AN EXPERIMENTAL INVESTIGATION OF THE EFFECTIVENESS

OF ASSERTION TRAINING WITH ALCOHOLICS

by

STEVEN M. HIRSCH, B.A.

A DISSERTATION

IN

PSYCHOLOGY

Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for

the Degree of

DOCTOR OF PHILOSOPHY

May, 19 75

c I V — \ (I -' *.

73

ACKNOWLE DGMENTS

I am sincerely grateful to my chairman. Dr. Jack L.

Bodden for his support and direction in this dissertation

project. Special appreciation and my deepest thanks go to

Dr. Robert P. Anderson for his constant encouragement and

concern. I greatly appreciated the statistical expertise

of Dr. Frank Lawlis and Dr. Douglas Chatfield and the

helpful critiques offered by Dr. Bruce Mattson. I am

also deeply indebted to Robert von Rosenberg, M.A., of

the Big Spring State Hospital, and Caren Phelan, Ph.D.,

of the San Antonio State Hospital, for their invaluable

efforts in making this study possible. I should also like

to extend a special note of thanks to the numerous profes­

sionals of the Texas Department of Mental Health and Mental

Retardation who gave so freely of their time and expertise

in making this project a reality. Finally, the granting of

research funds from the Texas Commission on Alcoholism, to

assist with data collection, was deeply appreciated.

11

TABLE OF CONTENTS

ACKNOWLEDGMENTS ' ii

LIST OF TABLES V

I. INTRODUCTION 1

Review of the Literature 6

Statement of the Problem . . . . . . . . . 54

II. METHODOLOGY 5 8

Subjects 58

Instrumentation . 5 8

Procedure 64

Control Group 65

Minimal Assertive Training Group 66

Assertive Training Group 6 7

Hypotheses 6 8

Statistical Analyses . . . . . 69

III. RESULTS 70

Demographic-Drinking Data Form 70

Hypothesis 1 73

Hypothesis 2 75

Hypothesis 3 76

Hypothesis 4 ' 82

IV. DISCUSSION . . . . . . . . . 84

Research Hypotheses 84

Theoretical Implications . 9 4

111

Programmatic and Research Implications . . 9 7

V. SUMMARY AND CONCLUSIONS 10 4

REFERENCES 10 8

APPENDIX . 115

A. Demographic Drinking Data Form 116

B. Rathus Assertiveness Scale . . . . . 120

C. Behavioral Assertiveness Test . . . . . . . . 122

D. Scoring Sheet for Behavior Assertiveness Test 125

E. Assertive Behavior Index 127

IV

LIST OF TABLES

1. Demographic-Drinking Data 70

2. Correlation of RAS Scores of Alcoholics with Five Variables 74

3. Rathus Pre-Posttest Means and Standard Deviations for Treatment and Control Groups 75

4. Analysis of Covariance Results of RAS Scores for Treatment and Control Groups . . . . 76

5. Response Latency Means and Standard Deviations for Treatment and Control Groups 77

6. Analysis of Variance Results of Response Latency Scores for Treatment and Control Groups 77

7. Response Duration Means and Standard Deviations for Treatment and Control Groups in Seconds 78

8. Analysis of Variance Results of Response Duration Scores for Treatment and Control Groups 79

9. Rater's Means, Standard Deviations and Inter-Rater Reliability of BAT 80

10. Average Total Assertiveness Means and Standard Deviations for Treatment and Control Groups 80

11. Analysis of Variance of Mean Assertiveness Scores on the BAT for Treatment and Control Groups 81

12. ABI Means and Standard Deviations for Treatment and Control Groups 82

V

CHAPTER I

INTRODUCTION

Alcoholism has long been a major health problem in

this country, directly affecting an estim.ated nine million

individuals and their families. It has become increasingly

evident that legal sanctions and moral pressures have not

provided an adequate remedy for this problem. Moreover,

it is estimated that treatment of alcoholic individuals

with present techniques would fill every existing hospital

bed and require the full-time services of every physician

in the Nation. Consequently, solution of this problem must

ultimately lie in the development of improved prevention

and treatment techniques.

In recent years there has been a movement av/ay from

a generalized "psychotherapy" treatment for emotional,

intellectual, or behavioral disorders in favor of a more

specific approach v/hich matches treatment and therapist

with client and problem. Clients who benefit from therapy

regardless of the type of treatment they undergo, often

state that they have become less inhibited, more outspoken,

and able to stand up for their rights. Most recently, a

growing number of clinicians (Blake, 19 65; Martorano, 19 74;

Miller, Hersen, Eisler, & Hilsman, 1974) have recognized

that various procedures subsumed under the heading of

assertive training can be effectively utilized to help

alcoholic clients achieve a greater degree of self-esteem

and emotional freedom.

The therapeutic technique called assertive training

has been in existence for about 15 years (Wolpe, 1958).

Practitioners have customarily employed it in one-to-one

therapeutic situations with persons for whom nonassertive-

ness or unassertiveness was part of their problem. Over

the last 3 or 4 years behavior change agents have expanded

the technique to the treatment of groups of unassertive

individuals. It is this modality of treatment which is

the subject of the present research.

Assertive behavior refers to all socially acceptable

expressions of personal rights and feelings (Wolpe &

Lazarus, 1966). It is the action of declaring oneself,

of stating, "this is who I am and what I think and feel."

Being assertive includes the honest, appropriate and rela­

tively straightforward expression of such negative feelings

as anger, dissatisfaction, and resentment, as well as the

expression of such positive feelings as love, affection,

and praise. Alberti and Emmons (19 70) perhaps summarized

it best when they said that assertive behavior is that

"which enables a person to act in his own best interest,

to stand up for himself without undue anxiety, to express

his rights without denying the rights of others" (p. 7).

The rationale for training in assertion is presented

by Wolpe (19 69) as follows:

Assertive training . . . is required for patients who in interpersonal contexts have unadaptive anxiety responses that prevent them from saying or doing V7hat is reasonable and right . . . suppression of feeling may lead to a continuing inner turmoil which may produce somatic symptoms and even patho­logical changes in predisposed organs. (p. 61)

A variety of component techniques including advice, exhorta­

tion, coaching, social modeling, role playing, behavioral

rehearsal, graded exercises, and homework assignments are

used during the full course of assertive training. Although

Wolpe emphasized the principle of "reciprocal inhibition" in

his earlier writing (Wolpe, 1958), in his more recent writ­

ings (Wolpe, 19 69, 19 70) the importance of operant condi­

tioning (reinforcement and shaping) during assertive

training has been acknowledged.

Increased assertiveness, therefore, is assumed by

Wolpe to benefit the client in two v;ays:

1. It is thought that more assertive behavior will

instill in the client a greater feeling of well-being. This

proposal is related to the "reciprocal inhibition" view that

'assertive behavior is very similar to deep muscle relaxation

in being physiologically antagonistic to anxiety.

2. It is assumed that, by behaving in a more assertive

manner, the client will bring about positive changes in be­

havior of others toward him, and thus be better able to

achieve significant social (as well as material) rewards

and, in fact, control his own destiny. In theory, therefore,

the individual achieves both internal and external positive

feedback from appropriate assertive responses.

The assertive training approach to psychotherapy is

based on a response acquisition model of treatment (McFall

& Twentyman, 1973). Within this model, unadaptive behaviors

are interpreted in terms of the absence of specific response

skills. The therapeutic objective of assertive training

with alcoholics is to provide clients with direct training

in precisely those interpersonal and social skills deficient

in their response repertoire.

Very little attention is given to eliminating exist­ing maladaptive behavior; instead, it is assumed that as skillful, adaptive responses are acquired, rehearsed and reinforced, the previous maladaptive responses will be displaced and will disappear. (McFall & Twentyman, 19 73, p. 199)

As positive rewarding behaviors are practiced and developed,

they will compete with and eventually replace alcoholic

avoidance and escape.

In contrast to psychodrama (Corsini, 1966; Moreno,

1946) or attitude change research (Elms, 1969) where the

focus is on the modification of abstract emotional or

cognitive processes, assertive training procedures are

r

designed to modify definite observable behaviors. Asser­

tive training, in effect, permits clients to simulate

problem situations and to practice new ways of responding

without concern or anxiety for the immediate real-life

consequences of their experimental behavior. Only after

clients develop proficiency and confidence in their

responses do they attempt to transfer them to real-life

situations.

Clinical evidence (Miller et al., 1974) suggests that

interpersonal situations requiring an alcoholic to respond

assertively (i.e., direct expression of personal rights

and feelings) are often stressful and frequently lead to

drinking episodes. Alcoholics whose drinking behavior

temporarily has been controlled' are unlikely to remain

abstinent for long if they lack the behavioral competencies

for securing gratification while sober. Therapeutic atten­

tion, therefore, is most profitably directed tov/ard building

up a repertoire of coping techniques for use in problem

situations in the community. Given m.ore effective and re­

warding means of dealing with environmental demands, indi­

viduals will have less need to resort to self-anesthetization

against everyday life experience.

To date, the vast majority of experimental studies in

assertive training has dealt with an unassertive, nonclient,

college population. Despite the fact that these studies

have reported very encouraging results, therapeutic outcome

studies using other populations have yet to be conducted.

Only within the last year have several pilot studies

(Miller et al., 1974; Martorano, 1974; Burtle et al., 1974)

indicated that assertive training may be a viable treatment

procedure with a significant proportion of alcoholics. This

proposed research project represents the first controlled

experimental investigation of the effectiveness of assertive

training with an alcoholic population.

Review of the Literature

History and Background of Assertive Training

Prominent theoreticians and practitioners. Andrew

Salter (1949), in his book Conditioned Reflex Therapy, was

the first to propose a learning theory-based view of asser­

tive and nonassertive behavior. Although he did not use

the term "assertive behavior," but instead the "excitatory

person," Salter seemed to have described what existential­

ists call the "self-actualized person" and what later

behavior therapists call the "assertive" or "emotionally

free" person. He claimed that such an individual is direct

and acts without restraint while his opposite, the "inhibi­

tory person," acts under restraint and suffers from "consti-

pation of the emotions" (p. 47). Salter also felt that "the

code of the inhibitory personality is to suppress the gut

and inflate the brain, . . . they do not fight for emotional

rights. They refuse to fight others, and end up by fighting

themselves" (p. 48). He probably represented one of the

first behaviorists concerned with moving the emotional3-y

deprived person toward new behavior that represents freedom

from past emotional imprisonment.

Salter contended that all individuals were born with a

preference for excitation (freedom and honesty in emotional

expression). Adhering to a Pavlovian model, Salter con­

tended that the cause of neurotic behavior was the develop­

ment of inhibitory personalities through the frequent

pairing of this natural excitation with adversive stimuli

(e.g., physical or verbal parental censure). He believed

that this generalized and pervasive inhibitory trait could

be unlearned through (a) employing logical directive to

convince the client of the disadvantages of inhibitory

behavior, and (b) having the client practice excitatory

exercises, e.g., "feeling talk," "facial talk," expressing

contradictory opinion, the use of "I," agreeing when

complimented, and improvising.

While Salter was the first published proponent of

assertive training, V7olpe was using assertive training at

the time Salter's book appeared. Encouraged by Salter's

writings, it was Wolpe and his followers who had the

greatest impact in describing and popularizing the asser­

tive training technique we see in clinical practice today.

Crucial differences between the approaches of Salter and

Wolpe are as follows:

8

1. Wolpe does not assume every client is primarily

in need of assertive training.

2. Whereas Salter views assertiveness (excitation)

as a generalized trait, Wolpe views it as being specific

to situations and emotions.

3. Wolpe is more concerned with the possible inter­

personal consequences (especially negative) of assertive

acts.

Wolpe's (1958, 1969, 1971) main focus, however, was

upon the reduction of anxiety. Assertive responses, like

sexual and relaxation responses, are inhibitory of anxiety

in his well-known classical conditioning model of learning.

Cure of social anxiety (which inhibits appropriate social

expressions) involves a counter-conditioning of the anxiety

of eliciting the inhibited assertive response through

persuasion, instructions, and cognitive restructuring.

The augmented emotions accompanying assertiveness inhibit

the anxiety response habit, and the motor act of assertion

is increased in frequency by its reinforcing consequences

in daily life (e.g., control of a social situation).

Today much of Wolpe's explanation of behavior change

via reciprocal inhibition is in question (Lang, 1969;

Meichenbaum, 1971; Spence, 1971), but his pioneer work in

assertive training is still important. His basic treatment

techniques can be found in the repertoire of many therapists

(a) encouraging the outward expression of feelings of anger,

etc.; (b) directly instructing the client how to act asser­

tively; and (c) rehearsing the needed behavior with the

client.

Another pioneer in training people in assertive

behavior is Arnold Lazarus. He stressed the idea of "emo­

tional freedom" of which assertiveness is a subunit.

Training in emotional freedom implies the recognition and appropriate expression of each and every affective state . . . assertive behavior will denote only that aspect of emotional freedom that concerns standing up for one's rights. (1971, p. 116)

In Behavior Therapy Techniques (1966), he and Wolpe described

three techniques which La2;arus has continued to elaborate

upon in subsequent writings (Lazarus, 1966; Piaget & Lazarus,

1969): (a) behavior rehearsal, (b) graded structure, and

(c) cognitive restructuring.

In behavior rehearsal the client repeatedly practices

assertive responses, such as telling a mother-in-law to stop

being so obtrusive, until he is relatively comfortable in

saying the words. If the target behavior is so anxiety

provoking that the individual would be unlikely ever to

perform it, Lazarus recommends "rehearsal desensitization."

Some less threatening behavior in a graded hierarchy of

assertive situations is rehearsed first until the client

is successful in its performance. Then he moves closer to

practicing the target behavior itself and ultimately to

performing it in his real-life situation.

10

Lazarus (19 71) believed that using a graded structure

of tasks in therapy can be therapeutic itself. Structure

gives a person a knowledge of what leads to what and of

the steps in the solution of a problem. It allows him to

attend to and concern himself with the "right things" in

therapy. The graded aspect lets a person have a sense of

achievement and fulfillment that results from a step-by-

step progress. Much like programmed learning, graded

structure allows the client to get a sense of mastery over

his problem via active participation, immediate reinforce­

ment and feedback concerning results.

Unassertive persons (as well as others) cause unneces­

sary grief for themselves because of faulty anxiety-arousing

conceptions regarding the social situation at hand. They

repeatedly tell themselves that they cannot handle a situa­

tion, or that they are inferior persons and so doom them­

selves to maladaptive behavior.

The bulk of therapeutic endeavors may be said to center around the correction of misconceptions. The people who consult us tend to view innocuous events as extremely noxious, and may disregard objectively noxious situations. Therapy often strives to show people how to separate subjective from objective dangers. Thereafter, the emphasis is on avoiding or coping with objectively hazard­ous events while ignoring the innocuous situations. (Lazarus, 1971, p. 165)

The first "do-it-yourself" manual of assertive training

was written by Alberti and Emmons (19 70) for both profession­

als and laymen. The authors' premise was that men are

11

created equal and so have a right to personal dignity

without untoward fear in social interactions. They pro­

posed to help their readers develop a more adequate reper­

toire of assertive behaviors in order that these persons

can have a greater range of choices of self-fulfilling

behaviors in a number of situations. This freedom to choose

responses should enable a person to escape from his con­

stricted pattern of behaviors and move to a fuller human

life.

Alberti and Emmons distinguished for their readers

aggressive, assertive, and nonassertive behaviors. The

aggressively behaving person enhances himself and achieves

his goals at the expense of others. He chooses for others,

deprecating him and leaving him hurt and without achieving

his goals. The nonassertive person is inhibited, self-

denying, and so does not achieve his goals. He allows

others to choose for him and ends up being anxious, hurt,

and full of suppressed rage. The assertive person behaves

in a self-enhancing way to achieve his goals. He makes

choices for himself and he allows the other person to pursue \

his goa ls as well.

Alberti and Emmons distinguished between a situation-

ally nonassertive person and a generally nonassertive one.

The former exhibits behavior that is generally adequate

but is ineffective in specific situations. The generally

t "' »

12

nonassertive person usually has low self-esteem, feels

inadequate, and is anxious in most all social situations.

These individuals may require special psychological help ^^

in addition to assertive training.

The authors prescribed a 14-step assertive training

procedure. First, clients receive an explanation from the

therapist of why they have a right to be assertive. There­

after, they undergo a problem-solving process which incor­

porates role playing, behavior rehearsal, and modeling. The '

therapist provides feedback and reward to the client in a

graded structure procedure similar to the one suggested by

Lazarus (19 71). Alberti and Emmons suggested instructing

clients in the principles of self-reinforcement so that

behaviors will be maintained and generalized in the absence

of the therapist. They also recommended the use of treatment

in a group, citing such advantages as being able to share

with others who have the same problems (cf. Yalom, 19 70, on

"Universality"), helpful modeling from others in the group,

a more realistic social learning environment, and increased

feedback.

Manuel Smith (19 75) has recently published a book on

systematic assertive therapy which is entitled When I Say No,

I Feel Guilty. The author has a somewhat different slant

toward assertive training than what Alberti and Emmons have

described. Smith appeared to use less modeling, coaching.

13

and behavior rehearsal, but instead he tended to focus on

the verbal content of an assertive message. He described

a number of training dialogues which cover many everyday

situations such as returning purchases, handling criticism,

asking for a raise, meeting new people at parties, and

learning how to say no without anger, fear or guilt. The

main focus of the book was directed toward the teaching of

verbal techniques such as "broken record," "fogging,"

"negative assertion," etc. with the basic notion of helping

clients learn how to out manipulate the manipulators.

Techniques of Assertive Training

Assertive training techniques are not as standardized

as many other^behavior therapy techniquesJ nor have they

been carefully investigated. Assertive training takes on

many different forms, but is generally applied in the fol­

lowing manner. Generally, a first step in teaching asser­

tive behavior requires that the client be aware of how he

comes across (awareness = response - ability). When the

therapist, using situations which are drawn from the

client's life, carefully distinguishes among assertive,

aggressive, and nonassertive behavior, the client usually

realizes quickly that he has an assertive problem. The

therapist must explore all areas of difficulty to determine

the controlling factors--the situational and personal vari­

ables that raise anxieties and decrease the client's ability

14

to behave assertively. Obtaining a good clinical history

is often an important part in determining those areas

where the patient is having problems.

