on another occasion, would you seek out (check only one): This same counselor No preference A different counselor FIGURE 3 Opinion of Future Counselor Choice

PAGE 49

37 and a total score. The scores are composed of eight positively-scored items and eight negatively-scored items per scale. . . The items of the Relationship Inventory were judged by a panel of five client centered counselors ( Barret t-Lennard , 1962). All but one item was placed in either the positive or negative category by all the judges. Odd-even internal consistency coefficients based on a sample of 40 patient judges ranged from .82 (p< .01) to .93 (p < .01) with a mean of .86 (p< .01), compared to therapist coefficients ranging from .88 (p<.01) to .96 (p<.01) with a mean of .92 (p^.Ol). Test-retest correlations over a four week period based on a sample of 36 college students rating a close, long-standing personal relationship they felt was stable and unchanging were .89 (p^^ .01) on an empathic understanding scale, .84 (p<( .01) on a level of regard scale, .90 (p<^ .01) on an unconditionality of regard scale, .86 (p<^ .01) on a congruence scale, and .95 (p< .01) total. Mills Si Zytowski (1967) found that test-retest correlations under the directions of estimating their mother's feelings about them (N = 79) were .90 (p<; .01) for accurate empathy (Empathic Understanding), .74 (p< .01) for level of regard, .80 (p<.01)for unconditionality of regard, and .88 (p< .01) for congruence, all significant (p< .01). These four scales were reported to account for approximately 70 percent of the variance loading on a single factor. They

PAGE 50

06 subsequently recommended that the instrum.ent be used to assess the total relationship rather than components of it. This point is further supported by the recommendation that the use of the subscales of the relationship seemed unjustified, again on a factor analytic basis (Lanning & Lemons, 1974). See Appendix D for a copy of the BarrettLennard Relationship Inventory and a tabulation scoring sheet for the Inventory. Specific Design Consideration Precautions Cautions were exercised, in accordance with recognized research design techniques, to minimize any intruding factors which could jeopardize the validity of the research (Campbell £z Stanley, 1963). Hypothesis 1 is of the "One. Group Pretest-Posttest Design" (Campbell & Stanley, 1963). The problem of factor.s other than the treatment (compatibility) accounting for differences in outcome among subjects (anxiety change) is not present in this study to any large degree. The short time span between the first and second testing of anxiety used to determine the anxiety change protected against the effects of other intervening variables accounting for observed differences in outcome. The use of two different forms of the instrument used to measure anxiety helped protect against the effect of the first test influencing the second. The nature of the instrument was such that "learning" from one testing to the next was not likely to occur. The fact that the

PAGE 51

39 meaning of the instrument was not likely to change based on an observer's interpretation and that its use over a short time span had been previously demonstrated as appropriate (Hodges & Spielburger, 19S9 ; Zuckerman , Lubin, Vogel, & Valerius, 1904) protected against problems inherent to the instrument. All levels of anxiety change and compatibility were used in the study, thereby eliminating error from the use of only extreme scores. Clients were assigned to counselors on a rotation basis and interruptions in counseling were minimized by hanging a "do not disturb" sign on the counselor's door and having all telephone calls held. V/hatever effect the measurement of anxiety from the first testing to the second testing might have on the change in anxiety is not known. Introduction of the anxiety measuring instrument as an aid to help the subjects focus on their feelings was considered to minimize this possible source of extraneous variation. No reports of adverse reactions were observed by participating counselors. In addition to the precautions noted above,' the possible effects of the subject's age, race, discrepancy from the required passing score, feeling about her entrance examination performance, the im.portance she placed on being admitted to the program, the length of the interview, the number of years of schooling the subject had completed, and the particular counselor she saw in the interview were treated as additional variables in the analysis of the data so that possible effects of those factors could be determined.

PAGE 52

40 Hypotheses 2 and 3 were of the "Static-6roup Comparison" design except that rather than comparing a group which received no treatment with a group which did as the design provides, this study compared differing levels of the treatment (compatibility). To this extent the design of the study with respect to Hypotheses 2 and 3 was also considered analogous to the "Recurrent Institutional Cycle Design" (Campbell &, Stanley, 1963) as the experimental group was exposed to the various treatments (differing levels of compatibility) over a period of time. The major difficulty with this type of design is xhat the nature of the situation might influence the completion of the measurement device. A quiet place to complete the instrument and assurance of the subject's anonymity were used as precautions against social or personal pressures to rate the interview positively or negatively or to indicate a particular preference for a counselor on another occasion. Procedures Counselor Training Prior to the beginning of the study, the five counselors participated in a three hour training session with the writer. During the session the counselors and the researcher discussed the procedures to be followed and the use of the instruments they were to administer. A standard format for the procedural parts of the interviewwas agreed upon by all of the counselors. The counselors

PAGE 53

41 practiced the use of the instruments with each other and were able to conduct two simulated interviews with each other until a standard method for administration of those instruments was achieved. The "Counselor Guidelines'' statement (see Appendix C) based on the format developed during the counselor training session was distributed to each counselor. Each counselor was also requested to complete the Fundamental Interpersonal Relationship Orientation Behavior instrument . Each was asked to complete it according to the directions with an indication that it was needed as a general descriptive measure. The counselors were not apprised of the specific nature of the research. They were told that the project would best be served if they behaved in their usual manner and that it was not their counseling style which was being investigated. All counse lors ' agreed to participate under these conditions and to follow the Counselor Guidelines as distributed . To protect against differing levels of facilitative conditions generally offered by the participating counselors being introduced as an uncontrolled source of variance in the outcome of the study, each counselor agreed to participate in a control study separate from the study of the rejected applicants. The counselors each agreed to present several female clients he had seen for the first time for 30 minutes or more to his secretary. He told the client that the secretary had a questionnaire for her to

PAGE 54

42 complete to assist the college in improving its services. The nature of the counseling session was not specified except that it could not be one in which the counselor was required to tell the client that she did not meet entrance requirements for a particular program or curriculum. The writer provided each counselor's secretary with Barrett-Lennard Relationship Inventory questionnaires and instructed him or her to ask female clients to be brought in by the counselor to complete one questionnaire. The client was encouraged to be honest in her responses on the questionnaire items and was told that neither she nor the counselor would be identified on the questionnaire. Initial Data Collection Each female applicant (potential subject) to the Associate of Science Registered Nursing program was required to schedule the reading and mathematics sections of the Comprehensive Test of Basic Skills. Each applicant was allowed to choose her own test date from a list of prescheduled testing dates. She was told how long the testing' session would take. Twenty minutes beyond the tim,e normally used to administer the Comprehensive Test of Basic SKills was included in that statement to allow time to gather data required for this study. " When administration of the entrance examination had been completed, the initial data gathering instrument was distributed (see Appendix A).

PAGE 55

43 Also distributed was a written statement about the research entitled "Consent Form" which had been approved by the University of Florida's Committee for the Protection of Human Subjects (see Appendix B). As applicants to other health related programs were also taking the Comprehensive Test of Basic Skills during thesetesting sessions, all persons taking the test were asked to complete the data gathering instruments and to read and sign the consent form. Emphasis was placed on the point that the information gathered in this research would have absolutely no effect on admission to the programs for which students were applying. The data gathering instrum.ent included spaces for theapplicants to indicate name, the program applied for, age, sex, race, telephone number, highest level of schooling completed, the Level of Importance Indicator, the item indicating perception of performance on the Comprehensive Test of Basic Skills, and the Fundamental Interpersonal Relationship Orientation Behavior. It was possible that age, race, highest level of schooling completed, the level of importance tlie person placed on being admitted to the program, and the feeling about examination performance items would have a potential bearing on the client's response to the main treatment and they were gathered as a way of providing caution against confounding from unknown sources. A copy of the data gathering instrument can be found in Appendix A.

PAGE 56

44 All applicants were thanked for their cooperation in this research effort and assured that their specific responses to the questionnaire would be confidential. Name and telephone number information requests were explained as necessary if follow-up were to become desirable. Upon departure all applicants were handed a sheet thanking them again for their cooperation in the research and instructing them about making an appointment to get the results of the Comprehensive Test of Basic Skills and reassuring them of the confidentiality of the research information (see Appendix B) . The Study Interviews The instruction sheet given applicants upon completion of the Comprehensive Test of Basic Skills and the first phase of the data collection for this study instructed them to call the researcher's secretary to m,ake an appointment to see a counselor about their Comprehensive Test of Basic Skills results. The procedures in which the secretary would participate were discussed thoroughly with her prior to the beginning of the interviews. These procedures were considered practical by her and were practiced until they were done consistently with five consecutive surrogate subjects. The secretary was furnished with a roster of applicants taking the Comprehensive Test of Basic Skills after each testing session. Those applicants to the Registered

PAGE 57

45 Nursing program who had not passed the Comprehensive Test of Basic Skills at the required 12.0 grade level and had completed the initial data gathering instruments were noted on each roster as appropriate for an appointment with one of the five participating counselors. These applicants are hereinafter referred to as subjects. Those applicants not so identified, including applicants to other health related programs, were scheduled for an appointment with the writer or another nonpart icipating counselor. The secretary maintained a master roster of counselor availability and made an appointment for each subject to see one of the participating counselors on the basis of the first appointment time available to the subject after she called for an appointment. Subjects were not told by the secretary whether or not they had passed the Comprehensive Test of Basic Skills but were informed xhat the counselor v.'ould have that information for them at the scheduled appoitnment. Subjects were informed that the counseling session would take approximately an hour. Subjects' files were available to the counselor for reference during the interviews. Counseling Interview Seating arrangement for the interview placed the counselor at his desk with the subject at the side of the desk. This provided space for review of the subject's folder and space for the subject to complete necessary

PAGE 58

46 data collection forms. This arrangement should have had neither a positive nor negative effect on subject anxiety as reported by Myers (1969). The secretary allowed no interruptions and a "do not disturb" sign was placed on the door to further insure privacy and continuity in the interview. While getting settled for the counseling interview, the counselor informed the subject that the interview would last approximately 45 minutes and that the secretary would want to see her for about five minutes when the interview was over. An easily read clock was within clear view of the client during the interview. No attempt was made to terminate the interview prematurely simply to meet the time frame stated unless other obligations of the counselor required it. In that case the counselor would, at the end of the interview, invite the client to make another appointment. By looking at but without opening the subject's folder, the counselor informed the subject that she had not passed the Comprehensive Test of Basic Skills (based on a coded notation on the tab of the folder just below the subject's name). At this point, ,the first function of the interview had been completed. In accordance with the standard procedure stated in the Counselor Guidelines and training received prior to the beginning of the study, the counselor immediately asked the subject .to complete the "Feeling Word Check List #1" while hestudied her test protocols. This administration of the Feeling Word Check '

PAGE 59

47 List was considered the initial (pretreatment ) measure of state anxiety used in the computation of data to test Hypothesis 1. The counselor recorded the time the Feeling Vs'ord Check List #1 was completed on the Process Notes form (see Appendix C) provided in the subject's folder. This notation is used in the computation of data included to test the hypotheses. Once form #1 of the Feeling Vvord Check List was completed the counselor placed it aside and continued with the interview in a manner of his own choosing. He was required to include each of the purposes of the interview, which were the following: -To discuss with the subject the implications of her scores for possible future admission to the program. -To discuss with the subject her educationalvocational goals in light of the current situation. -To discuss with the subject any other matters which seemed important within this general context of not meeting admission requirements for the Registered Nursing progra.m. V/hen both subject and counselor agreed that the discussions had reached a natural termination point, the subject was asked to complete Feeling Word Check List #2. The counselor recorded the time Feeling Word Check List P2 was begun on the Process Notes form. This notation was used in computation of data to test the hypotheses. The score on Form ^2 was the posttreatment measure of state

PAGE 60

48 anxiety and was used in the computation of data to test Hypothesis 1. The counselor then escorted the subject back to the secretary and returned to his office. He made notes of events, impressions and feelings on the Process Notes form which he believed to be of importance as he had counseled with the subject. The secretary asked the subject to complete a short opinionaire , ostensibly as the final part of the college's attempt to gain insight inxo how to improve its services to students and prospective students. The subject was seated in a quiet area to complete the Interview Rating Scale which was used to test Hypothesis 2. One additional item was added to the Interview Rating Scale (#19) and was used to test Hypothesis 3. A copy of the Interview Rating Scale including the added item can be found in Appendix F.

PAGE 61

CHAPTER IV RESULTS AND DISCUSSION Confidence Le\-el Parameters The .05 level of confidence was established as the necessary indication that the results obtained in statistical procedure did not occur by chance. Control Study for Facilitative Conditio ns Generally Offered by Participating Counselors The scores obtained on the Barrett-Lennard Relationship Inventory completed by the control study clients represented the level of facilitative conditions offered by participating counselors. The scores ranged from 46 points to 117 points. The mean was 113.30 points with a standard deviation of 32.72. Table 1 presents the means and standard deviations computed for the facilitative conditions variable for the entire group of control study clients and the subgroup of clients seen by each counselor. The analysis of variance for the facilitative conditions variable indicates that the level of facilitative conditions generally offered to female clients by the participating counselors is not significantly different from counselor to counselor as measured by the Barrett-Lennard Relationship Inventory (Barrett-Lennard, 1962). A comparison of the age of the clients seen by the participating counselors as a part of the control study 49

PAGE 62

50 TABLE 1 Facilitative Conditions Generally Offered to Female Client by Counselors in Separate Control Studv Total Score on the Barrett-Lennard Relationship Inventory Group f.lean SD Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 117. 38 111 . 88 114 . 88 114 . 25 ICS . 13 26.28 28 .18 30. 81 32 . 15 46 . 19 8 8 8* 8 8 Entire Group 113 . 30 ^2 .72 40 Analysis of Variance Source Between groups Within groups Sum/Squares 616 . 93 8 39217 . 850 df Mean Square 4 154.234 90 1120.512 1376 NS Total 39217 . 8 50 ^Counselor #3 did 10 control study interviews while the other counselors did 8. The highest and lowest scores attained by Counselor ^^3 were dropped in this computation to balance cell size.

