F... Iflllllllflllllllllljllfllll\lJllllllll

 

This is to certify that the

thesis entitled

Effect of Reduction of Physiological Anxiety
on Counseling Students' Empathy and Anxiety

presented by

Richard Walter Ober

has been accepted towards fulfillment
of the requirements for

Ph. D. . Counseling
degree 1n

 

 

,7 , ‘
lg’// 0 Q git/'5‘: 7%

Major professor

 

Date /5' "3:" 75'

0-7 639

 

OVERDUE FINES ARE 25¢ PER DAY _
PER ITEM

Return to book drop to remove
this checkout from your record.

 

 

 

 

 

 

EFFECT OF REDUCTION OF PHYSIOLOGICAL ANXIETY

ON COUNSELING STUDENTS' EMPATHY AND ANXIETY

By

Richard Walter Ober

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Counseling, Personnel Services,
and Educational Psychology

1979

ABSTRACT

EFFECT ON REDUCTION OF PHYSIOLOGICAL ANXIETY
ON COUNSELING STUDENTS' EMPATHY AND ANXIETY

By

Richard Walter Ober

The purpose of this study was to measure the effectiveness of Auto-
genic Training (AT) and electromyograph (ENG) biofeedback using the
frontalis muscle in reducing counselor verbal and general anxiety and
in increasing empathy. Twenty—two master's and two doctoral counseling
students who were beginning the first practicum in their respective
programs volunteered for this study. Fifteen of the master's students,
those in rehabilitation counseling and placement, received a .5 bonus
on their final grade for their participation in the study.

The subjects were separated into high and low levels of anxiety
on the basis of a frontalis EMG anxiety score which was obtained in the
following manner. Each subject's EMG output was measured for a six-1
minute period while he or she watched a videotape of a role played
client who made intentionally anxiety provoking statements. A second
reading was taken during an eight-minute baseline period. This latter
score was subtracted from the former score for each subject and the
scores were rank ordered. The median score served as the dividing point
between high and low levels of anxiety.

Subjects in the high and low levels of anxiety were randomly
assigned to three treatment groups. The first group received 25 minutes
of EMG biofeedback and three 25—minute sessions of taped AT relaxation
instructions per week. The second group listened to 80 to 85 minutes

of audiotaped information in counseling and psychology per week and met

Richard Walter Ober
with this researcher for 10 to 15 minutes each week to discuss this
material or other issues. This treatment was intended as a placebo.

The third group received no treatment. The treatments were given for a
four-week period.

The outcomes were measured by Carkhuff’s Empathic Understanding
scale (EU), the Judgments of Verbal Anxiety scale (JVA), and the IPAT
Anxiety Scale Questionnaire (ASQ). Ratings for the first two scales
were taken from subjects' interviews with coached clients. The data
were analyzed by a two—way Analysis of Variance of the posttest. There
were high and low levels of anxiety for the first factor, and three
levels--relaxation treatment, information treatment, and no treatment--
for the second factor. The following planned comparisons were used to
test for main effects: relaxation treatment was compared to information
treatment; relaxation treatment was compared to no treatment; and the
high anxiety level was compared to the low anxiety level. Two complex
comparisons measured interaction effects. The first compared the relax-
ation and information groups to high and low levels of anxiety. For the
second, the relaxation and no treatment groups were compared to both
levels of anxiety. Only one of the above, the comparison of the relaxa-
tion and treatment groups for the EU scale, approached significance
(p less than .063). The four measures--the EMG anxiety score, the EU,
JVA, and the ASQ--were intercorrelated. Only one of these, the correla-
tion for the JVA and ASQ, was significant (.428, p less than .023).

There are several possible reasons for the failure to find signifi-
cance in this study. The procedure for establishing high and low levels
of anxiety may not have accurately measured the subjects' levels of

anxiety. While subjects in the relaxation group achieved some degree of

Richard Walter Ober
relaxation, the treatment may not have been powerful enough to effect the
hypothesized change. The small sample size would have required rela-
tively high F ratios to achieve statistical significance, and one or two
extreme individuals in the relaxation group may have adversely effected
the means of scores on the outcome measures. Finally, it is possible
that the combined effect of instruction during regular praccicum supervi-
sion and continual exposure to clients during practicum resulted in
increased counselor empathic communication and reduced anxiety, such that
significant differences among treatment groups could not be measured.

The hypothesis which motivated this study, that relaxation will
improve the performance of anxious counselors, was still seen as plau-
sible and worthy of further inquiry. Future research could focus on
understanding the nature of counselor anxiety, its measurement, its spe-
cific effects on performance, and procedures by which it can be reduced.
Attention should be given to employing intensive designs to measure the

effectiveness of treatments.

This volume is dedicated with love to my wife, Martha,
who had to live with me while I undertook this study

and to my son, Matt, who had to live without me.

ii

ACKNOWLEDGMENTS

I would like to express my gratitude to the members of my committee
who were so helpful and supportive throughout my program of study.

First and foremost is Bob Winborn, the thinking man's pragmatist, whose
expert piloting through the shoals and channels of academia were inval-
uable. I would like to thank Jim Engelkes for his thoughtful criticism
(and subsequent improvement) of this study and his support during my job-
seeking efforts. I am grateful to Herb Burks, whose words of encourage-
ment seemed to come when they were most needed. A1 Kirk was most helpful
in guiding my selection of psychology courses and in providing support
during the refinement of the proposal for this study.

There are two other persons from whom I have greatly benefited dur—
ing the past three years. Bruce Burke helped me believe in my capabili—
ties and in the rightness of my judgment during several critical periods.
Norm Kagan has taught me that what people learn about themselves, by
themselves has potentially greater impact than the interpretations and
insights which we may give them.

I owe all of these men a great deal. This is not, however, the type
of debt one repays, but is an obligation which is fulfilled by future
service to colleagues, students, and clients.

Thanks are also due to doctoral colleagues, coached clients, and
raters who helped with this study. My greatest thanks go to my N of 24,

those practicum students who volunteered to serve as subjects.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . .
Chapter

I. PROBLEM, RATIONALE, AND RELATED RESEARCH . . .

Problem . . . . . . . . . . . . . . . . . . .

Theory and Related Research . . . . . . . . .

Rationale . . . . . . . . . . . . . . . . .

Process of Anxiety . . . . . . . . . . . .

Overview . . . . . . . . . . . . . . . .

Implications for this Study . . . . . . .

Research on Anxiety in Counseling . . . . .

Study of Anxiety . . . . . . . . . . . .

Reduction of Anxiety . . . . . . . . . .

Implications for this Study . . . . . . .

Biofeedback, Electromyograph, and Autogenic

Biofeedback . . . . . . . . . . . . . . .

Electromyograph . . . . . . . . . . . . .

Autogenic Training . . . . . . . . . . .

II. EXPERIMENTAL DESIGN AND METHODOLOGY . . . . . .

Subjects . . . . . . . . . . . . . . . . . .

Procedures . . . . . . . . . . . . . . . . .

Introduction . . . . . . . . . . . . . . .
Establishment of High and Low Anxiety Levels

iv

Training

Page
vii

viii

11
15'
17
17
20
22
25
25
27
27

27

Chapter
II. Treatments . . . . . . . . . . . . .
Relaxation Treatment . . . . . . .
Information Treatment . . . . . . .
No Treatment Control . . . . . . .
Outcome Interviews . . . . . . . . .
Measures . . . . . . . . . . . . . . .
Empathic Understanding . . . . . . .
Judgments of Verbal Anxiety . . . . .
Raters . . . . . .-. . . . . . . . .
IPAT Anxiety Scale Questionnaire . .
Ethical Issues . . . . . . . . . . . .
Hypotheses . . . . . . . . . . . . . .
Experimental Design and Analysis of the
III. ANALYSIS OF RESULTS . . . . . . . . . . .
Description of Results . . . . . . . .
Conclusion . . . . . . . . . . . . . .
IV. SUMMARY, DISCUSSION, AND CONCLUSION . . .
Summary . . . . . . . . . . . . . . . .
Discussion . . . . . . . . . . . . . .
Results Approaching Significance . .
Possible Sources of Nonsignificance .
Theoretical Basis of the Study . .
Formation of Levels of Anxiety . .
Sample Size . . . . . . . . . . . .
Treatments . . . . . . . . . . . .

Sex of Coached Clients . . . . . .

Page
32
32
35
36
36
39
4O
41
41
42
44
45
46
48
48
55
56
56
57
S7
58
58
59
6O
6O

62

Chapter
IV. Effect of Practicum and Supervision . . . .
Outcome Measures . . . . . . . . . . . . .
Implications for Future Research . . . . . .
Conclusion . . . . . . . . . . . . . . . . . .
APPENDICES
Appendix

A. PRESENTATION TO PRACTICUM GROUPS FOR SOLICITATION
OF VOLUNTEERS O O ‘0 C O O O O C C C O O O O I O O

B. SCRIPT OF VIDEOTAPE OF ROLEPLAYED CLIENT . . . .
C. AUTOGENIC PHRASES . . . . . . . . . . . . . . . .
D. CONTENTS OF AUDIOTAPES . . . . . . . . . . . . .
E. INSTRUCTIONS TO COACHED CLIENTS . . . . . . . . .
F. RELEASE OF INFORMATION FORM . . . . . . . . . . .
G. JUDGMENTS OF VERBAL ANXIETY . . . . . . . . . . .

REFERENCES 0 O O O O O O O O O O O O O O O O O O O O O 0

vi

Page

62

63

64

66

67

7O

73

76

78

79

80

86

3.2

3.3

3.4

3.5

LIST OF TABLES

Demographic Characteristics of the Sample . . . . . . . .
Means and Standard Deviations of Cells, Treatment

Groups, and Levels of Anxiety for the Outcome

Measures 0 O O O O O O O O O O I I O O O O O O O O O

Intercorrelation of Measures . . . . . . . . . . . . . .

Analysis of Variance for the Empathic Understanding
scale I O O O O O O O O O O O I O O O O O O O O O O O 0

Analysis of Variance for the Judgments of Verbal
Amiety scale 0 O O I O O O O O O O O O O I O O I O O 0

Analysis of Variance for the Anxiety Scale Questionnaire.

vii

Page

26

49

50

51

53

54

LIST OF FIGURES

Figure Page
2.1 Difference in EMG Microvolt Measurements Obtained

from Videotape Period Minus Baseline Period . . . . . . 31

2.2 Experimental Design . . . . . . . . . . . . . . . . . . .. 46

viii

CHAPTER I
PROBLEM, RATIONALE, AND RELATED RESEARCH

Problem

The goal of counselor education programs is the teaching of counsel-
ing skills and the development of counselor attitudes which will facili-
tate client functioning during the counseling process. Training in
specific response techniques, i.e., stylized counselor responses to
client statements, often comprises a significant portion of master's
level counselor education. While such skill training should not be seen
as the substance of counseling (Mahon & Altman, 1977), it does give the
counseling student some of the techniques needed to foster a caring rela-
tionship with the client and to maximize client self—exploration, goal
definition, and initiation of the change process. Consequently, much
research has been done regarding the effectiveness of teaching these
response skills. This study was undertaken to add to our understanding
in this area by assessing the effectiveness of a specific treatment,
Autogenic Training in conjunction with biofeedback, in reducing the
anxiety of students enrolled in a counseling practicum. The effective-
ness of this treatment was assessed by measures of empathy, verbal anxi-
ety, and general anxiety.

It is particularly important for counselors to communicate empathy.
The ability for the counselor to convey to the client that the latter is

understood, that someone shares in his or her present affective experience,

2
is seen by some as a necessary, if not sufficient, factor in the change
process (Rogers, 1961). Others may not accord it this importance, but
certainly feel that the counselor's communication of empathy and the
client's apprehension of it are facilitative in the counseling process
(Carkhuff, 1969).

Anxiety has long been identified as a source of interference in the
learning and performance of a wide range of behaviors (Gaudry & Spiel-
berger, 1971). Its negative effect on counselors' learning and perform-
ance of specific response skills has long been hypothesized (e.g., Bandura,
1956; Bergin & Solomon, 1970). Much recent research in counseling and
psychology has focused on the effect of anxiety on counselor performance
and the assessment of various treatment procedures to reduce anxiety.

The results of these research studies, however, have been mixed. For
example, Miller (1973) found that systematic desensitization was effec-
tive in reducing counselor trainee anxiety on one outcome measure, the
State Anxiety Inventory, but not on another, the Multilevel Measure of
Interpersonal Behavior. Carter and Pappas (1975), on the other hand,
found no significant differences for a multivariate analysis of anxiety
measures between one group which received systematic desensitization
treatment and another which received a placebo treatment. Likewise,
Graub (1975) found that subjects who received Autogenic Training were
rated higher on some dimensions of the Counselor Verbal Response Scale
than a no treatment group, while Fontaine (1976), using alpha wave train-
ing, found no significant difference between treatment and control groups
on Truax's accurate empathy ratings. Some of the factors which may have
contributed to the differences in these findings include unreliable out-

come measures, limited understanding of the construct of anxiety,

3
non-additive effects of one treatment on several types of anxiety, and
lack of initial identification of the levels of subjects' anxiety. These
and other studies will be discussed in detail in a following section con-
taining a review of related literature.

Given the importance of the communication of empathy in the counsel-
ing process, the interference of anxiety in counselors' verbal perform—
ance, and the availability of techniques which reduce anxiety, it would
be beneficial to devote further research efforts to study in this area.
Appropriate considerations in future designs might include the identifi-
cation of highly anxious counselors on a specific variable and the provi-
sion of treatment which corresponds to that variable.

This chapter includes a general overview of the design of this study
and a discussion of theory and related research. A discussion of proce-
dures used in this study, biofeedback with the frontalis muscle and Auto-

genic Training, is offered in the latter section.

Theory and Related Research

The theoretical and research bases for the procedures used in this
study are reviewed in this section. The rationale for the study is
presented to indicate those areas for which extended discussion is appro-
priate. The specific areas covered include the process of anxiety,
previous research in the nature and treatment of counselor anxiety, and
the implications of these findings for the present study. Theory and
research which are related to the relaxation treatment used in this
study-EMG biofeedback and Autogenic Training--are also covered.
Rationale

The purpose of this study was to measure the effectiveness of phys-

iological relaxation on the increase of empathy and the reduction of

4
verbal and general anxiety of counseling practicum students. The sub-
jects were divided into high and low levels of anxiety by electromyograph
readings which were taken while they watched an anxiety provoking video-
tape. They were divided into high and low anxiety groups because it was
hypothesized that the relaxation treatment would have its greatest effect
on individuals in the high anxiety group; this hypothesis could be tested
by identifying highly anxious subjects before treatments were adminis-
tered.

Subjects were randomly assigned to three treatment groups. The
first group received biofeedback in combination with Autogenic Training,
the second received a placebo treatment, and the third received no treat-
ment. The outcomes were measured by an empathy rating scale, a verbal
anxiety scale, and a standardized anxiety measure. Ratings for the first
two measures were taken from subjects' interviews with coached clients.

This study also attempted to avoid some of the criticisms of
research offered by Goldman (1976, 1977). ‘First, the number of subjects
was not large, thus enabling quality treatment to be provided. 'Second,
real people, if not real clients, were used in the procedure during which
outcome data were obtained, thereby making this study more similar to the
actual counseling situation. Third, the outcome criteria represented
important counselor behaviors, the increase of empathic communication and
the reduction of anxiety. It was also felt that improving counselor per-
formance by changing counselor behavior was an important issue (Loesch,
Crane, & Rucker, 1978). Finally, subjects in the relaxation treatment
group personally experienced the process and benefit of the relaxation

procedure (Hector, Elson, & Yager, 1977).

Process of Anxiety

 

Overview. Current research and theoretical positions do not present
'a unified model of the process of anxiety. The following description
represents a conciliation of diverse theoretical viewpoints to provide a
reasonable, workable model as a basis for this study.