A major goal of assertive training is building a

personal belief system which v/ill help the client to support

and justify his acting assertively. This is important so

that the client believes in his right to act assertively

even when he may be unjustly criticized, and can counteract

his own irrational guilt that often occurs as a result of

having asserted himself. An important part of this belief

system concerns the patient's acceptance of certain basic

interpersonal rights. Didactic explanations are very useful,

explaining the differences between nonassertive behavior,

aggressive behavior, and assertive behavior.

NON-ASSERTIVE BEHAVIOR

Actor

Self-denying

Inhibited Does not achieve desired goal(s)

ASSERTIVE BEHAVIOR

Actor

Self-enhancing

Expressive May achieve desired goal(s)

AGGRESSIVE BEHAVIOR

Actor

Self-enhancing at expense of another

Expressive Achieves desired goal(s) by hurting others

Allows others to Chooses for self Chooses for others choose for him

Hurt, anxious Feels good about Depreciates others self

15

Acted Upon

Guilty or angry, depreciates actor

Achieves desired goal(s) at actor's expense

Acted Upon

Self-enhancing expressive

May achieve desired goal(s)

Acted Upon

Self-denying, hurt defensive, humiliated

Does not achieve desired goal(s)

(Alberti & Emmons, 1970, p. 24)

In the initial stages of assertive training, the

therapist stresses the negative effects (frustration,

resentment, lack of satisfaction) of patient's nonassertive

behavior, hov; such behavior is learned and how more appro­

priate assertive responses can be conditioned with resulting

feelings of personal well-being, relief, and interpersonal

satisfaction. This procedure is an attempt to change the

client's attitude toward his behavior. The therapist briefly

describes the method and rationale of assertive training and

discusses some of the techniques that will be used.

A commonly used technique in assertive training is

behavior rehearsal (V7olpe, 1969). This is a procedure which

requires the client to act out relevant interpersonal inter­

actions. Initially the client demonstrates behavior which

is typical for him in a given situation. The therapist then

supplies corrective feedback, support, and verbal reinforce­

ment. By repeated practice, the client reduces his anxiety

and develops a sense of mastery over interpersonal situations

which previously caused difficulty. The therapist used the

16

principle of successive approximations to help shape the

client's behavior. Role rehearsal, in which the client

models the behavior of the significant other and the

therapist assumes the client's role, may also be employed.

Clients in a group receive feedback from each other and

from the therapist on their practice behaviors. They will

encourage and support one another's progress. The use of

video-tape feedback can be a particularly powerful technique

in changing client's behavior.

In a lengthy interaction, the therapist should

determine that each segment can be mastered by the client

with little or no anxiety before proceeding to the next

segment of interaction. In an interaction involving the

expression of negative feelings, it is helpful to have the

client begin with a relatively mild response. He should,

however, be given stronger responses in case the initial

response is ineffective. In rehearsing the expression of

negative feelings, it should be pointed out that feedback

pertaining to annoying or hurtful behaviors on the part of

the other person is far superior to personal attacks, which

are often irrelevant and have the effect of backing the

other into a corner. Clients are asked to try target

behavior(s) in their real life settings and report the,

success or failure in their next session. Needless to

say, the client must be considered potentially capable

•I ^

17

of completing the assignment within prescribed time limits

(Fensterheim, 1972).

Modeling is a technique based on principles of

imitative learning. When used in combination with feed­

back and instruction, the necessity of individuals dis­

covering the most effective response through trial and

error is eliminated. In his research, Bandura (1969, 1971)

had established that modeling and guided participation is

one of the most effective methods of modifying anxiety-

based avoidance behavior. According to this method, the

therapist models behavior for the client and preferably

has others model it as well. The client then practices

the model's behavior with coaching from the therapist in

a graduated sequence of behaviors from the easiest to the

most difficult. Favorable reinforcement categories are

established to maintain the new behaviors. Kanfer and

Phillips (1970) clarified this observational learning

process further by noting that the observer of modeled

behavior benefits most from the modeling if (a) information

presented by the model is explicit, (b) it occurs during

acquisition, and (c) the model increases his own effective­

ness over trial blocks.

Alberti and Emmons (19 70) have indicated that a group

atmosphere is effective and possibly superior to individual

treatment in assertive training. A group provides consensus

' 18

for what behavior is appropriately assertive in a situation,

multiple models, and massive social reinforcement. Con­

frontation between group members on here-and-now behaviors

may provide opportunities for practicing assertive behavior

"in vivo." Groups of extremely withdrawn individuals may

benefit from nonthreatening warm-up exercises, e.g., greet­

ings, free association, exchanging compliments, positive

self-statements, small talk. The client should practice

specific assertive acts, such as making eye contact, using

the pronoun "I," or using "feeling talk," both in and

outside the treatment situation (Fensterheim, 1972).

Server (19 72) felt that assertive training techniques

have customarily focused on modifying explicit verbal

messages while neglecting nonverbal variables. He stated

that "any experienced clinician is well aware of the fact

that what ineffective persons often lack is not knowledge

or courage but a command of style" (p. 173). Serber found

it useful to break down nonverbal behavior into the fol­

lowing specifics: (a) loudness of voice, (b) fluency of

spoken words, (c) eye contact, (d) facial expression,

(3) body expression, and (f) distance from person with whom

one is interacting. During training sessions Serber felt

it is best to concentrate upon only one nonverbal variable

at a time.

In the course of training, clients usually do a fair

amount of assertion for the sake of assertion. By the end

19

of training, hopefully they have a natural assertiveness

which they can choose to exercise if they wish. When

people know that they have the necessary skills to assert

themselves, they frequently feel less of a need to do so.

When they decide not to assert themselves, it is because

they choose not to and not because they are afraid.

In summary, the unassertive individual leads

anything but a free and spontaneous existence. He lacks

the necessary interpersonal skills with which to cope

effectively. Assertive training has been developed as a

therapeutic technique to help these individuals. It con­

sists of a number of techniques such as modeling, coaching,

role playing, behavior rehearsal, graded structure, and

reinforcem.ent. These methods are used to train a person

directly in increasing his behavioral repertoire and

reducing his discomfort in dealing with his fellow human

beings. Modeling, behavior rehearsal, guided participa­

tion, and related techniques have enjoyed success in the

modification of a number of problem behaviors (Bandura,

1971; Bandura, Blanchard, & Ritter, 1969). Until recently,

however, controlled empirical studies of the usage of these

techniques for the modification of unassertive behaviors

have been scarce. The following section will present a

review of those studies which have been reported.

•?/• II

20

Controlled Research on Assertive Training

The effects of assertive training and its component

techniques have been examined in a number of group analogue

and clinical outcome studies. Lazarus (1966) conducted one

of the initial studies in the area. He compared the effi­

cacy of four 30-minute sessions of behavior rehearsal,

direct advice, and nondirective therapy with respect to

improvement in the management of specific interpersonal

problems. All the subjects in the experiment were clients

of Lazarus. The results revealed that behavior rehearsal

was almost twice as effective as direct advice and the

nondirective treatment procedure fared v/orst of all. The

outcome measure of effectiveness was a statement by the

patient that he or she was behaving adaptively in the area

which had previously been the problem. Lazarus administered

the treatment for the three groups and acknowledged "the

possibility of experimenter bias," but he argued that the

superiority of behavior rehearsal is predicted on a

theoretical basis.

Friedman (1968, 1971) conducted an analogue study in

which the following treatments were compared: modeling,

modeling plus role playing, directed role playing, impro­

vised role playing, assertive script, and nonassertive

script. In a treatment period of 8 to 10 minutes per

subject, Friedman shov/ed the subject how to assert himself

21

when harassed by another student in a scene set in a

library. The teaching device was a script containing

the dialogue of the interchange between students.

Primary measures, used pre- and posttest in Friedman's

study were the Action Situation Inventory (ASI) and a Sum

Assertion score (SA). The former was a questionnaire

which consisted of 10 possible threatening behavioral

situations and six alternate reactions to them. Subjects

were to rate the reaction which would best describe their

behavior in the setting (ratings from "not assertive" to

"very assertive"). The SA was a behavioral m.easure con­

sisting of ratings by judges on 2 4 categories of the

subject's behavior and an audiotape interaction between

him or her and an accomplice of the experimenter. The

subject was given examples of assertive behavior and told

to act assertively in the ensuing interaction. He also was

told that he was being tape recorded, watched through a

one-way mirror, and being rated. A stressful role play

social situation was then created in which an accomplice

of the experimenter entered a room and obnoxiously inter­

fered for 6 minutes with a task on which the subject was

working--solving a cube puzzle. The judges later scored

the subject's behavior in one of 15 categories. Subjects

were included in the experiment who made six or fewer

assertive responses on this measure.

-^mHf0''' -''•mmmmm^mamaKS,.'•

22

There were five treatment conditions and a control

group in Friedman's study: Directed Role Playing, in

which the subjects rehearsed audibly the role of the

harassed student in the presence of an accomplice of the

experimenter who was the instigating student. Subjects

read their responses from a prepared script. The Improvised

Role Playing condition was the same as the previous one

except that the subjects improvised the harassed student's

responses instead of reading them. In the Live Modeling

condition, subjects observed models enacting the role of

harassed and instigating students. Students in the Model­

ing and Directed Role Playing condition combined the

activities of the Live Modeling and Directed Role Playing

subjects already described. In the Assertive Script condi-

tion subjects simply read the script. The control group

read scripts not involving assertive behavior.

Differential effects for treatment were demonstrated

only on the behavioral measure (SA). ASI results proved

to be insignificant. According to the SA data, subjects

in all treatm.ent conditions increased in mean assertiveness

scores over those of the controls; however, the Modeling

and Directed Role Playing group subjects, as predicted,

improved most. The second most powerful treatment proved

to be Live Modeling. Interestingly enough, even subjects

who simply read scripts portraying assertive responses

improved significantly in assertiveness.

23

In his discussion, Friedman (19 71) claimed that the

covert rehearsal responses generated by the assertive

information conveyed to the subjects in the modeling,

directed role playing, and assertive script conditions

were presumably used as internal cues for verbal assertive­

ness and thus account for most of the changes in these

three conditions.

Wlien m.odeling was combined with directed role playing, the hypothesized mediating variables to account for the efficacy of modeling and directed role playing (covert perceptual cognitive images, covert rehearsal, overt rehearsal) may summate; and consequently foster greater changes in asser­tiveness than in either of these conditions separately or in the assertive script condition. (1971, p.'l66)

One intriguing aspect of this study is that marked

behavioral changes were obtained vzith only 10 minutes of

treatment. Investigations of behaviors more difficult to

execute are needed before Friedman's conclusions can be

accepted v/ithout reservations.

The major effort to study assertive training pro­

cedures experimentally has been undertaken by McFall and

his colleagues at the University of Wisconsin (McFall &

Lillesand, 1971.- McFall & Marston, 1970; McFall & Twentyman,

1973). Their basic treatment techniques have been a stan­

dardized audiotaped, semi-automated laboratory analogue of

behavior rehearsal. They have employed a consistent

paradigm across studies varying the specific training

24

methods. The target behavior was limited to saying "no"

to unreasonable demands. This has greatly limited the

generalizations which may be drawn from their findings,

but allows for a focused, more precisely measured core

variable.

McFall and his associates are the only researchers in

the area of assertive training who have attempted to system­

atically determine which components of assertive training

are more effective.

Tv7o opposing research strategies can be taken in investigating psychotherapy techniques. On the one hand, an investigator can start with a fully developed treatment system and proceed to dis­mantle it (Lang, 1959), systematically eliminating individual treatment components and measuring the associated decrements in treatment effects. In this manner, the relative contribution of each component to the total system can be assessed. The dismantling strategy is most appropriate for structured and established systems composed of a number of treatment components.

Alternatively, an investigator can follow a constructive strategy. In this case, he starts by isolating the most fundamental or theoretically significant treatment component and assesses its effect on behavior. Then, building on this base, new treatm.ent components are systematically added to determine-whether they significantly enhance the fundamental treatment effects.

Effective components are retained. This con­structive strategy is most appropriate for studying complex, nonsystematic techniques that are loosely organized around a central treatment mechanism, e.g., behavioral rehearsal. One advantage of the constructive strategy is that it can lead to the development of an empirically based technique which is ultimately more powerful than the one initially giving rise to the research. (McFall & Marston, 1970, p. 302)

•WF

25

The first study of a series (McFall & Marston, 1970)

compared the effects of behavior rehearsal (with and without

tape-recorded feedback) with placebo therapy and a no-

treatment control. Forty-two volunteer, nonassertive,

college students were assigned to four groups: Behavior

Rehearsal, Behavioral Rehearsal with Feedback, Placebo

Insight Therapy, and Waiting List Control. The subjects

in the two behavior rehearsal groups and the placebo therapy

group then received four 1-hour treatment sessions over a

2 to 3 week period, while no-treatment control subjects

merely went on the waiting list. The results revealed that

the two behavioral techniques were significantly better than

the two control procedures on behavioral (semi-automated

role playing task), self-report (Wolpe-Lazarus Assertive-

ness Questionnaire), psychophysiological (pulse rate) and

in vivo (resistance to high pressure telephone salesmen)

measures of assertion. The addition of performance feed­

back appeared to enhance improvement but not at a signifi­

cant level. It should be noted that significant differences

were found on these "measures only when data from the two

experimental and two control conditions were combined and

then compared. The two more global self-report measures

used in the study, the Taylor Manifest Anxiety Scale and

the Fear Survey Schedule, showed no significant effects due

to treatment. The authors stated that this was not

.w

26

particularly surprising since these measures are not

specifically related to assertive behavior. Thus, behavior

rehearsal subjects did not indiscriminately report a general

reduction in problem behaviors, only that they had learned

to behave in a more assertive manner.

The significant results reported in the McFall and

Marston (19 70) study are important for several reasons.

First, the treatment procedures used were not selected

because they were expected to yield maximal levels of be­

havior change. Secondly, they were not being advocated as

fully developed clinical techniques. Rather, they were

studied to determine whether they represent a reasonable

foundation upon which to base a behavioral treatment

approach.

In a subsequent study, McFall and Lillesand (1971)

compared the relative efficacy of three training conditions:

(a) rehearsal of assertive responses (with response play­

back) plus symbolic verbal modeling and therapist coaching,

(b) covert rehearsal (with no response feedback) plus

modeling and coaching, and (c) assessment-placebo control.

The subjects for this experiment were obtained from an

introductory psychology class and had to meet the following

criteria for participation in the study: (a) they rated

themselves as having a significant problem saying "no" to

unreasonable requests, (b) they expressed an interest in

27

participating in a clinic where they would learn to refuse

imposing requests, and (c) they earned a low assertiveness

score on the Conflict Resolution Inventory, a 35-item

inventory of refusal behavior specifically developed for

this study. Out of more than 400 students, only 33 met

the three screening criteria. The subjects were randomly

assigned to one of the three treatment conditions, with 11

subjects in each condition. All subjects including controls

were seen individually for two experimental sessions 1 week

apart. In the first two conditions, subjects listened to

tape-recorded modeling of the refusal responses in addition

to hearing the narrator describe the situation and coach

them in what would make a good assertive response in the

situation.

As in the McFall and Marston (19 70) study, the results

of the McFall and Lillesand study showed that both experi­

mental groups evidenced greater pre-post changes on self-

report and behavioral measures than the control group. In

addition, subjects in the covert rehearsal group generally

showed the most pronounced changes in both self-report and

behavioral laboratory measures. The results are surprising

at first glance. One might have suspected the overt re­

hearsal group to perform better since there was more partic­

ipation on their part (they had to say something). The

authors suggested, to the contrary, that an overt response

28

could make the training experience adversive and therefore

to be avoided. The playback of an inept response might

punish a trainee. Perhaps the most important results of

the study are that although the rehearsal-modeling-coaching

therapy analogue was far from a clinical technique, non­

assertive subjects who received only 40 minutes of such

training improved dramatically in their assertive-refusal

performance on the Conflict Resolution Inventory and the

behavioral-laboratory measures.

In a third study McFall and Twentyman (19 73) reported

on four separate experiments assessing the relative contri­

bution of rehearsal, modeling, and coaching to assertive

training. The aim of the first experiment reported by

McFall and Twentyman was to systematically dismantle McFall

and Lillesand's (1971) treatment program while assessing

any associated decrement in overall treatment effects.

Seventy-two nonassertive college students, as measured

by their scores on the Conflict Resolution Inventory,

were assigned randomly to one of six treatment groups.

There were 12 subjects per group with a balanced distri­

bution of sexes across the groups. The six treatment con­

ditions examined in this study were: (a) Rehearsal,

Modeling, and Coaching; (b) Rehearsal and Modeling; (c)

Rehearsal and Coaching; (d) Rehearsal only; (e) Modeling

and Coaching; and (f) Assessment Control (i.e., no rehearsal,

29

no modeling, and no coaching). All subjects, including

controls, were seen individually for two 45-minute experi­

mental sessions 1 week apart. The authors found that the

strongest contributor to behavior change as measured by the

Conflict Resolution Inventory was the coaching element of

the treatment period. Covert Rehearsal brought about the

greatest improvement according to the behavioral test.

Modeling had little effect according to the results of

either the CRI or the behavior role playing test. Re­

hearsal and coaching operated in an additive fashion to

improve behavior performance.

In the second experiment reported by McFall and

Twentyman (19 73) , 90 unassertive college students were

assigned randomly to one of nine experimental groups within

a three by three factorial design. In addition, 14 highly

assertive subjects (those who scored as highly assertive

on the Conflict Resolution Inventory) were selected as

a "super-assertive" control group. Because this second

experiment was an analysis of the relative improvements

associated with different combinations of treatment per se,

the design did not include a no-treatment control group.