PAGE 63

51 was made with age of the clients seen by the counselors in the actual studj' interviews. The ages of the clients in the control study ranged from 17 years to 51 years. The mean age was 23.92 years with a sxandard deviation of 8.31 years. Table 2 presents the means and standard deviations computed for the entire control study group and the subgroup seen by each counselor. The analysis of variance of the ages of the control study clients shows that the mean ages of the control group clients and the subgroup of control group clients seen by each counselor were not significantly different. Table 2 also shows the results of a t_ test for group means which shows the control study client mean age was not significantly different from the subject sample mean age. A comparison of the length of the control study interviews was made with the length of the actual study interviews. The length of the control study interviews ranged from 15 minutes to 45 minutes. The mean length of the control study interviews was 28.95 minutes with a standard deviation of 7.45 minutes. Table 3 presents the means and standard deviations for the length of the control study group and the subgroup seen by each counselor. The analysis of variance of the length of the control study interviews shows that the mean interview length for the group and the subgroup seen by each counselor is not significantly different. Table 3 also shows the results of a t test for group means which shows the mean length

PAGE 64

52 TABLE 2 Control Study Client Ages Group Mean SD Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 All Clients 27.50 21 . 87 20.25 24. 87 25. 12 23.92 13.72 6 .79 3 .45 7 .45 6 .62 8.31 8 8 8 8 8 40 Source Between groups V/ithin groups Analysis of Variance Sum/ Squares df 310.475 4 2420.131 35 Mean Square 77 . 61 69 . 14 1 . 1253 NS Total 2720 .131 39 t Test Control Study Clients Subject Sample Mean 23.92 23.03 SD 8.13 6.19 SEM .98 .64 Value oft . 609 p < .01

PAGE 65

53 TABLE 3 Length of Control Study Interviews Group Mean SD N Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group 25 . 75 27. 50 30.00 31.50 30. 00 27.95 3.99 3.99 0.00 9.95 7.4; 8 8 8 8 8 40 Anal3rsis of Variance Source Sum/Souares df Mean Square Between groups 226.271 4 56.56 Within groups^ 1105.502 35 31.58 Total 1331 . 773 39 1 .79 NS Mean SD SEM t Test Control Group Clients 28.95 7.43 1.18 Subject Sample .27.87 18 . 89 1 . 94 Value of t .404 p < .01

PAGE 66

54 of the control study intei-views is not significantly different from the mean length of the study interviews. The control study group was composed of five black and 35 white clients. Tab],e 4 presents the absolute frequency of black and white control study clients for the entire group and for the subgroup seen by each counselor. It also shows the percentage of the entire group or subgroup falling into each category. A Chi-Square test was done to compare the control study racial group balance with the subject sample racial group balance. Table 4 shows that the Chi-Square is significant beyond the .01 level indicating that the control study group is not of the same racial balance as the study sample. Although it can be said that the facilitative conditions generally offered to female clients arenot significantly different Crom counselor to counselor, the small number of black clients seen as part of the control study does not allow that assurance to be made with respect to black clients seen by the counselors. Subject Pool Applicants to the Registered Nursing program who did not pass the Comprehensive Test of Basic Skills at the required 12.0 grade level on or after August 19, 1975, and continuing through November 22, 1976, became mem.bers of the subject pool. A total of 277 took and did not pass the Comprehensive Test of Basic Skills at the necessary level during this period. There were 7'^ of that number

PAGE 67

TABLE 4 Control Study Racial Group Membership Group Black White Total Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire Group N 1 1 2 1 0 c ,0 14 14 33 14 0 13 N 7 7 6 7 8 35 £7 lO 86 87 67 86 100 N 8 8 ,8 8 8 40 10 100 100 100 100 100 100 Subject Sample Control Study Group Chi Square Test Black N % 31 33 i: White N 64 67 35 87 Chi Square value (2df) X-^ = 34.95 (p< .05 = 3.841)'

PAGE 68

56 who did not complete the initial data collecting instruments at the time of the administration of the Comprehenisve Test of Basic Skills, leaving a pool of 20^. Ninety-five subjects made appointments with a counselor to get their test results, were seen by t;,e counselors participating in this study, and completed all data gathering instruments, thereby becoming the sample of subjects for this study. A comparison of the subjects who participated in this study and of the 112 subject pool members who did not participate was made to determine whether the subjects of this study represented the broader group. Table 5 presents the com.parative information. The t_ tests shown in the table demonstrate that the subject group and the nonparticipating subject pool group are not able to be distinguished from one another with respect to their discrepancies from the required passing score on the entrance examination, their ages, the level of importance they placed on being admitted to the program, and the number of years of schooling they had completed. Chi-Square tests shown in Table 6 suggest the groups were not significantly different in racial group membership and their feeling about their entrance examination performances. The subjects who participated in this study, therefore, cannot be differentiated from the total population of applicants who failed to meet the required passing score on the entrance examination.

PAGE 69

57 TABLE 5 Comparison of Participating Subjects and Non-Participating Subject Pool Members Part icip ants Xcn -Participants Value of ]t Mean SD N Discrepancy 24.04 21.91 95 22.15 20.34 112 from passing .64 score p = . 52 (two-tailed) Age 23.03 6.19 94 23.73 7.48 112 -.72 p .47 • . (two-tailed) Importance of Being Adiritted 6.13 . 85 95 6.18 .82 108 -.42 p = .67 ( two-tailed) Years of 12.61 .93 95 12.79 1.24 112 Schooling ^ Completed -1.13 p = .26 (two-tailed)

PAGE 70

o8 TABLE 6 Comparison of Participating Subjects and Non-Participating Subject Pool Members Part icipants Row % Non -Part icipants N Row % B] ack White 31 57.4 64 41.8 23 42.6 89 58.2 Chi Square Value (1 df) X = .145 p @ . 05 = 3 . 841 Felt passed examination "No feeling" about examination performance Felt not passed examination 56 47.5 19 45.2 20 44.4 62 52.5 23 54.8 25 55.6 Chi Square Value (2 df ) X" = 3.298 . p @ .05 = 5 . 991

PAGE 71

59 Independent Variables Subjects Age . The subjects' ages ranged from 17 years to 50 years. The mean was 23.032 years with a standard deviation of 6.193 years. Table 7 presents the means and standard' deviations computed for the age variable for the entire sample of subjects and for the subsample of subjects seen by each counselor. The results of the analysis of variance for the age variable show that the mean ages of the subjects in the entire sample and the subgroup of subjects seen by each counselor were not significantly different. Race. Sixty-four of the subjects identified themselves as white while 31 identified themselves as black. Table 8 presents the absolute frequencies of black and white subjects for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category and the results of a Chi-3quare test showing that the subgroups of subjects seen by each counselor are not sufficiently out of balance to indicate that the subgroups are likely to be from a different population of subjects. Years of Schooling . The number of years of schooling completed by the sample subjects ranged from 10 years to 16 years. The mean was 12.611 years with a standard deviation of .926 years. Table 9 presents the means and standard deviations computed for the years of schooling

PAGE 72

60 TABLE 7 Age of Subjects Group Mean SD N Counselor 1 24 . 300 7 .658 20 Counselor 2 23 . 105 4 .421 19 Coun sp 1 or 3 22.222 o R o n 18 Counsel or 4 24 . 400 p o . i 20 Counselnr 5 20 . 706 C\'~i A 17* Entire Group 23.032 6 193 94* Analysis of Variance Source Sum/Squares clf Mean Square F Between groups 173 .4648 4 43 3662 1 . 1374 Within groups 3393.4609 89 38 1288 NS Total 3566. 925S 92 *The age of one subject was not obtained, on N shown . Computations based

PAGE 73

61 TABLE 8 Subject Racial Group Membership Group Black White Total c /o c /O N 10 Counselor 1 S 30 12 70 20 21 Counselor 2 3 16 16 84 19 20 Counse lor o 4 22 14 78 18 19 Counselor 4 9 45 11 55 20 21 Counselor 5 7 39 11 61 18 19 Entire Samole 31 33 64 67 95 100 Chi Square Test Black White Sum Dev^/Exp (black plus white) Value of X'^ (p< .05, df = 4) 5.627 (p< .05) .9 .488 Total Exp Obs Dev^ Exp Obs Dev^ Counselor 1 6.5 8 2 . 25 13.5 12 2 . 25 20 Counselor 2 6.2 3 10.24 12 . 8 16 10.24 19 Counselor 3 5.9 4 3 . 61 12.1 14 3.61 18 Counselor 4 6.5 9 6 . 35 13.5 11 6 . 35 20 Counselor 5 5 . 9 7 1.21 12.1 11 1.21 18 Sum: Dev2/Exp 3 . 792

PAGE 74

62 TABLE 9 Number of Years of Schooling Completed by Subjects Group Mean SD "N Counselor 1 12.75 1.21 20 Counselor 2 '12.95 0.97 19 Counselor 3 . 12.33 0.6S IS Counselor 4 12.60 0.S2 20 Counselor 5 12.39 0.73 IS Entire group 12. Gl 0.93 95 Source Between groups Within groups Total Analysis of Variance Sum/Squares df 4.8135 4 75.7764 90 80.5898 94 Mean Square F 1.2034 1.4292 .0.8420 NS

PAGE 75

variable for the entire sample and the subsarnple seen by each counselor. The results of the analysis of variance for the years of schooling variable show that the mean number of years of schooling completed by subjects in the entire sample and the subsarnple of subjects seen by each counselor were not significantly different. Importance Placed on Becoming Admitted to Program . The point at which the level of importanqe indicator was marked was scaled graphically to the nearest half point and was labeled as the subject's l6\'el of importance score. The marked level of importance scores ranged num.erically from a low of three to a high of seven. The possible scale scores ranged from one to seven. The mean was 6.13 with a standard deviation of 0.85. Table 10 presents the mieans and standard deviations computed for the level of importance variable for the entire sample of subjects and for the subsarnple of subjects seen by each counselor. The results of the analysis of variance for the level of importance variable show that the mean level of importance checked by subject^ in the entire sample and the subsample of subjects seen by each counselor are not significantly different. Feeling about Entrance Examination Perf ormiance . Subjects were asked to place an "X" next to one of three statements which best reflected hov; they felt about their performance on the Comprehensive Test of Basic Skills. Fiftysix subjects indicated they felt they had passed, 19 indicated they hon-estly had no feelings either way. and 20 indicated

PAGE 76

64 TABLE 10 Importance Placed on Becoming Admitted to Program Group Mean . SD N Counselor 1 6.20 1.11 20 Counselor 2 6.03 0.98 19 Counselor 3 6.03 0.79 18 Counselor 4 6.28 0.64 20 Counselor 5 6.11 0.72 18 Entire group '6.13 0.85 95 Analysis of Variance Source Between groups Within groups Sum/Squares clf 0.9167 4 67.6840 90 Mean Square 0 . 2292 0 , 7521 F 0.3047 NS Total 68.6057 94

PAGE 77

they felt they had not passed. Table 11 presents the absolute frequencies by response category for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category and the results of a. Chi-Square test showing that the subgroups of subjects seen by each counselor are not sufficiently out of balance to show the subgroups lively to be from a different population of subjects. Discrepancy from Necessary Passing Score . The number of points by which each subject failed to meet the required 12.0 grade level on the Comprehensive Test of Basic Skills was labeled the subject's discrepancy score. The discrepancies ranged from one point to 89 points. The mean was 24.04 points with a standard deviation of 21.91 points. Table 12 presents the means and standard deviations computed for the discrepancy score variable for the entire sample and the subsam.ple seen by each counselor. The analysis of variance for the discrepancy score variable shows that the mean discrepancy scores for the entire sample of subjects and the subsample of subjects by each counselor were not significantly different. Subj ect-Counselor Compat ibi 1 i ty The subject-counselor dyad compatibility index values as measured by the Fundamental Interpersonal Relationship Orientation Behavior ranged from a minimum of two to the maximum possible of 18. The mean value was 12.505 with a standard deviation of 3.727. Table 13 presents the means

PAGE 78

66 TABLE 11 Subject Feeling About Entrance Test Performance Group Pass Don t Know Didn t Pass N CI 10 N c /c N r .0 Counselor 1 li OO 5 4 O f\ 2U Counselor 2 14 74 4 21 1 5 Counselor 3 11 61 1 6 6 33 Counsel or 4 14 70 2 10 4 20 Counselor 5 6 33 7 39 5 28 Entire Sample 56 59 19 20 21 Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Chi Square Test Pass Exp Obs Dev2 11.8 11 .64 11.2 14 7.84 10.6 11 .16 11.8 14 4.84 10.. 6 6 21 . 16 Don't Know Didn't Pass Exp Obs DeyS Exp Obs Dev2 4 5 1 . 4.2 4 .04 . 3.8 4 .04 4 1 9 3.6 1 6.76 3.8 6 4.84 4 2 4 4.2 4 .04 3.6 7 11.56 3.8 5 1.44 Sum: Dev-^/Exp 3.176 Sum: Dev2/Exp .'All) Value of (p^ .05, df = 12) 6.539 13.636 (p < .05) 21 . 026 3.921

PAGE 79

0/ TABLE 12 Discrepancy from Passing Score on Entrance Examination Group Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire group Mean 26 .80 18. 05 26 . 22 20 . 70 2S. 83 24 . 04 SD 19 . 09 24.6 3 16 . 87 25.62 21 .91 N 20 19 IS20 18 95 Source Between groups IVithin groups Analysis of Variance Sum/Squares df Mean Square F 155.8672 4 388.9668 0.8033 43579.9805 90 484.2219 NS Total 45135. 8477

PAGE 80

68 Group Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 TABLE 13 lubject -Counsel or Compatibility Mean 14. 000 13 . 526 9.611* 10 . 850* 14 . 500 SD 2.000 1 . 867 4.461 4.660 2 . 093 N 20 19 IS 20 18 All dyads 12 . 505 3 . 727 95 Analysis of Variance Source Between groups Within groups Total Sum/Squares df 341.6846 4 964.0654 90 1305.7500 94 Mean Square S5 . 4211 10. 7118 7 . 9745 (P< .01) *Signif icantly different, from each other and the eroup mean through the application of Tukey's test for significant gap (Edwards , 1954) . s i

PAGE 81

and standard deviations computed for the compatibility index valaes for all of the subject-counselor dyads and for the subsample of dyads involving each of the counselors. The results of the analysis of variance of the compatibility index values indicate that the mean compatibility index for the subsample of subject-counselor dyads involving Counselors #3 and ^^4 was lov/er than for the sub j ect-counselor dyads of the other three counselors. This lowered mean compatibility index for Counselors -3 and ^4 did not preclude their inclusion in any further analysis as the specific compatibility of the subject-counselor dyad was treated on a case-by-case basis with each of the outcome measures. It did mean that there would be a lowered possibility for the compatibility of those dyads to show as strong a relationship to the outcome measures.It also suggested that each counselor should be considered as a special subgroup in preliminary data analysis procedures used to discover interaction effects. Length of the. Interview The length of each study interview in minutes was determined by computing the difference between the time the subject completed the first measurement anxiety taken immediately after the subject was told she had not passed the Comprehensive Test of Basic Skills at the required level and the time she began the second measurement of anxiety at the conclusion of the interview.

PAGE 82

70 The length of the interviews ranged from six minutes to 96 minutes. The mean length was 27.874 minutes with a standard deviation of 18.893 minutes. Table 14 presents the means and standard deviations computed for the length of interview variable for all of the subject-counselor dyads and for the subsample of dyads involving each of the counselors. The analysis of variance for the length of interview variable demonstrated that Counselors #1 and #4 engaged in interviews of significantly greater mean length than did the other three counselors. The possibility that longer or shorter interviews might have differential effects on outcome measures suggested that each counselor should be considered a separate subgroup when examining preliminary data for interaction effects. Dependent Variables Anxiety Change Change in client anxiety over the course of the interview was determined by subtracting the measured anxiety level of each subject at the end of the interview from her • measured anxiety level at the beginning of the interview for each subject. As the anxiety change value is derived, an examination was made of the scores on the first and second anxiety measures and a ;t test for paired measures was done to determine the validity of the anxiety change values. Tables 15 and ]G show the mean and standard deviation values for the first and second anxiety measures

PAGE 83

71 Tx\BLE 14 Length of Interview Group Mean SD N Counselor 1 28.000* S . 820 20 Counselor 2 23.105 8.110. 19 Counselor 3 18.333 4.863 18 Counselor 4 52.300** 26.274 20 Counselor 5 / 15.167 5.227 18 All Groups 27.874 18.893 95 Analysis of Variance Source Sum/Squares df Mean Square F Between groups 16910.0430 4 4227.5078 22.8589 Within groups 16644.5195 90 184.9391 (p<;_.01) Total 33554.5625 *Signif icantly different at the .05 level through the application of Tukey's test for stragglers (Edwards, 1954). **Signif icantly different at the .05 level through the application of Tukey's test for significant gap (Edwards, 1954).