Anxiety consists of four major components: the stimulus, the neuro—
physiological response, the cognitive or phenomonological process, and
the behavioral or expressive aspect (Spielberger, 1972). There is gen-
eral agreement that a specific stimulus is the initial step in the pro-
duction of anxiety, although the capacity for a stimulus to provoke
anxiety will vary from person to person. It is also widely acknowledged
that the behavioral or expressive aspect is the end result of the process
of anxiety. There is, however, considerable disagreement among theorists
regarding the relationship between the cognitive process and the physio-
logical response. It is not clear whether a person's cognitive appraisal
of a stimulus results in the physiological response or if the physiologi—
cal reaction to the stimulus leads to the cognitive awareness of the
emotional reaction.

Lazarus and Averill (1972) have stated that emotion follows from the
subject's appraisal of any situation. The organism perceives, identi-
fies, assesses, decides, and compares for appropriate emotional response,
£222 responds emotionally. Izard (1972) believed that an individual's
reaction to anxiety-provoking stimuli occurs much too quickly to permit
this process of evaluation. Some situations require an immediate emo-
tional reaction. It was also Izard's contention that physiological
responses are preparations for adaptive action, the result of our evolu-

tionary past. Accordingly, physiological responses are more immediate,

6
more enduring, and, in essence, more instinctive than cognitive responses.

Izard argued for the influence of the physiological on the cognitive,
which has a reciprocal effect on the physiological.

Once out on the intrapsychic scene, it seems highly probable

that emotion will influence the subsequent stages of appraisal

(cognitive assessment of stimulus conditions) and all other

organismic processes and activities. It seems highly unlikely

that a person can complete an appraisal without some inter-

action between cognition and emotion (1972, p. 65).

Moreover, he viewed cognition as enabling the organism to further identify
objects or events as appropriate sources of anxiety. Spielberger (1972)
and Schachter (1964) agreed with Izard that emotional states have both
cognitive and physiological components, but they viewed the cognitive as
as having a stronger effect on the physiological.

Implications for this Study. For the above reasons, physiologically
measured anxiety was the focus of this study. Physiological responses
also have the advantages of greater ease and reliability of measurement,
and are amenable to modification by a wide range of techniques. Anxiety
was defined as a "sociopsychophysiological phenomenon experienced as a
foreboding, dread, or threat to the human organism whether the threat is
real or imagined dangers, the source of which may be conscious or uncon-
scious" (Lesse, 1970, p. 13).

No attempt was made to identify the source of subjects' physiologi-
cal anxiety. It would be difficult to differentiate between anxiety
which was rooted in enduring personality characteristics, in situational
variables, or in the stress of modern living which people generally

experience. Moreover, the treatment for all of these would essentially

7
be the same for the population of interest, a program of relaxation.

The contemporary person rarely is placed in the situation where the
real alternatives are fight or flight. Social conditioning through the
cognitive processes mentioned above can, however, produce similar reac-
tions to situations which threaten psychOIOgical safety. This is the
counselor trainee's situation. He or she has invested time, financial
resources, and self-esteem in succeeding as a counselor. In addition to
these self-generated pressures, the counselor trainee is also concerned
about supervisor evaluations, grades, and a host of other issues related
to his or her interaction with clients. Thus, initial counseling experi-
ences would appear to elicit anxiety often in the beginning counselor.

Dynamic conceptualizations of anxiety were not considered-in this
study. The anxiety experienced by the subject was viewed as the result
of reactions to specific stimuli in the counseling situation (Bandura,
1977) and, hence, were modifiable through direct intervention.

The outcomes of this study were measured by ratings of excerpts of
interviews for verbal anxiety and empathy and by a standardized test for
anxiety. Verbal behavior is in part neuromuscular behavior, and as such
may be influenced by general relaxation efforts. The results of other
studies have demonstrated a relationship between a reduction of anxiety
and an improvement of verbal behavior on the dimensions of interest.

Research on Anxiety in Counseling

 

The counseling research in this area may be divided into two catego-
ries, those studies which were designed to gain an understanding of the
nature and effects of counselor anxiety, and those which measured the
effects of treatments intended to reduce anxiety. They shall be con-

sidered in respective order, followed by a section relating this previous

research to the present study.

Study of Anxiety. Russell and Snyder (1963) used an analog to study

 

the relationship between counselor experience and type of reaction to
hostile and friendly coached clients. They hypothesized (a) that both
experienced and inexperienced counselors would be more anxious when coun-
seling hostile rather than friendly clients and (b) that the inexperienced
counselors would be more anxious than the experienced counselors with both
types of clients. Three of the four measures of anxiety-~eyeblink rate,
Clients' Scale for Counselor Anxiety (CSCA), and Judgments of Verbal Anxi-
ety (JVA), a rating scale of counselor verbal anxiety-~supported the first
hypothesis. Palmar sweat measures were in the expected direction but were
not significant. Only one measure, eyeblink rate, supported the second
hypothesis, although the JVA nearly reached the level of significance.
The authors concluded that client hostility, not counselor experience,
was a factor in the degree of anxiety manifested by the counselor. The
only relationships among the measures were between eyeblink and the CSCA
and eyeblink rate and the JVA.

The above study yielded several important findings. The first is
the presence of measurable anxiety during the experimental manipulation.
The second is the lack of concordance among measures, even among
physiological measures like eyeblink rate and palmar sweat. It is
important for the purposes of this study to note that the more sensitive
measure of anxiety, eyeblink rate, and the measure of verbal anxiety,
were positively related.

In an attempt to demonstrate the relationship between anxiety and
the development of empathy, Pennscott and Brown (1972) administered the

Taylor Manifest Anxiety Scale once per term over a three-term year-long

9
period to master's level counseling students. Subjects' scores on the
Taylor Manifest Anxiety Scale decreased significantly over the one-year
period. They correlated these scores with Truax's measure of accurate
empathy, taking one rating at the beginning and one at the end of the
second term. The correlation between the measures was .015. The authors
suspect that one reason for the lack of relationship between the two
measures was due to the measurement of empathy over one term only.

Roberts (1977) studied the relationship of physiological anxiety
(as measured by the Palmar Finger Sweat Bottle), a rating of non-verbal
anxiety, and a self—report of empathy, warmth, fear, and anxiety to coun-
seling graduate students' verbal response skills. These response skills
were measured by Truax's accurate empathy, nonpossessive warmth, and
genuineness scales. Videotaped client actors played seductive, angry,
aggressive, and rejecting roles. He found that subjects' verbal response
skills and physiological anxiety increased significantly when the client
behaved in a seductive manner; subjects' verbal response skills
decreased and nonverbal anxiety increased when the client was angry and
aggressive. This supports previous research indicating that anxiety can
affect verbal response skills. No significant differences were found
between subjects' verbal response skills and self-report of anxiety.
Physiological, behavioral, and self-report anxiety measures were found
to assess three different reactions in the subjects.

Bowman, Roberts, and Giesen (1977) used skin conductance, heart rate,
and subjects' estimates of their physiological responses to measure
anxiety under different counseling conditions. It was found that coun-
selors were more anxious when counseling a coached client than when read-

ing. Subjects could predict with moderate accuracy how much subjective

10
anxiety they would experience, and physiological estimates of anxiety
taken shortly before the treatment conditions had moderate predictive
value for the amount of anxiety subjects would experience while actually
in the experimental situation. There was no relationship between physio-
logical and subjective measures of anxiety, but the two physiological
measures were related.

An important conclusion of this study was the presence of at least
two anxiety systems-—the subjective and the physiologica1-—which worked
in an independent but parallel manner. Furthermore, while Russell and
Snyder (1963) demonstrated the anxiety—producing potential of counseling
hostile clients, this study demonstrated the anxiety-producing potential
of counseling sessions generally.

Bowman and Roberts (1978) partially replicated the previous study.
They compared the anxiety which counselors manifested during a counsel-
ing session with that manifested during a conversation. The findings
supported the previous research, that one can predict to a moderate
degree the amount of physiological or self-reported anxiety one will
experience in actual counseling sessions.

This study also demonstrated that counselors experienced as much
anxiety during a conversation as during a counseling session. Actually,
the two situations were not that dissimilar; the clients were coached
clients, and the conversations were held with people whom the counselors
had never met. Apparently the two situations were perceived by the sub-
jects as highly similar.

Bowman and Roberts (1979) recently replicated the above study but
used inexperienced counselor trainees and included a baseline period in

the design. The results were mixed: they found a significant increase

11
in heart rate during both conversation and counseling periods. Self-
ratings and skin conductance data indicated that the subjects experienced
greater anxiety during the interview than either during baseline or
conversation periods.

Reduction of Anxiety. One of the earliest studies which attempted to

 

reduce counselor anxiety was done by Lesh (1970). He hypothesized that
the practice of Zen meditation would reduce anxiety and permit the sub-
jects to attend to inner, spontaneous, empathic feelings. The subjects
were counseling graduate students; three groups, one treatment and two
control, were established in an unidentified manner. The subjects in the
treatment group practiced zazen meditation (a method of developing the
Zen openness to awareness) as a group for one-half hour per day for a
four-week period. The results indicated that the treatment group
improved in their empathic ability as measured by the Affective Sensi—
tivity Scale, one control group remained about the same, and the other
decreased in empathic ability. Although it was not stated, pretests were
apparently used along with other unspecified instruments which measured
openness to experience and self-actualization.

The shortcomings of this study include uncertain formation of groups,
questions regarding analysis, lack of a placebo treatment, and unspeci-
fied instruments. Yet the initial success of relaxation-related treat-
ment in the improvement of empathy was encouraging.

Monke (1971) employed systematic desensitization using a standard
hierarchy to provide relaxation treatment to school counseling master's
students. Treatment and no treatment group differences were significant
for self-report using the State-Trait Anxiety Inventory, but were not

significant for heart rate, skin conductance measures, and Judgments of

12
Verbal Anxiety ratings. The significant findings in the self-ratings could
have been due to placebo effects. The author did not identify pretreat-
ment levels of anxiety or the dimensions of anxiety under consideration.

Miller (1973) compared the effect of systematic desensitization and
no treatment in reducing the anxiety of counseling students in a begin—
ning practicum. The treatment group showed significant reduction of
anxiety on one outcome measure, the State Anxiety Inventory, but not on
the other, the Multilevel Measure of Interpersonal Behavior. The no
treatment group showed no significant differences between pretests and
posttests on either measure. Since the design lacked a placebo treatment,
one cannot be sure whether the significant difference obtained by the
systematic desensitization group was due to attention or some other non-
treatment effect. There is a possibility that the treatment subjects
were seen in groups, not individually. Furthermore, the apparent use of
gain scores may have masked changes in the control group's performance
on the outcome measures.

Fry (1973) sought to examine the effect of systematic desensitization
on Carkhuff's core-conditioning training (empathy, respect, genuineness,
and concreteness). In this study, it was assumed that any anxiety which
counselors experienced in interpersonal situations was the result of
their own lack of nurturing by significant others. If their fear of
intimacy could be reduced, effective interpersonal functioning could be
increased.

The subjects, human service workers enrolled in an evening course,
were matched on relevant factors and were assigned to treatment and no
treatment groups. The treatment groups were desensitized for emotionally

positive tone of voice, positive facial expressions, steady eye contact,

13
and posture denoting physical and emotional proximity. The outcome
measures, the Carkhuff scales measuring the core conditions mentioned
above, were taken on ratings of a role-played interview, an actual inter-
view, and subjects' responses to five stimulus expressions. The results
indicated that the treatment group scored significantly higher than the
no treatment control group. The treatment group subjects reported
significant anxiety reduction.

There are several methodological difficulties with this study.

Since no placebo was used, one cannot be sure whether the improvement

of the treatment group was the result of special attention from the exam-
iner. It is curious to note that the treatment group achieved higher
outcome ratings on the responses to stimulus expressions. That is, their
responses to the non-personal measure were higher than the control group
ratings, despite the fact that they were desensitized for interpersonal
stimuli.

Another difficulty with this study is the use of matching to form
"equal" groups. Subjects were matched on the Affective Sensitivity Scale,
the Wisconsin Relationship Orientation Scale, the Taylor Manifest Anxiety
Scale, and demographic variables. They were not matched on a pretest on
outcome measures, although they appear to have been roughly equal on
those dimensions. It is possible, however, that non-random assignment
resulted in the placement of subjects into the treatment group which evi—
denced greater gains than the no treatment group due to the interaction
of the treatment and individual factors. In any case, the procedure used
to assign subjects to treatment groups was not appropriate for the Analy-
sis of Variance statistical test.

Graub (1975) found that autogenic training was successful in reducing

14
the verbal anxiety of therapist trainees and in improving some perform-
ance dimensions using the Counselor Verbal Response Scale. The study did
not identify highly anxious trainees. The differences between the treatment
and 1“) treatment groups were not maintained over a four-month period.

Fontaine (1976) used volunteer graduate students in counselor educa-
tion and social work, assigning them to five groups: real alpha wave
training (AWT) with empathy instructions; bogus AWT with empathy instruc-
tions; real AWT with relaxation instructions; bogus AWT with relaxation
instructions; and a no treatment control group. The outcome measure for
empathy was the Truax accurate empathy rating of recorded interviews; the
outcomes for anxiety were measured by the Galvanic Skin Response (GSR)
and self-report.

The following results were obtained: GSR was significantly higher
for bogus AWT than for real AWT; the real AWT group was rated signifi-
cantly higher on accurate empathy than the bogus AWT group; the correla-
tion between anxiety measures and empathy was negative but was not
significant. However, no significance was obtained between any of the
experimental and no treatment groups on the three criterion measures.

Carter and Pappas (1975) employed three groups of graduate counsel-
ing students in their study. One group received systematic desensitiza—
tion to stressful counseling situations. A second received a placebo
treatment consisting of discussion of feelings and behaviors stemming
from interpersonal anxiety in other than counseling situations. A third
received no treatment.

A multivariate analysis, using six measures, was employed to test
the effect of treatment. There was a significant treatment difference

between the two treatment groups and the control group. There was no

15
significant difference between the two treatment groups. The reason for
these results may be easily seen, as the placebo group could have served
as a therapeutic device for the reduction of interpersonal anxiety.
Individual counseling is, if anything, an interpersonal relationship.
The four measures which accounted for the significant differences in the
multivariate analysis were extraneous overall movement, small movement,
speech disturbance, and a self-rating of state anxiety.

Using 39 graduate counseling students, Walters (1978) measured the
effect of skin temperature biofeedback relaxation, relaxation alone, and
no treatment on three measures. The outcome measures were the
Recognition-Assessment Empathy Inventory, the Affective Sensitivity
Scale, skin temperature taken at three periods, and the Rotter Internal-
External Control of Reinforcement Scale. The only significance was found
for the Rotter scale, with the relaxation alone group rated higher on the
Tranquility, Elation, and Composite scores of that instrument. Again,
highly anxious subjects were not identified beforehand, and the gains of
some highly anxious subjects may have been minimized by non-anxious sub-
jects. Moreover, the author might have selected another physiological
response (like EMG) which has a more direct relationship to anxiety.

Implications for this Study. Several trends in the research reviewed

 

above are important for the present study. The first is the relation-
ship between physiological anxiety and verbal activity. Russell and
Snyder (1963), Graub (1975), Carter and Pappas (1975), and Roberts (1977)
found a negative relationship between quality of verbal behavior and
anxiety. This seems reasonable inasmuch as the process of speech is a
combination of neuromuscular and cognitive factors. The relationship

between anxiety and empathy was also important for the present study.

16
Fontaine (1976) found they were negatively, but not significantly,
related. Fontaine (1976), Lesh (1970), and Fry (1973) all found that
relaxation treatment resulted in an increase in levels of empathy.

These findings have two implications for the present study. The
first is the suggestion that if anxiety interferes with verbal behavior,
then the reduction of anxiety may be related to improvement in the qual-
ity of verbal behavior. The second implication is the relationship
between the reduction of anxiety and increase in empathy ratings. Accord-
ingly, the reduction of anxiety may result in an increase in the empathy
ratings of counselor responses.