One of the experimental factors in this experiment

was the type of treatment received. The three treatment

conditions were: (a) Covert Rehearsal plus Modeling plus

Coaching; (b) Covert Rehearsal plus Coaching; and

-iaBP-' .rr

30

(c) Covert Rehearsal only. The second experimental factor

was the amount of time allowed to elapse between the end

of treatment and the first follow-up test of transfer. The

three time intervals studied in this experiment were:

11 days, 18 days, and 25 days. The results of the Conflict

Resolution Inventory Assertion Measure, Behavioral Role

Playing Assertion Test, Extended Interaction Test, and a

Waiting Room Behavior Test provided corroborative evidence

that the assertive training program consisting of rehearsal

and coaching produced positive treatment effects, but that

the modeling component when combined with the rehearsal and

coaching components added little if any to the treatment

effects. The evidence was mixed with respect to whether

there was transfer of training from laboratory to extra-

laboratory situations, although treated subjects did perform

as well as "super-assertive" controls on one of the follow-

up calls attempting to measure assertiveness.

In a third experiment reported by McFall and Twentyman

(19 73), the authors further evaluated the role of modeling

in assertive training and the possibility that a combined

covert-overt mode of rehearsal might be superior to either

of the two modes. The subjects consisted of 4 8 unassertive

college students as measured by the Conflict Resolution

Inventory. In addition, subjects were regarded as eligible

to participate only after they had demonstrated inadequate

31

assertive-refusal behavior by acquiescing to a pretreatment

administration of an in vivo, unobtrusive telephone test of

refusal behavior which had been developed in a pilot study.

The results of this third experiment indicated that the

previous failure of the modeling component to add to the

treatment effects of rehearsal and coaching probably was

not a function of the particular characteristics of the

models employed. In addition, the results clarified McFall

and Lillesand's (19 71) previous finding that covert rehearsal

without response playback was superior to overt rehearsal

with response playback. The present experiment found no

difference between covert and overt rehearsal when the play­

back variable was deleted, thereby suggesting that the play­

back variable probably produced the earlier differences.

In the fourth and final experiment reported by McFall

and Twentyman (1973), the continued lack of modeling effects

was further investigated. The subjects for this experim.ent

consisted of 54 unassertive undergraduates. The experi­

mental procedure was essentially the same as in the preceding

studies with the exception that, in addition to auditory

modeling, an audiovisual modeling treatment procedure was

developed.

The results of this study demonstrated that the addi­

tion of the visual component in the assertive training

program failed to enhance treatment effects in any detectable

.fT

32

manner. In addition the results showed some evidence that

the treatment effects transferred from laboratory to real

life situations.

In summary, the work of McFall et al. has demonstrated

that behavior rehearsal and therapist coaching are the most

powerful contributors to behavior change in unassertive

subjects. The behavioral training programs investigated by

McFall and his associates were only experimental analogues

of a fully developed clinical procedure. Nevertheless, the

procedures produced substantial improvement in the refusal

behavior of nonassertive subjects. McFall has admitted

that the target situation is a limited one and that general­

ization of results must be restricted.

All the studies by McFall and his associates have

been carefully planned and executed, using extremely

innovative and sophisticated measurement procedures. The

drawbacks are the restricted nature of the assertive behav­

ior studied and the audiotaped type of treatment which is

somewhat removed from real-life. It is encouraging, however,

that such a brief, semi-automated treatment should produce

behavior change, even with such a restricted range of

behaviors.

Several researchers have attempted to apply assertive

training procedures to a nonstudent client population.

Eisler, Hersen, and Miller (1973) compared the relative

33

effectiveness of videotaped modeling versus a practice-

control group versus a no-treatment control group. Psychi­

atric subjects in the modeling condition were exposed to a

videotaped model who was trained with respect to appropri­

ate verbal and nonverbal responses in five interpersonal

situations. The subjects in the practice-control condition

received an equal number of trials but without the benefit

of videotaped modeling. The results of the study showed

that significant pre-post differences on five of the eight

components of assertiveness were obtained for the modeling

condition when compared with the two control procedures.

The authors concluded that repeated exposure to difficult

situations does not change behavior since the practice

control group did not show significant differences from the

no-treatment control group.

In a subsequent study (Hersen, Eisler, Miller, Johnson,

& Pinkston, 1973), unassertive psychiatric patients matched

on age, years of education, diagnosis, and self-reported

assertiveness were assigned to one of five conditions with

10 subjects in each group. The five groups were: (a) test-

retest control group, (b) practice-control, (c) instructions,

(d) modeling, (e) modeling plus instructions. The authors

used a self-report measure of assertiveness and a behavioral

measure of assertiveness as the outcome criteria. The

behavioral measure consisted of five standardized inter-

M^«?-«»'S».l

34

personal situations which the subjects were asked to respond

to as they were being videotaped. The pre and post video­

taped responses were rated by two judges. The results

indicated that the modeling plus instructions group was

equal or superior to the instructions alone or modeling

alone groups on five of the seven components of the behavior

measure. There were no significant differences between the

practice-control and the test-retest groups, thus further

confirming results obtained in the Eisler, Hersen, and

Miller (19 73) study. There were no pre-posttest differ­

ences on the Wolpe-Lazarus Assertiveness Questionnaire

(self-report measure) among any of the groups. Similar

findings were reported by McFall and Marston (1970), and

only when they combined and compared the results of their

two experimental and two control conditions were these

differences significant. Hersen, Eisler, Miller, Johnson,

and Pinkston (19 73) stated that the failure of the self-

report measure to show significance is accounted for by

an "attitudinal lag" which they feel has been noted in

other behavioral contexts. The authors of the last two

reported studies felt that the significant results obtained

have greater clinical relevance than some of the v7ork done

by McFall and his associates, since they used unassertive

psychiatric patients rather than college students. Finally,

it should be noted that the studies by Eisler, Hersen, and

35

Miller (1973) and Hersen, Miller, Johnson, and Pinkston

(19 73) are somewhat at variance with the studies by McFall

and Marston (1970), McFall and Lillesand (1971), and McFall

and Twentyman (1973). The work of McFall and his colleagues

appears to demonstrate that modeling is the least effective

of the behavioral techniques used in assertive training,

while the work of Hersen and Eisler and their associates

shows that modeling can be quite effective in the remedia­

tion of unassertive behavior. Certainly, further research

on the relative effectiveness of different components of

assertive training is needed before an accurate determina­

tion can be made.

Instrumental skill training for unadaptive social

behavior was also used in a study involving reinforcement

of the expression of anger in anger-provoking situations.

Doering, Hamlin, Everstine, Eigenbrode, Chambers, Wolpin,

and Lackner (19 62) claimed that reinforcement of angry

replies to various anger-provoking situations during a pre­

test discrimination task served to cause an increase in

intensity of anger in posttest responses. It seems more

likely that the experimenter established a "set" in their

subjects to "respond with anger when you see a similar item

on the posttest," Simple instructions might have been a

more efficient procedure than this roundabout "reinforce­

ment method."

"•HHIi^^Hi^^HVSr" IVRT^T^^^JKA.fT

36

Wagner's (196 8b) subsequent study is better designed

and deserves more attention. Using the same kind of sub­

jects as in his previous study, he emphasized the con­

sequences of the assertive act, looking at the effects of

reinforcement and punishment of the role playing of anger-

expressing behavior on later posttest behavior. On a

pretest each subject made tape-recorded responses of what

he would say to the other person in several anger-provoking

situations described by the experimenter. Posttesting

involved more of the same. Training consisted of having

each subject role play, in a psychodrama setting, his

responses to several hypothetical anger-arousing situations.

The reinforced groups found that the other person in the

role play submitted to his anger and apologized for his

provocation. The punished group received anger retalia­

tion from the other person to the extent that the subject

"lost the battle," A control group role played other non-

anger- arousing scenes. Results showed a significant

increase in anger between assessment for the reinforcement

group but not for the others. Surprisingly, the punish­

ment group did not decrease in anger verbalization. Either

the punishment was not intense enough, practice at anger

expression somewhat innoculated them against the retalia­

tion, or they did not take the experiment that seriously

and so did not perceive the situation as very threatening

or punishing.

^ ..mmmiK^^^msmfmimi^Ayv W:-'-

37

In summary, the above studies point towards the

importance of reinforcing approximation of appropriate

assertive behavior during the treatment of the unasser­

tive patient. Moreover, the importance of the patient

undergoing a successful "assertive behavior" when applying

what he has learned in treatment to his natural environment

is underscored.

Assertive Training in Groups

A relatively recent modification of the traditional

assertive training repertoire has been a movement to use r

these techniques in areas beyond the confines of the

traditional one-to-one therapy situation. Perhaps influ- .:'

enced by the effects of the curative forces seemingly

operative in group therapy (Yalom, 19 70), behavior thera­

pists have attempted to extend assertive training to

treating groups of clients. As yet, little research has

been reported on this phenomienon. The few studies are

reported here,

Hedquist and V7einhold (19 70) examined the efficacy

of two methods of behavioral group counseling for socially

anxious and unassertive college students. Subjects were

used who scored high on the SR Inventory of Anxiousness

(SR-I) and the AS (Unassertiveness Scale (AS) of the

Guilford-Zimmerman Temperament Survey. Target behaviors

vfm-:wm:^'TV;m^mi^i'f

38

were any of the 14 items on the SR-I that subjects indi­

cated were problem areas for them. The criterion was the

frequency of verbal assertive responses made by each sub­

ject outside the group in situations they had identified

as distressing. This was operationally defined as an

"overt verbal response that required the subject to initiate

a social action or reaction with another person or persons

(the verbal assertive response)" (p. 239). A novel feature

of this study was the use of an "interpersonal diary" as an

assessment device. In this "diary" each subject recorded

the day, time, place, and person (s) with whom he/she emitted

any one of the responses from his or her selected list of

target behaviors. The measure of treatment effectiveness

was the total number of responses for each treatment group

of subjects taken collectively.

Subjects in the two treatment groups were matched on

SR-I and AS scores, and so far as possible, by target be­

havior from the "diary" with a control subject in a problem-

solving discussion and diary-keeping group. The same two

counselors co-led both treatment groups. Treatments were:

(a) Behavior Rehearsal Group which underwent behavior

rehearsal, some problem-solving procedures, feedback from

group members, direct modeling, and coaching from a thera­

pist; (b) Social Learning Group in which counselors modeled

problem-solving behaviors and saw to it that the subjects

!'?^?!'5^^W

39

kept the ground rules of honesty, responsibility, helpful­

ness, and action. An analysis of variance of a number of

assertive responses over the 6 weeks of treatment showed

that there were no significant differences in the number of

recorded assertive responses between the two treatment

groups, but that both groups did produce significantly more

verbal assertive responses than did the matched controls.

Follow-up data gathered 6 weeks after completion of treat­

ment using the diaries as the measure, revealed that there

were no statistically significant differences among the

three groups at that time although the trend still favored

the two treatment groups.

It is noteworthy to mention that the demand character­

istics of the situation may have confounded the treatment

effects producing the initial posttest results. The sub­

jects were students in an education course participating

in "treatment" as a part of a lab project. The students

may have wished to perform well for class and then returned

to "normal" when the class term was over. Subjects were

explicitly asked and expected to improve their assertive

behaviors. Secondly, although scoring relatively high on

the measures of anxiety and unassertiveness employed,

siabjects had not requested therapy. Third, the target

behaviors in this study often involved social interactions

and may have been somewhat superficial situations. One

40

might wonder just how seriously the students took the

behavior change situation.

The idea of choosing self-admitted target complaints,

the use of self-monitoring in the form of a diary, and

assertive training in a group are noteworthy additions

to the assertive training repertoire. Unfortunately, the

effort of the authors seemed diluted on what was primarily

a classroom exercise in which the demand characteristics

of that situation may have accounted for the results

obtained.

Rathus (19 72) compared the efficacy of assertive

training and a discussion procedure with unassertive

undergraduate females using a group treatment approach.

Subjects received seven sessions of treatment in each of

the two groups. In the assertive training group subjects

practiced tasks derived from Salter's (1949) "excitatory

exercises." In the group discussion condition subjects

met with the experimenter and reviewed the nature, acqui­

sition, and elimination of fear. It was hypothesized that

the subjects receiving assertive training would report and

exhibit more assertive behavior, and experience greater

reduction in fear during social confrontation than subjects

in the control groups. Pre-post differences on the Rathus

Assertiveness Scale indicated that assertive training

subjects reported significantly greater gains than discussion

£*>

41

or no-treatment control subjects. However, the study had

two methodological problems. First, subjects v;ere drawn

from the experimenter's own classes. Second, the experi­

menter served as therapist for both experimental groups

and the possibility exists of experimenter bias. Addi­

tionally, judgments of overall assertiveness made by inde­

pendent judges failed to differentiate the groups at a

statistically significant level. In a subsequent study,

Rathus (1973a) compared the effects of assertive training,

placebo therapy, and no treatment. The assertive training

group consisted of 2 4 undergraduate college v7omen who met

weekly seven times for hour-long sessions in v/hich they

observed videotaped-mediated assertive models and practiced

nine types of assertive responses. Subjects receiving the

placebo treatment also met weekly seven times for hour-long

sessions in which they observed videotaped-mediated models

and engaged in between-sessions assignments. During each

session the v/omen viewed tapes discussing or showing system­

atic desensitization to coramon social fears. Subjects in

the assertive training group evidenced significantly greater

pre-post changes on the Rathus Assertiveness Scale than

those in the placebo and no-treatment conditions. In addi­

tion, independent judgment of subjects' overall assertive­

ness, based on audiotaped question-and-answer sessions,

confirmed the superiority of the assertive training group.

42

The assertive training subjects were drawn from the experi­

menter's classes and treatment for both experimental con­

ditions was provided by the same experimenter.

Weinmen, Selbert, Wallace, and Post (19 72) conducted

an extended clinical outcome study in which three types of

therapeutic approaches were administered to hospitalized

chronic schizophrenics. The major treatment goal involved

increasing interpersonal assertiveness and interaction in

these patients, A total of 6 3 subjects were involved in

the study. Dependent measures selected for study were the

behavior and critical situations scale (objective behavioral

assessment of assertiveness), an 18 item anxiety question­

naire, and a shortened form of the Fear Survey Schedule II,

Subjects in the experiment were carefully matched on several

criteria and assigned to the following three treatment groups;

socioenvironmental therapy, systematic desensitization, and

relaxation therapy. Subjects in the systematic desensiti­

zation group received 36 one-half hour sessions over the

3 month period, but without systematic desensitization.

Socioenvironmental therapy (conducted over a 3 month period)

consisted of 5 weekly group activities directed towards

promoting social interaction. Subjects were divided into

older (48-67) and younger (20-42) groups for purposes of

data evaluation. The results indicated that older patients

in the socioenvironmental therapy condition demonstrated

0

43

the greatest pre-post changes in assertiveness. No differ­

ences were obtained for the younger group of patients.

Decreases in self-reported anxiety occurred for all three

groups irrespective of patient's age. V7einman and his

associates conclude that "the treatment of choice for the

older chronic schizophrenic remains socioenvironmental

therapy" (p. 252). The results of this study must be

interpreted with caution for several reasons. First, one

drawback of the study is that relaxation and systematic

desensitization are not the most ideal controls and are

generally not considered applicable to psychotic patients.

Second, the socioenvironmental therapy outlined in the

study was not a direct training procedure designed to teach

precisely those social skills needed by patients. Therefore,

it may be inappropriate to directly compare the socio­

environmental therapy used in this experiment with assertive

training procedures as generally outlined by the major

authors of assertive training.

Martinson and Zerface (19 70) investigated the differ-

,ential effectiveness of individual psychological counseling,

a social program of arranged interaction with college

females and a delayed treatment control for male students

evidencing a fear of dating. The individual counseling

group involved an average of three sessions and was described

as being "eclectic." In the arranged interactions group.

44

male subjects met once a week for 5 weeks with female sub­

ject volunteers, "interested in improving their social

skills," No treatment was given to the delayed treatment

control group. Results of this study show that the

arranged interactions group significantly decreased fear

of dating (verbal report) and increased actual dating

frequency when compared to the other two groups. Martinson

and Zerface underscored the fact that a "sem^-structured

social program . , , which included no contact with a pro­

fessional counselor proved more effective than professional

counseling for young men characterized by a fear of dating"

(p. 39),

Working from a self-reinforcement framework, Rehm and

Marston (196 8) attacked the problem of unassertive hetero­

sexual behavior and social anxiety. Their goals were to

effect a positive change in self-concept, reduce social

anxiety, and increase the approach to the dating fear

situation. The authors used the following techniques to

achieve their goals: (a) graded structure of difficult

situations; (b) behavioral goal setting for each individual;

(c) instructions to use lower evaluative criteria for rein­

forcement, namely, to use behavioral acts as goals and

bases for evaluation, not the feeling of confidence or

anxiety reduction; (d) self-monitoring; (e) self-reinforce­

ment for behaviors (point-giving); and (f) a group context.

• 45

In comparison with a nondirective therapy group and a no-

therapy control group, the self-reinforcement group showed

greater reduction in anxiety on several measures of verbal

report of discomfort in heterosexual social situations.

They also reported more dates with girls and evidenced

more verbal output on a situational test.

The research by Rehm and Marston reinforces the notion

that the subjects can be directly trained in behavioral

skills. Along with the other studies, reported in this

section, they demonstrate the efficacy of assertive train­

ing in a group context. In addition, these studies inter­

ject another approach to assertive training, allowing the

client to manage much of the behavior-change process on

his own. Training and goal setting, problem-solving, and

self-monitoring enable the unassertive person to take more

control of his life and handle new situations without

relying on a therapist.

Assertive Training with Alcoholics

Alcoholism primarily represents a learned pattern of

behavior rather than a manifestation of a particular type \

of predisposing underlying pathology (Vogel-Sprott, 1967;

Bandura, 1969). The physiological and psychological effects

of ethanol are very important in any discussion of alcoholism. '

Certainly the stress-reducing properties of alcohol make it

a potent positive rainforcer. Conger (1956) applied the

46

established behavioral principles of Hull to the problem

of chronic excessive drinking. Like other forms of

behavior, drinking alcohol was viewed not as a unique

manifestation, but as a response among many in the reper­

toire of an organism. Conger hypothesized that the exis­

tence of tension energizes the response, and the relief

of tension provided by alcohol reinforces the drinking

response. After a person becomes physically dependent on

alcohol, he is compelled to consume large quantities of

liquor both to alleviate distressing physical reactions

and to avoid their recurrence. Since the ingestion of

intoxicants promptly terminates physiologically generated

adversive stimulation, i.e., shakes and DT's, drinking

behavior is automatically and continuously reinforced.

