Rose1977.pdf

This article was downloaded by: [Carnegie Mellon University] On: 02 February 2015, At: 22:32 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Scandinavian Journal of Behaviour Therapy Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/sbeh19

Assertive Training in Groups: Research in Clinical Setting Sheldon D. Rose

a

a

School of Social Work University of Wisconsin Published online: 26 Jul 2010.

To cite this article: Sheldon D. Rose (1977) Assertive Training in Groups: Research in Clinical Setting, Scandinavian Journal of Behaviour Therapy, 6:2, 61-86, DOI: 10.1080/16506073.1977.9626686 To link to this article: http://dx.doi.org/10.1080/16506073.1977.9626686

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever

or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Nordisk Tidskrift f'dr Beteendeterapi, 6, 61-86, 1977 Scandinavian Journal of Behaviour Therapy, 6, 61-86,1977

Assertive Training in Groups: Research in Clinical Setting

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

SHELDON D, ROSE

School of Social Work University of Wisconsin

Under auspices of the Interpersonal Skill Training and Research Program a series of seven research, projects have recently been completed in which the effectiveness of assertive training in groups with various clinical and professional populations have been evaluated. These populations inlcuded young adults, women, social workers and the elderly. The programs for assertive training included a variety of pro­ cedures such as modeling, coaching, overt behaviour rehearsal, covert rehearsal, group feedback, buddy contacts, selected readings, homework assignments, and contingency contracts. Data were collected on outcome by means of assertion inventories and audiotaped role play tests, and on process by means of observa­ tion of interaction, attendance, completion rate of assignments, and self report at the end of each meeting. Results on the role play tests tended to support the efficacy of assertive training for all the populations, however, strongest results were obtained for the women and young adults and the weakest for the elderly. Evidence also supported the use of contingency contracts as a means of increasing the rate of assignment completion.

Social anxiety and difficulty in relating effectively to others is a problem shared by a large part of the population. This discomfort is often found linked with other problems such as delinquency, school and work absenteeism, alcoholism, depression, and loneliness. In order to Reprints can be obtained from Sheldon D Rose, School of Social Work, University of Wisconsin, 1*25 Henry Mall, Madison, Wise 53706.

61

facilitate social relational skills, a program has been developed which teaches these skills directly. In this program clients are taught such skills as standing up for one's rights, meeting people and making new friends, conversing and interviewing, and expressing their feelings. There is evidence that as a person learns such skills the intensity of social anxiety is often reduced (Wolpe, 1 9 7 3 ) . This skill training approach to treatment is called assertive or interpersonal skill training and refers to a set of procedures by which

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

clients are trained in behavioral areas in which they are deficient. These prodedures include primarily modeling, coaching, and behaviour rehearsal, although other procedures are often added. Not only is there considerable anecdotal evidence for use of this approach, in recent years many experiments have demonstrated the efficacy of various proce­ dures within this approach in modifying specific behaviours related to social anxiety and unassertiveness. (See for example Eisler, Hersen, and Miller, 1973; Galassi, Kostka, and Galassi, 1975; Hersen, Eisler, Miller, Johnson, and Pinkston, 1973; Hersen, Eisler, and Miller, 197^; MacDonald, Lindquist, Kramer, McGrath, and Rhyne, 1975; McFall and Lillesand, 1 9 7 1 ; McFall and Marston, 1970; McFall and Twentyman, 1973.) Although most of the above experiments were conducted with individu­ al subjects, it would seem.that social skill deficits can be most effec­ tively treated in social situations involving more than one client. The group can serve as a protected laboratory to practice skills which must eventually be carried out in the community. The group provides clients with a variety of social situations and potential roles that each can play. Moreover, in role played modeling and rehearsal the clients pro­ vide each other with an assortment of different types of roles, in the position of leader, therapeutic partner, or consultant to one's fellow clients. Several studies have made use of the group context for asser­ tive training experiments (Shoemaker and Paulson, 1976; Rathus, 1972; Lomont, Gilner, Spector, and Skinner, 1969; Hedquist and Weinhold, 1970; and Rimm, Hill, Brown, and Stuart, 197*+; Rimm, Keyson, and Hunziker, 1971) although in none of these was explicit use of the group made. To make use of the group, the therapist must initially strive to 1