The results of these previous studies have further implications for
the design of the present study. The first is the need to identify the
anxiety level of subjects in order to assess the differential gains of
high and low anxiety groups. While a large percentage of beginning coun—
selors will most likely feel some anxiety during a counseling session, it
is doubtful whether such anxiety in all cases will be sufficiently high
to be significantly reduced by the treatment. By separating subjects into
high and low anxiety groups, the researcher may measure the gains of the
high anxiety group which would contain the subjects for whom changes in
the desired direction would be expected. Accordingly, the design of this
study separated subjects into high and low anxiety levels, as described
in Chapter II.

A fourth implication for this design is the desirability of specify—
ing the type of anxiety the researcher wishes to reduce and a treatment
corresponding to that anxiety. Those studies based upon this approach
more frequently found significant differences in the hypothesized direc-

tion. For example, Fry (1973) used systematic desensitization to reduce

17
interpersonal anxiety and found that the treatment group scored signifi-
cantly higher on Carkhuff's core-conditioning responses than the no treat-
ment group. Also, Graub (1975) used AutOgenic Training to reduce physio-
logical anxiety and found that the treatment group scored higher on the
CVRS than the no treatment group. The present study, for this reason,
has focused on the reduction of physiological anxiety by a combination of
electromyograph biofeedback and Autogenic Training.

The final implication for the present study which has been taken
from this review is the inclusion of a placebo group. Of the eight treat-
ment studies examined, only three contained groups which received placebo
treatment. One yielded no significant differences (Fontaine, 1976) among
treatments. Another found that the placebo and treatment groups were
significantly lower on the anxiety measure than the control group, but no
significant differences were found between the treatment and placebo
groups (Carter & Pappas, 1975). The results of a third study indicated
significant improvement on the criterion measure for the placebo group
alone (Walters, 1978). It was therefore seen as important to include a
placebo in this study to measure the effects of factors other than spe-
cific treatment which would otherwise be confounded with treatment.
Biofeedback, Electromyograph, and Autogenic Training

This section addresses the theoretical and research bases for the
employment of these procedures in this study.

Biofeedback. Biofeedback was selected as a treatment procedure

 

because of its proven effectiveness in reducing anxiety (Brown, 1974,
1977; Blanchard & Epstein, 1978; Hassett, 1978; Pelletier, 1977). The
process of biofeedback has several basic components. Some physiological

aspect of the subject is measured, and this measurement is immediately

18
and constantly conveyed back to the subject. The subject is then able by
various means to alter the measurement in the desired direction. An
example may be seen in a simple procedure by which finger temperature can
be raised. An ordinary thermometer is taped to a subject's finger. By
repeating certain phrases (e.g., ”My hands feel warm and relaxed") the
subject can raise the finger temperature in about 15 or 20 minutes (Brown,
1977). The subject's perception of the increase in temperature is infor—
mation (feedback) that the process is effective. Typically, however, the
apparatus is much more complex. The physiological function is usually
amplified electronically to produce a signal which is convenient for the
subject to use, like a tone, light, digital display, or dial reading.
Whatever the equipment, the procedure is the same; the signal informs the
subject of his or her progress in altering the physiological state.

Biofeedback may be performed simply by the subject's monitoring of
the physiological measurement and "trying to make it change.” This
change can often be facilitated by the use of imagery, suggestion, or
relaxation.

Biofeedback has been practiced in recognizable form since the early'
1950's (Gaarder & Montgomery, 1977). During the 1960's it gained wide—
spread attention, particularly because of its use in assisting individuals
to achieve higher states of consciousness through monitoring electro—
encephalograph (EEG) readings.

Research efforts continue to refine treatment procedures and compile
empirical support for its effectiveness in treating a wide range of human
problems. The physiological functions which can be monitored and "fed
back" to the subject and the problems for which they are used include the

following (Blanchard & Epstein, 1978): skin temperature as a treatment

19
for migraine headaches and circulatory problems; heart rate for cardiac
disease; blood pressure for hypertension; electroencephalogram (EEG) for
seizure disorders and as a method of achieving altered states of con—
sciousness; and electromyograph (EMG) for the reduction of muscular ten-
sion and the rehabilitation of damaged muscles. The EMG biofeedback
procedure employed in this study will be discussed further in the follow—
ing section.

While the procedure for biofeedback is straightforward, the psycho-
physiological mechanism by which change is effected is not clearly under-
stood. Brown (1977) felt that several components may be involved. The
first is operant conditioning, in which the change effected in the phys-
iological measures is reinforcing to the subject. This reinforcement
results in the increased probability that the behavior will occur again.
The second is stress reduction. Biofeedback is used to reduce anxious
feelings which result in increased muscle and visceral tension. The
subject becomes aware of feelings of relaxation and can discriminate among
different levels of stress and relaxation. Feedback may be given on a
variety of physiological states. The third component is the cognitive
mechanism. The subject can, on the basis of biofeedback information,
alter a signal in the desired direction. The awareness of the current
state is the essential factor in producing change.

While all of the above factors may play some role in biofeedback,
they are clearly not independent. But the question has yet to be answered:
what is the physiological process of biofeedback? How does the provision
of information result in the change of a bodily state? Some evidence
supports the contention that voluntary control can be extended to the

autonomic nervous system (Gaarder & Montgomery, 1977), that is, one may

20

be taught to exert the same type of control over his or her heart as one
has over the biceps. While others question the studies which suggest.
this (Hassett, 1978), biofeedback can be an effective clinical tool
regardless of our lack of understanding of how it works (Brown, 1977).

Biofeedback has served another important function in addition to its
clinical application. It has served as the meeting point of diverse,
even opposed theoretical perspectives. As Brown has stated (1977), the two
major foci in the treatment of illness were traditional medicine (surgery
and drugs) and analytically based treatment. The use of biofeedback has
bridged this mind-body polarity by offering cogent evidence that mental
processes can be employed to alter physical states. The person not only
receives treatment but, in a sense, becomes it. The use of biofeedback
has also facilitated the unlikely rapprochement of those concerned with
external events and those concerned with internal states. For example,
the achievement of an internal state can be measured by the EEG, and the
behavioral effects of that state, such as creativity, can also be mea-
sured. One may carry this thesis further by arguing that biofeedback also
brings together East and West. Eastern practitioners of meditation have-
claimed the control of bodily functions like heart rate (Green, Green, 8
Walters, 1970). Western technology provides equipment which informs one of
the success of bodily control. Thus, the practices of Eastern Masters are

supported and augmented by physiological measurement.

‘Electromyograph. The physiological function which was selected for

 

use with biofeedback in this study was the electromyograph (EMG). This
measurement represents the amount of muscular activity in a specific
muscle group. The muscle selected for this study was the frontalis. The

frontalis extends from just above the eyebrows over the top of the head to

21
a point about two thirds of the way down the back of the skull.

EMG indicates the degree of muscle activity. All muscles are com-
posed of a number of muscle fibers.' Groups of fibers form single motor
units, and many motor units make up muscles as we know them. Neural
impulses are delivered from the brain through the spinal column to indi-
vidual motor units. The impulses then travel to the many muscle fibers.
The action of each muscle fiber is discrete. It contracts about one-half
its length when fired by a neural impulse. The stimulation of many
muscle fibers results in the contraction of the muscle. The more fibers
that are stimulated, the greater the contraction. The EMG measures the
neural impulses which are delivered to a number of muscle fibers. The
unit of measurement for this electrical activity is called "micromohs."

The measurement of the EMG is accomplished by the careful placement
of silver-chloride electrodes on the forehead after the skin is cleansed
(Hassett, 1978). One electrode is placed about one and one-half inches
above each eyebrow. A third electrode, the ground, is placed over a
honey portion of the forehead, such as in the center near where the scalp
begins. The ground serves as an electrical reference point. It measures
the amount of extraneous electrical impulses emanating from within the
subject, like the pumping action of the heart, or outside the subject,
like fluorescent lights, television monitors, or computer terminals. By
subtracting the amount of electrical energy measured by the ground (which
measures no EMG's) from that which is measured by the other two elec-
trodes, one can obtain the EMG output from that muscle. This signal is
then amplified and displayed in some fashion to the subject.

The employment of the frontalis muscle EMG in biofeedback has sev-

eral advantages. This muscle is sensitive to anxiety precipitated by

22
threatening interpersonal and social stimuli, and EMG reduction is the
easiest type of biofeedback to learn (Brown, 1977). Furthermore, the
frontalis is a direct expressor of stress within the neuromuscular system
and is convenient to use. Since it is not a postural muscle (i.e., does not
require use when the subject is standing, sitting, or reclining) the posi—
tion of the subject is not critical (Gaardner & Montgomery, 1977).

EMG biofeedback has been used to treat headache, subvocalization, and
stuttering (Blanchard & Epstein, 1978). Brown (1977) lists 27 emotional
and physiological disorders which may be treated with this procedure.

The effectiveness of EMG biofeedback relaxation has been examined in
several recent studies. Townsend, House, & Addario (1975) found that EMG
biofeedback and relaxation subjects were rated lower on anxiety than a
control group. Canter, Kondo, & Knott (1975) showed that subjects who
received EMG biofeedback and relaxation were rated lower on self and
therapist ratings of anxiety than those who received progressive relaxa-
tion. These results supported the use of EMG biofeedback in this study.

Autogenic Training. Autogenic Training (AT) was selected as the

 

relaxation procedure for use in this study. AT was developed by Schultz
in the early thirties; he based his original studies on work done by Vogt
(Schultz & Luthe, 1959). AT was developed from and as an alternative
to hypnosis, which has the disadvantage of subject inactivity and the
potential for dependence on the hypnotherapist. AT does provide the sub-
ject with the sensations of heaviness, warmth, and resultant relaxation
which one experiences in hypnosis.

There are three elements in AT (Luthe, 1963). These are: (a) the
subject's assumption of a comfortable position with minimal internal or

external disturbance, (b) the subject's repetition of autogenic phrases,

23
and (c) the maintenance of a passive mental attitude. The phrases are
offered by an actual person or recording and follow a specific sequence.
They describe feelings of warmth, heaviness, relaxation, and comfort in
various parts of the body. The subject is instructed to repeat the
phrases to himself—-not to "try" to relax, but to "let it happen."

AT, as described by Schultz and Luthe (1959) can be done in three
stages. The first stage consists of focusing on bodily feelings of
warmth, heaviness, and relaxation. These are called the "standard
exercises." Once these have been mastered, a series of "meditative
exercises" may be performed. These focus on visual imagery such as
colors, objects, and, finally, abstract concepts. The third stage,
"special exercises," enables one to become aware of one's own feelings,
until one ”communes with the unconscious" (Brown, 1974, p. 130). Only
standard exercises were used in this study.

There is a potential problem in using AT. It is possible that sub—
jects may experience ”autogenic discharges" when using any one of these
types of exercises. These discharges occur because the state of relaxa—
tion and consequent lowered cortical control permits the brain to receive
a "flood of downward, helter-skelter, ungoverned motor impulses" (Brown,
1974, p. 131). The subject may experience shaking of the whole body,
muscle twitching, or feelings of floating or detachment. These reactions
are not serious if the individual is in good mental and physical health,
and they cease with the termination of the exercise.

AT has been selected for use as a relaxation technique in this study
for several reasons. First, it does not require the specific instruction
and close supervision needed for progressive relaxation to be effective.

Second, it demands less "effort" than progressive relaxation; that is,

24

one listens to and repeats to oneself the autogenic phrases. The tedium
and discomfort of progressive relaxation are avoided. Third, many of the
subjects in this study had previous exposure to progressive relaxation in
a counseling strategies course. Such previous experience could present a
confounding variable. Fourth, AT is more in keeping with current, popular
relaxation techniques, and therefore had potential for attracting greater
student interest in the study.

Both biofeedback and AT are effective when used alone. Biofeedback
does not require the subject to do anything but be aware of information
about the physiological state he or she is trying to change and, by a
"nebulous process of intentions and decisions" (Brown, 1977, p. 6),
effect that change. AT can be effective without information on the phys-
siological state (Luthe, 1963). Yet by combining both, one can perhaps

enhance the achievement of increased relaxation and reduced anxiety.

CHAPTER II

EXPERIMENTAL DESIGN AND METHODOLOGY

Subjects

Volunteers for this study were sought from students enrolled in
their first counseling practicum in the Department of Counseling, Person-
nel Services, and Educational Psychology, College of Education, Michigan
State University during the Spring 1979 term. This researcher presented
the purpose and procedures of this study (as described in Appendix A) to
several practicum groups; 24 agreed to participate. The number of stud—
dents who volunteered from the different counseling programs are: two
master's students in agency counseling; 15 master's students in rehabili—
tation counseling and placement; five master's students in school coun-
seling; and two doctoral students in urban counseling. Those volunteers
in the rehabilitation practicum were offered a .5 bonus on their final
grade in the practicum if they participated in this study. For example,
a subject who received a grade of 3.0 for the practicum course would
receive a 3.5 for their final grade after the addition of the .5 bonus.
No other individuals were given incentives. -

The subjects for this study were not randomly drawn from a defined
population. For this reason, demographic information which describes the
sample on relevant characteristics is offered in Table 2.1, thus permit-
ting the identification of the population to whom the results of this

study may be generalized.

25

26

Table 2.1

Demographic Characteristics of the Sample

W
W

 

 

 

 

 

 

 

Sex Males Females
7 17
29% 71%
Age Under 25 26-30 31—35 36-40 Over 40
5 8 3 1
21% 33% 29% 13% 4%
i = 31.34 S.D. = 8.13
Marital Status Single Married Widowed Separated Divorced
10 7 1 2 4
42% 29% 4% 8% 17%
Ethnicity Caucasian Hispanic
22 2
92% 8%
Counseling Program Master's Doctoral
Agency Rehabilitation School Urban
2 15 5 2
8% 63% 21% 8%
Number of Credit Hours _ Master's _Doctoral
Completed, End of X S.D. X S.D.
Spring Term, 1979 32.73 8.5 38.5 2.5
Years of Paid Work Master's Doctoral
Experience for All _ _
Subjects X S.D. X S.D
In Counseling 1.23 2.61 5.0 3.0
In Human Services 5.27 5.6 7 5 3.5
(includes above)
Previous Experience in None Minimal Moderate
Relaxation 16 3
67% 21% 13%

 

27

Procedures

Introduction

 

This section includes the method by which subjects were divided into
high and low anxiety levels, the types of treatments used in the study,
and the method by which outcome data were obtained.

Throughout these procedures, all subjects were seen individually,
thereby maintaining the individual as the unit of analysis and providing
greater statistical power. It is possible that subjects, particularly
those who received different experimental treatments, discussed their
various experiences and expectations with one another. This is undesir-
able for it could weaken the steps which had been taken to keep the
individuals independent. It was, however, unavoidable.

In an attempt to reduce the bias in experimental treatment which
might have been introduced by practicum supervisors, no supervisors were
'informed of the participation of any of their students in this study.
Two of the supervisors served on the researcher's dissertation committee
and were well acquainted with the nature of the study, but had no know-
ledge of individual participation. At the end of the study, however, a
list of participating rehabilitation counseling placement students was
given to the coordinator of that practicum to enable those students to
receive the .5 bonus on their final grade.

Establishment of High and Low Anxiety Levels

 

This study included a procedure whereby the physiological anxiety
which subjects experienced during a simulated counseling session was
used to establish high and low anxiety levels. As discussed in Chapter
I, previous studies have failed to identify subjects' levels of anxiety

before treatments were administered. In these studies it is likely the

28
relaxation treatments were more beneficial for highly anxious subjects;
yet, analysis strategies were not included to measure their gains. By
assessing each subject's level of physiological anxiety, the performance
of high and low anxiety groups could be assessed on the outcome measures.
This procedure had a further advantage of making groups more heterogenous
on this variable. The design had greater precision for both of these
reasons.

A similar procedure was employed by Chapman (1974) in his research
on anxiety. He used electromyograph (EMG) as a blocking variable in a
study measuring EMG output during variations of observation. The sub-
jects were matched and assigned to three groups on the basis of previous
EMG readings. The analysis strategy did not, however, include a test of
main effects for the EMG levels.