Although this analysis may seem simplistic, it has provided

the basis for a great deal of scientific investigation as

evidenced by an excellent review of 8 8 studies by Cappell

and Herman (1972). They found that the tension-reduction

hypothesis of alcohol has been tested in a wide variety of

experimental situations, in a number of different species

and with varying degrees of sophistication in experimental

methodology and design. Much of the data is equivocal and

sometimes contradictory, although supportive results were

found in the area of conflict and experimental neurosis

(Cappell & Herman, 1972).

/

47

From a social learning point of view, alcoholics are

people who have acquired, through differential reinforce­

ment and modeling experiences, alcohol consumption as a

widely generalized response to aversive stimulation (Vogel-

Sprott, 1967). Therapeutic attention would therefore be

most profitably directed toward reducing the level of

aversive stimulation experienced by individuals and toward

eliminating alcohol stress responses either directly or,

preferably, by establishing alternative modes of coping.

Given more effective and rewarding means of dealing with

environmental demands, individuals will have less need to

resort to self-anesthetization against everyday experiences.

Research by Gross and Carpenter (19 71) indicated that

226 hospitalized alcoholics scored significantly lower than

the general population on the E and H factors of the Sixteen

Personality Factor Questionnaire. A person who scores low

on the E factor tends to give v/ay to others, to be docile,

and to conform. He is often dependent, submissive, and

overly accommodating. Similarly a person who scores low

on the H factor tends to be shy, withdrawing, cautious,

restrained, retiring, a "wallflower." He tends to be slow

and impeded in speech and in expressing himself. These two

factors seem to be closely related to the concept of asser­

tiveness and unassertiveness. Further research needs to be

conducted to determine if in fact alcoholics are more

assertive than the general population.

48

Drawing on 10 years of research into the psychological

meaning of alcohol consumption, McClelland, David, Kalin,

and VTanner (19 72) have developed a theory of alcoholism

which in some respects gives added validity to the idea of

using assertive training v;ith alcoholic patients. They

attempted to explain som.e of the variations in rates of

heavy drinking in terms of their pov er concern theory. For

example, why should middle-aged men drink more heavily than

younger or older men? They answered rhis question with the

following explanation.

Generally speaking, more strength and assertive­ness is expected of mi.ddle-aged men with heavy family and work responsibilities than of either younger or older men. Yet this increase in respon­sibility is associated with a regular physical decline in potency both in the sexual and aggres­sive senses of the term. What is more likely than that men faced with high demands for assertiveness and a lessening capacity should turn more often to the artificial sense of increased potency that drinking produces? (p. 296)

McClelland and his associates reported on a pilot

attempt to help alcoholics by socializing their power needs.

Although they arrived at their theory of drinking behavior

quite independently of a behavioristic model, it is interest­

ing to note that several of their specialized treatmient

approaches utilized role playing in ccial and interpersonal

situations and other strategies which are not uncommon to

assertive training. Follow-up of four alcoholic subjects

showed that the socialization treatment program "was clearly

49

of some benefit." They called for additional research

comparing their theory of personalized power with other

treatment approaches to see which is more effective.

Many researchers have attempted to define the

"alcoholic personality." However, in 1950, Sutherland,

Schroeder, and Tordella reviewed the literature related

to the hypothesis that alcoholism is caused by personality

traits. They concluded that no satisfactory evidence had

been found to warrant the contention that persons of one

type are more likely to become alcoholics than persons of

another type. In 1957, Syme reviewed another 7 years of

psychological studies and concluded again that there was

no evidence of a definable alcoholic personality upon which

alcoholism is likely to develop. There are, though, certain

characteristics that do appear to be quite common in a

majority of alcoholics. Catanzaro (1968) has enumerated

the following characteristics: (a) anger over dependency,

(b) inability to express emotions adequately, (c) high level

of anxiety in interpersonal relations, (d) emotional imma­

turity, (e) ambivalence tovzard authority, (f) low frustra­

tion tolerance, (g) grandiosity, (h) low self-esteem, (i)

feelings of isolation, (j) perfectionism, (k) guilt, (1) com-

pulsiveness, and (m) sex-role confusion. Several of these

characteristics appear to be directly related to the need

for assertive training with alcoholics.

50 /

Inability to express emotions adequately is common in alcoholics, and is also rather prominent in people suffering from depression. Alcoholics are in general very sensitive people. Consequently, they tend to build up feelings of anger at even minor rejections or frustrations. In addition, they find it very hard to deal adequately with their great wealth of angry feelings. They often find it difficult to talk out their feelings, and therefore either hold them inside or explosively let them out in an argument or fight. One of the main aims of therapy is to help the alcoholic learn to express his feelings verbally. (Catanzaro, 1968, p. 16)

Alcohol's ability to reduce a high level of anxiety in personal relations has caused Dr. Jellinek to dub alcohol "a social lubricant." Many alcoholics, early in their disease, use al­cohol as a drug for calming anxious and insecure feelings which arise at social gatherings. Thus, as their inability to deal with people effectively becomes more pronounced, they need increasing amounts to blot out their increasingly unpleasant reality. (p. 17)

Feelings of isolation are the natural outgrov7th of his inability to get along with people. As the alcoholic continues drinking, his behavior and con­versation become less acceptable to those about him, and consequently, his family and friends begin iso­lating him from their social circle. (p. 18)

-The three above mentioned characteristics appear to

show that assertive training procedures may be very useful

in developing social and interpersonal skills in alcoholics

By providing alcoholics direct training in precisely those

interpersonal and social skills lacking in their behavior,

it is hypothesized that they will have less need to resort

to alcohol.

Miller, Hersen, Eisler, and Hilsman (1974) attempted

to design a study to evaluate the role of stressful inter­

personal encounters on the etiology of alcohol abuse.

51 V

They hypothesized that when confronted with interpersonal

situations necessitating assertive behavior, alcoholics

would increase their rate of responding to obtain alcohol

on an operant task. On the other hand, it was hypothesized

that social drinkers would not show this relationship. The

subjects for this experiment consisted of eight alcoholics

and eight nonalcoholics hospitalized at a Veteran's Admin­

istration Hospital.

The social stress condition consisted of everyday life

experiences in which the subjects were to act assertively.

After the subjects responded, the experimenter would tell

the subjects that their performance was very poor compared

to other patients and that they apparently have let people

boss them around and that they don't stand up for their

rights. The subjects were then confronted by two experi­

menters with their social, parental, or vocational inade­

quacies. Pulse rate was obtained immediately prior to and

following these interactions. Both alcoholics and social

drinkers significantly increased their pulse rates as the

function of the stress conditions. The results of the study

revealed that exposure to interpersonal encounters requiring

assertive responses increased alcohol consumption in chronic

alcoholic subjects. By contrast, social drinkers did not

exhibit increased alcohol consumption under stressful

conditions.

52^

These data corroborate findings . . . in that alco­holics do not seem to have a lower tolerance for social stress than non-alcoholics. Rather, as a function of his prior experience, the alcoholics have learned to respond to stressful situations by consuming alcohol whereas the non-alcoholic has learned a variety of more adaptive responses (e.g., being appropriately assertive). (Miller, Hersen, Eisler, & Hilsman, 1974, p. 71)

Martorano (19 74) suggested that if alcoholics were

given an alternative behavior to drinking as a vehicle for

emotional expression, they might be able to decrease their

drinking. In an unique experiment, Martorano set out to

investigate the interrelationships among alcohol consump­

tion, mood, and social contact in four male alcoholics.

V7ithin each of three periods (Baseline, Experimental Period

I, and Experimental Period II), 6 days of drinking were

programmed during which subjects could buy alcohol, social

interactions, or closed circuit TV communications time with

points that they had previously earned at the operant console

or by participating in assertive and nonassertive training

tasks. The results showed that assertive training benefited

alcoholic subjects by increasing their social desirability

and activity and by decreasing resentment built up through

suppression of anger. The study also suggested that the

positive changes experienced by these men from assertive

training were not maintained once the drinking began. Fol­

lowing assertive training, the men were more likely to

externalize'anger and less likely to feel guilty about it.

53

Since this study utilized a within-subjects, small N design, \ /

the author felt the results of the study should be inter­

preted with caution.

Burtle, Whitlock, and Franks (19 74) in a pilot research

project with 16 women alcoholics, indicated that training

and techniques of social interaction played an important

role in the restoration of self-esteem and as a means of

helping women alcoholics maintain their sobriety. Follow-up

questionnaires returned by 10 subjects indicated that some

of the gains from the behavioral program were maintained

after 16 weeks. It is difficult to determine the extent to

which assertive training positively affected the outcome

since the design of the study did not utilize a control

group and assertive training V7as confounded with other be­

havioral techniques; however, the results are encouraging.

Eisler, Miller, Hersen, and Alford (19 74) reported the

effects of assertive training on marital interaction. Three

couples were videotaped while discussing their marital con­

flicts before and after husbands received training in asser­

tive expression. In the context of role-played interpersonal

encounters, assertive training consisted of instructions,

behavior rehearsal, coaching, and feedback. In all three

cases, behavioral tests revealed substantital improvement

in the husband's assertiveness. In two of the three cases,

increased assertiveness produced marked changes in the

54 ,/

couple's marital interactions. One of the most interesting

aspects of the author's article was the fact that one of the

husbands had a 6-year history of heavy sporadic drinking

following episodes of intense marital conflict. Several

issues involved in the marital discord were: the discipline

of his 21-year~old retarded daughter, the amount of time

he spent at home with his wife, and frequent arguments about

whether he could have a few beers to relax in the evening

without becoming inebriated. Breath alcohol levels taken

weekly for 6 weeks prior to assertive training ranged from

.01% to .20% with a mean of .08%. For 6 weekly intervals

following training, the levels ranged from .00% to .04% with

a mean of .02%. Thus, it appears that training the husband

to be assertive in simulated marital encounters did general­

ize to actual marital interactions. Not only did changes

in the husband's assertive behavior transfer to the inter­

action with his V7ife but also they led to a decrease in the

amount of alcohol he consumed. Although this finding

relates to only one case, it is indicative of the potential

effectiveness of assertive training with alcoholics.

Statement of the Problem

Purpose and Scope of the Study

The study had several major objectives. The first

objective was to delineate and better define the parameters

of unassertiveness-assertiveness in alcoholics as compared

55

to normals. The second objective of the study was to

demonstrate that assertive training can be an effective

technique in modifying unassertive behavior in alcoholics.

The design of the study not only attempted to show that

unassertive thoughts and attitudes can be changed, but

that the actual behavior of alcoholics in situations

requiring assertive responses could be modified.

Basically, the study attempted to determine whether

alcoholics can be trained to behave more assertively

following an assertive training program. Because this

study was the first in a planned series of experiments, it

did not answer several important questions. The first of

these questions relates to the issue of transfer of train­

ing once the patient has returned to the community. The

second and possibly the most important question is whether

assertive training in fact helps alcoholics maintain their

sobriety. Although this question will not be answered in

the present study, data regarding type, frequency, and

quantity of alcohol consumed were gathered in the hope of

doing a 1 year follow-up on all subjects (see Appendix A).

Importance of the Study

The significance of this study was twofold. First,

it dealt with the nation's third largest health problem.

In this country, the age-old problem of excessive drinking

is taking a disturbing new turn for the v'orsc. In the last

56

10 years, per capita consumption of alcohol in the United

States has increased approximately 26% (Chafetz, 1974).

More and more of the nation's young people are becoming

addicted to the most devastating drug of all—alcohol.

About 1 in 10 of the 95 million Americans who drink are

either full-fledged or at least serious problem drinkers.

The dollar cost of alcoholism may be as much as 25 billion

a year, much of it from lost work time in business, industry,

and government (Chafetz, 1974).

The second major importance of the study was that it

represented a movement av7ay from a generalized treatment

philosophy for alcoholism to a more specific approach which

matches treatment and therapist with client and problem.

It is generally recognized by practitioners in the field

of alcoholism that since multiple factors seem to contribute

to alcoholism, multiple approaches are necessary to promote

human well-being and to enhance human dignity in order to

prevent and treat alcoholism (Chafetz, 1974). Recognizing

the wealth of therapeutic approaches available today for

alcoholic patients, treatment staffs are studying the best

way to assign patients to treatment modes most suited to

their psychological and physical needs. An example of this

viewpoint is a study conducted by Kissin, Platz, and Su

(19 72) entitled "Selective Factors in Treatment Choice and

Outcome in Alcoholics." After an initial social interviev;

•-y:\ -^^ff^\

57

and psychological testing, over 450 alcoholic men were

assigned randomly to one of four treatment conditions. In

general, the investigators found that the success rate of

the treatment condition was directly correlated with the

number of treatment options offered to the patient.

If we conclude that the general population of alcoholics includes different types who would do better in different kinds of treatment, then the more types of treatment offered to a random group, the greater the overall success rate should be. This hypothesis was bcrne out, (Kissin, Platz, & Su, 1972, p, 782)

Kissin's study suggested that an eclectic approach

tailored to individual needs is likely to yield the greatest

benefit to persons abusing alcohol and if treatment programs

are to be truly effective for a large population of their

clients, then specialized and individualized treatments need

to be developed and applied.

b.-..:

CHAPTER II

METHODOLOGY

Subjects

The population from which subjects were selected for

this study was defined by the following criteria: chronic

state hospitalized alcoholics, at least 7 years of school­

ing, scoring in the lower 50th percentile on the Rathus

Assertiveness Scale, nonpsychotic, with at least low

average level of verbal functioning. Chronic alcoholism

here implied a relatively long (a year or more) period of

regular alcohol abuse, with the manifestation of withdrawal

symptoms upon termination of drinking episodes. One hundred

and two subjects at two state hospitals were randomly

assigned to the three treatment groups with 34 subjects

in each group. Another and separate group of 123 state

hospitalized alcoholics was randomly chosen as the stan­

dardization group for the self-report measure used in this

study.

Instrumentation

Four instruments were used; they were a demographic-

drinking data form, a self-report in nature, a behavioral

measure, and an unobtrusive rating of assertiveness.

Demographic-Drinking Data Form (DDDF). The DDDF

(see Appendix A) v7as developed to secure data relevant to

58

59

a description of the sample as well as to provide baseline

data regarding type, quantity and frequency of alcohol

consumed. This was important for several reasons. First,

it gave relevant data as to the populations to which the

results may be generalized (external validity). Secondly,

the data would be extremely useful in determining which

treatment significantly affected the quantity and frequency

of alcohol consumed should a follow-up be done at a later

time.

Rathus Assertiveness Scale (RAS). Rathus (1973b) " • • • • - I - - - T - T l I I I I • « 11 11 I I J» I I I I I . . I I I. •• _ . . I l l . • • .1 . • _ . . . . . . t J i M

developed a 30-item scale (see Appendix B) that is based

in part on assertive questions previously used by Wolpe

and Lazarus (1966) and Wolpe (1969). Each statement is

rated with a Likert Scale ranging from +3 to -3. The

possible total score can range from -90 to +90. Test-retest

reliability of the RAS was determined by administering the

instrument to 6 8 undergraduate college men and women ranging

in age from 17 to 27, and then retesting them after 8 weeks

passed. The mean pretest score was .2941; the standard

deviation was 29.121. Mean posttest score was 1.6176,and

the standard deviation was 2 7.6 319. A Pearson product-

moment correlation coefficient was run between respondents'

pre- and posttest scores yielding an £ of .7782 (£<.01),

indicating moderate to high stability of test scores over

a 2-month period. Split-half reliability yielded an r of

Vl9f

60

.772 3 (£ < .01) suggesting the quality measured by the RAS

possesses moderate homogeneity.

The validity of the RAS was established by comparing

self-reported RAS scores to two external measure of asser­

tiveness. In the first validity study, 6 7 subjects who had

taken the RAS were rated on a 17-item rating schedule by

someone they knew well, A Pearson product-moment correla­

tion coefficient was then computed. RAS scores correlated

significantly (£ < ,01) with each of the scales comprising

an assertiveness factor, RAS scores thus serve as a valid

indicator of respondents' assertiveness in terms of the

impression they make on other people (Rathus, 1973b),

Another index of RAS's validity was determined by

comparing subjects' RAS scores with ratings of their re­

sponses to five taped interpersonal situations requiring

an assertive response, A Pearson product-moment correlation

coefficient was computed between the RAS scores-and the

scores from the audiotaped sessions yielding an r of .7049

(£ < .01). Thus, RAS scores are also valid in terms of

impartial raters' impressions of the behavior that subjects

report they would exhibit in specific social encounters

(Rathus, 19 73b).

Behavior Assertiveness Test (BAT). Suggested by

similar devices used by McFall and Marston (1970), McFall

and Lillesand (1971), McFall and Twentyman (1973), Friedman

ifi«1

61

(1968), and Eisler et al. (1973), the BAT aimed to assess

the subjects' reactions to role-played threatening or irri­

tating social or interpersonal situations involving com­

binations of (a) standing up for one's rights, (b) expressing

one's feelings honestly and directly, and (c) showing anger

in a provoking situation. Subjects were presented with 10

tape-recorded stimulus situations (see Appendix C) requiring

assertive responses. They were instructed to respond to

each situation as if it were actually happening to them.

The siibjects' responses were tape-recorded by a

second tape recorder and analyzed at a later date by three

raters. ' The subjects' responses were scored in terms of

response latency, response duration, and on a six point

scale of overall assertiveness (see Appendix D) utilized

by Weiskott (19 75). Since Weiskott showed that the rating

scale of assertiveness had high inter-rater reliability

(ranging from .95 to .97), it was not deemed necessary to

establish inter-rater reliability prior to the actual use

of the instrument in the study. The three raters were given

an intensive 2-hour training session to acquaint them with

the use of the scoring instrument. The possible scoring

for each of the 10 situations ranged from a 6 (a very asser­

tive response) to a 1 (a totally unassertive response). The

total assertive score on the BAT could thus range from 60

(perfectly assertive in each situation) to a low of 10

(extremely passive and unassertive in each situation).