make the group attractive to its members; to create group situations

62

requiring social competency on the part of the members; to create a va­ riety of group roles which members can play; to delegate gradually the responsibilities of leadership; to present situations in which members function as consultants and partners in the therapeutic endeavor; and to control excessive group conflict. Interpersonal Skill Training and Research Program The Interpersonal Skill Training and Research Program was developed

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

to evaluate the effectiveness of such assertive training and other so­ cial skill groups in clinical settings with various populations as well as to provide opportunity for group leadership, observation, and gene­ ral research training. In addition to the use of assertive training procedures, group process was to be dealt with systematically. Since 1975 there have been seven projects completed under the auspices of Interpersonal Skill Training and Research Program in the area of asser­ tive training; all but one of these occurred

in the context of small

groups. 1) evaluation of outcome data of five assertiveness training groups with a clinical population of young adults (Rose, 1 9 7 5 ) ; 2) comparative study between assertiveness training and behavioural discussion groups with young adults (Schinke and Rose, 1 9 7 6 ) ; 3) comparative study between assertive training groups, group desensitization, discussion groups, and a test-retest control for women (DeLange, 1 9 7 6 ) ; h) comparative study between assertive training and discussion train­ ing with the elderly (Berger and Rose, 1 9 7 7 ) ; 5) comparative study among problem solving, discussion and assertive training groups for the elderly (Toseland, 1 9 7 7 ) ; 6) comparative study between assertive training and discussion train­ ing for social workers; and 7) multiple-baseline study of the effects of contingency contracting on the completion rate of behaviour assignments in six assertive training groups (Rose, 1977b). Three additional projects for which data are now being collected are: 8) the effects of various intervention procedures on leadership be-

63

haviour in assertive training groups 9) comparative study between assertive training and placebo control group for institutionalized elderly 10) social skill training in six groups of third, fourth, and fifth grade children. This paper will briefly report on the assertive training program used in the completed projects, the research procedures, the research design,

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

and the results.

PROJECT ONE: Assertive training for young adults In order to discover whether assertive training in groups was appli­ cable in social treatment agencies with typical clinical populations programs were organized in five different agencies (Rose, 1 9 7 5 ) . Each group used a similar treatment package. Twenty-four clients in groups ranging in size from k-8 persons met weekly from 8-10 sessions for two hours. The members were almost equally divided among males and females, and were from diverse socio-economic backgrounds. Data Collection The data collection procedures used varied somewhat from group to group; most included a diary kept by the members, personality invento­ ries filled out before and after treatment, such as the Rathus Asser­ tiveness Schedule (Rathus, 1 9 7 3 ) , the California Personality Inventory (Gough, 1 9 5 7 ) 5 "the Situation Reaction Inventory of Anxiousness (Endler and Okada, 1 9 7 5 ) , and the Willoughby Personality Inventory (Wolpe, 1 9 7 3 ) , and audiotaped behaviour role play test which was also given prior to and following treatment. The audiotaped behavior role play test consist­ ed of a number of social situations requiring an assertive response. Most of these situations were obtained from the clients themselves when they were recruited. In the testing situation each situation was read aloud to each client who then responded as if she or he was in that si­ tuation. The responses were recorded on audiotape and later coded for assertiveness. Of the above tests only the RAS was given in all five groups.