In this study 22 of the 24 subjects were seen individually for the
measurement of physiological anxiety. They were informed that the pur-
pose of the procedure was to separate subjects into high and low anxiety
groups. The procedure consisted of measuring subjects' EMG output during
an eight-minute baseline period and during a six-minute period while the
subjects watched a videotape of a client. This procedure is described in
detail below.

The EMG device employed was a J&J model M55 B, used with non-
disposable silver-chloride electrodes, disposable adhesive discs and a
standard conductive gel. The skin was prepared by first abrading the
three forehead electrode contact areas with a commercially available skin
cleanser containing aluminum oxide particles. This preparation caused
slight discomfort as it removed the topmost layer of skin cells, but it

provided excellent impulse conduction. These contact areas were then

29
cleansed with alcohol and the electrodes were attached. Recleansing and
reattaching were performed if electrode placement was not satisfactory.
Electrode placement was easily checked by the skin resistance readings on
the J&J M55.

Each subject was individually placed on the EMG device for an eight-
minute period to obtain a baseline reading. Subjects were instructed to
sit quietly and "see if the needle goes down," that is, to watch the
EMG output dial. No specific relaxation instructions were given. This
researcher then left the room for the eight-minute period. The mean EMG
output was computed for this period by a J&J Model LSC 150 digital
integrator.

After the baseline measurement was taken, the subject was informed
that he or she would be viewing a stimulus videotape. The following
instructions were given:

In just a minute you will be watching a videotape of a man who

is roleplaying a client. I want you to imagine that you, a

counselor, are seeing this client for the first time, alone,

in a private office, in a counseling setting which makes sense

to you. You have no information about him. I want you to

attend to what he says to you. You don't have to respond to

him. His statements are not necessarily related to one another,

but generally follow in the order in which they might occur in

an actual counseling interview. Some of the things he says

may be seen by some as anxiety provoking.

The videotape contained 11 short scenes of a 30-year—old man who
"faded in,” made a statement, then "faded out." The client's statements

included questions about the subject's competence and caring, references

30
to his life experiences and repugnant attitudes, ambiguous aspirations
for intimacy with the subject, and threats of violence toward the sub-
ject. Appendix B contains the script of this videotape. The "client," a
Ph.D candidate in counseling psychology, was chosen for his lack of
contact with master's counseling students and his acquaintance with video—
materials which were designed for affective stimulation. The script
for these 11 scenes was written by this researcher and was adjudged to
be anxiety producing by several doctoral students in counseling, who were
all experienced counselors.

The videotape was approximately six minutes long. This researcher
sat behind the subject and recorded the mean EMG output, as shown on the
scorekeeper, for each minute in the six-minute period. The subject saw
neither the scorekeeper nor the EMG dial. At the end of the videotape
period, the subjects were given general feedback on the EMG levels for
baseline and videotape periods. Each subject was asked about his or her
current affective state to insure that the vignettes had not been sever-
ally disturbing to anyone. The subjects were told that this researcher
would be making decisions regarding who would be included in the study
and would contact them within one week.

The EMG output measured during the baseline period was subtracted
from the EMG output obtained during the videotaped period. This
yielded a score for each subject indicating anxiety as measured by
the EMG. The frequency distribution of these scores is shown in
Figure 2.1. Subjects above the median were identified as the high
anxiety group, those below the median as the low anxiety group.

Both high and low levels were randomly assigned to three experimental

groups.

31

moiumm ocfifiommm was“: mofiuom
ommuoomfi> Scum nonwmuno mucoEouammmz u~o>ouofix 02m a“ mmocouowufin .fi.~ ouswam

mu~o>onofiz

m a m . N g o as

ntppr-nhmlpph-p.-nh— nub-prlphnsbthblphrPlnppbPPPLb-bbbbLL-npbb

I

 

H cmumoz

no.2 casuaa>mn ecaccmam

 

mm.~ :mmx

Aouanbalj

 

32

The two subjects who failed to keep appointments for this initial
phase were placed at the bottom of this ranking of the index of anxiety.
This placement of the two subjects in the low anxiety level was care-
fully considered. A central purpose of this study was to measure a
change in those subjects who were highly anxious. By assigning those
two subjects to the low anxiety group, one could be sure of the charac-
teristics of the high anxiety group, i.e., that all subjects in that

group were high on the variable of interest.

Treatments

 

The three experimental treatments included in this study were a
relaxation group which received a combination of biofeedback and Auto-
genic Training, a placebo group which received information in counseling
and psychology, and a no treatment control group. Each treatment pro-
cedure is described in detail below.

Relaxation Treatment. The subjects in the relaxation group were

 

seen individually by this researcher once per week for four weeks. As
subjects came for the first session, they were again told that the pur-
pose of the study was to measure the effect of various experiences on
counselor performance. The subjects in this treatment group were
informed that the researcher was interested in the effects of relaxa-
tion using biofeedback and Autogenic Training (AT). They were also
informed of the method of obtaining outcome data. Taped ratings of
their interviews with coached clients and scores on a standardized
anxiety test would be collected. The confidentiality of all information
was restated.

Each session consisted of placing the subject on the EMG device for

approximately 25 minutes. Skin preparation followed the procedure

33
described previously and care was taken to maintain consistent electrode
placement from session to session.

A minus threshold with a sound signal was used during each session.
As the subject relaxed, the sound which represented the EMG output level
of the frontalis muscle was lowered in tone. Once the EMG output was
reduced to a certain level, the sound shut off. This served as a rein-
forcer for the subjects' relaxation efforts. The EMG range and thres-
hold level were established individually for each subject and were
generally lowered in subsequent sessions. The threshold level was
adjusted to the point where the sound was off approximately 25% of the
time. This allowed the subject to receive information regarding his or
her progress in relaxation during the session, yet allowed some rein-
forcement for the termination of the sound. For the first session,
subjects listened to the AT relaxation tape which was made by this
researcher (Appendix C). For subsequent sessions no tape was used.

Subjects in this group were asked to listen to the AT audiotape and
practice relaxation three times per week in addition to the weekly bio-
feedback sessions. Thus, each subject received a total of one and
one-half hours of treatment per week for four weeks.

Subjects were informed that some individuals have experienced
feelings of anxiety and discomfort when they used AT. They were
instructed that if they began to feel anxiety, tingling, or any other
type of discomfort, they were to stop the relaxation sessions and check
with this researcher. These feelings, however, were not reported by any
subject.

All subjects in this group received all four biofeedback sessions.

During each session, this researcher asked about progress in the

34
practice relaxation sessions. All reported that the practice sessions
were progressing well. Many subjects commented on some aspect of the
practice sessions, such as when the best time was for them to listen to
the relaxation tapes, problems they had in practicing relaxation, or
other effects of relaxation, e.g., reduced test anxiety. These comments
led this researcher to believe that the subjects were indeed listening
to the tapes as requested. Furthermore, subjects had lower levels of
EMG output as measured by reduced thresholds with each session. This
indicated that relaxation was effective.

A biofeedback procedure was incorporated into the treatment to
provide a method of monitoring subjects' prOgress in relaxation. The
inclusion of such a procedure enabled this researcher to be more confi-
dent that subjects were actually performing the autogenic exercises. In
a study comparing autOgenic training and progressive relaxation on
physiological outcomes, Madden (1976) suspected that the subjects' lack
of motivation, hence limited practice of procedures, led to a lack of
significant differences between treatments. The limited use of bio-
feedback procedures in the present study attempted to control for this
possibility.

The primary focus of this treatment was AT. To include EMG bio-
feedback as a full treatment could, perhaps, have improved the precision
of this study by intensifying treatment. It would, however, also have
limited the generalization of the results to other counselor education
programs because of the expense of the necessary equipment. Since AT
would require at most a cassette recorder, many more counselors, super-
visors, and counselor educators would be able to employ this procedure

if it proved effective in reducing counselor anxiety.

35

Information Treatment. Subjects in the information group were seen

 

individually once per week for four weeks. They were informed that the
purpose of the study was to measure the effect of various experiences on
counselor performance. These subjects were specifically told that this
researcher was interested in the effects of helpful information in coun-
seling and psychology on their performance on the outcome measures.

This "helpful information" consisted of eight audiotapes of presen—
tations by significant figures in counseling and psychology (Appendix D).
These presentations covered general issues, such as Kohlberg's theory of
moral development and Skinner's discussion of the complexity of rein-
forcers in modern life. The subjects were also told that since they had
had courses in specific techniques in counseling, the tapes to which
they would listen were more abstract and theoretical. They were
apprised of the method of obtaining the outcome date. Taped ratings in
interviews with coached clients and scores on a standardized anxiety
test would be collected. The confidentiality of the information
gathered during the study was reaffirmed.

Subjects were asked to listen to the tapes while resting in a quiet
place. During their weekly meetings with this researcher they discussed
whatever they wished of the previous week's tapes and received the next
week's tapes. The subjects usually commented on some aspect of the
tapes. This suggested to this researcher that the subjects were listen-
ing to them. During these sessions other issues were often discussed,
such as their experiences in the practicum placement. These sessions
lasted approximately 10 to 15 minutes. Since they listened to 80 or 85
minutes of tapes per week, the total time of treatment of a subject in

this group was about one hour, 35 minutes per week. All subjects in

36
this group were seen for the required number of times.

Information treatment was included in this design to expose a group
of subjects to conditions which were similar to those of the relaxation
group, yet did not contain any specific treatment which could have
produced positive effects on the outcome measures. It was, therefore,

a placebo group. These subjects, like those in the relaxation group,

saw this researcher once per week and listened to audiotapes for about an
equal period of time. Yet they were not given specific relaxation
instructions, nor did the presentations contain any specific information

which could improve counselor performance on the outcome measures.

No Treatment Control. The third group was a no treatment control

 

group. These subjects were informed that they would not be asked to do
anything during the treatment period but would be contacted in several
weeks to arrange for an interview with a coached client and take the
anxiety questionnaire.

Outcome Interviews

 

At the end of the treatment period, all subjects arranged for a
session when all the materials needed for the outcome measure were
gathered. This consisted of the completion of the Anxiety Scale Ques-
tionnaire and an interview with the coached client. Empathic Under-
standing and Judgments of Verbal Anxiety were taken from this interview.
This section describes the preparation and procedures for the interviews
with the coached clients.

The coached clients were all women, a 19—year-old junior majoring
in nursing, a 21-year-old senior majoring in communication, and a
23-year-old senior also majoring in communication. These unpaid

volunteers were recommended to this researcher by a doctoral candidate

37
colleague who had supervised them during an independent study. Women
coached clients were preferred because most of the subjects were women,
and there was less opportunity for qualities of heterosexual interaction
to present a confounding variable during the outcome interview. More-
over, if male coached clients were used, their portrayal of the same
role played by the women coached clients might have had a different
impact on the subjects. If the role were modified for male clients,
further variability might have been introduced. For these reasons, it
was decided that only women coached clients would be used, and thereby
a standard client stimulus would be maintained. The limitations of this
procedure are discussed in Chapter IV.

The coached clients met with this researcher as a group for approx-
imately two hours to devise and refine their role. During this training
period, each was interviewed by a doctoral student while the others
observed. These interviews were critiqued for the coached clients'
ability to elicit the desirable responses from the counselors and for
the consistency of the role among coached clients. A description of the
role played by the coached clients is contained in Appendix E.

An additional coached client was used for an emergency situation.
One regular coached client cancelled for a period when she was scheduled
to be interviewed by four subjects. Inasmuch as the end of the term was
approaching, this researcher thought that it was advisable to locate
another coached client and complete the outcome interviews for those
subjects. Since neither of the other two coached clients was available,
this researcher recruited a paid volunteer from a class on interpersonal
communication. She was given a 10—minute orientation to her role and

task as a "client."

38

Subjects signed up for available interview times with the coached
clients. As this researcher did the actual scheduling, an attempt was
made to keep one coached client from getting a disproportionate number
of interviews. While assignments were not made in a truly random man-
ner, they were made without regard for any factor other than subject and
coached client availability.

When clients came in for interviews, they were asked to read and,
if they approved, to sign a release of information form (as shown in
Appendix F), to provide demographic data, and to complete the anxiety
questionnaire. Some subjects had already completed the first two tasks.
All had been previously informed of the purpose and nature of the inter-
view, but were given additional instructions at that time:

You are going to interview this client for a 10-minute period.

She is a coached client, that is, not a real client, and she

will present a standard client problem. Use the time to help the

client explore her current concerns, paying particular attention

to any feelings she may have which are associated with those con-

cerns. You do not have to explain the purpose of counseling, just

focus on the client's difficulties and her feelings about them.
Individual audiotapes were made for each of these sessions, and a pre—
viously assigned random number was affixed to each of the audiotapes.

During the-process of obtaining the outccme interview audiotapes,
four doctoral students in counseling also had interviews with coached
clients to obtain tapes for assessing rater validity for the empathy
scale. These interviews were randomly interspersed among the subjects'
interviews. This researcher asked these counselors to give examples of

two good and two poor interviews. The coached clients had been told

39
that they would be interviewed by four doctoral students and were
informed of the rationale for this procedure, but were not told who
these counselors were beforehand. The coached counselors followed the
same procedures as the subjects during these interviews. All tapes were
identical, and were identifiable only by a random number.

The raters were informed that they would be rating several tapes as
a check on their validity, but were not informed of which tapes these
were. The scores given to these tapes by the raters corresponded to the
level of empathy which the doctoral students were asked to demonstrate,
and their ratings approximated those which this researcher gave to these
tapes.

.All of the coached clients were apprised of the confidential nature
of the subjects' performance during the interview. As this researcher
later listened to the interviews, his impression was that each coached
client had performed well and that a consistent role was maintained by
all.

Subjects were given an opportunity to indicate whether they wanted
information about the results of the study and feedback on the interview.
When all interviews were completed, those subjects in the information
and control groups who could be contacted were invited to experience one
session on the biofeedback apparatus. Four subjects accepted this offer.
This offer was made after subject participation in the study was no
longer required; it was felt that to make this offer sooner would imply

that one treatment was better than another.

Measures
Three measures were used to assess the outcomes of this study. The

first two are Carkhuff's Empathic Understanding scale and Russell and

4O
Snyder's Judgments of Verbal Anxiety. Since both of these scales
‘require the rating of counselor responses, the rating procedures are
discussed after the scales themselves are described. The Anxiety Scale
Questionnaire, the third measure, is covered in the final section.

Empathic Understanding

 

The first measure which was used in this study was the Empathic
Understanding scale (EU; Carkhuff, 1969). This is perhaps the most
widely used instrument in counselor education. It requires that each
counselor response receive a one to five point rating, depending on the
level of empathy communicated to the client. It has demonstrated con-
siderable capacity to discriminate between the empathy levels of various
groups in the expected directions. Moreover, several of the studies
which were previously reviewed indicated that EU could measure increases
in empathy which were apparently effected through relaxation treatment.
This was important for the purpose of this study.

The widespread use of the EU by counselor educators has not been
without criticism. Gormally and Hill (1974) have offered cogent recom-
mendations for rater selection and rating procedures which were followed
in this study.

First, the raters had been purposely selected for their lack of
experience in counseling. Raters who are experienced counselors or
psychotherapists often tend to "punish" those they are rating by giving
a lower rating than warranted to a marginal response. Second, inter-
rater reliability was checked several times during the rating. Third,
tapes were randomly presented for rating and both raters worked on
approximately the same tapes at the same time, thus equalizing fatigue

and practice effects. Tapes were identified only by a random number,

41
and all tapes were identical.

Judgments of Verbal Anxiety

 

The second measure employed was the Judgments of Verbal Anxiety
(JVA; Appendix G), developed by Russell and Snyder (1963). This scale
measured counselor anxiety by rating the presence of 16 indicators of
counselor anxiety in counselor statements. These 16 indicators are:
direct statement of anxiety; avoidance of interaction; direct interrup-
tion; impersonal or premature interpretations; unnecessary reassurance;
disapproval of client behavior; intellectualization; inappropriate per-
sonal references; failure to finish sentences; repetition of words or
phrases; stuttering; blocking or lack of fluency; poor voice quality;
apology for counselor error or deficiency; inappropriate laughter; and
anxiety as revealed by some behavior not otherwise specified. Russell
and Snyder obtained excellent results in using this instrument to dis-
criminate between anxious and non-anxious counselors in their study,
obtaining significance at the .001 level.