62

The response latency for each subject was determined

by totaling the length of time between a bell sounding

(indicating a subject could respond) and when the subject

actually began responding. Each of the 10 role-played

situations allowed the subject a maximum of 20 seconds in

which to make his response. The response duration was

determined by totaling the actual amount of time the

subject spoke during each of the situations.

Since all the previous behavioral role-playing tests

had been constructed for either college students or psy­

chiatric populations, it was necessary to construct a

behavioral test which was appropriate for an alcoholic

population. The items for the BAT were selected in several

ways. First, items which had appeared in other behavioral

role-playing tests were used if they were appropriate for

the population of subjects being tested. For example,

situations involving college roommates were excluded.

Secondly, a sample of 45 alcoholics in the state hospital

were asked to write down those situations in their life in

which they had the most difficulty being assertive. Those

situations which appeared most frequently were included in

the study. The following is an example of one item which

appeared in the final version of the BAT.

Narrative: Imagine that this morning you took your

car to a local Exxon station and you

63

explicitly told the mechanic to give you

a simple tune-up. The bill should have

been about $20. It is now later in the

afternoon and you're at the station to

pick up your car. The mechanic is walking

over to you.

Mechanic: "Okay, let me make out a ticket for you.

The tune-up was $12 for parts and $8 for

labor. Uh, grease and oil job was $6.

Antifreeze was $5. Uh, $4 for a new oil

filter. And uh, $5 for rotating the

tires. That's $40 in all. Will this be

cash or charge?"

Bell sounds and subject begins responding.

20 seconds

Stop if you have not finished responding!

Assertive Behavior Index (ABI). This rating instrument

(see Appendix E) was specifically developed for this study.

The instrument attempted to measure a number of verbal and

nonverbal components of assertiveness such as eye contact,

facial expression, body movement, loudness of voice, and

fluency of voice. Each of the eight variables was scored

on a five point scale ranging from -2 to +2. The negative

end of the scale represented the unassertive aspect of the

64

variable while the positive end represented the aggressive

aspect of the variable. The center point (zero) represented

the appropriately assertive aspect of the variable. Possi-

bile scores ranged from -16 to +16.

All subjects, after completion of their 2 weeks in

the study, were individually interviev7ed regarding their

plans for returning home after discharge from the hospital.

These interviews generally lasted about 15 minutes and were

conducted by psychologists, alcoholism counselors, and

social workers not directly related to the study. Upon

termination of the interview, the staff member completed

the ABI, rating the patient's behavior during the previous

interview.

Procedure

Data collection for the investigation was accomplished

during a 5-month period from November, 19 74 through March,

19 75. Prior to the start of the actual study, the Rathus

Assertiveness Scale was randomly given to 123 alcoholics at

three state hospitals. The results of this testing were

tabulated and the standardization/normalization data were

used to spotlight unassertive subjects for inclusion in the

study and to test the first hypothesis. The mean Rathus

Assertiveness score was calculated for this sample of 12 3

alcoholics and found to be -5.04. This information was

subsequently used.to identify patients for inclusion in the

f-wi'-^y;: •^mmmammmmmmm^-"' ''"'•*' -^ws^sm.»i

65

second phase of the study. During a 4-month period,

almost all alcoholic patients at Big Spring State Hospital

and San Antonio State Hospital were administered the RAS.

All subjects were detoxified and free from major physical

problems when the RAS was administered. If a patient

scored -5 (50th percentile) or below on the RAS and met

the other criteria for inclusion in the study, he was

randomly assigned to either one of the two treatment groups

or the control group. The 50th percentile was an arbitrary

cut-off point, but it did help assure that subjects included

in the study tended toward unassertiveness, at least as

measured by the Rathus Assertiveness Scale,

The trainers for the treatment groups consisted of

four Masters level psychologists and counselors, two from

each state hospital. All treatments were conducted by

co-trainers. Prior to the start of the study, the four

trainers were brought.together for an intensive 3-day

workshop on assertive therapy.

Control Group

The subjects in the Control group were allowed to

participate in all alcoholism unit treatment and activities

except those involving assertive training. Subjects in the

Control group were posttested with RAS, BAT, and ABI after

2 weeks. Assertive treatment was not given to the Control

66

s u b j e c t s a f t e r t e r m i n a t i o n of t h e s t u d y s i n c e fo l lov7-up

d a t a w i l l be g a t h e r e d .

Minimal Assertive Training Group * ' " * ' " * * " " •• I • I l i a , • P i l l —T— I • . , • • W B i f c u M . i m i n i ami i — ^ • • • ^ ^ • l ^

Subjects in the Minimal Assertive Training group, in

addition to the regular treatment program, received 2 hours

of didactic presentation and group discussion on hov7 each

person could behave more assertively. This treatment group

was an open-ended group, that is, clients came into the

group and stayed for two sessions. The size of the group

fluctuated from week to week as new clients entered and

other clients who met the criterion of two sessions left.

The first session was an elaboration of the problem. Each

subject explored the situational determinants of his non-

assertiveness, distinguishing when he/she could perform or

could not perform satisfactorily and specifying the critical

factors. Toward the end of the session, the therapist

offered the following general interpretationsof subject's

inhibitions: "You are too concerned with what other people

think of you. You must stop worrying about making everyone

like you and stand up for your rights!" This theme was

elaborated in terms of each subject's particular problem

and each subject was then encouraged to overcome his/her

unassertiveness in daily life activities.

The second session came at least 1 week later and

continued to explore the same basic theme. Successful

67

and unsuccessful attempts made by patients during the week

were discussed and analyzed. Part of the session was

devoted to a didactic lecture on the unassertive-assertive-

aggressive continuum, helping each patient realize situa­

tions when he/she could act more assertively. Subjects

were given a common sense explanation of why they behaved

unassertively and they were taught how to discriminate

between assertive and unassertive behavior. They were

strongly encouraged to behave assertively. At the end of

2 weeks, subjects were posttested with the RJ-iS, BAT, and

ABI.

Assertive Training Group

The subjects in this treatment group received, in

addition to the regular treatment program, 10 hours of

assertive training over a 2-week period. This treatment

group was also an open-ended group. Clients entered the

group and stayed for a total of 10 sessions. The size of

the group fluctuated from day to day as new clients

entered and other clients who met the criterion of 10

sessions left. The counselors leading this group utilized

all the components of assertive training: modeling, coach­

ing, role playing, role reversal, instructions, behavior

rehearsal, feedback, graded-structured exercises, and

homework assignments. Each subject was asked to discuss

situations in his/her own life which were proving

m^

68

problematic. Much of the focus of this group was to develop

skills in precisely those aspects of unassertiveness in

which the client was having difficulty. Subjects were

given a theoretical rationale for their treatment and they

were strongly encouraged to begin behaving assertively.

At the end of 2 weeks, siobjects were posttested with the

RAS, BAT, and ABI,

Hypotheses

In accordance with the previously stated objectives

of this study, there were four major hypotheses specifically

considered in this experiment. The hypotheses are stated

in the null form,

1. There is no difference between the assertiveness

of alcoholics and that of "normals," as expressed

by scores obtained on the Rathus Assertiveness

Scale.

2. There is no difference between pre- and post-

assertiveness scores, for all the groups as mea­

sured by the Rathus Assertiveness Scale,

3. There is no significant difference between groups

regarding the effectiveness of the treatments as

measured by elements of the Behavioral Assertive­

ness Test (response latency, response duration,

and total assertiveness scores).

69

4, There is no significant difference betv7een groups

regarding the effectiveness of the treatments as

measured by the Assertive Behavior Index.

Statistical Analyses

The first hypothesis was tested by means of a t test

between the standardization group of alcoholic subjects on

the RAS and a group of college students ("normals") reported

by R.thus (19 73b) . The second hypothesis V7as tested by

using an analysis of covariance design. The third hypothe­

sis was tested by a Kruskal-Wallis one-way analysis of

variance by ranks for response latency, follov7ed by the

Mann-Whitney U Test. Response duration and total assertive­

ness scores were analyzed by a one-way analysis of variance.

The fourth hypothesis was tested by using a one-way analysis

of variance design. The analyses of the second, third, and

fourth hypotheses V7ere followed by Duncan's Multiple Range

Test (Kirk, 1968). Additional data from the RAS were analyzed

via correlational methods to help determine if age, sex,

number of admissions, length of stay, and educational level

affect assertiveness scores on the RAS.

CHAPTER III

RESULTS

For purposes of clarity, this chapter is divided into

five sections corresponding to the Demographic-Drinking Data

Form and the four stated hypotheses of the study.

Demographic-Drinking Data Form

Relevant means and percentages on the data gathered

from the Demographic Drinking Data Form (DDDF) are presented

in Table 1 for the entire sample of 102 alcoholics involved

in the assertive training component of the research project.

TABLE 1

DEMOGRAPHIC-DRINKING DATA

1. Sex: Male = 85% Female = 1 5 %

2. Mean Educational Level = 11.04

3. Mean Age = 4 7

4. Marital Status: 9% Never Married 2 7% Married Living with Spouse 6% V7idowed

44% Divorced/Annulled 14% Separated

5. Mean number of times clients had been hospitalized for alcohol-related problems during previous 30 days = .66

6. Mean number of days clients had been in state hospitals at the time of the RAS pretest = 8.13 days

70

71

TABLE 1—Continued

7. Mean number of days clients drank during previous 30 days = 20,08 days

8. Drinking Status:

How long since clients' last drink?

cx:

5 8% 1-6 days 25% 1-2 weeks 10% 3-4 weeks 7% 5-8 weeks

9. Drinking Frequency (previous 30 days):

Beer Wine

29% 6%

18% 9% 6%

32%

Eve ry day 5-6 days a 3-4 days a 1-2 days a Less often weekly None

week week week than

12% 5%

18% 4% 7%

54%

Every day 5-6 days a 3-4 days a 1-2 days a Less often weekly None

week week week than

..Liquor

23% 9%

18% 11% 11%

28%

Every day 5-6 days a 3-4 days a 1-2 days a Less often weekly None

week week week than

10. Drinking Quantity (typical day in which clients drank)

Beer V7ine

13% 3%

12% 14% 22% 4%

32%

6 quarts or more 5 quarts 4 quarts 3 quarts 1-2 quarts 1-3 glasses None

--"

4% 6%

14% 15% 7% •

0%

54%

5 fifths or more 3-4 fifths 2 fifths 1 fifth 2 water glasses or 3-5 wine glasses 1 water glass or 1-2 wine glasses None

72

TABLE 1—Continued

Liquor

9% 4 pints or more 6% 3 pints

20% 2 pints 20% 1 pints 7% 8-10 shots or drinks 6% 5-7 shots or drinks 2% 3-4 shots or drinks 2% 1-2 shots or drinks 2 8% None

11. Behavioral Aspects of Drinking (during previous month)

Number of times drunk

15% None 23% 1-4 29% 5-10 33% More than 10

Longest period between drinks

59% 12 hours 41% Less than or more 12 hours

Longest period of continued drinking

14% Less than 32% 6-12 hours 54% More than 6 hours 12 hours

Niomber of days had a drink upon awakening

24% None 5% 1-4 24% 5-10 47% More than 10

Meals missed because of drinking

15% None 7% 1-4 14% 5-10 64% More than 10

Drinking alone or with others

1% None 24% Mostly with 35% Sometimes alone, others sometimes with

others

40% Mostly alone

Number of memory lapses or "blackouts"

31% None 25% 1-2 20% 3-5 24% More than 5

Number of times had the "shakes"

26% None 16% 1-2 17% 3-5 41% More than 5

73

TABLE 1—Continued

Number of nights had difficulty sleeping

30% None 5% 1-2 16% 3-5 49% More than 5

Niamber of quarrels with others while drinking

65% None 10% 1-2 12% 3-5 13% More than 5

Drinking while on job/during daily activities

39% No 61% Yes

Days of work missed/days of inactivity because of drinking

36% None 8% 1-2 9% 3-5 47% More than 5

Hypothesis 1

The mean and standard deviation for the normalization/

standardization sample of 123 alcoholics on the RAS were

respectively -5.04 and 22.81. A t test was conducted to

evaluate differences between this random sample of alcoholics

and a sample of 6 8 college students (normals) reported by

Rathus (19 73b). The mean and standard deviation for the

sample of college students were,respectively, .29 and 29.67.

The results of the analysis were not significant, t (118) =

1.36, £ > .05. The null hypothesis was accepted that there

is no significant difference between the assertiveness of

alcoholics and that of "normals" as expressed by scores

obtained on the Rathus Assertiveness Scale.

Additional analyses of the data were done to determine

if age, sex, number of admissions, number of days in

74

hospitals, and the number of years of education signifi­

cantly correlated with the Rathus Assertiveness scores in

the sample of 12 3 alcoholics. The results are reported

in Table 2.

It is evident from Table 2 that both the number of

days hospitalized and the number of years of education are

significantly and positively correlated with the total

scores on the Rathus Assertiveness Scale. Thus, subjects

tend to be more assertive the longer they stay in the state

hospital and also those subjects with higher educational

levels tend to be more assertive.

TABLE 2

CORRELATION OF RAS SCORES OF ALCOHOLICS WITH FIVE VARIABLES

Variables r df t

Age with Total Score .135 121 1.58

Sex with Total Score .100 121 1.11

Number of Admissions with Total Score .055 121 .60

Number of Days Hospitalized with Total Score .200 121 2.25*

Number of Years of Education with Total Score .209 121 2.35*

*£ < .05 level of significance.

75

Hypothesis 2

The second hypothesis was concerned with the impact of

the assertive treatments on the self-report measure (Rathus

Assertiveness Scale). The RAS means and standard deviation

for the groups are presented in Table 3.

TABLE 3

RATHUS PRE-POSTTEST MEANS AND STANDARD DEVIATIONS FOR TREATMENT AND CONTROL GROUPS

RAS Pretest RAS Posttest Group

Mean SD Mean SD

Assertive Training

Minimal Assertive Training

Control

-22.32 12.99

•21.94 12.27

-19.11 14.09

2.52 24.86

-13.41 26.09

-16.70 24.26

*N = 34 'for each group.

The results of the analysis of covariance for RAS

scores are reported in Table 4.

Duncan's Multiple Range Test was used to carry out

pairwise comparisons across groups. The Assertive Training

group scored significantly higher than the Minimal Assertive

Training and the Control groups (p < .01). Both the Minimal

Assertive Training and Control groups were statistically

similar to each other. The null hypothesis of no difference

between groups was therefore rejected. The data v/ere

76

supportive of significant changes for the Assertive Training

group on the RAS while the other groups remained essentially

unchanged.

TABLE 4

ANALYSIS OF COVARIANCE RESULTS OF RAS SCORES FOR TREATMENT AND CONTROL GROUPS

^ Sum of ^ j . Mean „ Source ^ df ^ ^ F Squares — Squares — .•••^.p^.-i •-. — • > — ^ ^ u..i.,i , — 1 ^ , „ • •>-•.. W f • ....^Lam^-..^.—••••• •^-WM.i.i... — .1 .1 .1.1 • • I.— mn^ .. ...•••••—• ^. „—.*»ii. .. ••• —,..•• ..i i. • • • — • w ••*•••«•'•• m... . i.* i i i - i w « . »

Group 9026.58 2 4513.29 9.588*

Covariate (Pretest Scores) 16195.16 1 16195.16 34.405*

Error 46130.59 98 470.72

Total 69522.07 101

*£ < .0001.

Hypothesis 3

The third hypothesis was concerned with the impact of

the assertive treatments on the Behavioral Assertiveness

Test. Since the BAT was composed of three elements, each

is reported separately.

Response Latency

Prior to computing the analysis of variance for

Response Latency, each subject's total latency score in

seconds was transformed by a logarithmic transformation

(x' '-^ log-,f x) to achieve greater normality. The means

77

and standard deviations of the transformed scores are

presented in Table 5.

TABLE 5

RESPONSE LATENCY MEANS AND STANDARD DEVIATIONS FOR TREATMENT AND CONTROL GROUPS

Group Mean SD

Assertive Training

Minimal Assertive Training

Control

1.2445

1.3347

1.4156

.1867

.2086

.2477

*N = 34 for each group.

The results of the analysis of variance for Response

Latency scores are reported in Table 6.

TABLE 6

ANALYSIS OF VARIANCE RESULTS OF RESPONSE LATENCY SCORES FOR TREATMENT AND CONTROL GROUPS

Source

Group

Error

Total

*£ < .006.

Sum of Squares

.498

4.618

5.116

d£

2

99

101

Mean Squares

.249

.046

F

5. 34*

78

Duncan's Multiple Range Test was used to carry out

pair-wise comparisons across groups. The Assertive Train­

ing group had a significantly shorter Response Latency than

the Control group (p < .01), but not significantly shorter

than the Minimal Assertive Training group. The means for

the Minimal Assertive Training and Control groups were not

statistically different from each other. The null hypothe­

sis of no difference between groups was rejected and the

data were supportive of the effectiveness of Assertive

Training in reducing the Response Latency of subjects on a

role-played test of assertiveness over that of a Control

group,

Response Duration

The means and standard deviations for Response Duration

are presented in Table 7.

TABLE 7

RESPONSE DUR7\TI0N MEANS AND STANDARD DEVIATIONS FOR TREATMENT AND CONTROL GROUPS IN SECONDS

Group Mean SD

Assertive Training

Minimal Assertive Training

Control

8 8 . 1 1

6 5 . 1 1

5 8 . 88

3 4 . 0 9

3 4 . 3 6

3 0 . 8 2

*N = 34 for each group.

79

The results of the analysis of variance for Response

Duration are reported in Table 8.

TABLE 8

ANALYSIS OF VARIANCE RESULTS OF RESPONSE DURATION SCORES FOR TREATxMENT AND CONTROL GROUPS

Source ,^^^ °^ df /^^^ F Squares — Squares —

Group 16122.59 2 8061.29 7.34*

Error 108704.59 99 1098.02

Total 124827.18 101

*£ < .001.