PAGE 84

TABLE is First Anxiety Mc?asure Group Mean SD N Counselor 1 11.000 4.554 20 Counselor 2 • 11.000 3.559 19 Counselor 3 11.111 2.720 18 Counselor 4 11.250 3.508 20 Counselor 5 11.111 3.612 18 All groups 11.095 3.579 95 Analysis ox Variance Source Sun/Squares df Mean Square F Between groups 0.8423 4 0.2106 0.0157 Within groups 1203.3062 90 13.3701 NS Total 1204.1484

PAGE 85

TABLE 16 Second Anxiety Measure Group Mean SD N Counselor 1 ' 7.800 4.099 20 Counselor 2 7.263 3.297 19 Counselor 3 9.222 3.209 18 Counselor 4 8.200 2.285 20 Counselor 5 8.444 ' 4.090 18 Entire group 8.168 3.441 95 Analysis of Variance Source Sum/Squares df Mean Square F Between groups 39.6680 4 9.9170 0.8313 Within groups 1073.6406 90 11.9293 NS Total . . 1113 . 30S6 94

PAGE 86

74 for the entii-e sample of subjects and the subsamples of subjects seen by each counselor. The analysis of variance for those variables shov/s that the mean anxiety score for the entire sample of subjects and the subsample of subjects seen by each counselor did not differ significantly. The anxiety change values for the sample ranged from minus ten to 12. The mean anxiety change was found to be 2.905 with a standard deviation of 3.573. Table 17 shows the mean and standard deviation values for the entire sample of subjects and subsample of subjects seen by each counselor. The analysis of variance of the anxiety change variable shows that the mean anxiety change value for the entire sample of subjects and the subsample of subjects seen by each counselor did not significantly differ. The results of a t_ -test for mean differences between paired measures shows that the mean difference in measured anxiety over the course of the interview was significant across the pool of subjects. Rating of the Interview A rating of each in"cerview was taken using the Interview Rating Scale. The scores on the Interview Rating Scale ranged from 0 to 38. The mean score was found to be 28.579 with a standard deviation of 9.313. Table IS shows the means and standard deviations computed for the Interview Rating Scale scores for the entire sample of subjects and the subsamples seen by each counselor. The analysis

PAGE 87

TABLE 17 Anxiety Cnange Group • Mean SD N SxNK Range p = .05 Counselor 1 3.200 3. 750 20 3.37 Counselor 2 3.737 2. 864 19 3.70 Counselor 3 -1 ^ o n 766 18 3.94 Counselor 4 o . Uow o . 790 20 Counselor 5 2.667 3. 964 IS . 3.94 Entire group 2.905 3. 573 95 2.82 Analysis of Variance Source Surn/ squares df .Mean So^uare F Between groups 4 9.8005 0.7598 Within groups , 1160.9453 90 12.8994 NS Total 1200. 1475 94 t Test First Anxiety Measure ' w i t h Second Anxiety Measure First Measure Second Measure Mean 11 . 095 8. 168 SD 3.579 3.441 Diff 2 . 905 SD/Dif f 3 . 573 Value of t 8.02 P < .01

PAGE 88

76 TABLE IS Interview Rating Scale Group Mean SD Counselor 1 Counselor 2 Counselor 3 Counselor 4 Counselor 5 Entire group 31 . 250 26 .474 28. 333 29 . 750 26 . 778 28 . 579 7.629 12.607 8. 139 6 . 3.82 10 . 773 9. 313 N 20 19 18 20 18 95 Analysis of Variance Source Between groups Vv'ithin grouDs Sum/Squares clf 313.8867 4 7839.3633 90 Mean Square 7S .4717 87 . 1040 F 0 . 9009 .KS Total 3153 . 2500

PAGE 89

77 of variance of the results on the Interview Rating Scale shows that the mean rating of the interview scores for the entire sample of subjects and for the subsample of subjects seen by each counselor did not significant.! \ differ. Choice of Counselor for Future Occasion An indication of eacn subject's choice of counselor, should seeing a counselor at some unspecified future time seem appropriate, was obtained. Subjects were asked to check one of three response categ"ories--to see the same counselor, no preference, or to see a different counselor. Table 19 presents the absolute frequencies by response category for the entire sample of subjects and for the subsample of subjects seen by each counselor. It also shows the percentage of the entire sample or subsample falling into each category. Testing of the Hypotheses The study was undertaken to discover if a relationship exists between certain outcomes of stressful counseling interviews and several variables associated with those interviews. Ninety-five subjects completed all phases of the study. They were seen by one of five counselors. Three measures of outcome were chosen for examination. They were as follows: 1. Change in subject anxiety over the course of the interview. 2. A rating of the interview by the subject.

PAGE 90

78 TABLE 19 Preference for Particular Counselor for Future Occasion Group Same No Preference Different ,0 N % N a 10 Counselor 1 IS Counselor 2 13 Counselor 3 11 Counselor 4 16 Counselor 5 14 90.0 ' 2 6S.4 6 61.1 6 80.0 4 77.8 4 10.0 31 .6 33.3 20. 0 22 . 2 0 0 1 0 0 5.6 0 0 Entire Group 75 , 8 22 2: 1.1 Chi Square Value (4 df ) X2 .90 p @ .05 = 9.48S

PAGE 91

79 3. An indication of the subject. ' s choice of counselor if seeing a counselor again should seem appropriate. The primary focus of the study was to determine if . the subject-counselor reciprocal compatibility in the area of affection (Schutz, 1960) was correlated with the measures of outcomes. Other variables which were thought to be related to the measures of outcome were also considered. They • included the following: 1. The age of the subject. 2. The racial group to which the subject belonged. 3. The number of years of schooling the subject had completed. 4. The importance of becoming admitted to the program as reported by the subject. 5. The subjec't's feeling about her performance on the entrance examination. 6. The discrepancy from the required passing score on the entrance examination. 7. The length of the interview. Preliminary Analysis of the Data . A series of Pearson product moment correlations were computed between each of the outcome m.easures and each of the independent variables for each of several subgroups of subjects. The subgroups were black subjects, subjects who reiwrted "no feeling" about entrance examination performance, and subjects who reported they felt they

PAGE 92

had failed the entrance examination. This '.vas done Co inspect for the possibility that these subgroups of subjects might show a difference in their relationship to an outcome measure on one of the independent variables which might not appear when the entire sample of subjects was considered as a group. Subjects who saw Counselors #2, #3, #4, or #5 were also considered subgroups for this same reason. This use of "Counselor" as a subgroup was necessitated by the fact that the client-counselor compatibility scores ana the length of the interviews conducted by each counselor were not comparable as demonstrated in the analyses of variance of those variables., The subgroups of white subjects, subjects who expressed the feeling they had passed the entrance e.xamination, and subjects seen by Counselor -1 were not considered subgroups for this purpose as their scores on independent variables were automatically considered in the general data analytic procedures used to test the hypotheses. The intercorrelations for each subgroup (see Appendix G) were inspected for correlations between independent variables and outcome measure. An interaction effect was considered to exist v.-hen a significant correlation between an independent variable and an outcome measure w-as found for a particular subgroup. An interaction term was considered not to exist, however, when an even number of subgroups of similar nature, e.g., Counselor r2 and Counselor both showed a correlation between an independent variable and

PAGE 93

81 an ouccome measure and the signs of the correlation coefficient were in opposite directions, thereby having offsetting effects. Neither was an interaction effect considered to exist when the number of subjects in a subgroup fell to seven or below, as statistical analyses based on such a small number were considered unsound. Interaction terms were computed for the interaction effects noted by application of these criteria. Eleven such interactions were found to exist among the independent variables. The five with respect to the subject's anxiety change over the course of the interview were as follows: -Reporting "no feeling" about performance on the entrance exa;:iinatioii combined wixh the discrepancy from the required passing score (r = .39, p = .05, N 19) . -Reporting "no feeling" about performance on the entrance examination combined with the importance of becoming admitted to the program (r = -41, • p^ .05, N 19) . -Having seen Counselor ^^3 combined with the length of the interview (r = .46, p< .05. N = 18). -Having seen Counselor 7~5 combined with the clientcounselor compatibility score (4 = .43, p ^ . 05 , N = 18) . -Having seen Counselor fo com.bined with the . . importance of becoming adm.itted to the program

PAGE 94

82 (r = .46 , p < . 05 , M = 18) . The eight interactions with respect to the rating of the interview by the client were as follows: ' -Being black combined with the length of the interview (r = . 40 , p < . 05 , N = 31 ) . -Being black combined with the discrepancy from the required passing score on the entrance examination (r = -.47, p .01, N 31) . -Being black combined with reporting "no feeling" about performance on the entrance examination (passed or not passed) (r = -.32, p < . 05 , N = 31). -Reporting "no iueling" about performance on the entrance examination combined with the discrepancy from the required passing score (r = .45, p ^ . 05 , . N = 19) . -Reporting "no feeling" about performance on the entrance examination com.bined with the importance of being admitted to the program ( r = . 71 , p < . 01 , N 19) . -Reporting feeling the entrance examination not passed combined with the discrepancy from the required passing score (r = -.60, p<.01, N 20). -Having seen Counselor =2 combined with the discrepancy from the required passing score (r = -.73, p< Oi , N 19) .

PAGE 95

83 -Having seen Counseloir4 combined with the importance of becoming admitted zo the programi (r = .48, p ^ . 05 , N = 20) . The four interactions with respect to the subject's choice of counselor for some future occasion were as follows -Reporting "no feeling" about perform.ance on the entrance examination combined with the importance of becoming admitted to the program (r = -.75, p < .01, = 19) . -Having seen Counselor p2 combined with the subject's age (r = -.57, p< .01, N = 19). -Having seen Counselor 4^4 combined with the subjectcounselor compatibility (r = . 53 , p< .01, N = 20). -Having seen Counselor -72 combined with the discrepancy from the required passing score (r =-.43. p< .05, N =19). Interaction terms consistent with these interactions of independent variables were computed and were used as additional independent variables in the analyses of data with respect to the dependent variables. Hypothesis 1 Clients in higher compatibility client-counselor dyads will show greater reduction in anxiety over the course of a stressful interview than will clients in lower client-counselor compatibility dyads .

PAGE 96

84 Procedure . A forward stepwise multiple regression procedure was performed with tne subject's anxiety change as the dependent variable. The independent variables were as f ol lows : 1. Subject-counselor compatibility 2. Age of subject 3. Being black 4. Years of schooling 5. Importance of being admitted 6. Reporting "no feeling" about examination perform.ance 7. Reporting feeling entrance examination not passed 8. Discrepancy from passing score 9. Length of interview 10. Reporting "no feeling" about examination performance combined with discrepancy from, required score 11. Reporting "no feeling" about exramination performance combined with importance of being admitted 12. Saw Counselor #3 combined with length of interview 13. Saw Counselor #5 com^bined with subject-counselor compatibility 14. Saw counselor #5 combined with im.portance of being admitted. Findings . Table 20 presents a summary of the results of the regression analysis with respect to anxiety reduction scores as the dependent variable. The table shows only the first six steps of the analysis as . the significance of the

PAGE 97

85 o o Id o 0 •H H 0 u o o s 0 ! — I fn K O •H >, +J 0 ^ 'h ^ -P Co C ^ 0 0 O •r-i -a < > O rH S CtS -H O ci t-l > Td cc; 0 05 0 0 C " 0 0 -n-. +-> 0 HOW in o o • o CO o o CO o V c (M r-! o O cc O) a CO 00 00 rH CO VI LO CD u J CD CO r~~i L' J 1 rH r— i + + + 1 ! 00 CO CO CO CD SM o o CSj (M 1 — 1 rH CO CO o CD ,• — ^ u J CO o CM CO , — ! r— t r— 1 CD iTN <->H nT C-l / — ^ J /— s o 1 ^/*^ en m ro \.' J -P •H •p +^ M — < P ^3 o 0 0 £ c o T— 1 o C Jh 0 CO >, V c p P H 0 •H c 1—1 c 0 C 0 c o 'o •H m > •H ^ •r-1 c 1 O 0 r— 1 rH U 1 — : +j •H 0 0 0 0 r^ -P 0 P x: > 0 Cm 0 c 0 Cl CD 0 % ^ to C *' J Ch cm 0 -P C « c r-t P G K -p 0 S3 0 r-: ?i 0 0 >^ K) 0 ^ -H -P Ch " -P ca CO LO CD

PAGE 98

86 multiple correlation coefficient did not reach the .05 level when the seventh independent variable was entered into the analysis. Inspection of Table 20 reveals that the single best predictor of anxiety reduction was the importance the subject placed on being admitted to the program. It accounted for approximately 4.7 percent of the variance and created a multiple correlation coefficient of .216. Once the importance of being admitted was considered, the next most important factor realted to subject anxiety reduction was the subjectcounselor compatibility which contributed .048 to the multiple correlation and accounted for approximately 2.3 percent miore of the variance in the anxiety reduction scores of the subjects. Subjects in higher compatibility dyads did show greater anxiety reduction over the course of the interview than did subjects in lower compatibility dyads. The subject-counselor compatibility was not, however, the most important variable with respect to anxiety reduction and contributed only a small amount to subjects' anxiety reduction. The Pearson product moment correlation betv/een sub j ect -counselor compatibility and anxiety reduction for all dyads was found to be .18 (p<^ .05). It should be noted that subj ect -counselor compatibility combined with having seen Counselor nb met a criterion for becoming an interaction term with respect to anxiety reduction. It was possible that the strength of the relationship between compatibility and anxiety reduction in the case of one counselor ( r .43, p<.05) was strong enough to produce the appearance of a

PAGE 99

general effect. That interaction term did not enter the multiple regression analysis while the multiple correlation coefficient was significant or prior to the entry of subjectcounselor compatibility as a general effect. The general effect of subject-counselor compatibility would appear, therefore, to be stronger with respect to anxiety reduction over the course of the interview. The hypothesis that subjects in higher compatibility dyads did show greater anxiety reduction over the course of the interview than subjects in lower compatibility dyads was upheld. • . Discussion . The other four variables which entered the regression analysis before the multiple correlation ceased to be significant are also shown in order of entry in Table 20. As can be seen, all six variables accounted for only about 13.5 percent of the variance of the anxiety reduction scores and the last two to enter the regression model (reporting ''no feeling" about test performance and being black) were negatively related to anxiety reduction as they entered the multiple regression correlation analysis. This should not be interpreted as suggestive of withholding counseling from, black female applicants who do not meet entrance examination requirements or from female applicants not meeting entrance examination requirements who do not feel that they either passed or did not pass the examination. It does suggest that with those subjects there was an increase in anxiety over the course of the interview and that other

PAGE 100

88 methodologies might be more appropriate when anxiety reduction is the goal. Two of the six variables related to anxiety reduction involved the subject's indicating "no feeling" about her performance on the entrance examination. V/hen combined with greater discrepancy from the required passing score on the entrance examination there was a slight positive relationship with anxiety reduction, but when considered alone the relationship was negative though still small. This suggests that where the subjects reported that they did not have an internal feeling about having passed or not passed the examination and had not passed the examiination by greater amounts they experienced some anxiety reduction during the interview. IVhen subjects reporting "no feeling" about entrance examination performance are considered on that basis alone, however, there was a slight increase in anxiety over the course of the interview. It appears that anxiety increased under the stress of discovering that the entrance examination had not been passed for subjects as a whole but decreased slightly for subjects who had not passed by ^vider margins. That the length of the interview should have a positive effect on anxiety reduction was not surprising. The more time spent in a supportive relationship where subjects discovered they had not met entrance examination requirements did promote a reduction in anxiety over the course of that interview.

PAGE 101

89 Pearson product moment correlations and analyses of variance were computed for the first and second measures of anxiety and the predictor variables (see Appendix H) in an attempt to gain further understanding of the nature of anxiety change over the course of the interviews. What emerged was a low positive correlation between the first measurement of anxiety and the importance subjects placed on being admitted to the program (r = .30, • p<.01, N ^ 95). This indicated that those who placed more importance on being admitted to the program were more anxious on being told they had' not passed the entrance examination. It also mieant that those who placed higher levels of importance on being admitted to the program had a slightly greater theoretical opportunity for anxiety reduction. The fact that there was not a significant relationship between the second anxiety measure and the importance subjects placed on being admitted to the program confirn:s that the greater anxiety reduction experienced by those placing miore importance on being admitted to the program was a function of their higher levels of anxiety at the time the first measure was taken,. This study did not attempt to examine the issue of higher or lower initial anxiety levels, but the finding does suggest that the counseling process was more effective with subjects who had higher levels o L' initial

PAGE 102

90 anxiety or that processes intrinsic to those who placed more importance on being admitted to the program were at work. The practitioner m.ight wish to focus on rhose for whom being admitted is more important and other researchers may wish to examine the ramifications of this phenomenon more closely . Examination of the correlations in Appendix H also revealed that neg3,tive relationships existed between age a.nd the first anxiety measurement (r = -.31, p< .01, N 95) and between age and the second measurement of anxiety (r = -.19, p ^ . 05 , N = 95). Apparently older subjects experienced less anxiety at discovering they had not passed the entrance examination than younger subjects and their an.xiety levels at the Cxose of :he interview were also less than those of younger subjects. This discovery does not suggest any implication for anxiety reduction based on age, but it does raise an interesting question into the dynamics of anxiety in the type of stressful situation which w-as the setting for this study and a possible area for further research. It also suggests that younger subjects may be less well able to deal with the initial stress of discovering they are not admissible to a program of their choice and that practitioners may want to attend more closely to the stress created w'hen such events occur to younger people. . .