To this researcher's awareness, this instrument has been used in
only one other study (Monke, 1971). No significant differences were
found in that study using this or any objective measure.

Raters

Two women served as raters for these two instruments. One was 22
years-old and had recently received her bachelor's degree in social work.
She had had some interview training but no counseling preparation. The
other was a 20 year-old junior majoring in community development. She
had had no training in counseling or interpersonal communication. They
were paid for their efforts.

The raters were trained for approximately three hours before rating

42
began. Training consisted of an orientation to the study and the mea-
sures, and practice with rating tapes. After a statement was rated,
each shared her rating and the rationale for it. When both raters gave
the same scores to responses approximately 80% of the time and these
scores concurred with this researcher's interpretation of the scoring
criteria, the raters began actual tape rating. It should be noted that
raters were trained on portions of two subject interview tapes. When
actual rating began, however, the ratings made during training were
disregarded and those tapes were included with the others.

All rating was done within a four-day period. Both raters worked
separately but on the same tapes at approximately the same time.
Through this method, an attempt was made to equalize practice and
fatigue factors. Every counselor response for the 10-minute period was
rated. The raters were situated in the College of Education during the
times they were rating, and this researcher was available for the clari-
fication of issues and responses to questions. Interrater reliability
was measured by the Pearson Product Moment Coefficient of Correlation.
Correlations for both instruments were calculated when raters had com-
pleted their ninth and eighteenth tapes. When correlations were cal-
culated for 18 tapes, considerable divergence for both measures was
observed. The raters were retrained briefly, and interrater reliability
was subsequently increased. The final interrater reliability was .89
for Empathic Understanding and .73 for Judgments of Verbal Anxiety.

IPAT Anxiety Scale Questionnaire

 

The IPAT Anxiety Scale Questionnaire (ASQ; Krug, Scheier, & Cat-
tel, 1976) is a 40—item standardized group instrument which assessed

overt and covert symptoms of anxiety. The selection of items was based

43
on Cattell's research (1973) on personality factors. The items which
were included in this scale represented proportionally those factors
of which anxiety is a component. These factors include Tense/Relaxed;
Apprehensive/Self-Assured; Emotionally Stable/Emotionally Unstable;
Suspicious/Trusting; and Uncontrolled/Self-Controlled.

Test—retest reliabilities for this measure vary from .93 for inter-
vals of one week to .70 for those of two years. Split—half reliabili-
ties range from .78 to .92 when corrected to full test length. This
type of reliability is more important for this study because it demon-
strates the accuracy of the current assessment of a changeable character-
istic, anxiety. The validity of this instrument has been demonstrated
several ways. Its correlation with the pure anxiety factors is .90
average across several studies. Its correlation with clinical judgments
of anxiety is approximately .90 when these judgments are corrected for
low reliability. When its scores are correlated with those of other
anxiety measures and both are corrected for imperfect reliability, the
correlation approaches .90 (Krug et al., 1976). Moreover, this instru-
ment is highly resistant to response distortion due to social desir-
ability (Schmidt, 1969).

The ASQ has demonstrated the capacity to measure the reduction of
anxiety in several studies. Sherman, Mulac, & McCann (1974) and
Nicoletti (1973) showed that this instrument was able to measure the
effects of treatment procedures which reduced speech anxiety. Sherman
and Plummer (1973) reported that subjects who received generalized
anxiety reduction treatment scored lower on the ASQ than a control
group. These results indicate the appropriateness of this instrument

for this study.

44

Ethical Issues

 

Efforts were made throughout the conduct of this study to assure
that subjects' rights were protected under the ethical standards of the
American Personnel and Guidance Association (1976) and American Psycho—
logical Association (1973). The following points outline these efforts:

1. The confidentiality of the participation of each subject was
assured. No information regarding any individual was available to any
person other than this researcher, with the following exceptions. A
list of names of participating rehabilitation practicum students was
given to the practicum coordinator. This was necessary for him to give
each subject a .5 bonus on their final grade for their participation and
was done with the agreement of those students involved. Coached clients
were aware of subjects' performance during the outcome interview, but
had no knowledge of their score on any measure. Raters knew the subjects
only by random number. Both raters and coached clients were apprised
of the confidential nature of any subject's performance. In addition,
all subjects signed release of information forms (Appendix F) permitting
the use of data for analysis and a description of the sample. All
audiotapes were erased when the report of the study was completed.

2. Responsibilities of researcher and subjects were clearly stated
when subjects were first seen individually. These included time commit-
ment per week, number of weeks, the demographic information which would
be required, the nature of the outcome measures, the interview with the
coached clients, and confidentiality.

3. It was made clear to each subject that participation was volun—
tary and none of the information gathered would have any effect on their

grade for the practicum course. Those in the rehabilitation practicum

45

were informed that their participation by itself would qualify them for
a .5 bonus on the final grade.

4. The potential for harmful effects, though minimal, was present
with the practice of AT. These harmful effects included muscle twitch-
ing, feelings of floating or detachment, or shaking of the whole body.
Accordingly, subjects were apprised of this possibility and the appro-
priate action to take should negative effects occur.

5. Since the research hypotheses in the following section differ
from the expectations which were communicated to the subjects at the
beginning of the study, necessary deception was employed. A summary of
results, correcting the deception, was sent to all participants when the
report of this study was completed.

While the information treatment was not expected to reduce coun-
selor anxiety, it is suggested that the subjects in this group might
have benefited from the information contained in these tapes. They
included presentations by prominent, current thinkers in the field of
human behavior. It is possible that this information contributed to the
subjects' general understanding of those factors which shape our

behavior.

Hypotheses

 

The null hypothesis was stated thus: There is no significant dif—
ference in the ratings of Empathic Understanding, ratings of Judgments
of Verbal Anxiety, or scores on the Anxiety Scale Questionnaire among
high and low anxiety levels of counseling practicum students who receive
relaxation treatment, information treatment, or no treatment.

The research hypothesis, however, stated that the relaxation group

would have significantly higher EU ratings and significantly lower JVA

46
ratings and ASQ scores than the other two groups. Furthermore, it was
hypothesized that subjects in the high anxiety level would have signifi-
cantly lower EU ratings and significantly higher JVA and ASQ scores than
subjects in the low anxiety level. The final and most important hypo-
thesis is that the greatest differences among the three groups would be
in the high anxiety level. Since these subjects had been assigned to
that level because of their high anxiety, the relaxation treatment
should have its greatest effect on subjects in this level. It was hypo-
thesized that they would have significantly higher EU ratings and signif—
icantly lower JVS ratings and ASQ scores than the two other groups in

that level.

Experimental Design and Analysis of the Data

 

A 2 x 3 factorial design was employed, as shown in Figure 2.2. The
first factor, physiological anxiety, had high and low levels. This
design was employed to provide greater precision. The second factor
contained the experimental manipulations. These were relaxation treat-
ment, information treatment, and no treatment. Each criterion measure

was tested independently.

Figure 2.2

Experimental Design

 

 

Relaxation Information No
Treatment Treatment Treatment
High
Anxiety
Low

 

 

 

 

Anxiety

 

47

A two-way Analysis of Variance (ANOVA) of the posttest was selected
as the analysis strategy as it provides an appropriate test of signifi—
cance with simple random assignment (Campbell & Stanley, 1963). Planned
comparisons were used to test for significance among treatments, between
anxiety levels, and in interactions. In these comparisons the relaxa-
tion treatment group was compared separately to the information treat—
ment group and to the no treatment group. The review of previous stud-
ies provided sufficient reason to expect that if significant differences
were found between treatments, they would occur between the relaxation
treatment and the other treatment groups. The use of this procedure
provided greater power than the omnibus F test to reject the null hypo-
thesis if it were false.

All cells contained equal numbers of subjects. This permitted the
ANOVA to be robust for violation of the assumption of equality of vari-
ance. Since fixed effects were used, the test was robust for the viola-

tion of the assumption of normality.

CHAPTER III

ANALYSIS OF RESULTS

Description of Results

 

The results of the analysis of the data are reviewed in this chap-
ter. Each of the measures, Empathic Understanding (EU), Judgments of
Verbal Anxiety (JVA), and the Anxiety Scale Questionnaire (ASQ) are
covered in that order. The means and standard deviations for each outcome
measure for cells, treatment groups, and levels of anxiety are given in
Table 3.1.

The data were analyzed by a Control Data 6500 computer at the Com-
puter Center, Michigan State University. Results had to attain the .05
alpha level before they were considered statistically significant.

A two—way Analysis of Variance of the posttest with planned compar-
isons was employed as the analysis strategy. There were three levels of
the treatment factor, including relaxation, information, and no treat-
ment. The anxiety factor contained high and low levels.

The following planned comparisons were made for each of the outcome
measures to test for main effects: (a) a comparison of the relaxation
treatment group and the information treatment group; (b) a comparison
of the relaxation treatment group and the no treatment group; and (c) a
comparison of the high and low levels of anxiety across treatment groups.
Two complex comparisons were employed to test for interactions of treat—

ment and anxiety level. The first compared high and low levels of

48

49

Table 3.1

Means and Standard Deviations of Cells, Treatment Groups,
and Levels of Anxiety for the Outcome Measures

 

Level of Anxiety

High X 3.05
3.0. .28
Low 2 2.85
S.D. .49

Total Treatment
Group

Empathic Understanding

Relaxation
Group
E 3.38
S.D. .34
i 2.92
S.D. .69
i 3.15
S.D. .52

Information
Group

E 2.7

S.D. .19
i 2.73
S.D. .24
i 2.71
S.D. .22

 

No Treatment

Group
E 3.09
S.D. .3
i 2.91
S.D. .55
i 3.0
S.D. .43

 

Level of Anxiety

High 2 6.17
3.0. 4.53
Low X 7.58
5.0. 4.27

Total Treatment
Group

Judgments of Verbal Anxiety

Relaxation
Group

R 7.0

5.0. 6.49
i 9.38
S.D 5.76
i 8.19
S.D. 6.13

Information
Group

R 7.75
S D. 4.01
i 6.75
S D. 2.25
i 7.2

S.D. 3.13

No Treatment

Group
R 3.75
S.D. 3.09
i 6.63
5.0 4.8
i 5.17
S.D. 3.95

 

Level of Anxiety

25.67

High i
S.D. 10.03

Low X
S

Total Treatment
Group

Anxiety Scale Questionnaire

Relaxation

i
5

Group

27.5
.D. 10.15

Information
Group
E 24.25
3.0. 8.62
E 22.25
3.0. 10.05
E 23.25
5.0. 9.34

No Treatment

Group
E 25.25
S.D. 11.32
E 30.25
5.0. 7.04
E 27.75
S.D. 9.18

 

50
anxiety with relaxation and information treatments. The second compared
both levels of anxiety with relaxation treatment and no treatment.
Table 3.2 shows the intercorrelation of the measures used in this
study. Only one, a .428 correlation between the JVA and the ASQ, was

significant (p less than .023).

Table 3.2

Intercorrelation of Measures

Electro- Anxiety Scale Judgments of
myograph Questionnaire Verbal Anxiety

Emphatic r = .223 .288 -.176
Understanding p less than .16 .097 .217
n = -22 24 24
Judgments of r = —.298 .428
Verbal Anxiety p less than .089 .428
n = 22 24
Anxiety Scale r = .055
Questionnaire p less than .404
n: 22

 

The significant, positive correlation indicated that the ASQ and the JVA
measure, to a moderate degree, the same component of anxiety.

Table 3.3 contains the five planned comparisons for the EU scale.
The comparison of the relaxation and information groups approached the
.05 level of significance (p less than .063). One cannot claim that
these results indicate a trend in the hypothesized direction, however,
the comparison between the relaxation and no treatment groups did not
approach significance (p .497). There was no significant difference
between the high and low levels of anxiety (p less than .265).

First the complex planned comparison, relaxation treatment and

51

Table 3.3

Analysis of Variance for the Empathic Understanding Scale

Source of Variance df Mean Square F Ratio P Value

 

Paired Planned Comparisons

Relaxation Treatment and 1 .705 3.9 .063
Information Treatment -

Relaxation Treatment and .087 .481 .497
No Treatment

High and Low Levels of 1 .24 1.327 .265
Anxiety

Complex Planned Comparisons

Relaxation and Information 1 .159 .877 .361
Treatment and High and Low
Levels of Anxiety

Relaxation and No Treat— 1 .078 .433 .519
ment and High and Low
Levels of Anxiety

Error 18 .181

 

52

information treatment groups were compared to high and low levels of
anxiety; this comparison was not significant (p less than .361). The
second complex comparison of the relaxation and no treatment groups and
high and low levels of anxiety was also not significant (p less than
.519).

Table 3.4 contains the results of the planned comparisons for the
JVA scale. There were no significant differences between the comparison
of the relaxation and information groups (p less than .783), or between
the relaxation and no treatment groups (p less than .212). The means
of these groups correspond to one of the research hypotheses and apriori
reasoning; these are 8.19, 7.2, and 5.17 for the relaxation, information
and no treatment groups, respectively. One would expect that the real
treatment group would have the highest mean, followed by the placebo
group, then the no treatment group. The failure to find significance
might be related to the large standard deviations for these groups.
This would result in inflated mean squares, thus making significance
difficult to obtain. The comparison for the high and low levels of
anxiety was not significant (p less than .464). The complex comparisons
of the relaxation and information groups and high and low levels of anx-
iety were not significant (p less than .379), nor was significance found
for the complex comparison for the relaxation and no treatment groups
and the two levels of anxiety (p less than .915).

Table 3.5 shows the results of the planned comparisons for the ASQ.
The comparison of the relaxation and information groups was not signif-
icant (p less than .291), nor was the comparison of the relaxation and
no treatment groups (p less than .913). There were no significant dif—

ferences between high and low levels of anxiety (p less than .823). The

53

Table 3.4

Analysis of Variance for the Judgments of Verbal Anxiety Scale

 

Source of Variance df Mean Square F Ratio P Value

 

Paired Planned Comparisons

Relaxation Treatment and 1 1.688 .078 .783
Information Treatment

Relaxation Treatment and 1 36.0 1.672 .212
No Treatment

High and Low Levels of 1 12.042 .559 .464
Anxiety

Complex Planned Comparisons

Relaxation and Information 1 17.521 .814 .379
Treatment and High and Low
Levels of Anxiety

Relaxation and No Treat- 1 .25 .012 .915
ment and High and Low
Levels of Anxiety

Error 18 21.535

 

54

Table 3.5

Analysis of Variance for the Anxiety Scale Questionnaire

 

 

Source of Variance df Mean Square F Ratio P Value

 

Paired Planned Comparisons

Relaxation Treatment and
Information Treatment

Relaxation Treatment and
No Treatment

High and Low Levels of
Anxiety

Complex Planned Comparisons

Relaxation and Information
Treatment and High and Low
Levels of Anxiety

Relaxation and No Treat—
ment and High and Low
Levels of Anxiety

Error

18

96.333

1.0

4.167

24.083

3.25

81.222

1.186

.012

.051

.297

.372

.291

.913

.823

.593

.549

 

55
complex comparison of the relaxation and information groups and the high
and low levels of anxiety was not significant (p less than .593). The
second complex comparison of the relaxation and no treatment groups and

both levels of anxiety was also not significant (p less than .549).

Conclusion

 

It was concluded that the results of the analysis of the data sup-
ported the retention of the null hypothesis. Neither information nor
relaxation treatment were more effective than no treatment in reducing
anxiety or increasing empathy as measured by the respective scales.
Moreover, no significant differences were found between levels of anxi-
ety on these measures. There were no significant interactions.

The moderate positive correlation (.428) between the ASQ and the
JVA was statistically significant (p less than .023). This indicated that
both of these instruments measured some aspect of the subjects' anxiety.
This finding lends support to the validity of the JVA in the assessment of

anxiety.