Duncan's Multiple Range Test was used to carry out * -

pair-wise comparisons across groups. The Assertive Training

group spoke significantly longer on the BAT than either the

Minimal Assertive Training or the Control groups (p < .01),

The null hypothesis of no difference between groups was

therefore rejected. The Minimal Assertive Training and

Control groups were essentially statistically similar to

each other. The results of the data were supportive of the

effectiveness of Assertive Training in increasing the quan­

tity of verbalization on a role-played test of assertiveness

80

Total Assertiveness Scores

Inter-rater reliability data for the Total Assertive­

ness Scores on the BAT are presented in Table 9,

TABLE 9

RATER'S MEANS, STANDARD DEVIATIONS AND INTER-RATER RELIABILITY OF BAT

Rater Mean SD Pearson-Product Moment

Correlations

3 3 . 9 4

3 3 . 4 8

1 1 . 3 0

1 1 , 2 3

Rater 1 with Rater 2 £ = ,9 34

Rater 2 with Rater 3 r = ,9 36

35,43 12,73 Rater 3 with Rater 1 r = ,9 43

It can be seen from the data in Table 9 that there is

a high degree of inter-rater reliability for the Total

Assertiveness scores on the BAT. The Total Assertiveness

Score means and standard deviations for the groups are

presented in Table 10.

TABLE 10

AVERAGE TOTAL ASSERTIVENESS MEANS AND STANDARD DEVIATIONS FOR TREATMENT AND CONTROL GROUPS

Group

Assertive Training

Minimal Assertive Training

Control

*N 34 for each group.

Mean SD

4 2 . 9 1

3 2 . 1 7

2 7 . 5 5

1 1 . 4 0

1 1 . 9 4

1 1 . 6 8

81

Analysis of variance was computed for the Mean

Total Assertiveness Scores on the BAT and the results are

reported in Table 11.

TABLE 11

ANALYSIS OF VARIANCE OF MEAN ASSERTIVENESS SCORES ON THE BAT FOR TREATMENT AND CONTROL GROUPS

Source / ^ °f df /^^^ F Squares — Squares —

Group 4219.19 2 2109.59 23.75*

Error 8800.05 99 88.88

Total 13019.25 101

*£ < .0000001.

Duncan's Multiple Range Test was used to carry out

pair-wise comparisons across groups. Each of the groups was

homogeneous with itself. The Assertive Training group

scored significantly more assertive on the BAT than either

the Minimal Assertive Training or Control groups (£ < .01).

In addition, the Minimal Assertive Training group scored

significantly higher on the BAT than the Control group

(£ < .05). The null hypothesis of no difference between

groups was therefore rejected.

82

Hypothesis 4

The fourth hypothesis was concerned with the impact of

assertive treatments on the "in-vivo" ratings of assertive­

ness (ABI). The means and standard deviations of the ABI

are presented in Table 12.

TABLE 12

ABI MEANS AND STANDARD DEVIATIONS FOR TREATMENT AND CONTROL GROUPS

Group* Mean S£

Assertive Training -1.58 2.33

Minimal Assertive Training -4.23 2.96

Control -4.23 4.66

*N = 34 for each group.

Due to the possibility that the ABI did not achieve

an interval level scale, the Kruskal-Wallis one-way analysis

of variance by ranks was utilized to test the differences

between groups. The results of the analysis were significant,

H = 17.87, £ < .001. The analysis of variance was followed

by Mann-Whitney U Tests to determine which groups were sta­

tistically different. The Assertive Training group was

rated significantly more assertive than either the Minimal

Assertive Training or the Control groups (£ < .01). The

Minimal Assertive Training and Control groups were not

83

significantly different from each other. The null hypothe­

sis of no difference between groups was rejected.

CHAPTER IV

DISCUSSION

Data relevant to the research hypotheses were presented

in Chapter III. The results obtained were totally congruent

with V7hat had been expected. Discussion of these results

and their implications are presented in two sections: an

investigation cf the data regarding the research hypotheses

is followed by a presentation of the implications for fur­

ther research.

Research Hypotheses — i , n i « i » f j « « w , I — ^ w — M .11 <•••••>• in II • •• — • ^ » w < w .11 M lattmi •»

Hypothesis 1

The null hypothesis of no significant differences in

assertiveness between alcoholics and "normals," as measured

by the RAS, V7as accepted. In accepting the null hypothesis,

several considerations need to be kept in mind. First, the

comparison sample of college students (Rathus, 1973b) used

in his study in considerably younger than the mean age of

47 years reported for the present sample of alcoholics. No

data are yet available regarding the relationship of age

and assertiveness in the general population. In addition,

the problems inherently associated ' - h using a relatively

small sample of college students as representative of the

"normal" population speak for themselves. Similarly, there

are external validity problem.s associated with utilizing a

84

85

sample of chronic state hospitalized alcoholics as represen­

tative of the total alcoholic population.

Assertiveness as a variable is probably normally dis­

tributed in state hospitalized alcoholics. It is likely

that alcoholics do not differ greatly from nonalcoholics in

their capacity to assert themselves and are able to stand

up for their rights without undue anxiety. Some do have

problems being too aggressive or hostile while others are,

conversely, too passive or unassertive. It is this latter

group which was the major focus of this research project.

The fact that state hospitalized alcoholics did not signifi­

cantly differ from a comparison group in terms of a self-

report measure of assertiveness does not diminish the

importance of this research. The significance of this

study lies in its movement away from a generalized treatment

orientation for alcoholism to a more focused approach which

matches treatment and therapist with client and.problem.

Several issues surrounding the adequacy of the Rathus

Assertiveness S-cale as a measurement of assertiveness need

to be briefly discussed. First, the RAS is a relatively

gross measure of a person's self-reported attitude toward

his ov7n assertiveness. The extent to which a person's score

may be influenced by social desirability, response set and

the demand characteristics of the setting in which the RAS

is administered is unclear. Each of these factors may

86

contribute to lowered validity. Finally, there exists som.e

degree of disparity between a client's perception or self-

reported attitude of assertiveness and actual behavior as

noted in differences between the RAS, BAT and ABI. Cer­

tainly, these factors restrict the generalizability of the

reported results.

Apparently, age, number of admissions,and sex did not

significantly affect the RAS scores in the standardization

sample of 12 3 alcoholics. The general expectation that

women tend to score more unassertively than men did not

materialize for the sample investigated. It is not sur­

prising that assertiveness is significantly and positively

correlated with length of treatment (number of days hospi­

talized) , since a large component of many treatment orienta-

tions is to encourage clients to be more expressive with

their feelings and to be less inhibited and more outspoken.

The effects of ongoing treatment probably account for the

correlation between RAS scores and number of days hospital­

ized. The finding that clients with higher educational

levels tend to be m.ore assertive is interesting. Certainly,

the verbal skills learned in school should make assertive­

ness somewhat easier for many individuals. An alternative

explanation may be that the people with higher levels of

education had less difficulty reading the RAS and conse­

quently were able to answer the items more assertively.

87

Furthermore, clients with more education would be more

familiar with testing situations. It should be noted that

both the number of years of education and the number of

days hospitalized each account for only 4% of the variance.

Hypothesis 2

The second hypothesis was concerned with the effective­

ness of assertive treatment on RAS scores. Although the

three groups were essentially similar prior to the start of

treatment, only the Assertive Training group scored signifi­

cantly higher at the time of posttesting. In the course of

10 hours of assertive training, this group averaged one

standard deviation gain on the RAS. The results obtained on

this outcome measure clearly show the superiority of the

Assertive Training group in modifying the self-reported

attitudes of unassertive subjects.

Although the Minimal Assertive Training group showed

some pre-posttest gain on the RAS, it is somewhat surprising

that the group did not gain more than it did. One of the

major reasons for including the Minimal Assertive Training

group in the research design was to evaluate the effective­

ness of essentially just telling someone to be more asser­

tive and the impact this would have on a person's attitude.

It could also be argued that the small change noted for the

Minimal Assertive Training group is due to the implicit

88

demand characteristics of the experimental paradigm. In

other words, the pretesting experience plus talking about

assertiveness suggest that clients in the Minimal group

should show some improvement on the posttest. Furthermore,

the results from the previous hypothesis related to a slight

increase in assertiveness as a function of length of time

in treatment could also account for the slight pre-posttest

changes exhibited by the Minimal and Control groups. Cer­

tainly the results obtained on this measure are totally

congruent with the results reported by Rathus (19 73a) in

which his placebo treatment group showed a similar slight

increase in pre-posttest scores.

Hypothesis 3

The overall results on the Behavioral Assertiveness

Test clearly showed the superiority of the Assertive Train­

ing group in modifying unassertiveness in alcoholics. The

results of each element of the BAT will be discussed

separately.

Response Latency. The data on Response Latency

shov7ed that the Assertive Training group had significantly

shorter reaction times to the 10 tape recorded situations

on the BAT than the Control group. Although the Control

and Minimal Assertive Training groups did not significantly

differ from each other, only the difference between Control

and Assertive Training groups reached a significant level

89

(p < .01). It is apparent that the practice and training

received by the Assertive Training group allowed them to

respond to the task more quickly. The fact that the

Assertive Training group showed less hesitation in their

responses suggests that their replies on the BAT were

more accessible and readily available for use in situations

requiring assertive responses.

Response Duration. The subjects in the Assertive

Training group talked significantly longer than either the

subjects in the Minimal or Control groups. The data for

Response Duration are extremely supportive of the effective­

ness of Assertive Training in increasing the quantity of

verbalization on a role-played situational test of asser-

tiveness. The Assertive Training group averaged nearly 30

seconds more verbalization than the Control group. It

appears that the behavior rehearsal, modeling and coaching

which the Assertive Training group received significantly

affected the quantity of output on the BAT. Most situations

which require an assertive response cannot be adequately

handled by one- or two-v7ord responses, and persistence is

generally required in order to get one's point across.

The subjects in the Assertive Training group clearly demon­

strated their superiority in this area.

Total Assertiveness Scores. The superiority of the

Assertive Training group was again demonstrated, as the

90

results showed that the Assertive Training group was rated

significantly more assertive than either the Minimal or

Control group. In addition, the Minimal Assertive Training

group scored significantly higher than the Control group.

Although both treatment groups gained over the Control

group, the magnitude of change for the Assertive Training

group was dramatic.

It is interesting and somev7hat surprising that the

Minimal Assertive Training gained significantly on this

behavioral measure while not showing such changes on the

self-reported attitudinal measure (PJ S). The significant

increase for the Minimal Assertive Training group over the

Control group on the BAT and not on the RAS might be ac­

counted for in terms of the characteristics of the two

measures. The RAS is a self-report attitudinal measure

and may be less sensitive to change than behavioral mea­

sures. Similar findings were also reported by McFall and

Marston (1970) and Hersen et al. (1973). There appears to

be some evidence for an "attitudinal lag" on self-report

measures in comparison to more behaviorally oriented mea­

sures. Apparently, significant changes occurred for this

group with only 2 hours of treatment. The effectiveness of

discussion, didactic presentation and common sense explana­

tions of why one behaves unassertively can be important in

effecting change. The positive change for the Minimal group

91

is consistent with the results reported by Friedman (19 71),

m which marked behavioral changes were obtained with only

10 minutes of treatment.

The use of the EAT as an instrument for measuring

assertiveness in subjects gained validity from this research.

The high inter-rater reliabilities showed that simulated

real-life situations are consistently and reliably obtained

across different raters. It was readily apparent from lis­

tening to the tapes that the subjects became very emotionally

involved with the task. From a clinical point of view, the

BAT revealed much more information about the person than his

degree of assertiveness. It also highlighted other clinical

problems. • The manner in which a person dealt with signifi­

cant others in his environment can become much clearer in

listening to 10 minutes of the BAT than several hours of

interviews or self-report. Male-female, marriage, and

other interpersonal problems as well as the way in which

a person deals with them stand out clearly in listening

clinically to many of the tapes. The BAT takes a relatively

short amount of time to administer (10 m.inutes) and can pro­

vide a wealth of clinical information. After this research

project got underway, it became apparent that the use of

the BAT as a screening instrument would have been more

preferable than using the RAS.

92

The BAT has several features which make its use as a

screening instrument potentially more desirable than the

RAS. First, the client does not have to be able to read.

Secondly, the BAT provides more clinical information and

is usually well accepted by clients. Thirdly, the BAT

appears to tap actual behavior rather than self-reported

attitudes. The potential use of the BAT and similar

devices in treatment programs as a valuable diagnostic tool

cannot be underscored.

Hypothesis 4

The results of the Assertive Behavior Index clearly

demonstrated the superiority of the Assertive Training

group over either the Minimal Assertive Training or Control

groups. The data shov7ed that subjects receiving assertive

training behaved more assertively on a number of nonverbal

and verbal variables, such as eye contact, body language,

voice loudness, etc. While subjects in the Minimal Asser­

tive Training group showed a significant increase in asser­

tiveness over the Control group on the Total Assertiveness

Score of the BAT, they did not demonstrate this superiority

in an unobtrusive rating situation. In fact, the ABI means

for the Minimal and Control groups were identical. This

suggests that what the Minimal Assertive Training group

learned was a modification of the content of their responses

93

and not what Serber (1972) called a "command of style." He

defined lack of style as the inability to master appropriate

nonverbal, as well as verbal, components of behavior. Thus,

a person may say the right thing but come across in an

unassertive manner.

The feedback modeling, coaching, role playing and

behavior rehearsal received by the Assertive Training group

appears to have made a significant difference in the way

clients came across during the interviews in which they

were rated. The fact that this group of subjects was rated

more assertive has important significance. The 15-minute

interview was a real-life situation and the clients had no

idea that they were being rated. The fact that the ABI was

a nonobtrusive measure suggests that the cognitive or atti­

tudinal expectancy effect which could have been operating

in the RAS and BAT would be greatly minimized. In other

words, the clients were not aware that assertiveness was

being evaluated and there was no reason to expect the client

to respond according to his conception of how an assertive

person behaves. Although the time lag between the assertive

training and the interviews was relatively short (several

days), the data are suggestive of transfer of training in

assertive skills.

It should be mentioned that although the ABI raters

were not directly connected with the research study, the

94

possibility of "experimenter bias" cannot be ruled out.

All the raters were full-time staff members on the Alcohol­

ism Units and took part in the staffing of clients. It was

not always possible for the rater to rem.ain unaware of

which group a subject was in. The degree to which the

raters knew the client's treatment group and the extent

of bias that this knowledge may have had on the obtained

results cannot be adequately determined.

Theoretical Implications

In contrast to psychoanalysis and other more traditional

psychotherapies which focus on subjective behavior, i.e.,

thoughts and feelings, in order to effect change indirectly

in the client's overt behavior, assertive training focuses

directly on the client's overt behavior. The results ob­

tained in this study are quite supportive of behavior therapy

and tend to challenge some of the basic assumptions of tradi­

tional psychotherapies. In applying the assertive training

techniques used in this research, no attempt was made to

explore the genesis or dynamics of unassertiveness. In

addition, no effort was made to delve into the childhood

memories, unconscious processes, or client dreams. In fact,

reorganization of the client's personality was not a major

objective of the assertiveness training.

In a departure from "nondirective" psychotherapies,

in which a client is helped to understand the way he feels

95

and, consequently, is free to adopt more appropriate

behavior, assertive training follows through by providing

specific instruction, modeling, and behavior rehearsal.

Not only is a client helped to understand his situation,

but he is provided with specific techniques for the

development of alternative behaviors.

Behavior therapists conceive of psychiatric syndromes

as collections of faulty habits which can be best modified

if they become the direct focus of treatment. The role of

a behavior therapist is to help the client identify the

presenting problem in terms of observable behavior. After

a specific behavior pattern has been determined and the

conditions under which the maladaptive behavior is main­

tained, the behavior therapist selects the potentially

most suitable and efficient methods developed and validated

in the experimental laboratory (McFall et al. , 1973). As

demonstrated in this study, this approach can effect signif­

icant change within a relatively short period of time.

The assertive training techniques utilized in this

research project have some characteristics in common with

other types of behavior therapy groups. First, the asser­

tive training group was organized on the basis that a common

modification technique was applicable to all the members.

Secondly, the group V7as highly structured and goal-oriented

with the primary objective being the amelioration of

96

unadaptive anxiety responses in social and interpersonal

situations. Thus, the goal of the assertive treatment was

similar to what Wolpe and Lazarus discussed (1966) as the

goal for many relaxation and systematic desensitization

groups. The major differences between these approaches

lie in the fact that assertive training techniques stress

an active orientation and the ability to communicate satis­

factorily with others.

The assertive training procedures used in this study

also give added validity to the modeling work of Bandura

(1971) and to the skill deficit theories of McFall and his

associates. McFall views unassertiveness as a behavioral

deficit or lack of skill in otherwise normal people who are

over-polite or too civilized. Frequently, an individual is

quite able to engage in significant interchanges without

apparent anxiety, but consistently exhibits a woeful lack

of skill, resulting in failure to satisfy needs. Since the

major thrust of this study's assertive treatment was toward

skill building via coaching, modeling, and behavior rehears­

al, the significant results obtained tend to reinforce the

theoretical notions of McFall. Actually, the results of

this study extend the work of McFall and his associates in

the behavioral techniques which they studied, since this

experiment used a more chronic, nonstudent population.

Furthermore, McFall only used semi-automated experimental

97

analogues of more fully developed clinical procedures,

while the assertive treatment in the present study was more

comprehensive in its approach.

The findings of this study are consistent with the

"tension reduction" model of alcoholism (Cappell & Herman,

19 72). Drinking alcohol may be one way that stress-induced

tension can be reduced. V7olpe (1958) views assertive be­

havior as being physiologically antagonistic to anxiety and

tension. In the acting out of assertion, there is an

augmentation of the emotion at its core and the strength

this emotion thus acquired may enable it to inhibit the

concurrently evoked anxiety. Assertiveness may function

to reduce the tension build-up in stressful social situations

and consequently lower the need to use alcohol for tension

reduction. Certainly, the results of this study are con­

sistent with the previously reported work of Miller,et al.

(1974) and Martorano (1974) in which interpersonal variables

may represent significant factors in determining the alco­

holic's drinking as a function of stressful situations.