PAGE 103

91 Hypothesis 2 Clients in higher compatibility clientcounselor dyads will rate the counseling session more favorably than will clients in lower compatibility client -counselor dyads . Procedures . A forv/ard stepwise' multiple regression procedure was performed with che subjects' rating of the interview as the dependent variable. The independent variables were as follows: 1. Subject-counselor compat ibi 1 iry 2. Age of subject 3. Being black 4. Years of schooling 5. Importance of being admitted 6. Reporting "no feeling'' about examination performance 7. Reporting feeling entrance examination not passed 8. Discrepancy from passing score 9. Length of interview 10. Being black comxbined with length of interview 11. Being black combined with discrepancy from passing score 12. Being black combined with reporting "no feeling" about examination performance 13. Reporting "no feeling" about examination performance combined with discrepancy from passing score

PAGE 104

92 14. Reporting "no feeling" about examination performance combined with importance of being admitted 15. Saw Counselor #2 combined with discrepancy from passing score 16. Saw Counselor ^4 combined with importance of being admitted 17. Felt entrance examination not passed combined with discrepancy fromi passing score Findings . Table 21 presents a summary of the multiple regression analysis done with the client's rating of the interview scores as the dependent variable. The table shows all seventeen of the independent \ariables regressed in the order in which they entered the analysis. The multiple correlation coefficient was significant at the entry of the first variable and remained so through the inclusion of the last variable to enter. Inspection of Table 21 shows that subject-counselor, compatibility was the last variable to be entered into the regression analysis and accounted for an additional 2.9 percent of the variance of the rating of the interview scores once all of the other variables had been entered into the analysis. Subject-counselor compatibility was not found to be an independent correlate of the rating of the interview scores for the entire sample (r = -.03, p = .40). Neither did subject-counselor compatibility appear as a

PAGE 105

93 rH r— 1 rH CO 1 1 o o 00 C CD CD • o • CO lO V CO . . . V . V H c O Ci, r , — { 1 — 1 1 — 1 CQ r-i o o Cj CO C^^ CO 0 r— i ^' Q 'H 9h >i 1 — i to to Q) •H CO rH -P 1 > 1 1 1— -: r. rH O _p o 00 in t> h~4 CO CO o t> •rH lO U J LO '~' cs r~l <, > • H Q) C S< C 00 C3 1—1 00 rH (M o •r-1 rH ^ ^ J_i o CO •H > rH |. > ' 1^ t J , — ( CD < Sh Q CJ r-1 a 00 rH o CD CD uo 00 in uo •H P E E CO -p c o xs a; E rH a, O -H !h C Cu +^ a c c -P a o Ki rH C 1 r~ -P oni a: O o •H p E O OJ •H rH a—, -p •1— i ^H -a u a 4H CD ro j; tP C-H a; o o c o w c >; •H G >. -p •H E o 0 o 6 c o CD •H CO p •H w w O E ti r-i "o i CC 5 Ch be OJ a b£ CC 0 -p c OJ -p _j o >. •H •H 'rH rH p c Sh br. > O -p rC o Vi > M iH fi C/3 in c !h HI •H 1 a •H ci CO p a rH cr; tH JO CO

PAGE 106

94 0 c o 0 t3 br [in c3 W C Cti O •H >> -P 0 fn ?-i X) V M 0 0 o c •H <: > 0 c r— I Ci d -H 0 a 0 -d 0 C rH 0 ^ 0 -H i-i > 0 a P 0 w c m o CO • CO v-' CO CO 00 C<1 CO en CO +-> o 0 •I — ' |C3 w 0 c CO CO (X) CO o CO CD 00 CO CO r-H o o CO • CS w CO CO CD CO o I — I o o CD CO CO in o CO CO o c CO CO in 1 ! 1 c IT. to CD CD CO CO .-H o LO CO to o V CO CO CO o c r— ' 1—1 CT; C-1 V" o O CD CD CD CO c & 0 bj; •H 0 c > •H 0 4-1 0 f— 1 •H 0 0 0 -H 4-> C 0 0 o 0 S C c d +-> tr-. re H 1 — i t>; •J >^ 0 o O c fl 0 d 0 E C m -a r-l 0 rH 0 r 3 S n_ o 4-1 Xj 0 •H 0 (— b f— 0 be •H 'J: x; G C O Uj •H b£ 4J -H •P c: 75 0 c; 0 d 0 •H 0 a 0 CO 23 C<1 CO o CD CO CO CM o o CD CD If: 0 -P +^ 0 ,^ C G be d •H b.C -P C 5h O I— I tp a 0 !m 0 0 0 G d d rH 5 -Q ^1 C bC C H 0 CC ^ CI, JO CD i> CO

PAGE 107

=5 C o o I — i Ki 3 c d 0) Oh ra a -H -a u >, -p o worn CM 00 C2 o G •H CD Q O CO CM CO CD C D c o a ^' O Tj -P O rH Cj 'i-l 0) o X a. K a; c CO CO • CD o •d CO nH -H 0) d o •p m > p 0 ^ CO CD ^ ^ ^ 6 O CO d c • 1 o r-* iH CO 1— 1 -d • o o I> < > CO CO CM d lO , t> CD CD P Ih CO CO O d H > O P' CO CO c o o d o c o < Di CD O rH I O O CM C o O CD O 00 CO o o CD CD LO CO o lO CTj C^5 CO CC CD CD CD CD 5:' (D •H 1 > E O P C 1—1 p CD 73 (D P Q) a2 PI a pi 0 CO >= o G p> o -p P . c -p c •p p •P >> 0 •p o 0) f-i c> o o rP •p o o •P p" CD C 1 ^ 0) 'fcL O c3 C3 C.C •P p> pi C; c CJ -P H •P •H p bJ"; -P P •p CD ci o ^-^ E S 0 O •rn (D P •P G uQ g 0) O y, 0 C[) .1 •p ai Q) cp •P h3 (2; p., V o CO 00 CO o o LO CD l>

PAGE 108

96 significant correlate of the rating of the interview scores within any of the categorical subgroups. As one of a large number of variables, subject-counselor compatibility did contribute to the magnitude of the multiple correlation coefficient. To that extent it is of some use in predicting higher rating of the interview scores for subjects. On the basis of these findings, however, the hypothesis that subjects in higher compatibility dyads would rate the counseling session more favorably than subjects in lower compatibility dyads cannot be upheld. Discussion . Further inspection of Table 21 reveals that the first three variables to enter the regression analysis were interaction variables. Although their effect on the multiple correlation coefficient was to increase its magnitude, the fact that the correlation of each variable with the rating of the interview scores was negative indicates that the combinations of factors represented by those variables had the effect of reducing rating of the interview scores. All three of the variables involved the subject's discrepancy from the required passing score on the entrance examination. The combination of having seen Counselor 42 and having missed passing the entrance examination by a larger margin had the greatest impact on rating of the interview scores, accounting for nearly 16 percent of the variance of those scores. The effect was to rate the interview lower than when that combination was not present.

PAGE 109

97 Again in combination with greater discrepancy from the required passing score on the entrance examination, the subjects' feeling they had not passed the entrance examination accounted for about eight percent of the variance of the rating of the interview scores. Subjects reporting "no feeling" about their performance on the entrance examination and still in combination with having greater discrepancies from required passing score on the entrance examination • contributed another five percent of the variance on this outcome measure. The effect of these three interaction variables taken together as the three which entered the regression analysis first and in order, accounts for approxin:ately 29 percent of the total variance of the rating of the interview scores or about 60 percent of the variance accounted for by all of the variables used in the analysis. By contrast it can be seen from the sum.mary table that the discrepancy from the required passing score alone had very little impact in the overall analysis, entering as the sixteenth variable and accounting for only .4 percent of the variance in the rating of the interview scores. The relationship to the rating of the interview scores was negative and therefore consistent in that greater discrepancy from the passing score alone had a slightly depressing effect on the rating of the interview scores. Subjects who reported they felt they had not passed the entrance examination tended to rate the interviews more favorably, with this category of response accounting for

PAGE 110

9S about two percent of the variance of the rating of the interview scores. Subjects who reported "no feeling" about their performance on the entrance examination tended to rate the interviews less favorably, but affected the variance accounted for within the analysis by only about .2 percent. This cluster of variables carrying information about subjects' discrepancy from xhe required passing score, their feeling about their performance on the entrance examination, and having seen Counselor ^2 accounted for approximately 65 percent of the variance of the rating of the interview scores accounted for by all of the variables combined. All • but two percent of that variance was based on negative relationships of the independent variables and had the effect of producing lowered ratings of the interviews. The importance o.f becoming admitted to the program entered the regression analysis on the fourth step. It accounted for only about two percent of the variance of the rating of the interview scores and higher levels of im.portance were indicative of more favorable ratings of the interviews. When combined with subjects reporting "no feeling" about performance on the entrance examination, however, it accounted for about seven percent of the variance of the rating of the interviews and w^as associated with higher interview ratings. It would seem, therefore, that subjects who placed more importance on becoming aciinitted tc the program and reported "no feeling" about performance on the entrance

PAGE 111

99 examination rated the interviews more favorably than subjects who did not fall into that combined category. Black subjects as a whole showed no tendency to rate the interviews more or less favorably as a group (r = -.09, p = .19). There was a tendency for black subjects to rate the interviews less favorably when they had greater discrepancies from the required passing score and when they reported "no feeling" about performance on the entrance examination with those com.b inat ions of characteristics enterin the regression analysis at steps 9 and 11 respectively (see Table 21). Black subjects who participated in longer interviews tended to rate the interviews more positively with that interaction variable entering the analysis at step 7. The sum of the variances accounted for by variables which involved black subjects is approximately thi'oe percent, of which 1.5 percent is related to higher ratings and 1.5 percent related to lower racings of the interviews. It would appear that for black subjects, other characteristics which are associated '.vith particular subjects are more imioortant than being blacl; and that the effects of being black were offsetting in this study. There v/as a slight and significant tendency for older subjects to rate the interviews more favorably (r = .23, v^.Ol). As a variable, subject age entered the regression analysis on step 5 and accounted for about two percent of the variance in the rating of the interview scores.

PAGE 112

100 Although the number ol years of schooling entered the regression analysis at step 10 and contributed to the A^ariance in the rating of the interview scores, that contribution was only .2 percent. The Pearson product moment correlation of subjects' years of schooling with their ratings of the interview was not significant(r = .04, p = .34). Realistically, the number of years of schooling seems to have had no real impact on interview ratings.Hypothesis 3 Clients in higher compatibility clientcounselor dyads will indicate a preference for seeing the sam.e counselor again significantly more often than will clients in lower compatibility client-counselor dyads. Procedures . A linearized discriminant function analysis was performed on all of the independent variables with respect to the categorical outcome variable choice of counselor for a future counseling occasion. The independent variables were as follows: 1. Subject-counselor compatibility • 2. Age of subject 3 . Being black 4. Years of schooling completed ' 5. Importance of being admitted 6. Reporting "no feeling" about entrance exajTiinat ion 7. Reporting feeling entrance examination not passed

PAGE 113

101 8. Discrepancy from passing score 9. Length of interview 10. Saw Counselor #2 11. Saw Counselor 12. Saw Counselor #4 13. Saw Counselor f^'o 14. Saw Counselor P2 combined with age of subject 15. Saw Counselor f4 confoined with subject-counselor compatibility 16. Saw Counselor ^^2 combined with discrepancy from passing score 17. Reported ''no feeling" about test performance combined with importance of being admitted. The discriminant analysis procedure attempts to predict which category on the outcom.e variable should have been chosen by each subject. To accomplish this, the procedure computes a weight coefficient by which each subject's standardized score (z) on each predictor m.easure is multiplied for assignment into each of the possible dependent variable categories. The probability that a particular subject belongs in a particular category is then com.puted and the prediction made. In doing so, the procedure takes into account the category each person selected and that person's score on the predictor being considered. The coefficient thereby reflects the subject's actual performance on the predictor variables with respect to the category

PAGE 114

chosen within the dependent variable. When the data from this study were entered into the discriminant analysis procedure, the parameters of the procedure rejected the choice of "a different counselor" from the analyses as it was represented by only one case. All results of the analysis are based therefore on the statistical assumption that only two choices were available to the subjects, those of ''this same counselor" and "no preference" on the indicator for preference of a particular counselor for some future occasion. Findings . Inspection of Table 22 shows that the coefficients for the measure of subject-counselor compatibility for each choice category are of the same sign and nearly the same magnitude, indicating that the dyads' compatibility scores would not differentially predict between the two categories which were considered available as the analysis was performed. The hypothesis that subjects in higher compatibility dyads would choose the response "this same counselor" significantly more often than subjects in lower compatibility dyads cannot, therefore, be confirmed. A given subject was more likely to choose the category with the higher weight coefficient as each independent variable was considered (when coefficients are both negati\'e, the smaller absolute value is considered higher . ) Discussio n . Further examination of Table 22 reveals the following:

PAGE 115

103 TABLE 22 Cate^-or Leal Weight Coefficients Independent Variables Choice of Counselor for Future Counseling Same Jo Preference Sub jectcounselor compatibility .8834 Length of interview -.1399 Discrepancy from passing score -1.1534 Age of subject .2001 Importance of being admitted -.2074 Years of schooling completed .0358 Being black ' 3.765 "No feeling" about entrance examination -9.6486 Felt "did not pass" entrance examination 1.7520 Saw Counselor #2 5.5537 Saw Counselor =3 ' 7.3370 Saw Counselor 1=^4 • 6.0549 Saw Counselor -5 4.2864 Saw Counselor ~2 combined with age of subject .0048 Saw Counselor -4 combined with subject-counselor compatibility -.0'^99 .9461 -.2000. -.1672 .5892 . 0013 . 1590 1 .6105 7. 5867 .2616 16.2402 7.6686 11 .0627 4.73S0 -.4053 -.4302

PAGE 116

Table 22 Continued Independent Variables Choice of Ccanselor for Future Counseling Same No Preference Saw Counselor #2 combined with discrepancy from passing score -.0123 .0009 Reported "no feeling" about test performance combined with importance of being admitted .1321 -.0818

PAGE 117

105 1. Subjects with higher discrepancy scores from the required passing score were weighted as more likely to choose "no preference" for a future counseling occasion. 2. Subjects who placed more importance on being admitted to the program were weighted as more likely to choose "no preference." 3. Black subjects were weighted as more likely to choose "this same counselor." 4. Subjects who indicated "no feeling" about entrance examination performance were weighted as more likely to choose "no preference." 5. Subjects who felt they had not passed the entrance exam.ination were v;eighted as more likely to choose "no preference . " 6. Subjects who saw Counselor #2 were weighted as more likely to choose "no prefer once . 7. Subjects who saw Counselor were weighted as more likely to choose "no preference." The fact that 70 percent .of the subjects indicated a preference for seeing the same counselor for a future counseling occasion suggests that caution should be exercised in trying to interpret these findings. Only those variables for which there were clear differences in coefficient values have been noted as a guide for interpreting the table.