6k

Ongoing data about the group process was also collected in

most groups. This included data on participation (collected by obser­ vers), satisfaction (responses from the clients on a seven point sca­ le), productivity (the percentage of behaviour assignments completed), and attendance. Clients also kept self-report data on behaviours ac­ tually changed in the real world or stress situations successfully dealt with. Intervention

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

The major intervention procedures used in this project were model­ ing, coaching, behaviour rehearsal, covert rehearsal, homework assign­ ments, contingency contracting, and group feedback. In most groups re­ laxation training was also used. Leaders also worked on keeping the attraction of the group high, keeping a broad distribution of partici­ pation, and involving the members in leadership responsibilities. In a typical meeting, members brought in diaries kept during the week. Each would describe one social situation with which she or he was ha­ ving difficulty. The group members would ask for amplification. Sug­ gestions would be made as to how it might be done differently. Some­ one in the group would demonstrate to the given client how it might be carried off. Following discussion of the modeling event, the given client would rehearse the situation often with suggestions from a coach who sat behind the client. Other members would play the role of the significant other. The group would then discuss what the client did well and what she or he might have done somewhat differently. When everyone had

roleplayed at least one situation, members decided

on homework assignments such as keeping a diary, practicing relaxa­ tion, trying out a rehearsed behaviour in the real world, and contacting a "buddy" (their partner in the group). Some would attach contingen­ cies to the completion of the assignment. At the end of the meeting clients were given relaxation practice and asked to fill in an evalua­ tion of the meeting. (These procedures are described in detail in Rose, 1977a).

65

Results On the self report data, all but three persons reported change on at least one behaviour they worked on. The average was 1.6 behaviours changed per person. Of 2k persons completing the program, 21 showed positive changes on the Rathus Assertiveness Schedule. On all other instruments all but two clients showed changes in the desired direc­ tion. On the basis of those data, it was concluded that 21 persons

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

(86 %) successfully utilized the program. However, even though most people showed improvement, several still showed extremely low (nonassertive) scores. These people were referred for additional training. Although the results were generally positive, in the absence of a control group, one could not reject the hypothesis that some general effect (placebo) rather than assertive training brought about these shifts in behaviour and test scores. Moreover, the programs though similar varied as to length, recruitment procedures, content of pro­ gram, and a number of other relevant features. To deal with these problems, Project Two was designed.

PROJECT TWO: Assertive training and behaviour discussion The purpose of Project Two was to compare the effectiveness of com­ mon elements found in the clinical packages used in Project One to a discussion group condition under standardized conditions. Subjects we­ re recruited who were similar to the clients in Project One (Sehinke and Rose, 1 9 7 6 ) . All groups, with five to six members each, met for six weekly two-hour sessions. Subjects were randomly assigned to four experimental groups and three control groups, consisting of 21 and 1 5 , respectively. These groups were led by seven comparably trained gradua­ te students (who were somewhat more experienced than

those in Project

One). Assessment instruments included a Social Contact Monitoring Sheet, a self-report survey of assertive behaviours, the Rathus Assertive­ ness Schedule, and the California Psychological Inventory, Class 1 Scales. All were administered at pre-treatment, six weeks later at post-treatment, and three months later at follow-up. Audio-taped

66

behavioural role-plays were administered

at post-treatment and follow-

up, using stimulus situations gathered from the study sample at an initial telephone contact. The assertive training procedures were similar to the ones descri­ bed in Project One except that no relaxation training was used. In the placebo control (behavioural discussion) groups, the same type of situations that were rehearsed in the experimental group were dis­ cussed, members provided each other with feedback on how to do them,

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

but no rehearsal was permitted. The results of this comparative evaluation indicate that subjects receiving the experimental package were rated as significantly (p < 0.001) more assertive on components of the audio-taped behavioral role-play at post-treatment and follow-up, than subjects receiving the placebo-control condition. The three self-report instruments, the Social Contact Monitoring Sheet, the Rathus Assertiveness Schedule, and the California Psychological Inventory, show strong gains for sub­ jects in both conditions from pre-treatment to post-treatment and from post-treatment to follow-up, but there are no significant differences between the two conditions on these three measurements. Results were interpreted as lending support to the hypothesis that group assertiveness training is useful approach to treatment of cli­ nical population of young adults.