Programmatic and Research Implications

Recent developments in the application of learning

theory to therapy have been useful in designing behaviorally

oriented ward treatment programs, such as token economies

for the mentally ill and retarded. However, to date there

98

has been relatively little attempt to apply other techniques

of behavior therapy to state hospitalized alcoholics. A

partial explanation of this fact is that behavior modifi­

cation techniques are frequently associated in the public

mind with the mechanistic human engineering of Orwell's

19_84_. Behaviorally oriented programs such as token econo­

mies and adversive conditioning tend to reinforce the

stereotype of emphasis on external control. However,

clinical techniques developed and used by Wolpe, Lazarus,

and others, such as relaxation, desensitization and asser­

tive training, combined with instruction about the affects

of rewards and punishments upon behavior, can greatly

increase clients' control over their internal and external

environments.

Seldom have attempts been made concurrently to train

socially acceptable behaviors in the place of heavy drinking.

This study represents a significant step forward in develop­

ing a behavior modification program with an emphasis on

self-control. It has been shown in this study that the

assertive training method is more effective than a minimal

and a no-treatment control in inducing more assertive be­

havior in a sample of chronic state hospitalized alcoholics.

It appears that encouraging a client to demand his rights,

to express his feelings, and to be generally outgoing is

certainly a step in the direction of fuller functioning.

99

The repertoire of coping techniques mastered by the asser­

tive training groups should assist them in securing greater

satisfaction and gratification when they return home after

their hospital stay. Given more rewarding and effective

means of dealing with their environment, alcoholics should

be better able to maintain their sobriety.

The positive results obtained in this study also have

implications for the future development of alcoholism treat­

ment programs. First, the results suggest that if one type

of behavioral program (assertive training) can be effective

in the amelioration of a specific problem area, then other

types of specialized treatment techniques can be developed

and applied. If future alcoholism programs are to be truly

effective for a large proportion of clients, then other

specialized techniques must be researched and adopted. The

need for multidisciplinary staff utilizing a broad range of

proven treatment modalities appears to be the only effec­

tive way to treat chronic alcoholism.

Secondly, as more and more alcoholism programs apply

for alcoholism accreditation from the Joint Commission on

Accreditation of Hospitals, there exists a need for better

assessment techniques related to specific client problem

areas. The accreditation standards call for documentation

that individualized treatment plans are based on the

diagnostic assessment of a client's phychological-social

100

needs. The RAS, BAT, and ABI utilized in this study can

provide such documentation and justification. In addition,

the RAS, BAT, and ABI could be used in program evaluation,

also a requirement in the accreditation standards.

Finally, one aspect of the assertive training which

was not revealed in the reported statistics was the manner

in which the patients reacted to the Assertive Training

group. The clients frequently stated that the Assertive

Training group was their favorite group on the unit.

Clients responded well to the training and could readily

identify with the need for assertive training and understand

the relevance of the group to their lives. The quality of

the Assertive Training group was markedly different from

most of the other groups on the units. Clients were

interested and excited about being in the group. There

was a tremendous amount of "esprit de corps" among members

and the quantity and quality of interpersonal relationships

in the group remained consistently high. Members would

frequently volunteer to role-play situations and the amount

of "risking" behavior on the part of clients was, quite

frankly, extremely atypical for this chronic population.

It is noteworthy to mention that when the members of the

Assertive Training group at one hospital learned that the

group leader would be terminated at the conclusion of this

research grant, they started a petition for the hospital

to keep her on permc-inent Jy.

101

The major concern of this investigation was not with

the methodology of assertive training but rather in demon­

strating that a relatively short amount of training could

have significant impact on the assertiveness of passive

alcoholics. The comparisons between the Control group and

Assertive Training group were by far the more important in

this study, although if this study were to be replicated,

it might be appropriate to equate the number of hours of

treatment the Minimal and Assertive groups received. It

might be argued by some that the reason the Assertive

Training group did so much better than the Minimal Asser­

tive Training group was that its subjects received eight

additional hours of treatment. That possibility certainly

exists, but what is more important is the fact that the

Assertive Training group far exceeded the Control group.

Also related to this issue was the fact that the Minimal

Assertive Training group received no skill training. For

example, they had no opportunity to practice specific

assertive responses. Since the work of McFall and his

associates (1970, 1971, 1973) appears to indicate that

behavior rehearsal is the most powerful component of asser­

tive training, the lack of practice rather than the length

of treatment the Minimal Assertive Training group received

might be the important factor.

Several research questions related to the usefulness

of assertive training for alcoholic clients remain to be

102

investigated. This study was limited to demonstrating that

unassertive behavior can adequately be modified through

the component techniques of assertive training. It did

not address itself to the issue of transfer of training

once the client leaves the state hospital and returns to

the community. Do the assertive skills hold up and for

how long? What is the impact of an appropriately assertive

person returning to a family situation in which he had

previously been quite passive? These questions and others

need to be researched in order to more fully evaluate the

effectiveness of assertive training.

The most important aspect of assertive training with

alcoholics is undoubtedly the impact such training has on

the frequency and quantity of alcohol consumed. Does

assertive training lead to a decrease in the amount of

alcohol consumed? Does it help alcoholic clients maintain

their sobriety? This experimenter is hopeful that a 1-year

follow-up will help answer some of these questions. As of

now, they remain unanswered.

Another question regarding assertive training relates

to the types of clinical problems for V7hich these techniques

are appropriate. The focus of this research project was on

the amelioration of unassertive behavior in alcoholic clients

How can aggressive or hostile clients be helped? Are asser­

tive techniques useful in changing aggressive/hostile

103

b e h a v i o r i n t o more a p p r o p r i a t e a s s e r t i v e behav io r? There

d e f i n i t e l y e x i s t s a need fo r f u tu r e r e s e a r c h i n t o t h e s e

a r e a s . The f i n d i n g s of t he p r e s e n t s tudy a re indeed h i g h l y

e n c o u r a g i n g , b u t only on the b a s i s of con t inued i n v e s t i ­

g a t i o n and outcome s t u d i e s can one expec t t o develop an

e f f e c t i v e and e f f i c i e n t t r e a t m e n t for a l coho l i sm.

CHAPTER V

SUMMARY AND CONCLUSIONS

One of the major objectives of this study was to

delineate the parameters of unassertiveness in alcoholics

as compared to normals. A random sample of 12 3 state hos­

pitalized alcoholics were chosen as a standardization/

normalization group for the Rathus Assertiveness Scale

(RAS). The RAS is a self-report, paper and pencil measure

of a person's perceived assertiveness. This sample of 123

alcoholics was then compared to a group of college students

(normals). The results of the comparison revealed no

significant differences between assertiveness of state

hospitalized alcoholics and that of normals as measured

by scores obtained on the RAS.

Additional analyses of the data were done to determine

if-'age, sex, number of admissions to state hospital, number

of days in state hospital, and the number of years of edu­

cation significantly correlated with the RAS scores in

the sample of alcoholics. The results showed that alco­

holic clients tended to become more assertive the longer

they stayed in the state hospital and also that those

clients with higher educational levels tended to be more

assertive.

The second major objective of the study was to demon­

strate that assertive training can be an effective technique

104

105

in modifying unassertive attitudes and behavior in alcoholic

clients. The subjects for this part of the study were a

separate group of 102 chronic state hospitalized alcoholics.

An equal number of subjects were randomly assigned into a

Control group. Minimal Assertive Training group and an

Assertive Training group based upon a subject scoring in

the lower 5 0th percentile on the RAS. The Control group

received the regular state hospital treatment program. The

Minimal Assertive Training group, in addition to the regu­

lar state hospital program received two hours of didactic

presentation and group discussion on how each person could

behave more assertively. The subjects in the Assertive

Training group, in addition to the regular state hospital

program, received 10 hours of assertive training, utilizing

all the component techniques of assertion training such as

modeling, coaching, role playing, instructions, behavior

rehearsal and homework assignments. Clients in both the

Minimal and Assertive training groups were strongly en­

couraged to begin behaving assertively. At the end of

two weeks, all subjects were posttested with the RAS. In

addition, clients were also posttested with a tape-recorded

situational test of assertive behavior and an unobtrusive

rating scale of assertiveness.

The results of the research shov7ed significant

pre-posttest changes for the Assertive Training group on

106

the RAS, but not for the Control or Minimal Assertive

Training groups. On the tape-recorded situational test of

assertive behavior, between group differences revealed a

similar superiority by the Assertive Training group over

the Control group in terms of response latency, response

duration, and ratings of assertiveness. Clients in the

Assertive Training group spoke more rapidly, spoke longer,

and responded more assertively to the role-played threaten­

ing or irritating social and interpersonal situations they

encountered on the tape recorded task. In addition, the

Minimal Assertive Training group significantly improved

their assertiveness over the control group in their ratings

of assertiveness. Their improvement, however, was much

less dramatic than the Assertive Training group.

The final outcome measure used in the study consisted

of a rating instrument which attempted to measure a number

of verbal and nonverbal components of assertiveness, such

as eye contact, body posture, facial expressions, loudness

of voice, and voice rhythm. Each client V7as interviewed

regarding plans for returning home after discharge from the

hospital. The staff ratings of assertiveness showed signif­

icant improvement for the Assertive Training group over

either the Minimal or Control groups. The overall results

of the study clearly demonstrated that group assertive

training can be an effective therapeutic technique for

rx;)difying unassertive attitudes and behavior in alcoholics.

107

Although this study did not directly address itself

to the issue of tremsfer of assertive skills once a client

leaves the state hospital and the effect such skill train­

ing has on future alcohol consumption, the results are

extremely suggestive of the potential usefulness of asser­

tion training in the treatment of alcoholism. It is hoped

that by providing unassertive alcoholics with the social

and interpersonal skills necessary to cope more effectively

with their environment that they will have less of a need

to resort to self-anesthetization and escape from everyday

life experiences. If alcoholics are given alternative

behaviors to drinking as a vehicle for emotional expression,

they might be able to decrease their drinking and to lead a

more comfortable, rewarding life.

REFERENCES

Alberti; R., & Emmons, M. Your perfect right. San Luis Obispo: Impact Books, 1970". ' ''

Bandura, A. Principles of behavior modification. New York: Holt, Rmehart, and Winston, 19 69.

Bandura, A. Psychotherapy based upon modeling principles. In A. E. Bergin & S. L. Garfield (Eds.), The handbook o^ psychotherapy and behavior change. New York:"~JoFn Wiley and Sons, 19 71.

Bandura, A., Blanchard, E. B., & Ritter, B. The relative efficacy of desensitization and mtodeling approaches for inducing behavioral, affective and attitudinal changes. Journal of Personality and Social Psycholoqy, 19 69, 12, TTT^lT. "' • — ' ^~

Bates, H. D. , &• Zimjnerman, S. F. Toward the development of a screening scale for assertive training. Psychologi­cal Reports, 1971, 2_8, 99-107. ^

Bergin, A. E. The evaluation of therapeutic outcomes. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of Psy­chotherapy and behavior change. New York: John Wiley and Sons, 19 71. '' "

Blake, B. G. The application of behavior therapy to the treatment of alcoholism. Behavior Research and Therapy, 1965, 3, 75-85.

Blake, B. G. A follow-up of alcoholics treated by behavior therapy. Behavior Research and Therapy, 196 7, 5, 89-94. ' " """

Burtle, v., V7hitlock, D. , & Franks, V. Modification of low self-esteem in women alcoholics: A behavior treatment approach. Psychotherapy: Theory, Research, and Prac­tice, 1974, _11, 36-40.

Cappell, H. , & Herman, P. Alcohol ai J! tension reduction. Quarterly Journal of Studies on Alcohol, 19 72, 33, 33-64.

Catanzaro, R. J. Alcoholism: The total treatment approach. Springfield, 111.: Charles C ThoTiia's, 1968.

108

109

Chafetz, M. E. Alcohol and health: New knowledge. (USDHEW), preprint edition, June, 1974, NIAAA, Stock No. 1724-00399.

Conger, J. Alcoholism: Theory, problem, and challenge. II._ Reinforcement theory and the dynamics of alco­holism. Quarterly Journal of Studies on Alcohol, 1956, 17_, 2^6-36T, ' "

Corsini, R. Roleplaying in psychotherapy: A manual. Chicago: Aldine, 1966.

Doering, M., Hamlin, R., Everstine, L., Eigenbrode, C., Chambers, G., Wolpin, M., & Lackner, F. The use of training to increase intensity of angry verbalization. Psychological Monographs, 1962, 7£v37, V7hole No. 556).

Edwards, N. B. Case conference: Assertive training in a case of homosexual pedophillia. Journal of Behavior Therapy and Experimental Psychiatry, 1972, 2^ 55-63.

Eisler, R. M. , Miller, P. M., & Hersen, M. Components of assertive behavior. Journal of Clinical Psychology, 1973, 2^(3), 295-299. ~" "

Eisler, R. , Hersen, M. , & Miller, P. Effects of modeling on components of assertive behavior. Journal of Behavior Therapy and Experimental Psychiatry^ r9"73, £, 1-6. "" "^

Eisler, R. M., Miller, P. M., Hersen, M., & Alford, H. Effects of assertive training on marital interaction. Archives of General Psychiatry, 1974, 3£' 643-649.

Eisler, R. M., & Hersen, M. Behavioral techniques in family oriented crisis intervention. Archives of General Psy­chiatry, 1973, 2 ./ 111-116.

Ellis, A. Reason and emotion in psychotherapy. New York: Lyle Stuart, 1962.

Elms, A. Roleplaying, reward, and attitude change. New York: Van Norstrand, 1969.

Fensterheim, H. Behavior therapy: Assertive training in groups. In C. J. Sager & H. S. Kaplan (Eds.), Progress in group and family therapy. New York: Brunner/Mazel, TTTT.

110

Friedman, P. H. The effects of modeling and roleplaying on assertive behavior. Unpublished doctoral thesis. University of Wisconsin, 196 8.

Friedman, P. H. The effects of modeling and roleplaying on assertive behavior. In C. M. Franks (Ed.), Advances in behavior therapy. New York: Academic Press", 19 71.

Galassi, J., DeLo, J., Galassi, M., & Bastien, S. The college self-expression scale: A measure of assertive­ness. Behavior Therapy, 1974, £, 165-171.

Gross, W. F., & Carpenter, L. L. Alcoholic personality: Reality or fiction? Psychological Report, 1971, 28, 375-378. "

Guilford, J. P. Fundamental statistics in psychology and education. New York: McGraw-Hill, 1965.

Hedquist, F. J. , & V7einhold, B. K. Behavioral group counsel­ing with socially anxious and unassertive college students. Journal of Counseling Psychology, 19 70, r7, 237-242.

Hersen, M., Eisler, R. M., & Miller, P. M, Development of assertive responses: Clinical measurement and research considerations. Behavior Research and Therapy, 19 73, Ij., 505-521.

Hersen, M., Eisler, R., Miller, P., Johnson, M., & Pinkston, S. Effects of practice, instructions, and modeling on components of assertive behavior. Behavior Research and Therapy, 1973, 1_1, 443-451.

Kanfer, F., & Phillips, J. Learning foundations of behavior therapy. New York: John Wiley and~Sons, 19 70"

Kirk, R. E. Experimental design: Procedures for the be­havioral sciences. Belmont, Calif,: Brooks/Cole, 1968.

Kissin, B., Platz, A., & Su, W. H. Selective factors in treatment choice and outcome in alcoholics. In N. K. Mello and J. H. Mendelson (Eds.), Recent advances in studies of alcoholism. Washington, D.C.: U.S. Govern-inent Printing Office, 19 72.

Lang, P. J. The mechanics of desensitization and the labo­ratory study of human fear. In C. M. Franks (Ed,), Behavior therapy: Appraisal and status. New York: McGraw-Ki i r ; l9"69.

Ill

Lazarus, A. A. Behavior rehearsal vs. non-directive therapy vs. advice in effecting behavior change. Behavior Re­search and Therapy, 1966, £, 209-212.

Lazarus, A. A. Behavior therapy and beyond. New York: McGraw-Hill, 1971. ~~~

Lewinsonn, P. M. , Weinstein, M. S., & Alper, T. A. Behav­ioral approach to the group treatment of depressed persons: A methodological contribution. Journal of Clinical Psychology, 1970, 2^^, 525-532.

Lisansky, E. Clinical research in alcoholism and the use of psychological tests: A reevaluation. In R. Fox (Ed.), Alcoholism: Behavioral research and thera­peutic approaches. New York: Springer Publishing Co., 1967, 3-15.

Lomont, J., Gilner, F., Specter, N., & Skinner, K. Group assertion training and group insight therapies. Psy­chological Reports, 1969, Z5, 463-470.

Martinson, W. D., & Zerface, J. P. Comparison of individual counseling and a social program with nondaters. Journal of Counseling Psychology, 1970, Yl, 36-40.

Martorano, R. Mood and social perception in four alcoholics: Effects of drinking and assertion training. Quarterly Journal of Studies on Alcohol, 1974, 3_5.' 445-457.

McFall, R. M., & Lillesand, D. V, B. Behavioral rehearsal with modeling and coaching in assertive training: Assessment and training stimuli. (Mimeographed) University of Wisconsin, 1970.

McFall, R. M., & Lillesand, D. V. B. Behavior rehearsal with modeling and coaching in assertion training. Journal of Abnormal Psychology, 1971, 7_7/ 313-323.

McFall, R. M., & Marston, A. R. An experimental investiga­tion of behavior rehearsal in assertive training. Journal of Abnormal Psychology, 1970, 76_, 295-303.

McFall, R. M., & Twentyman, C. T. Four experiments on the relative contributions of rehearsal, modeling, and coaching to assertion training. Journal of Abnormal Psychology, 1973, 8^, 199-218.

112 McClelland, D. C , Davis, W. N., Kaline, R. , & Wanner, E.

The drinking man. New York: The Free Press, 1972.

Meichenbaum, D. Cognitive factors in behavior modification: Modifying what clients say to themselves. Paper pre­sented at Association for Advancement of Behavior Therapy. Washington, D.C., 1971.

Miller, A. G. Role playing: An alternative to deception? A review of the evidence. American Psychologist, 1972, 22, 623-636. ^

Miller, P. M. , Hersen, M., Eisler, R. M., & Hilsman, G. Effects of social stress on operant drinking of alco­holics and social drinkers. Behavior Research and Therapy, 1974, L2, 67-72. '

Moreno, J. Psychodrama. Vol. I. New York: Beacon House, 1946.