PAGE 118

CHAPTER V SUMMARY AND CONCLUSIONS Summary Five experienced white male counselors engaged in a total of 95 stressful interviews daring which a black or, white female applicant to a community college Associate Degree Registered Nursing program was informed she had not passed an entrance examination at the necessary level to qualify for admission to the program. The study was conducted in an already established counseling setting to which certain elements were added for the purpose of conducting the study. A consistent data gathering system was used before, during, and after the interviews. Each counselor was allowed to conduct the interviews in his own usua manner. A system for rotational assignment of subjects to counselors, privacy and continuity of the interviews, and annonymity of the subjects was used. The client-counselor interpersonal relationship was the general area of concern within which this study was conducted. A theory of interpersonal relationship needs was prepared by Schutz (1960) and was accompanied by a relatively easy to use instrument called the Fundamental Interpersonal Relationship Orientation Behavior. V^ithin this theoretical formulation, Schutz forwarded the concept 106

PAGE 119

107 of reciprocal compatibility and a mothod of determining the reciprocal compatibility of any pair of persons (dyad). Reciprocal compatibility was considered to exist when two persons interpersonal relationship needs were in complementary balance. The reciprocal compatibility of a client and a counselor in the area of warmth or affection seemed to hold relevance for counseling. This research attempted to assess the specific applicability and use. fulness of the concept of reciprocal compatibility in a stressful counseling situation. Three outcome measures were selected for use in this study. They were the client's anxiety reduction over the course of the interview, a rating of the interview by the subject immediately after the interview had been completed, and an 'indication by each subject of her preference for seeing the same counselor again should counseling seem appropriate in the future. Reciprocal compatibility was the independent variable of primary concern in this study. Other factors which were thought to have possible effects on the outcomes of the interviews were measured and treated as additional independent variables. Those factors were the length of the interview, the discrepancy between the subject's entrance examination score and the required passing score, the subject's age, the number of years of schooling completed by the subject, the importance the subject placed on being admitted to the program, the subject's race, and the subject's feeling

PAGE 120

108 about her performance on the entrance examination. The three hypotheses tested within the scope of this study were as f ol lows : 1. Over the course of the interview, clients in higher compatibility client-counselor dyads will show greater reduction in anxiety than will clients in lower compatibility client-counselor dyads. 2. Clients in higher compatibility client-counselor . dyads will rate the counseling session more favorably than will clients in lower compatibility . client-counselor dyads. 3. Clients in higher compatibility client-counselor dyads will indicate a preference for seeing the same counselor again significantly more often than will clients in lower compatibility dyads. Using a multiple regression analysis procedure, subject counselor compatibility was found to be related to anxiety reduction at a very low level. Subj i -counselor compatibility accounted for only about two percent of the variance in the reduction of anxiety scores and was surpassed by the importance the subject placed on being admitted to the program. A total of only about 13 percent of the anxiety reduction score variance was accounted for at a significant level by a combination of six of the independent variables. The analysis also showed greater anxiety reduction occurred among subjects who were seen longer by the counselor and among those subjects whose scores on the entrance

PAGE 121

109 examination were more discrepant from the required passing score and who expressed no feeling about their performance on that examination. Black subjects and subjects who expressed no feeling about their performance on the entrance examination (except those who had failed by wider margins) showed a slight increase in anxiety over the course of the interview. As the reduction in anxiety over the course of the interview was well distributed ar.d consistent across counselors, the inability of the independent variables to show more than minimal relationship to anxiety reduction shows that factors other than those m.easured and statistically treated were operating. One of the most obvious explanations is that the person's own defense system begins to bring down the anxiety as it would under any threatening circumstance. A research design which would allow for a waiting period of 20 to 30 minutes after receipt of threatening information for part of a treatment group and no such waiting period for a control group could provide some information about anxiety reduction in the face of tlireatening information. The investigation of the specific anxiety reducing techniques used by specific persons in a threatening situation could also provide useful information. Grater (1964) suggested that clients prefer warm-personal counselors for personal counseling and more pragmatic and action oriented counselors lor educational-vccai, ional counseling. The P'^-rce]) i !a.'Ii :

PAGE 122

110 not passing an entrance examination for a particular prograr could be seen either way ana was not assessed by this study An approach based on Grater's finding could be -useful in determining the concomitants of anxiety reduction in a "stressful" situation dependent upon the perception of the situation as a personal problem or a practical problem. A multiple regression analysis was done with respect to the subjects' rating of the interview. All of the variables, including a variety of combination variables indicative of specific effects occurring for particular subgroups, were included in the regression with the multiple correlation coefficient remaining significant. The major conclusion drawn from the analysis of the ratings of the interviews was that the study showed more about factors which were, related to lower ratings of the interviews than It did about what factors were related to higher ratings of the interviews, except as reverse inferences might be made. Subjects who failed to meet necessary passing scores on the entrance examination and saw one particular counselor, subjects who were either unsure of their examination performance or who felt they had not passed the examination, and combinations of those characteristics had the greatest negative impact on interview ratings. These factors represented over one-half of the variance accounted for by all of the factors. Subject-counselor compatibility was not supported as related to better ratings of the interviews.

PAGE 123

Ill The situation in which the subject received information about her inadmissibility to the nursing program was rated more negatively by subjects whoso performance on the entrance examination was poorer and whose prior perception of what her performance had been was less positive. Results also seemed to suggest that something specific to the nature of the relationship between the individual subjects and counselors, not measured within the scope . of this study, had considerable impact on the ratings of the counseling interviews. The overall results might well have been clouded by this isolated source of variability which was not indicative of the entire pool of interviews. Conclusions about results are, therefore, tenuous. There was evidence, nonetheless, that subjects whose scores on the entrance examination were more discrepant from required passing scores rated the interviews less favorably. A control study for the general level of facilitative conditions offered to female clients by the counselors who participated in the study showed comparability am.ong counselors. The significant effect of having seen a particular counselor on the. subject's rating of the interviews suggests that careful and precise control of counselor variables beyond those used in this study would be required. A discriminant analysis procedure was done to discover if subject-counselor compatibility was related to subjects indicating a preference fo return to the same counselor on some future occasion if counseling were considered appropriate.

PAGE 124

112 The hypothesis that compatibility would be an indicator of a greater frequency of an expression to that effect by subjects was rejected. The distribution of actual responses was so heavily weighted in favor of subjects indicating a preference to return to the same counselor that any conclusion would be highly suspect. Results did show a tendency for subjects who had not passed the entrance exam.ination by wider margins, who v/ere unsure of their examination performance, who felt they had not passed the exam.ination, who placed more importance on being admitted to the program, and who had seen particular , counselors to not indicate a preference for seeing the same counselor on another occasion. The tendency of subjects with these characteristics to not indicate a preference to see the same counselor on some future occasion suggests again that the style of the counselor may be important. If subjects who possess these characteristics see their failing to meet adm.ission requirements as a personal problem, their assignment to a counselor who perceives the situation as an educational-vocational problem and thereby behaves in a more pragmatic manner could create a situation which a subject might prefer to avoid in the future. Assignm.ent of clients to counselors based on the client's perception of the nature of the problem and the known style of the counselor as suggested by Grater (1964) would seemi to be an avenue worthy of exploration under stressful circumstances such as those

PAGE 125

113 encountered by the subjects in this study. The overvvheiming tendency of subjects to indicate a preference to see the same counselor on some future counseling occasion was not indicated by inspection of the results of the discriminant analysis procedure. This general tendency may simply be a matter of most subjects having been satisfied with the counseling interview and only some being dissatisfied as might be inferred from their lack of preference to return to the same counselor. Conclusi ons The need to better understand the effects of the interprsonal relationship between .a client and counselor was the original impetus for this study. Beutler (1973) suggested that the patient-therapist dyad should be considered "... as a separate treatment whose effectiveness depends not only on the individual characteristics of patients and ' therapists but upon characteristics of their mutual compatibility'' (p. 305). The specific intent of this study was to show that higher levels of reciprocal compatibility in the area of affection (Schutz, 1960) of the subject-counselor dyad would be positively related to subjects' anxiety reduction, subjects' ratings of the interviews, and a greater likelihood of subjects to indicate a preference to return to the same counselor for future counseling. If these relationships did exist, there would be implications for more advantageous

PAGE 126

114 assignment of clients to counselors. The reciprocal compatibility in the -^rea of affection of the subject-counselor dyad as computed from the subject's and the counselor's scores on the Fundamental Interpersonal Relationship Orientation Behavior (Schutz, 1960) was found to be positively related to anxiety reduction in a stressful interview. The relationship was at a low level (r = .18, p< .05) and accounted for only a small part of the variance in the anxiety reduction scores of the subjects (2.3 percent) when other factors were considered. The use of reciprocal compatibility between a client and a counselor as a practical device for client-counselor dyad formation seems unwarranted when subject anxiety reduction in a stressful interview is the goal. Subject-counselor reciprocal compatibility in the area of affection was found not to be related to the ratings of the interview's by the subjects. As a tool for assigning clients to counselors to produce more client satisfaction in a stressful counseling interview, this index of compatibility seemiS to have no relevance. Subject-counselor reciprocal compatibility in the area of affection was found not to be related to subjects indicating a preference to return to the same counselor in the future. Assigning clients to counselors for stressful interviews with the idea that clients who are more compatible with their counselors will report a preference to return to that counselor in the future and might actually

PAGE 127

115 do so seems unwarranted. Overall, the hope that the reciprocal compatibility in the area of affection of the subject-counselor dyad would contribute to a more effective method for assigning clients to counselors for stressful interviews was not realized.

PAGE 128

APPENDIX A

PAGE 129

I in 1 U { 1 ,J i « 1! 5 fU a Li ii u I = £ Z.. if 5 I I i I t I if! fl * I till ^ u h If n I .Ii f in Hi ^1 ^ !i '4 i It Ml li i 1 fiii ii iHH V i ii n U I . . . . I I u ^n. i i 9 f Li I 6 Ml II in 1 [ J i I if II I 1 1 = Ii J f Li ^ ^ f { n 1 : 1. !i in t\ n i\ n. i r 1= i £ J f 1 ill ; F i 1ft 3 I 2. lit 1 ^ I i I 5 If I It i ! 5 J. n i I i 1 fl 5= 117

PAGE 130

118 O ^: C r -1 (.1 c r-* t-i :* o > o Co L. c CI > t s c X 'S o O o v. c p o L> a. •H n: to 1 0. D >< a O C a. -J • I.; O ( . 1 0 c C 0 c r-l 'l o 7 CJ o 1, cr 6 1 H "X b t." (.: 0 o •H S 0 EH > 0 L. c: a) n 771 o ci C 4-> > ;: c n a c .i 0 t. r > E a I -II I I 4^ X t I £ • p •J 1: O r3 +3 ^

PAGE 131

APPENDIX B

PAGE 132

CONSENT FORM The purpose of the research In which you're being asked to participate is to discover if people's preferences for the way they interact with other people make any difference in how (hose interactions turn out. If these preferences do make any difference, we can then arrange for students and faculty to work together in combinations which best benefit the student Participation in the research will involve the following: 1. Completion of a questionnaire about how you like • to behave with other people and how much you like to do so. (The questions are stated in general terms and do not refer to any specific person). On the same form you'll be asked to give some information about yourself and about how important it is for you to get into the program you've applied for, and how you think you did on the Comprehensive Test of Basic Skills'. 2. Some of you will be asked to complete additional questionnaires later on. They will occur at the time you are inform.ed of your test results. None of the information gathered in the research will become part of your record at Santa Fe . None of the information will be seen by anyone other than the researcher and his assistants, and none of the information will be used in any way to influence your admission into any program for which you are now or may apply for. in the future. Results of this research will contain no specific information about anyone as an individual nor will information gathered from individuals be available to anyone except the researcher and his assistants. These research procedures are being done in addition to what we normally do as part of the testing and admission procedures for Health Related Programs. You will not be treated any differently by being I participant in this research than if you were not a participant in the research, except that you will be asked to provide some information and complete some questionnaires as outlined above. 120

PAGE 133

121 I, the undersigned, do understand the purpose of the above investigation project. The study proposed has been defined and explained to me by the investigator whose name is signed below an.d I agree to participate in this study. I understand that I may withdraw this consent at any time. Subject's Signature Date I, the undersigned, have defined and explained this study to the volunteer. Investigator's Signature Date V/itness Date

PAGE 134

122 INSTRUCTIONS TO APPLICANTS The results of the Comprehensive Test of Basic Skills will not be ready for seven to ten days. To get your results, please use the following ^uide: Program Applying For Dental Assistant Cardio-Vascular Technology Licensed Practical Nursing PiBgistered Nursing Respiratory Therapy Contact for Appointment Secretary Health Programs Counseling Office 377-5161 South Campus Extension 202-203 Nuclear Medicine Technology Radiologic Technology Health Related Programs 377-5161 South Campus Extension 208 Mr. Robert Short Thank you again for assisting us to gather information we^hope-will improve the learning climate\it Santa Fe . The information is con f iden tical and will not be used in any way to influence your admission lo any program. ^

PAGE 135

APPENDIX C i

PAGE 136

COUNSELOR GUIDELINES The study in which you are participating will best be served if you conduct the session in your own way , so please be yourse If . Three procedures* which v;ould not usually be a part of the counselingsession will be necessary, however. They are shown below in the listing of procedures and functions of the interview. Please follow the sequence of procedures in the order specified but do not in any other way restrict your usual style with respect to the functions of the interviews. Functions and Procedures CAUTION: Do not inform the student about whether or not she has passed the test until you are both in the office and seated, with you behind the desk and she at the side of the desk with some clear desk space between you. *Procedure 1 Address sub j ect " saying : " Subject ' s name , you didn't pass the test." Gently sliding the Feeling Word Check List onto the desk in front of the subject and laying a pencil on top of it, say, "I know this seems like a very awkward time to ask you to do this, but would you please check all of the words on this green sheet which reflect how you feel right at this very moment . It will only take you a few minutes to do this, and knowing' just how you feel right now will probably help you make the most of this session." (Optional, only if the subject does not begin to complete the check list): "Would you begin novv, please." Note: While she is filling out the Feeling Word Check List, please fainiliarize yourself with the materials in her folder making an effort not to distract her from completing the check list. When the check list has been completed remove it from view without inspection or comment, noting the time of completion of the Feeling Word Check List #1 in the space provided on the blue Process Notes sheet. 124

PAGE 137

125 Funct ion Function Func Lion : Note : *Procedure 2 Discuss her scores, as specifically as seems appropriate and the implications of the scores, the areas of strength and weakness reflected by them, etc.. and the implications of the scores for possible future admission to the program. Discuss her educational-vocational goals in light of the current situation to whatever extent and in whatever manner seems fitting and appropriate for you both. Discuss any other matters which seem of importance to you and/or her within the general context of not having met requirements for the Registered Nursing program. y/hen you both agree that the discussions have reached a natural termination point,' please complete Procedure #2. Ask her to complete the yellow form of the Feeling V/ord Check List reminding her to check the words that reflect her feelings right nov/. Note: ^Procedure 3 While she is doing this, please complete the carbonized counseling notes form included in the folder, note the time of the beginning of the yellow form (r-^2) of the Feeling'word Check List in the appropriate space on the Process Notes form, return test protocols to folder, etc., making an effort not to distract her from completing the check list. When the check list has been completed, remove it from view without inspection or comment, and give her a copy of\he counseling notes form. Escort her to the secretary in order to ' insure the complet i on of one more procedure required for the study which will be administered by the secretary. Please leave the subject with the secretary and return to your office and complete the Process Notes sheet. When the Process Notes sheet has been completed, please assemble the green and yellow forms of the Feeling Word Check List and the Process Notes form and staple them together being sure that your name and the subject's' name are on the Process Notes form.

PAGE 138

PROCESS NOTES As professional counselors, v.'e are all interested in what helps students, especially under difficult circumstances such as those under which these interviews are being held. Often our subjective observations, comments, feelings, intuitions are of as much value as any "objective" measures we might use. Please use this sheet to make note of anythin that happened which you feel, think, believe or otherwise consider to have been potentially important to the client, yourself, or the relationship between you with respect to the outcome of this interview. Counselor Student FV.'CL Date Fl (green) #2 (yellow)

PAGE 139

APPENDIX D I

PAGE 140

BARRETT-LENNARD RELATIONSHIP INVENTORY CODE: DATE; RELATIONSHIP INVENTORY -FORM 0S-:vI-64 Below are listed a variety of ways that one person may feel or behave in relation to another person. Please consider each statement with reference to your present relationship with your counselor. Mark each statement in the left margin, according to how strongly you feel that it is true, or not true, in this relationship. Please mark every one. IVrite in +3, +2, +1, or -1, -2, -3 to stand for the following answers: +3: Yes, I strongly feel that it is true. .+2: Yes, I feel it is true. +1: Yes, I feel that is probably true, or more true than untrue. -1: No, I feel that it is probably untrue, or more untrue than true. -2: No, I feel it is not true. -3: No, I strongly feel that it is not true. 6 He respects me as a person. He wants to understand how I see things. His interest in me depends on the things I say or do . He is comfortable and at ease in cur relationship. He feels a true liking for m.e . He may understand my words but he does not see the way I feel. Whether I am feeling happy or unhappv with myself makes no real difference to the way he feels about me . 12S

PAGE 141

I feel that he puts on a role in fi^ont with me. He is impatient with me. He nearly always knows exactly what I mean. Depending on my behavior, he has a better opinion of me sometimes than he has at other times. I feel that he is real and genuine with me. I feel appreciated by him. He looks at what I do from his own point of view. His feeling toward me doesn't depend on how I feel toward him. It makes him uneasy when I ask or talk about certain things. He is indifferent to me. He usually senses or realizes what I am feeling. He wants me to be a particular kind of person. I nearly always feel that what he says expresses exactly what he is feeling and thinking as he says it. He finds me rather dull and uninteresting. His own attitudes toward some of the things I do or say prevent him from understanding me . ° I can (or could) be openly critical or appreciative of him without really making him feel any" r^if _ ferently about me. He wants me to think that he likes me or understands me more than he really does. He cares for .me. Sometimes he thinks that I feel a certain way because that's the way he feels. He likes certain things about me. and there are other things he does not like.