PROJECT THREE: Assertive training, desensitization and discussion In the first two projects, clients of both sexes were recruited. Many of the women recruited for assertive training groups have ex­ pressed a desire to be in an all-women's group. On the basis of this ex­ pressed client interest, DeLange (1976) designed a project for women only to compare the effectiveness of assertive training in groups, group desensitization, a discussion control and an assessment only control. An additional purpose was to study the effect of the client's general level of anxiety on the treatment method. Also, in contrast to the first two projects, DeLange made use of semi automated treat­ ment procedures.

67

In order to generalize to the large numbers of assertive training groups being attended by women throughout the country, recruitment was specifically designed to increase probability of obtaining women from different age groups and occupations. One hundred twenty women were recruited by advertising, posters, newspaper stories, and word of mouth. Program and Test Development DeLange carefully adhered to the model recommended by Goldfried and

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

D'Zurilla (1969) for constructing a behavior role play test and evolving training situations for the treatment program. Thirty women kept seven day logs of all situations requiring an assertive response. They also described any other similar situation which might require an assertive response and which had previously occurred to them or others they knew. On the basis of these data, 50 vignettes and an accompanying question­ naire were constructed and submitted to hO women who responded to each situation in terms of its frequency of occurrence to them and. their degree of comfortableness and satisfaction in handling the situation. New situations were also requested. Thus a list of 8k situations were generated, some of which were eliminated if not sufficiently relevant or problematic. The remaining items were in turn submitted to the sub­ ject population who rated them in terms of satisfaction and comfortable­ ness with their responses, its importance for them to work on, and whether the situation ever happened to them. On the basis of the above criteria, items were selected which also fit into the following five categories: stating one's needs and initiating confrontation; handling criticism; requesting from others; interaction with authority figures; and initiating actions. For purposes of the group desensitization condition, test items were placed in a hierarchy of difficulty based on the degree of discomfort on the items. Response alternatives were elicited by submitting the remaining 62 items to a population not in the experiment, this resulted in 1 1 to 15 responses for each situation ranging from unassertive to aggressive. Expert judges rated the responses to the 1 1 test items on a 5 point scale from unassertive (ineffective) to most assertive (effective) and

68

wrote out criteria for making a judgement. On the basis of these judge­ ments a rating manual was devised. Research Design The 120 subjects were divided into two levels of anxiety, high and low, on the basis of scores on a test designed to measure trait anxiety, forming two equal groups. Each woman in these conditions was then ran­ domly assigned to one of four treatment conditions: systematic desensi­

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

tization, assertive training, discussion, and assessment-control. There were 30 women in each condition, 15 high anxiety and 15 low anxiety women. Within each treatment condition, women were arranged into groups of six, three high anxiety and three low anxiety, on the basis of their schedules of available meeting times. There were five groups of six wo­ men within each treatment condition. The design was counterbalanced for therapist; each of five highly trained female therapists was assigned one group in each treatment condition, randomly where possible within available time schedules. All groups were conducted simultaneously within the same six weeks using the same assertive situations in all groups in every treatment condition within the same week. Each person paid a $18.00 deposit (with several exceptions). The final evaluation instruments used were the behaviour role play test, the Assertion Inventory (Gambrill and Richey, 1 9 7 5 ) , the Rathus Assertion Schedule (Rathus, 1 9 7 3 ) , and the State Trait Anxiety (STAl), A-Trait (Spielberger, Gorsuch, and Lushene, 1970). All except the STAI were administered at pre- and post-treatment. The STAI is a self-report measure reflecting an individual's general anxiety level. The median of the scores on the STAI-A-Trait was used as the dividing point for high and low-anxiety subjects. The behaviour role play test consisted of 1 1 tape recorded items. Of these items, six were used in the training sessions, three others were in the same categories as those used in the training sessions but of different content (novel), and two were of different content and diffe­ rent general categories (novel-novel). This breakdown enabled testing for generalization within categories to untreated situations and across categories to situations in untreated categories. During the test each