CHAPTER IV
_SUMMARY, DISCUSSION, AND CONCLUSION

Summar

The purpose of this study was to measure the effect of relaxation
on counselors' communication of empathy, verbal anxiety, and general
anxiety. The subjects were 22 masters and two doctoral students in
their initial counseling practicum at Michigan State University. They
were initially divided into high and low anxiety levels by scores
obtained from frontalis muscle electromyograph (EMG) readings. To
obtain these readings, a baseline period EMG reading was subtracted from
an EMG reading taken while the subjects watched a videotape of a coached
client who intentionally made anxiety-provoking statements. The median
score of this variable was used to divide subjects into high and low
levels of anxiety.

The subjects in the high and low anxiety levels were randomly
assigned to three groups. The first group received approximately 30
minutes per week of frontalis muscle EMG biofeedback provided by this
researcher and three 25-minute sessions per week of audiotaped Autogenic
Training relaxation instructions. The information group was seen for 10
to 15 minutes per week by this researcher and listened to approximately
90 minutes per week of audiotaped information in counseling and psychol—
ogy. The third group received no treatment.

At the end of the four-week treatment period, each subject

56

57
interviewed a coached client for 10 minutes. Audiotapes of these inter-
views were used to obtain ratings of the two outcome measures, the
Empathic Understanding scale (EU), and the Judgments of Verbal Anxiety
scale (JVA). A third outcome measure, the Anxiety Scale Questionnaire
(ASQ), was administered to the subjects immediately after these
interviews.

Each outcome measure was analyzed by a posttest Analysis of Vari—
ance. There were no significant differences among treatment groups or
between high and low levels of anxiety for any of the measures. There
were no significant interactions. The planned comparisons of the relax-
ation and information groups for the EU scale approached the .05 level
of significance (p less than .063). The correlation between the ASQ and
the JVA was moderate, though significant (.428, p less than .023).

The subjects' performance on the outcome measures was compared for
any relationship to various demographic variables as described in Table
2.1. The specific variables which were examined were sex, age, years of
experience in human services and counseling, marital status, and coun-
seling program. While it is possible that there is a relationship
between any of these variables or a combination thereof in the larger

population, none was identified in this sample.

Discussion

 

Results Approaching Significance

 

Only one of the planned comparisons approached significance. The
mean of the Empathic Understanding scale for the relaxation treatment
group was higher than the mean for the information treatment group
(p less than .063). This could be interpreted as supporting the

research hypothesis in the direction of significance but for an

58
additional factor: there was no difference approaching significance
between the relaxation treatment and the no treatment groups on the same
measure. Therefore, any suggestion that relaxation may have been more
effective than information treatment in increasing EU ratings is
countered by the finding that no treatment was as effective as
relaxation.

Possible Sources of Nonsignificance

 

This section contains a review of possible reasons for the failure to
find significance in this study. Any one of these factors, alone or in
combination with others, may have been the source of lack of signifi-
cance. This discussion is also offered to indicate those areas where
the alleviation of methodological shortcomings may improve the design
of future research.

Theoretical Basis of the Study. A basic premise of this study was

 

that anxiety interfered with the communication of empathy. This postu-
lation can draw from two sources of support. The first is the widely
established relationship between anxiety and performance (e.g., Gaudey &
Spielberger, 1971; Krug et al., 1976), in which high levels of anxiety
usually interfere with performance of specific tasks. The reduction of
anxiety leads to improved performance. One may expect that is as true
of the communication of empathy as it is of other types of tasks.

' There is, moreover, further evidence to support the relationship of
the reduction of anxiety with the increase of empathy. A review of
several of the studies cited in Chapter I indicated that the reduction of
anxiety resulted in an increase in empathy. Fontaine (1976), for
example, found that a group of graduate students in counselor education

and social work which received alpha wave training (AWT) relaxation were

59
significantly higher on Accurate Empathy ratings than a similar group who
received bogus AWT. Fry (1973), using human service workers as subjects,
found that the group which received systematic desensitization to emo—
tionally positive client behaviors were rated higher on Carkhuff's rat-
ings of empathy, respect, genuineness, and concreteness than a no treat-
ment group. These results SUggest that further investigation of this
hypothesis is warranted.

Formation of Levels of Anxiety. The effectiveness of the methods

 

used to establish high and low levels of anxiety must be considered.

The procedure of subtracting the baseline frontalis muscle EMG from the
EMG taken while the subjects watched an anxiety-producing videotape to
obtain a leveling variable has not been used previously in counseling
research. The validity of this procedure could only be supported by
evidence that subjects were appropriately placed in high and low levels
of anxiety. Subjects in the high anxiety level of the relaxation group
making the greatest gains in the hypothesized direction would constitute
such evidence. Since no significant differences were found, the validity
of this procedure has yet to be established.

The above aspect of this study should receive further inquiry. The
use of the EMG output from the frontalis muscle as an index of anxiety
has many advantages, such as its sensitivity to threatening interpersonal
and social stimuli (Brown, 1977). Moreover, it is not susceptible to
response distortion as are some standardized tests. Further research
with frontalis EMG output could provide information regarding the valid-
ity of this procedure as a blocking or outcome variable. Some of the

factors which may have led to the lack of significance in using this pro-

cedure in this study include unreliability of measurement or inadequate

60
validity for its specific purpose.

This procedure was intended to separate subjects into high and low
anxiety levels. If it failed to accomplish this, one could not ade—
quately test the hypothesis that relaxation treatment would have its
greatest effect on subjects in the high anxiety group. This would,
however, not influence the test for the main effects for treatment. The
high and low anxiety levels were disregarded in the test for significant
differences among treatment groups.

There were no significant correlations of the EMG scores with either
of the other two anxiety measures or with the EU scale. These findings are
similar to those of some previous studies in which there has been little
relationship between physiological and other measures of anxiety (Bowman
et al., 1977; Monke, 1971; Roberts, 1977).

Sample Size. Small sample size is often cited as a possible reason

 

for lack of significance. This may have been a factor in this study
where the total sample size was 24, with eight per treatment group and
four per cell. There are two problems caused by small sample sizes.

The first is the reduced degrees of freedom which in turn requires a
higher F ratio to obtain significance. The second problem caused by a
small sample size is the increased likelihood that extreme subject varia-
tion on the outcome measure will reduce the mean treatment effects. In
this study, for example, the presence of one or two individuals in the
relaxation group who were remarkably nonempathic before treatment and
remained so despite treatment may have accounted for lack of significance
on that outcome measure.

Treatments. Informal researcher observation and reduced EMG bio-

 

feedback thresholds indicated that relaxation group subjects became more

61
relaxed as treatment progressed. It is possible, however, that the
effects of global relaxation were not generalized to counselor-client
interactions. If such were the case, a more specific technique of anxi-
ety reduction, such as progressive relaxation and systematic desensitiza—
tion, might have proven more effective in reducing counselor anxiety.
This treatment procedure has been effective in previous studies (Miller,
1973; Fry, 1973; Carter & Pappas, 1975).

Another concern is the length of the treatment period. Clients in the
relaxation group had a total of approximately 16 hours of treatment.

This might not have been long enough to maximize the effects of Autogenic
Training, and a longer period might have enhanced treatment. This could
be supported by the lack of significant differences among treatment
groups on the general anxiety measure, the ASQ. If the relaxation treat-
ment had been effective, a reduction of generalized anxiety as measured
by this instrument would have been expected.

There is also the possibility that the information treatment was not
an effective placebo. The information contained in the audiotapes might
have had a beneficial effect on the subjects' communication of empathy,
or might have in some manner reduced the anxiety which they experienced in
the counseling situation. For example, some of R0gers' statements might
have helped these subjects see themselves as persons who could aid others
in becoming more open to experience. Although this appears improbable,
it must be mentioned as a possible factor.

There are several difficulties in giving placebo treatments. For
one thing, it is difficult to provide a placebo which appears to be a
treatment, yet has no treatment effects (Campbell & Stanley, 1963). An-

other problem is the beneficial effect of participating in research,

62
regardless of the treatment given (Borg & Gall, 1976). The subjects'
belief that they are receiving something that should make them better
counselors may be sufficient to produce gains on the outcome measures.
Receiving treatments or simply being included in a research study may
cause the subject to attend to certain aspects of their performance, thus
effecting their rating on the outcome criteria.

A final concern related to the treatments provided in this study was
the possible effect of a single treatment-giver. One must ask whether
there were any qualities regarding this researcher, who administered all
treatments, which nullified any positive effects of treatment. Although
this is unlikely, it must be included as a possible consideration.

Sex of Coached Clients. The exclusive use of women as coached

 

clients for the outcome interviews represents a limitation of this study.
Although the rationale for this has been discussed, it is possible that
the use of another procedure to control for the sex of the coached client
would have yielded different results for the EU and JVA scales. While
this procedure probably did not contribute to the lack of significant
findings, the use of only women coached clients would limit the generali-
zations which could be made from this study.

Effect of Practicum and Supervision. Another factor which might

 

have contributed to the lack of significant findings was the effect of
practicum and supervision. It is possible that the combined effects of
instruction during regular supervision and continual exposure to clients
during the practicum resulted in increased counselor empathic communica-
tion and decreased counselor anxiety. These inadvertent treatments might
have had effects on all subjects such that the additional effects of

relaxation were minimized. Consequently, significant differences among

63
groups could not have been measured.

Outcome Measures. Consideration must be given to the appropriate—

 

ness of the outcome measures used in this study. Each measure will be
considered individually.

The EU scale has been used widely and successfully in counselor
education research (Hefele & Hurst, 1972). Its use as an outcome measure
in this study was further supported in that a variation of Carkhuff's
training model was used in teaching specific response skills to most of
the subjects. Furthermore, interrater reliability was very high. It
seems likely that if the treatments had had the hypothesized effect, this
instrument would have been capable of measuring differences.

The JVA scale was chosen for its previously demonstrated ability to
discriminate between counselors with high and low levels of verbal anxi-
ety (Russell & Snyder, 1963). The conditions of the previous study were
unlike those of the present study, however. Russell and Snyder used
coached clients who intentionally produced high and low levels of anxiety
by playing the roles of hostile and friendly clients. Using the JVA,
these researchers were able to measure differences between treatments and
levels of experience. In the present study, the coached clients were not
instructed to provoke anxiety in the subjects, but to play a normal
client role. Thus, the subjects' verbal anxiety might not have been
marked enough to be measured by this measurement.

The correlation between the JVA scale and the ASQ was moderate and
significant (.428, p less than .023). This indicated that the JVA scale
was measuring some aspect of anxiety which was also measured by the ASQ.

This finding lends support to the validity of this instrument. While

the JVA identified some level of anxiety and can achieve acceptable

64
interrater reliability, it did not detect any differences among treatment
groups or levels of anxiety.

The ASQ has the highest demonstrated validity and reliability of the
three measures used in this study (Krug et al., 1976). Previous studies
have indicated that the ASQ is capable of measuring the effects of relax—
ation treatments (Nicoletti, 1973; Sherman et al., 1974; Sherman & Plummer,
1973). As mentioned previously, the lack of effectiveness of relaxation
in this study as measured by this instrument was more likely due to
inadequate treatment than to problems with this measure.

Implications for Future Research

 

The theory which motivated this study--that the reduction of anxiety
will lead to improved counselor empathy and verbal performance--has occa-
sional support in previous research and may warrant consideration for
continued investigation. The following discussion includes recommenda-
tions for further research in this area taken from observations on this
study.

One approach to broaden understanding of this area would be to
devote further study to examining the effect of anxiety on counselor per-
formance as a precursor to studying the effects of treatments on reducing
anxiety. This has been the approach of Russell and Snyder (1963),
Roberts (1977), and Bowman and colleagues (Bowman & Roberts, 1978, 1979;
Bowman et al., 1977). Some of the areas which may be beneficial for fur-
ther inquiry are the types of measures which best assess counselor anxi-
ety over the course of counselor preparation, the effect of counselor
anxiety on interview behavior, and the characteristics which may be used
to identify levels of anxiety. Future researchers would be better able

to measure treatment effects with a more solid measurement basis.

65

A variety of physiolOgical measures have been used in previous
studies, including alpha waves (Fontaine, 1976), skin conductance (Monke,
1971), heart rate (Bowman et al., 1977), eyeblink rate (Russell & Snyder,
1963), and skin temperature (Walters, 1978). Recent research has indi-
cated that hormones which are produced as a result of stress and anxiety
can be measured in blood and urine samples (Mason, 1975). It may be
appropriate to assess idiographic physiological responses to anxiety
stimuli. For some individuals, anxiety may be best measured by EMG; for
others, heart rate may be the best indicator of anxiety. It seems likely
that the most effective procedures for the assessment of physiological'
anxiety would be to identify the most reactive modality for an individual.
The use of such measurements can add greatly to the body of knowledge
concerning the sources, effects, and treatment of anxiety. Of course,
the goal for counselor educators is the reduction of anxiety which
impairs counselor performance.

An increase in sample size will almost always be beneficial in
improving precision. However, consideration should be given to working
with anxious practicum students using an intensive design. Such a design
could include individualized treatment given over a longer period of
time, with the subjects themselves serving as controls. Subjects’ per-
formance on outcome criteria could be measured at several periods during
which they may or may not have received treatment.

While this study focused on empathy and verbal and general anxiety,
there are other important counselor variables which could be considered
in the measurement of the effects of relaxation on counselor performance.
These include the rating of nonverbal counselor behavior, client rating

of counselor performance, measurement of change in counselor personality

66
variables, and, in the case of real clients, the assessment of client
behavior such as depth of self-exploration or behavioral change. These
variables may assess changes which cannot be adequately measured by those

examined in this study.

Conclusion

 

This study failed to demonstrate that relaxation treatment resulted
in increasing counseling practicum students' empathy and decreasing their
verbal and general anxiety when compared to information treatment and no
treatment. Furthermore, there were no significant differences between
the two levels of anxiety on any of the outcome measures. Despite the
lack of significance, previous research has given some support to the
hypothesis that relaxation treatment improves counselor performance. The
recommendations discussed in the preceding section may be beneficial in

providing greater precision in future research in this area.

APPENDIX A

APPENDIX A

PRESENTATION TO PRACTICUM GROUPS FOR SOLICITATION OF VOLUNTEERS

I have asked to give me a few minutes
(name of practicum supervisor)

 

of your time to ask your help in some research I'm doing. I'm interested
in studying the relationship between practicum students' physiological
anxiety and counseling skills. Specifically, I would like to ask for
volunteers who are interested in trying biofeedback equipment. As a
result of the measurements which I will take, certain students will be
offered the opportunity to participate in a further experience to measure
the effectiveness of methods of relaxing and improving counseling skills.

I'd like to say a little about biofeedback. This is a procedure
whereby a person is given information on some physiological state of
their body—-their heart rate, the heat their body gives off, or the
amount of activity in a muscle group, for example. The person is given
this information in the form of a sound signal, a number, or the reading
on a dial. He or she is then able to raise or lower the signal in the
desired direction by various techniques. Biofeedback can be used to
reduce heart rate, increase the temperature in one part of the body, and
indicate when a person is more relaxed. It is remarkable how biofeedback
enables people to gain control over some physiological responses which
they previously were unable to control.

The specific equipment with which I am working is the electromyograph.

67

68
This is a device which measures the energy which is given off by a par-
ticular muscle group, in this instance the muscles on the forehead.
These muscles are particularly sensitive to tension and nervousness which
occur throughout the body.

What I am offering you is an opportunity to be placed on the EMG
device for about 15 minutes. The EMG is measured by electrodes which I
will place on your forehead. They stick with a little bit of adhesive
and there is little discomfort. For the first eight minutes I will
record your EMG level; for the last six minutes I will measure your reac—
' tion to the videotape of a client. This client will say things which
would tend to provoke anxiety in some counselors.

I am asking you to do this for two reasons. Biofeedback is becoming
very popular in a wide variety of settings such as education, counseling,
psychology, medicine, and rehabilitation. This is an opportunity for you
to experience this innovative technique._ Secondly, your participation
will provide me with data on how anxious counselors are with clients and
whether they can learn to relax.