PAGE 142

130 _2S. He does not avoid anything that is important for ovir relationship. 29. I feel that he disapproves of me. 30. He realizes what I mean even when I have difficulty in saying it. 31. His attitude toward me stays the sane; he is not pleased with me sometimes and critical or disappointed at other times. 32. Sometim.es he is not at all comfortable but we go on, outwardly ignoring it. 33. He just tolerates m.e . _34 . He usually understands the whole of what I mean. _35 . If I show that I am angry with him, he becomes hurt or angry with me, too, _36 . He expresses his true im.pressions and feelings with me. _37. He is friendly and warm with me. _3S. He just ta.kes no notice of some things that I think or feel. _39. How much he likes or dislikes me is not altered by anything that I tell him about myself. _40. At times I sense that he is not aware of what he is really feeling with me. _41. I feel that he really values me. . _42. He appreciates exactly how the things I experience feel to me. 43. He approves of some things I do, and plainly disapproves of others. 44. He is willing to express whatever is actually on his mind with me, including any feelings abou: himself or about me. 45. He doesn't like me for myself. At times he thinks that I feel a lot more strongly about a particular thing than I really do.

PAGE 143

"1 1 47. V.'hether I am in i^ood spirits or feeling upset does not make him feel any more or less appreciative of me. 48. He is openly himself in our relationship. 49. i seem to irritate and bother him. 50. He does not realize how sensitive I am about some of the things we discuss . 51. n'hether the ideas and feelings I express are "good" or "bad" seems to make no difference to his feeling toward me. 52. There are times when I feel that his outward response to me is quite different from the way he feels underneath. 53. At timies he feels contempt for me. 54. He understands me. 5o . Sometimes I am, more worthwhile in his eyes than I am at other times. 56. I have not felt that he tries to hide anything from himself that he feels with me. 57. He is truly interested in me. 58. His response to me is usually so fixed and auto' matic that I don't really get through to him. 59. I don't think that anything I say or do really changes the way he feels toward me. 60. 'iVhat he says to me often gives a wrong impression of his whole thought or feeling at the time. 61. He feels deep affection for me. 62. When I am hurt or upset he can recognize my feelings exactly, without becoming upset himself. 63. V.'hat other people think of me does (or would, if he knew) affect the wav he feels toward me. 64. I believe that he has feelings he does not tell me about that are causing difficulty in our relationship .

PAGE 144

132 BARRETT-LENNARD RELATIONSHIP INVENTORY SCORING SHEET 64 Item Forms Level of Empathic Unconcl. of Congruent Regard Understanding 'Regard Items Score Items Score Items Score Items Score 1 2 7 4 5 10 15 12 13 18 23 20 25 30 31 28 37 34 39 36 41 42 47 44 57 54 51 48 61 62 59 56 SUM A Items Score Items Score 9 . 6 . 17 14 21 22 29 26 33 38 45 46 49 50 53 58 SUM B' B = -B' A + B Grand Total All Scales Items Score 3 11 ' 19 27 35 43 55 63 Item.s Score 8 16 24 32 40 52. 60 64

PAGE 145

appe:\T)ix e

PAGE 146

CODE: .... NO: .... FEELING WORD CHECK LIST ^^1 DIRECTIONS: The words printed below describe different kinds of feelings. Please check the words w'hich describe how you feel RIGHT NOV; AT THIS VERY MOMENT. Some of the words may seem alike but we want you to consider each one and to check ALL of the words that describe your PRESENT feelings . "1 1 . "1 "1 r-l~L +" 1 1 /~i o "v^ 't' f~~^ /*! iiL,nLnearLeu 1 Q /~\T70'K*TirV~i o T >vi o r\ KjvKf i. Vv i 1 1J-tilt; U. 9 k i n d 1 y panicKy Q o . I o n e 1 y O 1 z 1 . furious A 4 . ,j oyf ul zz . peaceful r o . loving 23 . frightened b . jealous 24 . pleasant 7 . mad 25 . friendly o o . insecure 26 . ratt led Q ZJ . mean fretful 10. hopeless 28 . sad 11 . merry 29. fearless 12 . helpless 30. secure 13. miserable 31 . fearful 14. happy 32 . sent imental 15. nervous 33 . ea.sygoing 16 . grim 34 . serious 17. overconcerned 35. desperate 18 . gloomy . 36 . shaky 134

PAGE 147

135 37. cross 3S . solemn 39. cool 40. steady 41. contrary 42. tender 43. contented 44 . tense 45. complaining 46. ; terrified 47. cheerful 43. threatened 49. charming 50. _____ thoughtful 51. calm 52. unconcerned 53 . bitter 54 . uneasy 55. angry 56 . upset 57. agitated 58 . warm 59. afraid 60 . worrj'ing

PAGE 148

136 Code : No : FEELING WORD CHECK LIST n2 DIRE CTIONS: The words printed below desc ribe different kinds of feelings. Pi ease check the words which describe how you feel RIGHT ' NOW AT THIS VERY MOMEN' Some of the words may seem alike but we want you to consider each one and to check ALL of the words that describe your PRESENT feelings . 1 . lighthearted 19. gloomy lonely 20. panicky 3 . kindly 21. peaceful 4 . loving 22 . furious 5 . joyful 23 . pleasant 6 , mad 24. frightened 7 . jealous 25. ra 1 1 led o o . mean 26 . friendly 9 . insecure 27. sad 10 . merry 2S . fretful 11 . hopeless 29. secure 12 . miserable 30. fearless 13. helpless 31 . sentimental 14. nerv.ous 32. fearful 15. happy 33 . serious 16 . overconcerned 34 . easy-going 17. grim 35. shakv 18. overwhelmed 36 . desDer ate

PAGE 149

1 137 37. solemn 38 . cross 39. steady 40. cool 41. tender 42. contrary 43. tense 44. contented 45. terrified 46 . complaining 47. threatened 48. cheerful 49. thou;;htful 50. charming 51. unconcerned 52 . calrr. 53. uneasy 54 . bitter 55. upset 56 . angry 57. warm 58. ' -agitated 59 . worrying 60. afraid i

PAGE 150

138 SCORING KEY FOR AFFECT ADJECTIVE CHECK LIST The anxiety-plus words, scored 1 when checked are afraid shaky desperate tense fearful terrified frightened upset nervous worrying panicky The anxiety-minus words, scored 1 when not checked are; calm loving cheerful pleasant contented secure happy .' steady joyful thoughtful

PAGE 151

APPENDIX F

PAGE 152

INTERVIEW RATING SCALE Instructions It is essential that all ratings be made by you as honestly as possible. Your task is to rate the counseling experience you have just completed. Rate your experience in terms of "what is." not "what ought to he." Look at the following example which has been filled out to show you how to use the scale. Always Occasionally Never 1. The counselor is a nice person. The person who marked this thinks that his counselor is occasionally a nice person. You are to answer all the questions by placing an "X" on the line which best expresses what you feel ...oout the interview at the present time . Use any one of the five lines for your rating of each statement according to the extent it holds true in your own experience . Here are some hints to help you: 1. V/ork rapidly. There is no time limit, but do not spend much time on any one item. 2. Mark all items according to your feelings today. Now proceed to answer the questions on the following pages. REMEMBER : 1. Try to answer each question as honestly as vou can right now. 2. This is not a test. 3. The counselor will not know who has completed thisscale. 140

PAGE 153

141 INTERVIEW RATING SCALE Code : No : . ' Date . . , Items Scctle Always Occasionally Never 1. I felt that the counselor accepted me as an individual . I felt comfortable in my interview with the counselor . 3". The counselor acted as though he thought my concerns and problems ' were unim-portant to him. 4. The counselor acted uncertain of himself. 5. The counselor acted cold and distant. 6. I felt at ease with the counselor . 7. The counselor seem.ed restless while talking to me. 8. In our talk the counselor acted as if he were better than I . 9. I believe the counselor had a genuine desire to be of service to me. 10. The counselor was awkward in starting our interview. 11. I felt satisfied as a result of my talk with the counselor. (continued on other side

PAGE 154

142 Always Occasionallj'' Never 12. The counselor was very patient. 13. Other students could be helped by talking with a counselor. 14. In opening our conversation the counselor was relaxed. 15. I distrusted the counselor. 16. The counselor insisted on being always right. 17. The counselor gave the impression of feeling at ease . 18. The counselor acted as if he had a job to do and didn't care how he accomplished it. 19. If you should decide to seek out a counselor on another occasion, would you seek out (check one only): a IZU CO This same counselor No preference A different counselor

PAGE 155

143 SCORING GUIDE FOP, INTERVIEW RATING SCALE Item Always Occasionally Never Score 2 _1 3 -1 4 -1 6 2 1 7 -1 -1 8 _iL_ _:i 12 2 0 13 2 1 14 f__ ^ 15 -1 -1 16 ' -1 -1 17 ^ 3 0 IS -1 -1 -1 -1 -1 0 -1 -1 -1 0 2 0 12 1 1 2 0 -1 -1 0 12 -10 2 0 -1 -1 0 0 2 0 0 1 -1 -1 -1 0 -1 -1 0 -1 -1 0 2 -10 2 _0 -1 -1 -10 2 Total Score

PAGE 156

APPENDIX G

PAGE 157

CO < oo o « CO < > K 2". r-i Q P4 O o M <: o o r-l H 0) 0 E '-I O D 3 O O S cr CO i CD uO (M 0 H 1 O O r-l c i 11 II II II ^ o 1 1 ci g; o >* CO i i c 1 1 a' 1 0 -)_) j d) J 1 ' — f lO o C3 p bS) i II II i o 'I 'i i — 1 r-H I— 1 o r-HCTj • 1 II II I 11 It *n o' L. J /] r-l' O O O aa ^ 0 II II Uj \j \^ •H O '-^ > CO CD LO O ^ O O ^ O 0 -P G/ cr. rt 3 O O U O II II LO CC LT o rP C7) • II II 2= a o P P> 0) o p :3 X -a c G/ < K 10 CO o o • 11 II 2: a CD rin o 'y^ I cc in o rP CTJ • ii li ^5 a CC lO o I II II 7^ a O LO CNl O CD CM 00 r-3 CD I II 00 o in ^ r-^ CD • 2: a. in c m c^] CD LT ex: O CD • o • II n I II i> o ^ cc. L'^ -g^ CO in o O CD • rH CT) • I li I o (1) a p > 0 0 P 'Xj p c QJ r-p P" 5 p fp cl p 0 (— ( O V-i X H •p =p (— Pi QJ X3 0 M 7i x: rP 0 c p t— 1 o in .-H CD CD Ln o I II II 2: a ^ ^ O CD • ^ O I— c. • II II 1z. a CO II II Is a CM fp LO O M O +-> C c o G O tp c m •H 03 d 0 X! 4-> C O CD +-> 3 C O o CO -p c P o Tj C p •p 0 0 c 0 P P C p r: P 0 0 •p O 0 ^ to p X Pi 3 Cv d 0 •P j_> O o c P c 3 o 1— 1 o Pt 14 5

PAGE 158

146 fcc r— rH lO CO Q C r-1 CO C CO 'r-t O o CM CNl • o C-l w • d r-* . c o 1 II II II II 1 II II c/:) o c^; ^ 4-i ^: t.0 -P O C -H CO C -H P E C Tj 1—1 2; <; MM -P o 0 c o g; 0 -H o < 0 0) r— 1 ,Q b.C 03 c •H a -H 'i~< c y) CO ci > Crt i-J Pi j-> lU Q) < o C w e 1— I J 0 CT; 0 0 73 -' u < CO C Q CO 0 a i o ^ i CJ -o c r-i Q ;a -1K tc 0 ft C +^ C C 1— ' >. 4J o •H r— i P -ro i 0 M +J r-H -H 0 G -H <; (U Cfi C3 H => s PS ^3 0 C W C/D U C o CO u CO cj 0) E o o p c o rH C O CO II 11 •z o. I> o o CO r^i • 1 II II is a o o o O CI C2 O CO O C-J • I II II CM O CO Csl II p. II o I> o o 00 CNl • II II o o CO CM I II 2 CO tO CO O C<1 • o o o O CM • O O r-i C<1 C^j • i II II 000 C^^ • II II 00 — I O r-< 73 CQ • I 11 CD CO CD cc 0 0 CO CM 1 — i 1 11 II II II 1 II II 'Z p 'z 0. ?H c 0 u_ 0 tz c 0 d 0 c 0 •i-i k>l •H P" 5 rP +-> +-> > (D 0) 0 Cl P w. •H ;3 H CU •H V, 0 -p 0,' p 'd pi 0 c p 0 < cr: t— 1 1— i 0

PAGE 159

m CD C 0 rH 0 c/: H m C/3 •H E c> x; o ci CD -p |3 p -P S 1 — i •H CD 4-^ C -p C 0.1 r\ 6 0) i t— 1 0 c c a; -P 0 i — ( -P G c a o d 5 H 0 ca II !l 0 II II '3 m CO o I II II >^ -r-i -P -P 0 O X -a C 0 c 00 II II ^ c. II II r-i CC CC 1 — 1 I II I: fcX 0 > c x: ^ •H P" 0 pi p> t> CD O O II II (jj in oi : — I • I II II l> c O.. Tj O U C 0 c ;3 -p ^ ;3 0 0 0 'P W •n PI c rH O O p^ 'P O '

PAGE 160

148 0 •H ( — 1 CT!M C'. o CTl 0-1 0 1—1 o 1— 1 • CO rH r-t 0 O r; 1 II II II II II II O O >^ 00 <3 'z o 2 p o c b£ T3 aJ C 0) -P •ri -P CO O >x M +-1 1 — 1 CO C" o •H o !-i C\! rH • o 1 — i P 11 II 1] ll 1 1 1 u 11 i— 1 G G. '—' G in 1 '• O 1> O CD H o • CO rH rH fcr r-i • o tr. o. o CO o o P. o 1 — i Cj 1 II II 1 II 11 11 II 5 G •H 0. 'H O iM r-1 2; o rH r-, I II 25 LT cT; o M • I II II .00 CO C5 CO O rH . I II II 2 £1. CO CC' i:^ CO O rH . I II 11 G O p G G c 'p, > G P o 0 G G G 3 •P •H P" 0 !-> -p pi tr > G G O O G W •H G H P' y. tj p -P G P G G G P O Q < Qi KH p( a Pi G

PAGE 161

r; c o m w •H le CO Cfi pa •H +j •r-( o f-l c > p +-> (b I— I w CD c Ph 0 G o H 00 r—i V ^ CO CO 0 -P p C Q) 1 !l II 1 i: II 0 0 z J2 •i; cz c bj: c c z n CO c Z Q. < Pi CO , o II II z a ^ CO o LO I II II z c o •H cj; o > c x: ^ •H -p o p -p d c ci O n LO ^ o CO II II Z c ii z CM CO. CO o 00 II z II o c Xj O '-^ C C d 'H C P i-H 'm 3 O O O 0; H 4H T3 0 1^ ;3 ^ O 0