Neuman, D. Using assertive training. In J. D. Krumboltz & E. C. Thorenson (Eds,), Behavioral counseling: Cases and techniques. New York: Holt, Rinehart, and Winston, 1969, 433-441.

Patterson, R. L. Time-out and assertive training for a dependent child. Behavior Therapy, 1972, 2/ 466-468,

Piaget, G., & Lazarus, A. A. The use of rehearsal-desensitization. Psychotherapy, Research and Practice. 1969, 6_r 264-266.

Rathus, S. A. An experimental investigation of assertive training in a group setting. Journal of Behavior Therapy and Experimental Psychiatry, 1972, 2/ 81-86.

Rathus, S. A. Investigation of assertive behavior through videotape-mediated assertive models and directed prac­tice. Behavior Research and Therapy, 1973, 12' 57-65. (a)

Rathus, S. A. A 30-item schedule for assessing assertive behavior. Behavior Therapy, 1973, A_, 398-406. (b)

Rehm, L. P., & Marston, A. R. Reduction of social anxiety through modification of self-reinforcement: An insti­gation therapy technique. Journal of Consulting and Clinical Psychology, 1968, 3Tr'565-574.

113

Salter, A. Conditioned reflex therapy. New York: Capri­corn Books, 1949, 1961.

Serber, M.^ Teaching the nonverbal components of assertive training. Journal of Behavior Therapy and Experimental Psychiatry, 1972, 2/ 179-183.

Smith, M. S. When I say no, I feel guilty. New York: The Dial Press, 1975. ~"

Spence, J. Behavior therapy and the treatment of anxiety: The role of instrumental response. Paper given at American Psychological Association, Washington, D.C., 1971.

Syme, L. Personality characteristics and the alcoholic: A critique of current studies. Quarterly Journal of Studies on Alcohol, 1957, 12' 288-302.

Strum, I. E. Implications of role playing methodology for clinical procedure. Behavior Therapy, 1971, 2^, 88-96.

Strum, I. E. Behavioristic aspects of psychodrama. Group Psychotherapy, 1965, 12, 50-64.

Sutherland, E. H. , Schroeder, H. G., & Tordella, C. L. Personality traits and the alcoholics; a critique of existing studies. Quarterly Journal of Studies on Alcohol, 1950, 11./ 547-561.

Veldman, D. J. Fortran programming for the behavioral sciences. New York: Holt, Rinehart, and Winston,1967 .

Vogel-Sprott, M. Alcoholism a learned behavior: Some hypotheses and research. In R. Fox (Ed.), Alcoholism: Behavioral research and therapeutic approaches. New York: Springer Publishing Co., 1967, 46-54.

Wagner, M. K. Comparative effectiveness of behavioral rehearsal and verbal reinforcement for effecting anger expressiveness. Psychological Reports, 1968, 22, 1079-1080. (a)

Wagner, M. K. Reinforcement of the expression of anger through role playing. Behavior Research and Therapy, 1968, 6, 91-96. (b)

114

Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. Unobtrusive measures: Non-reactive research in the social science"sT Chicago: Rand McNally and Company, 1966.

Weinman, B., Selbert, P., Wallace, M., & Post, M. Inducing assertive behavior in chronic schizophrenics. Journal of Consulting and Clinical Psychology, 1972, 2i' 246-252.

Weiskott, G. Assertiveness, territoriality, and personal space behavior as a factor of group assertive training with a college population. Unpublished doctoral dis­sertation, The University of Texas at Austin, 1975.

Wolpe, J. Psychotherapy by reciprocal inhibition. Stanford: Stanford University Press, 1958,

Wolpe, J. The practice of behavior therapy. New York: Pergammon Press, 1969.

Wolpe, J. The instigation of assertive behavior: Tran­scripts from two cases. Journal of Behavior Therapy and Experimental Psychiatry, 1970, 1_, 145-151.

Wolpe, J. Neurotic depression: Experimental analog, clini­cal syndromes, and treatment. American Journal of Psy­chotherapy, 1971, 25.' 362-368.

Wolpe, J,, & Lazarus, A, A. .Behavior therapy techniques. Oxford: Pergammon Press, 1966,

Yalom, I. D. The theory and practice of group psychotherapy. New York: Basic Books, 1970.

Yates, A. J. Behavior therapy. New York: Wiley, 1970.

APPENDIX

A. Demographic Drinking Data Form

B. Rathus Assertiveness Scale

C. Behavioral Assertiveness Test

D. Scoring Sheet for Behavior Assertiveness Test

E. Assertive Behavior Index

115

116

APPENDIX A: DEMOGPAPHIC DRINKING DATA FORM

NAHE last first m.i

ADDRESS number and street

PHONE

city

/ /

zip county

BIRTHDATE month day year

SEX

EDUCATION

MALE FEMALE

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 (circle one)

1. What is client's present marital status Never married Married living with spouse

Separated Widowed Divorced/xVnnulled

2.

3.

4.

Occupation

How many times has cia-ent been hospitalized for alcohol-related problems during the past 30 days? Number

Hov7 many days has client been in the hospital? Number

5. DRINKING STATUS

A. How long has it been since client's last drink? 1-6 days 1-2 weeks

" 3-4 weeks 5-8 weeks 9-12 \.eeks 3-4 months 5-6 months Over 6 months

B. HOW many days did client drink during the past 30 days?

117

6. DRINKING QUANTITY AND FREQUENCY

A. About how often did client drink during the past 30 days:

BEER

Every day 5-6 days a week 3-4 days a week 1-2 days a week Less often than weekly None

WINE

Every day 5-6 days a week 3-4 days a v/eek 1-2 days a week Less often than weekly None

LIQUOR

Every day 5-6 days a week 3-4 days a week 2-2 days a week Less often than weekly None

(NOTE TO INTERVIEWER: Be sure to complete quantity for each type of drink.)

B. About how much did client drink in a typical day in which he drank?

BEER 6 quarts or more '5 quarts "4 quarts "3 quarts "1-2 quarts '1-3 glasses None

V7INE 5 fifths or more "3-4 fifths '2 fifths '1 fifth '2 water glasses or '3-5 wine glasses 1 water glass or '1-2 wine glasses None

LIQUOR 4 pints or more '3 pints '2 pints "l pint '8-10 shots or drinks '5-7 shots or drinks '3-4 shots or drinks '1-2 shots or drinks 'None

118

NOTE TO INTERVIEWER:

1 quart = three 11 oz. bottles (cans) or four 8 oz. glasses

1 fifth is a standard size bottle and is equal to about three 8 oz. water or six 4 oz. wine glasses. There are 5 fifths in a gallon or 2 1/2 fifths to a half gallon.

1 pint. = 16 oz. or just over ten 1 1/2 oz. shots. There are 2 pints in 1 quart and a little over 1 1/2 pints in a fifth.

7. BEHAVIORAL ASPECTS OF DRINKING

A. Number of times drunk

None 1-4 ^ 5-10 More than 10

E. Longest period between drinks

12 hours or more Less than 12 hours

C. Longest period of continued drinking

Less than 6 hours 6-12 hours More

than 12 hours

D. Number of days had a drink upon awakening

None 1-4 5-10 More than 10

E. Meals missed because of drinking

None 1-4 5-10 _More than 10

F. Drinking alone or with others

None Mostly with others Sometimes

alone, sometimes with other- _Mostly alone

G. Number of mem.ory lapses or "blackouts"

None 1-2 3-5 More than 5

119

H. Number of times had the "shakes"

l^one 1-2 3-5 More than 5

I. Number of nights had difficulty sleeping

None 1-2 3-5 More than 5

J. Number of quarrels with others while drinking

None 1-2 3-5 More than 5

K. Drinking while on job/during daily activities

No Yes

L. Days of work missed/days of inactivity because of drinking

None 1-2 3-5 More than 5

120

APPENDIX B: RATHUS ASSERTIVENESS SCALE

Age: Sex:

Approximately how many days have you been in the state hospital? Approximately how many times have you been admitted to the state hospital? Circle the last school grade you completed: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Directions: Indicate how characteristic or descriptive each of the following statements is of you by using the code given below. Please answer each question as honestly as you can,

a +3 very characteristic of me, extremely descriptive +2 rather characteristic of me, quite descriptive +1 somewhat characteristic of me, slightly descriptive -1 somewhat uncharacteristic of me, slightly non-

descriptive -2 rather uncharacteristic of me, quite nondescriptive -3 very uncharacteristic of me, extremely nondescriptive

1. Most people seem to be more aggressive and assertive than I am.*

2. I have hesitated to make or accept dates because of shyne s s.*

3. When the food served at a restaurant is not done to my satisfaction, I complain about it to the waiter or waitress,

4. I am careful to avoid hurting other people s feel-ings, even when I feel that I have been injured.*

5. If a salesman has gone to considerable trouble to show me merchandise which is not quite suitable, I have a difficult time saying "No."*

6. When I am asked to do something, I insist upon knowing why.

7. There are times when I look for a good, vigorous argument.

8. I strive to get ahead as well as most people m my position.

9 To be honest, people often take advantage of me, 10. I enjoy starting conversations with new acquaint

ances and strangers. ^^^^^^ 11. I often don't know what to say to attractive persons

of the opposite sex.* -,-, ^ v. • ^r-c. 12. I will hesitate to make phone calls to business

establishments and institutions.*

121

13. I would rather apply for a job or for admission to a college by writing letters than by going through with personal interviev7S. *

14. I find it embarrassing to return merchandise.* 15. If a close and respected relative were annoying me,

I would smother my feelings rather than express my annoyance.*

16. I have avoided asking questions for fear of sounding stupid.*

27. During an argument, I am sometimes afraid that I will be so upset that I will shake all over.*

18. If a famed and respected lecturer rr.akes a statement which I think is incorrect, I will have the audi­ence hear my point of view as well.

19. I avoid arguing over prices with clerks and salesmen.*

20. When I have done something important or worthwhile, I manage to let others know about it.

_21« I am open and frank about my feelings. 22. If someone has been spreading false and bad stories

about me, I see him (her) as soon as possible to "have a talk" about it.

23. I often have a hard time saying "No."* 24. I tend to bottle up my emotions rather than make a

scene,* 25. I complain about poor service in a restaurant and

elsewhere. 26. When I am given a compliment, I sometimes just

don't know V7hat to say.* 27. If a couple near me in a theatre or at a lecture

were conversing rather loudly, I would ask them to be quiet or to take their conversation elsewhere.

28. Anyone attempting to push ahead of me in a line is in for a good battle.

29. I am quick to express an opinion. '30. There are times when I just can't say anything.*

^Total score obtained by adding numerical responses to each item after changing the signs of reversed items.

*

Reversed item.

122 APPENDIX C: BEHAVIORAL ASSERTIVENESS TEST

1"

The purpose of today's procedure is to find out how you react to some, ordinary everyday situations that might happen outside the hospital. The idea is for you to respond just as if you were actually in that situation, whether at home, in a store or in a restaurant. For example, if I say that you are in a restaurant and the waitress approaches, try and imagine that you are really there.

I will narrate and describe a numuber of different situations After I have described a scene, one of the characters will speak to you. Ac: soon as this person is finished speaking, you will hear a bell. Immediately after hearing the bell,.I want you to respond to that person and say what you normally would say if you were in that situation. If you have any questions, be sure to ask them nov7 because once we start the person running the tape recorders will not be able to answer you.

(PAUSE)

Let's have one practice exercise to be sure you understand the instructions. Remember to respond as if you were really there in each situation.

(PRACTICE EXERCISE)

Narrator:

Wife:

You have just come home from a hard day's work dead tired. Your wife informs you that she has accepted an invitation for you both to visit some friends that evening. You are definitely not in the mood to go out.

"I just knew that you'd like to visit tonight, let's go right after dinner."

1, Narrator

Daughter

You are in the middle of watching your favor­ite TV show. Your daughter walks in and changes the TV channel as she does everytime you are watching TV.

"Let's V7atch this movie instead; it is sup­pose to be real good."

123 2. Narrator: This morning you took your car to the local

Exxon station and you clearly told the mechanic to give you a simple tune-up. The bill should have been around $20. It is now later in the afternoon and you are at the station to pick up your car. The mechanic IS walking over to you.

:l£2ll?-ili5_- "OK, let me make out a ticket for you. The tune-up was $12 for parts and $8 for labor. Uh, grease and oil job was $6. Antifreeze was $5. Uh, $4 for a nev7 oil filter, and uh, $5 for rotating the tires. That's $4 0. in all. Will this be cash or charge?"

- Narrator: You're in a restaurant with some friends. You order a very rare steak. The waitress brings a steak to the table which is so well done that it looks burned.

Waitress: "I hope you enjoy your dinner."

^' Nar.rator: You are in a crowded grocery store and are in a hurry. You had picked up one small item and you get in line to pay for it when a woman with a shopping cart full of groceries cuts in line right in front of you.

Woman: "You won't mind if I cut in here, will you? I*m late for an appointment,"

5. Nar.rator: Suppose you worked part-time in an office in the afternoon. At 4:35 one afternoon, as you v/ere looking forward to going home and antici­pating your evening out wich some friends, your boss asked you to work overtime for that evening.

2oss^; "Vyould you mind working overtime this evening?"

6. Narrator: You almost finished painting some furniture v/hen you suddenly realize that you won't have enough to finish the job. The salesman at the hardware store W-:ongly advised you that a quart would be sufficient. You get in your car and rush back to the hardware store to buy another pint and arrive a few minutes before closing.

124

Salesman: "I'm sorry but you will have to come back to­morrow. I'm trying to close up a few minutes

Narrator:

8

Spouse:

Narrator:

Mechanic

Narrator

early today."

You have just punished your child for his in­considerate behavior and told him that he must stay in his room for the rest of the afternoon. Your spouse feels sorry for him and tells him that he can go out and play,

"It's so nice outside; it is a shame to make him stay in his room,"

You left your car at 9:00 in the morning at the Goodyear Tire Dealer for a new set of tires. At 4:30 that afternoon you call to see if it is ready and they told you tliat it v/as. A friend, V7ho is going nearby, gives you a ride to the Goodyear Dealer and then leaves before you get your car, figuring that it would be ready.

"The blue Chevy? Nah, it's not ready. Come back tomorrow and we will have it ready for ya."

You go to a ball game with a reserved seat ticket. When you arrive you find a man has put his coat in the seat for which you have a reserved ticket. You ask him to remove his coat, and he tells you he is "saving that seat for a friend."

10

Man:

Narrator

Friend:

"I'm sorry that seat is saved."

You are talking with a friend who the night before embarrassed you at a party by reveal­ing a very personal and confidential story you had told her sometime ago in private,

"Sorry I embarrassed you, but I just had to tell them, I mean it was such a funny story and well, sometimes I think you are too darn sensitive."

125

APPENDIX D: SCORING SHEET FOR BEHAVIOR ASSERTIVENESS TEST

Subject's Name

Rater's Name

Duration of Response Total Score

Subject # Date

Response Latency Total Score

Response Latency

SCORING:

6 = A

Rating Scene # (circle

1 1 2 2 1 2 3 1 2 4 1 2 5 1 2 6 1 2 7 1 2 8 1 2 9 1 2

10 1 2

for Response appropriate #)

3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6 3 4 5 6

Total Score

very assertive response; the subject

Duration of Response

is quite con'

4 =

fident; shows appropriate emotions spontaneously; pursues his demand or initiation strongly, recogniz­ing reasonable limits; is insistent without being overbearing or self-righteous; his content is quite substantial or convincing.

A somewhat assertive response; the subject exhibits some confidence; shows guarded emotion; repeats a demand with some development of content; pushes his point, but not really enough; his content is moder­ately convincing.

A mildly assertive response; the subject simply states his demand or makes his initiation without much confidence or effect; the response, although assertive, is either very brief or very long, with the attitude of "trying to explain"; content is on the shallow side.

126

3 =? A mildly unassertive response; the subject shows some passivity; may induce some guilt in partner; is indirect in responding; content is on the shallow side; he is slightly hesitant; pursues his demand or initiation very awkwardly.

2 = A cover-up unassertive response; the subject anxiously repeats himself or i¥ jocular; gives phony, "hot-air" responses; finds some cause for agreement; shows a number of passive-aggressive signs; sounds like he's having difficulty handling the situation.

• ~ Totally unassertive response; the subject does not respond; gives an irrelevant response; is extremely passive-aggressive; completely fails to exercise his rights or initiate an interaction; agrees com­pletely with partner.

127

APPENDIX E: ASSERTIVE BEHAVIOR INDEX

Date

Interviewer's Name_ Patient's Name

Before the_interview, read over the various categories of behavior listed below. Do not fill out this form durina the interview but keep the categories in your mind as you inter­view the patient. At the end of the interview, please com­plete the form. (Be sure to rate the patient in each of the eight areas listed below.) You have five possible~responses to choose from in each C£*tegory. Place a check mark (/) in the column which you feel best describes or captures the patient's behavior during the interview. (Make only one check mark for each category.)

!• Eye Contact: avoids contact most of time avoids contact sometimes ^appropriate frequently" staring ^constantly staring

^' Facial Expression: ^ face rigid/no facial expression __somewhat rigid/lTttle facial expression ^appro­

priate somev7hat exaggerated/inappropriate excessive/exaggerated/ inappropriate

•' Body Posture: slouching somewhat slouching appropriate somewhat fixed/rigid fixed/

rigid

4. Body Orientation: body constantly positioned away from you body sometimes positioned away from you

sits facing you appropriately sometimes sits on edge of chair or leans too far forward _constantly sits on edge of chair or leans too far forward

5. Hand and Arm. Movements: no gesturing seldom • gesturing .appropriate, movement frequently

excessive/inappropriate constantly excessive/ inappropriate

6. Voice Modulation: constan-^1^^ too soft fre­quently too soft appropriate frequently too loud constantly too loud

7. Voice Rhythm: constantly too slow frequently too slow appropriate frequently too fast

constantly too fast

w^^

128

8. Overall Impression of Patient's Behavior during Interview; passive unassertive somewhat passive/unassertive assertive somewhat aggressive/hostile aggressive/hostile