As I mentioned previously, I will obtain data on the EMG reading for
all volunteers. From among these, some students will be offered the
opportunity to further participate in this study. They will be assigned
to one of three groups-one will receive relaxation training incorporat-
ing biofeedback, another will receive helpful information regarding
issues in counseling and psychology, and the third will not receive any
treatment. The effectiveness of these treatments will be measured by
ratings taken from participant interviews with clients which will be
arranged.

Any ratings taken in this project will not affect participants'

69
grades in this course. Individuals' data will be confidential and will
be known only to me. The time requirement for those who participate in
the second aspect of this study will be 1% to 2 hours per week for four
weeks. Most of this can be done on your own time: 30 minutes of this per
week will be scheduled with me at your convenience. Those in the group
which will receive no treatment will not be required to do anything dur-
ing this period. During the fifth week, I will ask all participants to
complete an anxiety questionnaire and to conduct a 10-minute interview

with a client whom I will provide.

APPENDIX B

APPENDIX B
SCRIPT OF VIDEOTAPE OF ROLEPLAYED CLIENT

Statement One
Well, I've had these problems for a long time. I've been talking
mostly to my friends but that never seems to help. Things just keep
building up and up--sometimes I just want to explode--sometimes I just

don't think I can go on any more. I hope you can help me.

Statement Two
I don't know how I got to be this way. Maybe it was my parents. My
father would get drunk and come in late and my mother would be lying in
the bed and he would ... well, you know. It was pretty horrible--I still

have nightmares. How can I get rid of these nightmares?

Statement Three
May be that's part of my problem with other men—-you know, wanting
to be close to them—~wanting to be very close with them. You, you're so
kind and understanding, I think you're the one who will be able to help

me.

Statement Four
I know a lot of this is really heavy. It's a lot to lay on someone
new to this business like you. But can you handle it? I mean, do you

know what I feel? I got all this stuff inside me. Do you really think

70

71
you can help me? Are my problems too much for someone who is just a

student?

Statement Five
I've noticed something about you that's a little, well, different.
I mean that's cool with me--people think I'm strange and I'm really okay.
But, you know, I just want to check out with you if the way you are
causes problems for people. Come on, you know what I mean. You can be

straight with me. Why should I care?

Statement Six
This isn't getting me anywhere! I've talked and talked and I've
still got these things inside me eating me up. What good can talking do?
How can that help me? How can talking help solve my problem? You're
supposed to be the counselor ... counsel me! Solve my problems! If you

can.

Statement Seven
Things really started getting bad in Viet Nam. I was a grunt--an
infantry man. It was my job to kill gooks. I was a good grunt too.
Course I was never straight. Every time I went out on a mission I got
high on anything I could smoke, snort or shoot up. It helped me do my
job better. "If they're gooks, blow 'em away," I used to say. Didn't
matter if they were on their side, on our side, or civilian. You can't

trust any of them.

Statement Eight
You know, I like having someone to talk to. It kind of takes some

of the load off. I like talking to someone who doesn't jump in with

72
advice or ask dumb questions or tell me to forget it and feel better. I

really like talking to you.

Statement Nine
Can we be friends? I hope so. I hope you can come out from behind
that counselor mask and really be a friend to me. You know I find you an
attractive person-strong, secure, safe to be with. What a shame to let
it melt away at the end of the session. Maybe we can get together and

talk some more later on. Would that be okay?

Statement Ten
Sometimes I'm not sure you like me or even want to help me. I get
this hunch that there's something about me that you don't like. And I
think that I know what it is. It kind of pisses me off, knowing that
you dislike me for that reason. I don't think it's very nice of you to
feel that way. You're supposed to be a caring person. I wonder how many

other poor bastards you screwed up because of your attitude.

Statement Eleven
I feel like I've done all the work here. I've talked about some
things that were really hard for me-really painful. There you are, not
saying a damn thing. Do you even care about me? Do you even know who I
am? You're starting to piss me off. I'd like to kick the shit out of
you. Give me one good reason why I shouldn't kick the living shit out

of you.

APPENDIX C

APPENDIX C

AUTOGENIC PHRASES

I feel very calm and quiet.

I feel very comfortable and quiet.

I am beginning to feel quite relaxed.
I am beginning to feel quite relaxed.

My
My
My
My
My

MY

feet feel heavy and relaxed.

feet feel heavy and relaxed.

ankles feel heavy and relaxed.

ankles feel heavy and relaxed.

knees feel heavy and relaxed.

knees feel heavy and relaxed.

hips feel heavy and relaxed.

hips feel heavy and relaxed.

feet, my ankles, my knees, and my hips all feel heavy and relaxed.
feet, my ankles, my knees, and my hips all feel heavy and relaxed.
stomach and the whole center portion of my body feel heavy and relaxed.
stomach and the whole center portion of my body feel heavy and relaxed.
hands feel heavy and relaxed.

hands feel heavy and relaxed.

arms feel heavy and relaxed.

arms feel heavy and relaxed.

shoulders feel heavy and relaxed.

shoulders feel heavy and relaxed.

hands, my arms, and my shoulders all feel heavy and relaxed.
hands, my arms, and my shoulders all feel heavy and relaxed.

neck feels heavy and relaxed.

neck feels heavy and relaxed.

jaws feel heavy and relaxed.

jaws feel heavy and relaxed.

forehead feels heavy and relaxed.

forehead feels heavy and relaxed.

neck, my jaws, and my forehead all feel heavy and relaxed.

neck, my jaws, and my forehead all feel heavy and relaxed.

whole body feels heavy and relaxed.

whole body feels heavy and relaxed.

breathing is getting deeper and deeper.

breathing is getting deeper and deeper.

I can feel the sun shining down on me warming the top of my head.
The top of my head feels warm and heavy.

The top of my head feels warm and heavy.

The relaxing warmth flows into my right shoulder.

My

right shoulder feels warm and heavy.

My right shoulder feels warm and heavy.

73

74

My breathing is getting deeper and deeper.

The relaxing warmth flows down to my right hand.

My right hand feels warm and heavy.

My right hand feels warm and heavy.

The relaxing warmth flows back up to my right arm.

My right arm feels warm and heavy.

My right arm feels warm and heavy.

The relaxing warmth spreads up through my right elbow into my right
shoulder.

My right elbow, my right shoulder feel warm and heavy.
My right elbow, my right shoulder feel warm and heavy.
The relaxing warmth flows slowly throughout my whole back.
I feel the warmth relaxing my back.

My back feels warm and heavy.

My back feels warm and heavy.

The relaxing warmth flows up my back and into my neck.
My neck feels warm and heavy.

My neck feels warm and heavy.

The relaxing warmth flows into my left shoulder.

My left shoulder feels warm and heavy.

My left shoulder feels warm and heavy.

My breathing is getting deeper and deeper.

The relaxing warmth flows down to my left hand.

My left hand feels warm and heavy.

My left hand feels warm and heavy.

The relaxing warmth flows back up to my left arm.

My left arm feels warm and heavy.

My left arm feels warm and heavy.

The relaxing warmth spreads up through my left elbow into my left
shoulder. '

My left elbow, my left shoulder feel warm and heavy.
My left elbow, my left shoulder feel warm and heavy.
The relaxing warmth flows to my heart.

My heart feels warm and easy.

My heart feels warm and easy.

My heartbeat is slow and regular.

My heartbeat is slow and regular.

The relaxing warmth flows down into my stomach.

My stomach feels warm and quiet.

My stomach feels warm and quiet.

My breathing is deeper and deeper.

My breathing is deeper and deeper.

The relaxing warmth flows down into my right thigh.

My right thigh feels warm and heavy.

My right thigh feels warm and heavy.

The relaxing warmth flows down into my right foot.

My right foot feels warm and heavy.

My right foot feels warm and heavy.

The relaxing warmth flows slowly up through my right calf, to my right
knee, to my right thigh.

My right leg feels warm and heavy.

My right leg feels warm and heavy.

My breathing is deeper and deeper.

My breathing is deeper and deeper.

75

The relaxing warmth flows down into my left thigh.
My left thigh feels warm and heavy.

My left thigh feels warm and heavy.

The relaxing warmth flows down into my left foot.
My left foot feels warm and heavy.

My left foot feels warm and heavy.

The relaxing warmth flows slowly up through my left calf, to my left

knee, to my left thigh.

My left leg feels warm and heavy.
My left leg feels warm and heavy.
My breathing is deeper and deeper.
My breathing is deeper and deeper.

The relaxing warmth flows up through my abdomen, through my stomach and

into my heart.

My heart feels warm and easy.

My heart feels warm and easy. .

My heart pumps relaxing warmth throughout my entire body.
My whole body is heavy, warm, relaxed.

My whole body is heavy, warm, relaxed.

I am breathing deeper and deeper.

I am breathing deeper and deeper.

My whole body feels very quiet and very serene.

My whole body feels very comfortable and very relaxed.
My mind is still.

My mind is quiet.

My mind is easy.

I withdraw my thoughts from my surroundings.

Nothing exists around me.

I feel serene, secure, still.

My thoughts are all turned inward.

I am at ease, completely at ease.

Deep within my mind I can visualize and experience myself as
I am comfortable and still.

My mind is calm and quiet.

I feel an inward peace.

I feel a new sense of well being.

I am breathing more and more deeply.

I am going to count to myself from ten to one.

As I approach one, I will feel more awake and alert.
When I reach one, I will be fully awake and fully alert.
Ten

Nine

Eight

Seven

I’m beginning to feel more alert. I open my eyes.
Six

Five

I'm beginning to move my arms, my hands, and my neck.
Four

Three

I'm fully alert.

Two

One

Now I am fully alert, fully refreshed.

My time for relaxation is over.

relaxed.

APPENDIX D

APPENDIX D

CONTENTS OF AUDIOTAPES

These audiotapes were transcribed from master tapes which are avail-
able from the Voice library, Main Library, Michigan State University,
East Lansing, Michigan.

Tape Number One
25 minutes

Bruno Bettelheim discusses the Freudian implications of children's
fairy tales.

Tape Number Two
55 minutes

B. F. Skinner discusses the manner in which sophisticated rein-
forcers shape social behavior and the benefits of applying learning
principles in the creation of a behaviorist Utopia.

Tape Number Three
30 minutes
R. D. Laing discusses the differences between perceived and actual
relationships among family members.
Tape Number Four
50 minutes
B. F. Skinner and Carl Rogers debate the relative merits of behav-
iorism and humanism.
Tape Number Five
30 minutes
Carl Rogers discusses the "new person," who is open to awareness and

assumes responsibility for his or her behavior, and is therefore able to
achieve self-actualization.

76

77

Tape Number Six
50 minutes

Rollo May and B. F. Skinner respond to audience questions regarding
authoritarianism and moral choices.
Tape Number Seven
45 minutes
Lawrence Kohlberg discusses his theory of moral development and the
need for moral education.
Tape Number Eight

30 minutes

R. D. Laing discusses problems in communication among family members
as a cause and consequence of emotional disorders.

APPENDIX E

APPENDIX E
INSTRUCTIONS TO COACHED CLIENTS

The role you are playing is that of a 21-year—old single woman who
lives on campus. She is currently an elementary education major but has
doubts about how much she enjoys this field and is not performing well
in her courses. Her problems are underscored by her friendship with a
roommate who has the same major. The roommate attends parties more
frequently than the person whom you are portraying but gets much better
grades. You are concerned about having made a mistake in majoring in
elementary education, yet are apprehensive about having to "go back,"
that is, having to change majors and perhaps having to attend school for
an additional year or more. The woman you are portraying has periods
of depression, doubt, and confusion. She is unsure about her career
goals and about her choice of a major.

Give a brief (several sentences) description of the person you are
role playing (in the first person, of course) at the beginning of the

session if the counselor asks for it. Emphasize your feelings more than

 

facts, but include some of the latter. Don't feel pressured into talking
by periods of silence. The counselor has responsibility for the direc—
tion of the interview. If the silence continues for a 90-second period,
you may restate the 1325 feeling you had. For example, you might say,

"I'm really confused/depressed/hurt now.”

78

APPENDIX F

APPENDIX F

RELEASE OF INFORMATION FORM

I have been informed of the nature of this study. I give the
undersigned researcher the right to use the audiotape of my inter-
view with the coached client and the anxiety questionnaire for
rating purposes. I am aware that any personal information obtained
during this study is confidential, and no information pertaining to

any participant will be shared with any other person.

 

(Signature of Student)

 

Richard W. Ober

79

APPENDIX C

APPENDIX G

JUDGMENTS OF VERBAL ANXIETY

Criteria and Examples for the Detection of Anxiety

in Counselor Interview Statements

Direct statements of anxiety (the counselor refers to subjective or
physiological expression of his own feelings of tension generated in
the interview).

Counselor:
Counselor:
Counselor:
Counselor:
Counselor:

You make me feel a little uncomfortable.

This interview is making me a bit tense.

I'm sweating a bit.

Talking about these things makes me feel uncomfortable.
You're pushing me kind of hard. [Also shows Criterion 6.]

Avoidance of the interaction between the two by the counselor asking
questions or making statements which serve to change the subject or
divert the discussion, the questions or statements introducing new
material which is irrelevant to the topic being discussed. (Asking
a question or introducing a new topic when it is requested or when
you would judge that the previous issue is exhausted would not con-
stitute an avoidance of the interaction by the counselor.)

Client:

Counselor:

Client:

Counselor:

Client:

Counselor:

Client:

Counselor:

School is enjoyable for me in general. It's just that
my grades are poor.
I would like us to look at your home life for a moment.

My advisor is making things miserable for me. He never
lets me take electives I want, but only things he goes
for.

Did you take the academic course or vocational course
in high school?

I feel that this school is like a country club, and
people really should come here to work.
Do you live in one of the dorms or in town?

My roommate is rather a weird guy. He and I don't get
along too well.

It is important for you to think about your progress
to date.

Direct interruptions (the client's verbalizations are interrupted

8O

81

by the counselor asking a question or making a statement of his own

whether it is pertinent or not.
ments of simple agreement like "Uh-huh," "Yeah,” etc.

The exception to this are state-
Simple agree-

ment includes such things as "I see," "Oh, yes," etc., and does not

"constitute an interruption.

However, longer statements of agree-

ment, e.g., "Oh, yes, I see what you mean" would satisfy this crite-

rion.

It is not an interruption if the counselor has been inter-

rupted and he continues his statement even though he interrupts the

client in the process.

The fact that he is continuing his own

statement will be indicated on the transcript by two hyphens [-—]
before his statement and by the fact that usually his statement will

begin with a lower-case letter.

[A direct interruption is indicated

on the transcript by the client's verbalization being followed by a

dash, --.])

Client:

Counselor:

Client:

Counselor:

Client:
Counselor:

Client:

Counselor:

Client:

Counselor:

Client:
Counselor:

Fraternity life is ridiculous. There's all this secret
stuff that --
Tell me something about your relationship with your

roommate 0

My brother is okay, but there are a lot of areas of dif-
ferences that bOther --

How about your relationship with your parents, your mom
and dad?

My advisor doesn't really seem to understand me --
You feel that he's really not able to see what your
problems are.

Well, I can't recall what --
You can't recall what you did know.

It seems to be a matter of disorganization.
things togeth ~-
Would you describe this as a feeling of panic?

I can't put

I get tense. It's a problem of -—
You say that you feel you tighten up during exams?

Impersonal interpretations or premature interpretations which block

the client's discussion.

[”Interpretation" is defined here as an

explanation by the counselor of the meaning or significance of what

the client has or has not done, said, or thought.]

(These interpre-

tations may be too deep for him to follow, they may be too deep and
too threatening, or they may be so impersonal that the client is

unable to relate them to his own situation with meaning.

The inter-

pretations may be correct or incorrect, but they must be imper-
sonal or premature in order to satisfy this criterion.)

Client:

Counselor:

I want to consider my family in a moment, but first I
think I ought to add that I find I tend to reject what
my advisor says without really thinking about it.

It seems as though you're rejecting things your advisor
says much like you would with your father.

82

Client: My dating relationships are rather good. I'm sure that
this isn't causing any of my problems.