PAGE 162

150 PS P CO <: S O o J < G 2: o o n < K o K O CO is O a > CD Si m CD s 0 O O -M o 0 H i CO CC o r— 1, i ' cc CO CC o o [/: O o 1— I o r— 1 • CO 1— 1 o • rt r" i 11 II II II 1 II 11 11 II 0 ii CO is a. Cu (D 1 O G -a 03 o CD I— 1 O 4-' +j CO CO CC cn TO CO r— i o c 1 — 1 o o — ' • •H E II II 11 11 11 11 11 II CD 1—1 CQ < a is a 'z -p C [> r-l CO O X X o :m CO CO CO r-t CO c •H o 1 — i co m r-H t-H ' o c.C <; o II II 11 11 II 11 11 II is Cm rt' •H 5 01 c. a o «^ CO o CO 00 CO cr. CO : — . CO CO GO 1 — 1 — ^ 1 — i r-i' 1— ! CM u 'X 5 O 1 II 11 1 i[ 11 11 11 II 11 •H o CO is c. is c Is a p. CD 00 c c:D CO o CO CO CO ^ r-l • r-i • II I! 2 a 11 II is a C^3 CC CO I—. rH • I II II is Q. — ' 1 C X2 -rl CO CC CD -AH r-l rH c C CI 11 :3 E O C CJ O 00 CC CSl 11 II is Ch o CD 03 ^ O rH • 11 11 is; o p -p CD O •H 13 C 0 c 5 0 CJ5 •H > 0 P 0 -P Pi -P HH c 0 Tj O 'n O C Zi -u o -P ci :3 o 0 m •H IH -C 0 ^ c ^ o I— i C tH II II I II II 2 c 'C U G Ir C_) 0 C 0 c o pi ^ S-i rH ci fn O 0 0 O 0 VM r. 0 -H un >-H >> rH c O' CO cS o u CD +j o 0) tp «Pi c o •H -p o d Jh Q; +-> C 0 c •H E 0 P 0 •a C 0 P> rH G, S 0 a cc cc 0 pi ^ O -H C -P' rj CC 0 ci > 0 o c tx-P 0 p 0 « CC o erf o cc H o r^ C H C

PAGE 163

'/! 'o O CD ( — ' c C/j •rH IT. D, •r-i ri g 4-> CD K> O !^-| C3 G -P -P 1 li II CD 'Si ^ CD P, C -P G CD Ch CD •H +J G fcj: CD O c O •H c rH 0) ri 1 0) -p CD ^— ( p 0 ;3 cp 11 II c p !h r\ 0 s p. O r-< Ki rCD cn 0 o u l-O CQ o I •1 II C H 0 w C a) 5; r;; o CD, >> -H Pi -P 0 0 CD U •p ::i +-> C 0 o < cx; CO I il II II CD CO ^ o (M o II II lO I II II CO O i II II 0 0 73 M c 0 0 C •P pi 0 -p 1-1 0 ^ Co P 0 c P cC o p< w •p P" 0 •p c p> -P 73 0 P US p C 1 p c 0 pi c PI o

PAGE 164

CD ri > O « o CO C CO o cv: • X2 o I II II "O c Qj +-> •H u a; -M c •H a 1— 1 o < C C CD tj:' r-l < O >. =: c cr. a; c G s « c p •H ?H C Q 'H tr. 05 O rH II II 1> O CO O CM O (M II II £3 a CC' O CO O CM • CO o C3 o II CO O 00 II II C O i-i CO C<1 • 2: Clin o CM C\J e5 c CO 1—' csi C; O rH CNl • 2; c DO C CO C CM • x: > CO 0 -p O C\l C) o CO c ft cr CM • o CM o C\] o c o; ^' II II 1 II II II II kJ c a ft ft >. +-> •H 1 — 1 o r— 1 •H o c^q o o CO o o 1 c I — 1 CM • CO CM LO M p r— 1 •r— 1 G Q +-> II II II II 1 II II ri ft s ft E 3 c 0) in 0 Ir. o X3 G o o (D E -H >> •H •r-i -P 1 — • pi fcX > ct ^ p o ^ c O K o •H 5 C -p •P 5 CD Cj :3 C U c 0 o < 1—1 0. G

PAGE 165

153 X3 3 C H +J a o u o r-l 0 M c O u rH CD C O c CD 3 CC Q) E O o o C Q O Cfl v. d C/3 CI. E -O C r-i' C;'. O CD 4-1 hi CD CD' CD E CD fo 0 G 2; CJ C3 23 (^5 CM ^ Cvl • I II i: n CO CM II II o c^j • I li II CS! o o !> CM I II II 'O c o CD' >i -H•H -P 4-1 +-> o > cc CD CJ C M O •H p CD •H -P 4-> 13 C < K c •H 1— 1 CM O Cj> • II II CD CD' be -J C C P -H Q) -P i-i ^ 3 a) o

PAGE 166

O -H pC5 C<1 00 o OO CO c CO r-H CO c o rH • CO rP CC cc 0 II II t 1 1 II II II " CO g G. O y bi Co 0 CO 00 CO •H -P cc o lO OD CC CO 1 — i CO 0 -P ^ • CO 1 — i CG •H pS II II II 11 1 II il £ o < :s a 0 0 00 0 o Ln o 0 •H CO •— t rP o •H 1 — i ;^ < o 1 II II li II II II o > ci P. r-< C •r-t a; C3 G CO l-O QJ 0 0) 5h 0 (M CO 00 CO rP CO CO 0 0 O E: 1— I • CSl 1—1 00 C2 Pi C c • 0 o •H u r-; > -p CO O o CO CO GO lO CO 1 — i t-l • r-' CO r-H 3 :_, )— ! II II II ii 1 11 II H M -P 1— ^. o ip H 1 ^ 0 O C0 p CO CC' o C<] 00 t — \ 0 CO lO h< 1 — 1 CO 1 — 1 0 1 — c C. < o II II i 11 II 11 ~. CO CJ s p. 2 0 G c o c o 0 0 0 0 Q 0 0 >> •H P -p 0 -p E pi 0 > P 0 0 o c c 0 s-l o •p +-I 0 p -p X -p -P 0 i0 tp G c H 0 .O < cr; 0 PH p< PI

PAGE 167

in =tfc o rH m o u c w •H M •H c r-H K CD 0) cjO (D a +J O M C 1— 1 0 -rf o 4-J E c — +-I C Q c o r— I r" •H CD II II r-l CO I— I II II £5 a o II II Ol 4-' 4-' O O c q; < Pi cc O II II 2: P. I II II oo I bS O > r-i 4-^ 0 4-^ 4-> ci ^ C Pi C y-t I II II CO [> o I II II 0 O b£. CJ ^ c O C 3 -H 0 -fJ r— 1 ctf t^ S O C 0 =H W •H E o G ;^ o c i— I Ci, '4H O

PAGE 168

156 CO CO G rH •H Ih > -P C 0 T3 C a dj 3 m CD E O O c; cn o o c II CO II II II 1 ii 11 11 1 II CI, w — » 1 1 01 4-1 -H M C r— o 0 * o CD £ ^ 11 II II II 1 II II p. &. C: i_) cc 1— Q ( — CD r-r Ci.. I It II 11 II 1 .1 II r "i o, ri 0) LO r-i Jc 1— 1 CD rH ^^ c o r— < CD c 0 CJ II 1 1 li II II li w a <^ c. •r-1 \> -p VlO ^ 1— 1 0 rH 6 r-1 CO CD — c >: -P •H 1 c T— 1 P' 0 c u CJ W II LO o II II I 11 il CD CO -nT C^l O CD • II II I II II 2i CDc a; o fci: a O P c o G a> C 3 •r-. >>-p P +-I CD ^ ^ -p > p 3 GJ 0 o M o O P^ Pi O Pl -a CU P Ij c o c P 0 0 o PH p< o

PAGE 169

c o o w •H Cfl (/} G, H E +^ C 0 ci C I II II l-O ^ o (X) o o r-l. O ] • 0) c c E a; -p CM +-3 o o r C. C r• II II LO C. ^" C r-1 • I II II t> ^ a O M I II II 12; o c o c o < d; 5 u o > -P -P o ^ z: -P o -P I— I d ^ r 0) P ^ TZl O ^ d G P O O PI G. 'iH C

PAGE 170

158 to I— i ci •H C •r^ 0) >^ c CL ci C •H I' +^ o H E h1 < 03 0 -r-i > GO < l; c— -p 2: s a >. C O c O T-, 1—' 0) c d 0) K 0 ci. C o O ^ £ S-i CO o •-3 •H U o 1— 1 00 < ^4 0 u < _r; > PS 4-> o c/; 0 J C 1— ( o < 1 0 9 1^' 4-1 ^ " 0 u: CO CJ 0) CD E O O +-> O CO r-l (M CO • I II II CM 11 II o — ' n I II II IS c CM O C -D I — ^ I — ^ I— I CM CO • II II co CO — ^ O CO • 2: a o o ^ II II o O '-^ M t-1 00 • II II 2; a CO CM CO' II II CO ^ o CO o 2 p. st< ^ O CO C O 2 C. CD Gi r-H CO O ^T) . II II CO CD r— I CO O CO • I II II CO vjl 1—1 ^ CO II II 2; c o c in O 00 • CO O O 00 00 • I II I. 0 0 c 0 ?0 G 0 C 3 •t—i >> -H -M 0 XJ M -t^ or. 0 > 0 0 O c o 0 'tT c/: •H 3 H -r-> 0 •i-j T3 -p +-> •o 0 !h C 0 C f-i C 0 o 1— 1 1— 1 o

PAGE 171

rH Co •rH fn > Qj C H -P C C 0 CL' o Q. Q) 1 Xj C h— 1 C -P O O H « C3 X P, H E ^ O p -P r« -P o •p-i m o Q) -p E Q tf-, -p C C 0 5-1 r c CO I II II 03 coO I II II c o CO V -r-i b£ OJ > H +-> G » LO O C-] II II CO ^ c c cj a; cc =! CD 'o 0 Jh :3 ^ G o 1-^ a'p o

PAGE 172

00 CD O CO CM ? — 1 CO c C\! ^03 rr c o 1 — 1 1 — 1 r_ 1 — 1 -1—1 o rH 1 II II 11 II 1 II II CQ is a 'Z, — fcX/ 1 i 0 CO CC' o 1. J r r-, t — ' If. CO rH lO r— 1 lC 0 • • x: 1 II II 1 II II II li is p. 2C O O " C: •f-t 4-1 O :o CO CD CD 1 1 G • o LO r— i in 0 CC, -p-i • II II 1 II II II II ^ X; 0) >— i 0 < £r. c K CO LO •H CO CD CO CD o LO CD CO o in 1 — i in O LO C . t o > 1 II II II 11 1 II II 'z. < Ft", rx 1— 1 CC 1 — ' C£ CO CD o 0 di o LO LO • K C 1 II II 1 II II i 11 i! U h-i GO o o O • H Q. ES CI, 1— 1 < ""^ C/O < '"-^ W c •r-i a; > cc cs O c cc C2 CO /-^ _J o iH o LO • • CO li !l II 11 11 II < o C a CO < >, , — 1 CD cr, cc C-l c; 1 — i LO CD rH 1 c ^ C 1 — 1 LO rH lO u 0 -P II II 11 II 1 II 11 •I — ; C CI S: *s z 0 0 CO CJ o 0 c 'cp 0 =C c O a O 0 >, -H H -P r-i E ^ +-1 'U Q > H C O o a c GJ X o •H P 0) •H 'H r-" p Til i) r5 c 6i <: p; 1—1 1— 1

PAGE 173

> 0 o Q, 0 C 0 S o o o C o •H 1— t C r-i r— ) CO C O ao c K 0 +J •r-i 4-> o -I-' 0 < o a; 0 c.r. CJ o H c J55 r-i o CO o •H >! a +c ^ c I o ^ J-i 1—1 -r^ o o ^ 0 M 03 •^ C & 3 E :r; O O K D O (M OG — ( I II (M i> C"; II o ^ r-i II 7, CNJ I I! II Q O II II OJ CD O r-II II 0-; o c II II 2 a c >i -r-l -P P 0 o —I ;3 00 n I II II co OO O CO c ^ • II II o CD o CD CT. o) CO II II C<1 C Q O nT rH I II II cr. o CO' O r• CO 00 G> I II II 5 0 •H 0 -H 0 -U +-> -J '-^ C !— 1 00 CO o I II i: Q :^ 2 C^. o uO (T; O I> r-l • II II C^Q I> o o 2^ & 00 [> CTi CO O rH • II II 2: a I II li O CD ^ O rH • 2 o c 0 0 CjC 73 o 0 -U 0 irf 0 O 0 •H •a 0 U O 1— 1 0

PAGE 174

CO t-' o o O o > >^ CO (Hi < P w CO CO < O W o I— ( CO CO < I — i tc ^ o CO C "J 1 r-l lO CO LT, o Co o 01 j i! II II II 1! II .-^ j a Cl, i C -H 1 , CSl 1 CT; C I CN] O 1— 1 o T—l o CO ^ C ! cvi eg • o :\-) O] o 1,1 o 1 II II II II II II CO 1 C0 1 rt c 0 CO l> +-> I> O CM CO i> o C Cl o eg O CQ -rC E II il II II ' II II E ^ 'w a; i-i 0 < Cl CO C'T ^ i00 O C**" ~ — ^ 0 CSl CSl 1 — ci t-^ be ' — i 1 H II .1 .1 11 M 1! II il <' C -p Ei c to G >. c C CO o d X S' 0 o o , c C E s-' CTj O (M Q o c CJj o X' 0 G 1-^ CO CD CM CQ C<] •H ?H C G CO II II 1 II II II ;i C-Cu r-i -C > tlj 0 CZ' C tC "C o c C<] CO o IM o cn; • M J o -^1 1 1 1 II II I] II 1 II > 2c Cl p. Q_ H 1 — ' ?H '1-1 1 0 x:' CO H Ti CD O f— o C CO CD O 0 0 1-^ r ] c\i • 1 — ; CVI CNl C X. ri II II 1 II II 1 II li ^ :3 e Cl C C 0 O H 3 G 0 < Pi CD •H CJj 0 > c s: u •H -p 0 -p -t-i VH d K O 0 0 'C 0 ^ 0 c = P 0 -P ci P-( ^ b 0 C H '-H a 0 M c ?^ c

PAGE 175

I APPENDIX H

PAGE 176

Correlations of First and Second Anxiety Measures with Predictor Variables First Anxiety Second Anxiety Measure Measure Subject -Counselor Compatibility Length of the Interview Discrepancy from Passing Score Age of Client Importance of Being Admitted Years of Schooling N P N P N P N P N P N P . 04 .15 = Q ^ . oo p = .08 . 11 .05 = 95 v -N Q ^ V D = .15 P .31 -.12 . 15 = 95 N 95 = .12 P .07 -.30 . 19 = 94 N 94 . 01 P . 03 .30 09 = 95 V 95 . 01 P .20 -.01 05 = 95 N 95 = .46 D . 30 164

PAGE 177

165 First and Second Anxiety Measures by Race Criterion FirstAnxiety Measure Groups Mean SD Black White Entire Sample 11 .032 11 . 125 11 .095 4 . 004 3 . 38S 579 31 64 95 Source Between Groups Within Groups Total Analysis of Variance Sum/ Squares df 0 . 1807 1203 . 9678 1204 . 1484 1 93 94 Mean Square 0 . 1807 12 .9459 .014 NS Criterion Second Anxiety Measure Groups lean SD Black White Entire Samole 8.32 3 8.355 8 .344 3 .468 3 . 178 3 .259 31 64 95 Source Between Groups Within Groups Analysis of Variance Sum/Squares df 0 . 0225 976 . 9697 1 93 Mean Square 0 .0225 10.7359 0021 NS Entire Sample 976 . 992:

PAGE 178

]66 First and Second Anxiety Measure by Feeling About Entrance Examination Performance Criterion Groups Felt Passed "No feeling" about performance Felt "not passed" Entire Sample FirsrAnxiety Measure \iean 10 .714 11 .474 11 .800 11 .095 SD 4 .021 1 .954 3 . 443 3 . 579 N 56 19 20 95 Source Between Groups Within Groups Total Analysis of Variance Sum/ Squares df 20 . 7815 1183 .3669 1204 . 1384 2 92 Mean Square F 10.3907 .8078 12.8627 NS Cr i ter ion Second Anxiety Measure Groups Felt Passed "^^"o feeling" about performance .Felt "not passed" Entire Sample Mean SD 8 . 167 8.263 8.900 8 . 344 3.261 2 . 446 3.959 3 . 259 56 19 20 95 Source Between Groups Within Groups Analysis of Variance Sum/Squares df 8 . 0071 968 .9851 2 92 Mean Square 4 . 0035 10 . 7665 .3719 NS Total 976 . 9922

PAGE 179

REFERExNCES Anderson, R. P., SAnderson, G. V. Development of an instrument for measuring rapport. Personnel and Guidance Journal , 1962, £1 ( 1 ) , 18-24. Barr, A. J., Goodni-ht , J. 'A., Sail, J. P., Helwig, J. T. A Users Guide to SAS 76 . Raleigh, N.C.: Sparks Press, 197S. Barrett-Lennard. G. T. Dimensions of therapist response as causal factors in therapeutic change. Psycho logical ?..onographs . 1962, 76 (43, Whole No. 562). Beckner, B. L. Counseling black students: Any room for whitey? Journal of Higher Educa tion, 1970, 41 (3) 630-637. — — Berzins, J. I., Friedman, W. H., & Seidman . E. Relationship of the A-B variable to psychological symptomatology and psychotherapeutic expectancies. Journal of Abnormal Psychology , 1969, 74 (1), 119-125. Beutler, L. E. The therapy dyad: Yet another look at diagnostic assessment. Journal of Personality Assessment , 1973, 37 (4), 303-308. Blocher, D. What can counseling offer clients? Implications for selection. In J. M. Whitoley (Ed.), Research in counseling: Evaluation and refocu s. Columbus 0 Merrill, 1967. Bordin, E. S. Implication of client e.xpectation for the counseling psocess . Journal of Counselin'^^ Psychology 1955, 2 (1), 17-21. ^ Brough . J. R. Sources of student perceptions of the role of the counselor. Personnel and Guidance Journal 1965, 43, 597-599. ' ~ ' — ' Campbell, D. T., & Stanley, J. C. Expermental and auasiexperimental designs for research on teaching. ' In N. L. Gage (Ed.), Handbook of research on teaching Chicago, II.: . Rand McNally, 1963~; Chaikin, A. L., Derlaga, V. J., Bayma, B., Shaw, J. Neoriticism and disclosure reciprocity. Journal of Consulting and CI in ica l Psychol ogy . 1975 43 (il s2-9Tr ^ / — ' 167

PAGE 180

168 Cohen, J. Multiple regression as a general data analytic system. Psychological Bulletin , 196S, 70 (6), 426-443. Cook, T. E. The influence of client-counselor similarity on change in meaning during brief counseling. Journal of Counseling Psychology , 1966, 13_, 77-81. D'Angelli, A. R. Changes in self -reported anxiety during a small group experience. Journal of Counseling Psychology , 1974, 2_1 , 202-205. Dreger, R. ' ! . , Sc Miller, K. S. Comparative psychological studies of negroes and whites in the United States. Psychological Bulletin .lonograph Supplement , 1968, to' (3 . pt . 2) . Edwards, A. L. Statistical Methods for the Behavioral Sciences . New York: Rinehart & Company, Inc., 1954. Ewing , T. N. Racial similarity of client and counselor and client satisfaction with counseling. Journal of Counseling Psychology , 1974, 21 (5), 446-449. Eysenck, H. J. Some comments on the relation between A-B • status of behavior therapists and success of treatment. Journal of Consulting and Clinical Psychology , 1975 43 ( 1) , 86-89. Feidler, F. E. A comparison of therapeutic relationships in psychoanalytic nondirect i ve , Adlerian therapy. Journal of Consulting Psychology , 1950, 1_4, 436-445. Ferreira, A. J. The intimacy needs in psychotherapy. American Journal of Psychoanalysis , 1964, 2A (2) 190-193. Fremont, T., Means, G. H., & Means, R. S. Anxiety as a function of task performance feedback and ex xraversionintroversion . Psychological Reports , 1970, 27^ (2), 455-458. Friedman, M. Dialogue and the "essential we." American Journal of Psychoanalysis , 1960, 20 ( 1 ) , 26-34. Garfield, S. L. Basic ingredients or common factors in psychotherapy? Journal of Consulting and Clinical Psychology , 1973, £1 (1), 9-12. Gassner , S. }! . Relationship between patient therapist compatibility and treatment effectiveness. Journal of Consulting and Clinical Psychology. 1970, 34_ (3) 408-414~ '

PAGE 181

1C9 Gelso, C. J. Effects of audiorecording and videorecording on client satisfaction and self-expression. Journal of Consulting and Clinical Psychology , 1973, 40 ( 3 ) , 455-461. Graff, R. W., Rogue, D., & Danish, S. Vocational-educational counseling practices: A survey of university counseling centers. Journal of Counseling Psychology , 1974, 2J, 579-5S0. Grater, H. A. Client preferences for affective or cognitive counselor characteristics and first interview behavior. Journal of Counseling Psychology , 1964, 11 (3), 248-250. Greenberg , R. P. Effects of presession information on perception of the therapist and receptivity to influence in a psychotherapy analogue. Journal of Consulting and Clinical Psychology , 1969, 33,425-429. Hansen, D. A. The indifferent intercourse of counseling and sociology. Journa] of Counseling Psychology , 1963. 10, 3-13. Hayes, C. V. The measurement of anxiety in sophomore nursing students using Zuckerman's AACL. Nursing Research , 1966, 15, 262-267. Hodges, W. F. Effects of ego, threat and threat of pain on state anxiety. Journal of Personality and Social Psychology, 1963, 8 (4), 364-372. Hodges, W. F. , & Spielberger, C. D. Digit span: An indicant of threat or state anxiety. Journal of Consulting and Clinical Psychology '. 1969, 33 (4T, 430-434. . . ' ' Izard, C. E. Personality similarity and friendship. Journal of Abnormal and Social Psychology , 1960a, 62 ( 1) , 47-51 . ^ ' ^ Izard, C. E. Personality similarity, positive affect, and interpersonal attraction. Journal of Abnormal and Social Psychology . 1960b, 61 (3), 484-485. Kiesler, D. J. Some myths of psychotherapeutic research and a search for a paradigm. Psychological Bulletin 1966, 65, 110-136. '

PAGE 182

170 Lanning, W. L., k Lemons, S. L. Another look at the factor structure of the Barrett-Lennard Relationship Inventory. Measurement and Evaluation in Guidance , 1974 , 6 (4) , 22S-231 . Linden, J. D., Stone. S. C, & Schertzer , B. Development and evaluation of an inventory for rating counseling. Personnel and Guidance Journal , 1965, 44, 267-276. Lorion , R. P. Patient and therapist variables in the treatment of low-income patients. Psychological Bulle tin, 1974, _S1 (6), 344-354, Mendelsohn, G. A. Effects of client personality and clientcounselor similarity on duration of counseling: A replication and extension. Journal of Counseling Psvcholoc£v, 1966, 1_3 (2), 288-234. Mendelsohn, G. A., & Geller, M. H. Effects of clientcounselor similarity on the outcome of counseling. Journal of Counseling Psychology , 1963, 10 (1), 71-77. Mendelsohn, G. A. & Geller, M. H. Structure of client attitudes toward counseling and their relation to client counselor sim.ilarity. Journal of Consulting Psychology, 1965, 29, 63-72. Mendelsohn, G. A., & Rankin, N. 0. Client-counselor compatibility and the outcome, of counselinc;. Journal of Abnormal Psychology , 1969, 74 (2), 157-163. Mickelson, D. J., £; Stevic, R. R. Differential effects of f acilitarive and nonf acilitative behavioral counselors. Journal of Counseling Psychology , 1971, 1^ (4), 314-319. Mills, D. H. , S: Zytowski , D. G. Helping relationship: A structural analysis. Journal of Counseling Psychology , 1967, 14 (3) , 193-197. Myers, R. K. Some effects of seating arrangements in counseling . (Doctoral dissertation, University of Florida. 1969). Nie, N. H. , Hull , C. II . , Jenkins, J. G. , Steinbrenner , K. , Bent, D. H. Statistical Package for the Social Sciences (2nd Edition). "IcGraw Hill, New York, 1975. Patterson, C. H. The selection of counselors. In J. .M . \Vhiteley (Ed.). Research in counseling: Evaluatio n and ref ocus . Columbus, 0.: Merrill, 1937.

PAGE 183

171 Pope. B., Nudler, S., Vonkorf f , M. R., & McGhee , J. P. The experienced interviewer versus the complete novice. Journal of Personali ty and Soc ial Psycholop-y 1974, 42 (2), 692^:^697: ' iii^^Pope B., & Siegman, A. W. The effect of theranist activity level and specificity on patient productivitv and speech disturbances in the initial interview. Journal of Consulting Psychology . 1962, 26 489. ~ Pope, B., , & Siegman, A. W. Effects of question specificity and anxiety producing messages on verbal fluency in the initial interview. J mirnal of Perso nality anri Social Psychology . 1965, 1 JT) , 522-530. — ~ Pope, B., & SiegTOan, A. W. Interviewer warmth in relation to interviewee verbal behavior. Journal of Consultinoand Social Psychol ogy. 1968, 32, 5S8-595"7 ~ Pope, B., Siegman, A. W . , & Blass , T. Anxiety and sneech m the initial interview. J ournal of Consultinoand Clinical Psychology . 1970, 3^ , 233-238. ' Razin A. M. A-B variable in psychotherapy: \ critical review. Psychological Bulletin . 1971, 75 1-21. Reimann, F. The personality structure of the analyst and Its influence on the course of teratment. ' America p Journal of Psychoanalysis . 19G8 , 28 (1), 69^79-; Rogers C. R. The necessary and sufficient conditions of therapeutic personality change. Journ al of Consulting Psychology , 1957. 2^(2), 95-103. — Russell, R. D. Black perspectives of guidance. Personnel and Guidance Journal . 1970, 4S (9), 721-728. "~ Sapolsky, A. Effects of interpersonal relationshio on verbal conditioning. Journ al of Abnor mal and' Clinical Psychology . 1960. 60 241-246^ ' — Sapolsky, A. Relationship between patient-doctor compatibility mutual perception, and outcor.,e of treatment. Journal of Abnormal Psychology . 1965, 70 (1), 70-76. Schutz, W. C. The interpersonal unde r'wor2d. PaloAlto Calif,; Science and Behavior, 1960. Sheiner, S The changing image of the psychoanalytic relationship.^^^ American Journal of Psy choana lysis_ . 1967,

PAGE 184

172 Shoben, E. S., Jr. Counseling and the learning of integrative behavior. Journal of Counseling Psychology , 1954. 1 (1), 42-48. Spielberger, C. D. (Ed.) Anxiety: Current trends in theory and research . New York: Academic Press, 1972. Stranges, R. J.. & Riccio. R. J. Counselee preferences for ' counselors : Some implications for counselor education. Counselor Education and Supervision , 1970, 10 (1), 39-4"5; St'rupp . H. H. Psvchotherapy . Annual Review of Psychology , 1962, 13, 445-478. Strupp, H. H. On the basic ingredients of psychotherapy. Journal of Consulting and Clinical Psychology, 1973a, 41 (1), 1-S. Strupp, H. K. The interpersonal relationship as a vehicle for therapeutic learning. Journal of C onsulting and Clinical Psychology . 1973b, ±1 (1), 13-15. Thomas, C. v; . Black-white campus and the function of counseling. The Counseling Psychologist, 1969, I (3), 70-73'. Truax, C. B. A scale for the measurement of accurate empathy. D iscussion Paper , No . 20 , Wisconsin Psychiatric Institute, University of V.^isconsin, 1961. Truax, C. B. A tentative scale for the measurement of unconditional positive regard. Discussion Paper , No. 23, Wisconsin Psychiatric Institute, University of Wisconsin, 1962a. Truax, C. B. A tentative scale for the measurement of therapist genuineness or self -congruence . Discussion Paper^ No. 35, Wisconsin Psychiatric Institute, Universi ty of Wisconsin. 1962b. Truax, C. B. An approach to unraveling the patient-therapist interaction. Journal of Counseling Psycholog y. 1963, 10 (3) , 256-263. Truax, C. B. Sel f -disclosure , genuineness, and the interpersonal relationshin . Counselor Education f-nd Super . vision, 1971, 10 (4), 351-354.

PAGE 185

1 ] 73 Trua.x, C. B. Therapist empathy, warmth, and genuineness and personality change in group psychotherapy; A comparison between interaction unit measures, time sample measures, patient perception measures. Journal of Clinical Psychology ,' 1966, 22 (6), 225-229. Truax. C. B., V.'argo, D. G., Frank, J. D., Imber, S. D., Battle, C. C. , Hoehn-Savic, R., Nash, E. H., L Stone, A. R. The therapist's contribution to accurate empathy, nonpossessive warmth, and genuineness in psychotherapy. Journal of Clinical P sychology. 1966 22 431-435. — — Underwood, B. J., Duncan, C. P., Spence . J. T., & Cotton, J. W. Elementary statistics . New York: Appleton, 1954. Walberg, II. J. Generalized regression models in educational research. Am e r i c a n E du c a t i o n a 1 Research Journal 1971 8 (1), 71-91. ' ' Whitehorn. J. C. Si Bezz , B. J. Further studies of the doctor as a crucial variable in the outcome of the treatment with schizophrenic patients. American Journal of Psychiatry . 1960, ri7_, 215-223^ Whiteley,, J. M., Burkhart, M . Q., Harway-IIerman , M . , S: V/hiteley, R. :.! . Counseling and student development. Annual Review of Psychology . 1975, 26, 337-366. Vi'illiams, R. L . , & Kirkland, J. The white counselor and the black client. The Counseling Ps ychologist 1971 2 (4) , 114-117. ^ V/olkon, G. H. , .Moriwaki , S., &V,'illiams, K. J. Race and social class as factors in the orientation toward psychotherapy. Journal of Counseling Psychology 1973 20 (4). 312-316. ' Zuckerman, .M . The development of an affect adjective check list for the measurement of anxiety. Journal of Counseling Psychology , 1960, 24, 457-46"27~ Zuckerman, M.. & Biase. D. V. Replication and further data on the validity of the affect adjective check list measure of anxiety. Journal of Co nsulting Psvcholoo-y 1962, 29, 291. (AbstF^TFtl ' " ~' Zuckerman. i Lubin, B. Normative data for the MAACL . Psychological Reports , 1965, 16, 438. Zuckerman, M . , Lubin, B., Vogel, L., k Valerius. E. Measurement of experimentally induced affects. Journal of Consulting Psychology . 1964, 28, 418-425.

PAGE 186

BIOGRAPHICAL SKETCH Douglas Gilmore Johnson was born in Minneapolis, Minnesota, on September 13, 1938. After graduating from Roosevelt High School in Minneapolis, he enrolled for one year at Macalester College in St. Paul, Minnesota, and subsequently transferred to the University of Minnesota where he received a Bachelor of Science degree in Industrial Education in 1961. He was employed as an Assistant Dean and Director of Union Program.s and Assistant Director of Housing at the State University of New York at Oswego from 1961 to 1964. He entered a master's degree program in counseling and guidance at the University of Florida in the fall of 1964 and received his Master of Education in that field in 1965. He entered the doctora,l programin counselor education the following year. He has been employed as a counselor at Santa Fe Community College since the college opened in 1966. Mr. Johnson is married to the former Judy Johnson and they have two daughters, Alison and .Meredich. 174

PAGE 187

I certify that I have raad 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 Education. / Theodore Landsman Chairman Professor of Counselor Education and Psychology 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 decree of Doctor of Education. / Richard J. Anderson Professor of Psycholoszy 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 Education. Adelbert G. Cranney , Jr. Associate Professor of English and Education ''^ This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Education. June, 1977 Dean, College of "Educa t ion Dean, Graduate School