Counselor: The fact that you deny the importance of this area so
quickly indicates that there is something wrong here.

Client: I'm in a lot of extracurricular activities, maybe too
many.

Counselor: The need to join a lot of things often shows insecurity
on the part of the person.

Client: And there's another area of difficulty too. My father
always tries to run my life.

Counselor: You really feel guilty about not going along with what
your father suggests even though you feel it's for your
own good.

Unnecessary reassurance. (Either too much reassurance or an
inappropriate attempt is made by the counselor to instill confidence
in the client regarding something he has or has not done, said, or
thought, or to indicate that there will be a favorable outcome to a
problem. There may be too great a use of suggestion, of the pres-
tige of the counselor, of persuasive arguments, or of the citation
of parallel cases in this attempt to reassure.)

Counselor: I'm confident you'll do very well when we get these
problems straightened out.

Counselor: You shouldn't feel guilty about this. It's normal to
have these thoughts.

Counselor: Other people have done things as bad.

Counselor: It's perfectly all right to feel that way. Many people
do.

Counselor: I'm sure you'll be able to work these problems out in
counseling.

Disapproval. (This includes unwitting disapproval wherein the coun-
selor either intentionally or unintentionally indicates that he dis-
approves of something the client has or has not done, said, or
thought. The client's behavior is revealed as being unacceptable in
some way to the counselor. Disapproval is shown through content and
sometimes voice. Usually content will be the most important except
in cases of sarcasm.)

Counselor: That certainly wasn't a very smart thing to do.

Counselor: Let's go to something a little more interesting.

Counselor: That's all very well, but there doesn't seem to be much
point to it.

Counselor: Let's get to the heart of the matter.

Counselor: You've really gotten your life mixed up, haven't you?

Counselor: I doubt if this topic will lead to any real solution of
your problems.

Counselor intellectualizing and not responding to the client's feel-
ings. ["Intellectualizing” is defined here as the intentional or

 

83

unintentional analysis of a problem or discussion of it in purely
intellectual terms to the neglect or exclusion of feelings or pract—
ical considerations.]

Client: My brother and I don't get along well.
Counselor: Relationships with siblings often are disturbed and can
lead to problems in other areas, even school.

Client: I get frustrated because I can't remember.
Counselor: This seems to be an example of the power of repression.
[Also shows Criterion 4.]

Client: My real trouble is with tension on exams.
Counselor: This is a common problem and its roots typically are
found in poor study habits. [Also shows Criterion 4.]

Client: I was really mad at the guy.
Counselor: At certain times one's feelings are disturbed by the
behavior of others.

Client: I've even gotten the feeling that some of my profs have
it in for me.
Counselor: Things do not go smoothly in your courses.

Introjection of references by the counselor to his own life or
becoming too personally involved with the client. (There is a loss
of objectivity and the counselor may step too far out of his role
toward that of a normal friendship with the client. Whether inten—
tionally or unintentionally, the counselor relates his own situation
to the client by his verbalizations to too great a degree so that
the counseling relationship may become less effective.)

Counselor: That happened to me once too.

Counselor: I worked through a problem like this, and I think you
can too.

Counselor: Oh, I had that course from Sutherland.

Counselor: My department has made me take some courses I didn't
want .

Unfinished sentence. (The counselor's thought is never completed.
Another thought might then be expressed, the old thought might be
approached in a radically different way, or silence might follow.
The exception to this would be unfinished sentences caused by a
direct interruption by the client. [Unfinished sentences are repre-

sented in the transcripts by three underlining dashes, _I_ _ ,
whereas an interrupted sentence would be followed by a dash, -—. If
there is a pause involved also, dots will be added, _ _ _...])

Counselor: You wanted
Counselor: You went to-liv; with your dad and you had _ _ _ You
lived in New York with him.
Counselor: Oh, you can't recall what you did know or what you
... It's just hard for you to remember the material.

 

10.

11.

12.

13.

84

Repeating words or phrases. (This occurs when his own full words or
phrases [either parts or full] are repeated by the counselor in
close sequence within a sentence. Also included under this crite-
rion are modifications of words or phrases which closely follow the
initial verbalization. The exception to this would be repetitions
which seem to be the result of direct interruptions by the client.
The repeated words do not have to follow each other directly. As
the definition states, they can be in close sequence within a
sentence. This does not mean that any word repeated within a sentence
is a repetition. Most of them will come up right next to each other
either as the same form or a modification thereof. Repeated "yes,"
"no," and "uh-huh" do not constitute repetitions nor do others of
the same general nature. Also if a person is interrupted and he
repeats after the interruption what he said immediately before, it
is not scored as 10. However, anything repeated on the same side

of the interruption is a 10.)

Counselor: Have you, have you tried writing these before .
[Also shows Criterion 9.] ->—

Counselor: You feel, you feel that it would be good for you.

Counselor: Was this common to all courses, common to them?

Counselor: Do quite a few of your instructors depend, depend on
subjective exams?

Counselor: You worked for two summers in a, an office.

Stuttering. (This occurs when the counselor repeats parts of words
or when a word is left unfinished at the end of a sentence or phrase.
The exception to this would be a word lefE_;nfinished due to a
direct interruption by the client, and this is shown on the tran-
script by, £.g., pret --. If something is a stutter, it is not
counted itself as a repeated word, but only as a stutter. However,
if there is :3 word connected with it that is repeated, then both
Criteria 10 and 11 are applicable. A stutter is followed by three
underlining dashes, , if it is the last word in the sentence.
However, it should 031; he scored as 11 and not 3 also.

Counselor: You ne-, never thought of that before?
Counselor: Oh, I'd say it's pret-, pretty important.
Counselor: He even sta-_ _ _...

Counselor: Now that is unusf_ _ _
Blocking. (This occurs when the counselor gropes for the proper
expression and indicates hesitation. There is a lack of fluency in
his speech.)

Counselor: Well, in addition to this (uh) let me ask (uh) if this
has always been a problem?

Counselor: (Uh) I think eventually you might hope to find some of
the (uh)...answers.

Counselor: Have you (uh)...tried writing these out ahead of time?

Voice quality. (The counselor's intonation reveals anxiety, there is
tremulousness, his voice "cracks," and so forth. Voice quality for
#13 refers strictly to voice characteristics such as tremulousness,

 

14.

15.

16.

85
cracking, etc., and must never be influenced by content.)

Anxiety as revealed by some criterion not defined in the list of
criteria and examples.

Apologizing to the client for a supposed fault in technique, not
being of sufficient assistance, etc.

Inappropriate laughter. (If this occurs in the middle of a client
statement, it should be scored on the next counselor unit number.
This criterion is not appropriate when the client has said some-
thing amusing or is revealing amusement himself at the same time.)

REFERENCES

 

 

REFERENCES

American Personnel and Guidance Association. Ethical standards casebook
(2nd ed.). Washington, D.C.: Author, 1976.

 

American Psychological Association. Ethical principles in the conduct
of research with human participants. Washington, D.C.: Author,
1973.

 

Bandura, A. Psychotherapist's anxiety level, self-insight, and psycho—
therapeutic competence. Journal of Abnormal and Social Psychology,
1956, 52, 333-337.

 

Bandura, A. Social learning theory. Englewood Cliffs, N.J.: Prentice
Hall, 1977.

 

Bergin, A., & Solomon, 5. Personality and performance correlates of
empathic understanding. In J. T. Hart (Ed.), New directions in
client centered therapy. Boston: Houghton Mifflin, 1970.

 

 

Blanchard, E. B., & Epstein, L. H. A biofeedback primer. Reading,
Mass.: Addison-Wesley, 1978.

 

Borg, W. R., & Gall, M.D. Educational research (2nd ed.). New York:
David McKay, 1976.

 

Bowman, J. T., & Roberts, G. T. Counselor anxiety during a counseling
interview. Counselor Education and Supervision, 1978, 11, 205-212.

 

Bowman, J. T., & Roberts, G. T. Counselor trainees' anxiety during
counseling. Journal of Counseling Psychology, 1979, 26, 85-88.

 

Bowman, J. T., Roberts, G. T., & Giesen, J. M. Counselor trainee
anxiety during the initial counseling interview. Journal of Coun-
seling Psychology, 1977, 25, 137-143.

 

 

Brown, B. B. New mind, new body. New York: Harper & Row, 1974.

 

Brown, B. B. Stress and the art of biofeedback. New York: Harper &
Row, 1977.

 

Campbell, D. T., & Stanley, J. C. Experimental and quasi-experimental
designs for research. Chicago: Rand McNally, 1963.

 

 

Canter, A., Kondo, C. Y., & Knott, J. R. A comparison of EMG feedback
and progressive muscle relaxation training in anxiety neurosis.
British Journal of Psychiatry, 1975, 127, 470-477.

 

86

87

Carkhuff, R. R. Helping and human relations. New York: Holt, Rinehart,
& Winston, 1969.

 

Carter, 0., & Pappas, J. P. Systematic desensitization and awareness
treatment for reducing counselor anxiety. Journal of Counseling
Psychology, 1975, 22, 147-151

 

 

Cattell, R. B. Personality by mood and questionnaire. San Francisco:
Josey-Bass, 1973.

 

Chapman, A. J. An electromyographic study of social facilitation: A
test of the "mere presence" hypothesis. British Journal of
Psychology, 1974, 62, 123-128.

 

Fontaine, J. H. Alpha wave training effects on level of empathy in
helping profession trainees (Doctoral dissertation, University of
Hawaii, 1975). Dissertation Abstracts International, 1976, 36,
7867A.

 

Fry, P. S. Effects of desensitization treatment on core conditioning
training, Journal of Counseling Psychology, 1973, 29, 214-219.

 

Gaarder, K. R., & Montgomery, P. S. Clinical biofeedback: Approcedural
manual. Baltimore: Williams & Williams, 1977.

 

Gaudry, E., & Spielberger, C. D. Anxiepy and educational achievement.
Sidney, Australia: John Wiley, 1971.

 

Goldman, L. A revolution in counseling research. Journal of Counseling
Psychology, 1976, 23, 543-552.

 

 

Goldman, L. Toward more meaningful research. Personnel and Guidance
Journal, 1977, 55, 363-368.

 

Gormally, J., & Hill, C. E. Guidelines for research on Carkhuff's
training model. Journal of Counseling Psychology, 1974, 21,

 

Graub, S. M. An exploratory experimental study of the effects of auto-
genic training on therapist trainees' anxiety and performance
(Doctoral dissertation, McGill University, 1974). Dissertation
Abstracts International, 1975, 35, 6455A. (National Library of
Canada at Ottawa) .—

 

 

Green, E. E., Green, A. M., & Walters, E. 0. Voluntary control of
internal states: Psychological and physiological. Journal of
Transpersonal Psychology, 1970, 2, 1-26.

 

 

Hassett, J. A primer of psychophysiology. San Francisco: W. H. Free-
man, 1978.

 

Hector, M., Elson, S. E., & Yager, G. G. Teaching counseling skills
through self—management procedures. Counselor Education and
Supervision, 1977, 11, 12-22.

 

 

88

Hefele, T. J., & Hurst, M. W. Interpersonal skill measurement: Preci-
sion, validity, and utility. The Counseling Psychologist, 1972,

 

Izard, C. E. A variable combination of interacting fundamental emotions.
In C. D. Spielberger (Ed.), Anxiety: Current trends in theory and
research (Vol. 1). New York: Academic Press, 1972.

 

Krug, S. E., Scheier, I. H., & Cattell, R. B. Handbook for the IPAT
Anxiety Scale. Champaign, 111.: Institute for Personality and
Ability Testing, 1976.

 

 

Lazarus, R. S., & Averill, J. R. Emotion and cognition with special
reference to anxiety. In C. D. Spielberger (Ed.), Anxiety: Cur-
rent trends in theory and research (Vol. 2). New York: Academic
Press, 1972.

 

 

Lesh, T. V. The relationship between Zen meditation and the development
of accurate empathy (Doctoral dissertation, University of Oregon,
1969). Dissertation Abstracts International, 1970, 30, 4778A.
(University Microfilms No. 70-9450) _—

 

Lesse, 5. Anxiety: Its components, development, and treatment.
New York: Grune & Stratton, 1970.

 

Loesch, L. 0., Crane, B. B., & Rucker, B. B. Counselor trainee
effectiveness: More puzzle pieces? Counselor Education and Super-
vision, 1978, 17, 195-204.

 

Luthe, W. Autogenic training: Research and applications in medicine.
American Journal of Psychotherapy, 1963, 13, 174-195.

 

Madden, J. M. A comparison of the physiological effects of two relaxa-
tion procedures: Aut0genic training and progressive relaxation
(Doctoral dissertation, University of Rochester, 1975). Disserta-
tion Abstracts International, 1976, 31, 467B. (University Micro-
films No. 76-14, 769)

 

Mahon, B. R., & Altman, H. A. Skill training: Cautions and recom-
mendations. Counselor Education and Supervision, 1977, 11. 42—50.

 

Mason, J. W. Emotions as reflected in patterns of endocrine integration.
In L. Levi (Ed.), Emotions: Their parameters and measurement.
New York: Raven Press, 1975.

 

Miller, R. W. The reduction of anxiety in counselor trainees using
systematic desensitization (Doctoral dissertation, University of
Northern Colorado, 1972). Dissertation Abstracts International,
1973, 33, 3317B. (University Microfilms No. 73-295)

 

Monke, R. H. Effect of systematic desensitization on the training of
counselors. Journal of Counseling Psychology, 1971, 18, 320-323.

 

 

89

Nicoletti, J. A. Anxiety management training (Doctoral dissertation,
Colorado State University 1972). Dissertation Abstracts Inter-
national, 1973, 33, 6089B. (University Microfilms No. 73-13,058)

 

Pelletier, K. R. Mind as healer, mind as slayer. New York: Delta, 1977.

 

Pennscott, W., & Brown, D. F. Anxiety and empathy in a counseling and
guidance institute. Counselor Education and Supervision, 1972,
.2, 257-2610

 

Roberts, W. E. The assessment of anxiety with the counselor relationship
skills of accurate empathy, nonpossessive warmth, and genuineness
(Doctoral dissertation, Kansas State University, 1977). Disserta-
tion Abstracts International, 1977, 38, 640A. (University Micro-
films No. 77-16,972)

 

Rogers, C. R. On Becoming a person. Boston: Houghton Mifflin, 1961.

 

Russell, P. D., & Snyder, W. U. Counselor anxiety in relation to amount
of clinical experience and quality of affect demonstrated by
clients. Journal of Consulting Psychology, 1963, 21, 358-363.

 

Schachter, S. The interaction of cognitive and physiological deter-
minants of emotional state. In L. Berkowitz (Ed.), Advances in
experimental and social_psychology (Vol. 1). New York: Academic
Press, 1964.

 

 

Schmidt, H. 0. Relationship between IPAT Self-Analysis and the Marlowe-
Crowne Personal Reaction Inventory. Psychological Reports, 1969,

23, 361-362.

 

Schultz, J. H., & Luthe, W. Autogenic training (Vol. 1). New York:
Grune & Stratton, 1969.

 

Sherman, A. R., Mulac, A., & McCann, M. J. Synergistic effect of
subjective and behavioral dimensions of speech anxiety. Journal of
Consulting and Clinical Psychology, 1974, 42, 819-827.

 

 

Sherman, A. R., & Plummer, I. L. Training in relaxation as a behav-
ioral self-management skill: An exploratory investigation.
Behavior Therapy, 1973, 4, 543-550.

 

Spielberger, C. D. Anxiety as an emotional state. In C. D. Spielberger
(Ed.), Anxiety: Current trends in theory and research (Vol. 1).
New York: Academic Press, 1972.

 

Townsend, R. E., House, J. F., & Addario, D. A comparison of biofeedback-
mediated relaxation and group therapy in the treatment of chronic
anxiety. American Journal of Psychiatry, 1975, 132, 517-519.

 

Walters, E. D. Skin temperature biofeedback training for empathic
sensitivity in counseling students (Doctoral dissertation, Kansas
State University, 1977). Dissertation Abstracts International,
1978, 38, 5261A. (University Microfilms No. 7800829)