69

woman responded to each item as if she were in that situation. The re­ sponse was tape recorded. After each response, the subject stated her anxiety level and her satisfaction with her response. Treatment Condition Each treatment used six situations within a specified category per session, with the exception of the first session which contained only

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

three. Every group, regardless of treatment condition, worked on the same six situations in the same week of sessions. All situations were presented by means of an audio-tape recorder. After a woman volunteered or agreed to work on a situation the As­ sertive Training procedures were as follows: Via audiotape: 1. presentation of the situation 2. coaching The narrator discussed the situation in terms of the task to be done and generally how to handle it 3. model one A female voice responded to the situation k. coaching The narrator pointed out what the model had done and gave more instructions 5. model two A second female voice responded giving an alternative approach Via therapist: 6. group discussion of the models The emphasis was on these models only as alternatives. A woman in the group could offer another way of approaching it and the woman rehearsing it could incorporate these approaches or use her own 7. behavior rehearsal The taped situation was again played and the designated S respond­ ed to it 8.

self-evaluation and self-reinforcement The subject reported her anxiety level and satisfaction verbally and recorded them on her record card; she told the group what she

70

liked about her response and what she would like to change. She gave herself points on a clicker counter from one to three (l = making a response; 2 = improvement over what you might have done before, fairly satisfied; 3 = really feel good about response, the way I would like to do it) also recording them on her card. 9. group feedback Members and therapist told the client specifically what they liked about the response and gave a suggestion for change.

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

10. repeat practice, steps 7~9 1 1 . repeat practice again, if so desired (DeLange, 1976, pp. U6-kT) Women were told that they could practice as often as they felt it nece sary to feel comfortable about what they had performed. Therapists were advised to give everyone an opportunity to respond and to facilitate specific positive and negative comments from every member. The highly audible clicker, used for giving oneself points, enabled the group mem­ bers and therapist to respond positively to the number of points a wo­ man gave to herself. In the group desensitization condition the therapist demonstrated relaxation, trained the group in visualization of scenes, and taught the members how to report anxiety level. Within each session the members were first put into a state of deep muscle relaxation; and hierarchial items were presented (3 times each). Six scenes only were visualized each session. After desensitization, members discussed relaxation, vis­ ualization,

the arousal of tension, and any questions and comments the

group or therapist wanted to make regarding the desensitization proce­ dure

a common clinical practice.

In the placebo-control (discussion) condition,women listened to the same situations via an audio-tape. Approximately every 15 minutes a new situation was played by the therapist. Women were instructed to think of how they would normally respond and what feelings were elicited. The emphasis was on understanding what one felt, why assertive behavior was inhibited, and what had occurred in the past to cause that reaction. Women shared the feelings aroused by the situation and related similar past experiences. They endeavored to identify the emotions felt, to specify what the current blocks to handling themselves in assertive sit71

uations

were, and to explore the reasons, past and present, that caused

their current feelings. Therapists were instructed to treat this group as a support group. Any specific suggestions by group members as to what to say or how to act in a situation were minimized, with the thera­ pist refocusing the discussion on feelings and causes of inhibition. Women in the assessment control were randomly selected to wait for the second round of groups. These women completed all pre- and post-

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

tests in the same week that women in the other treatment conditions did. These women were then debriefed and given six weeks of treatment with­ out cost. They were allowed to select the type of techniques used with their group. The complete findings are too detailed to be reported here. In gene­ ral the data supported the hypothesis that assertive training was the most effective procedure. Women given Assertive Training improved their performance on the behavior role-play test significantly more than any other treatment condition. On the two self-report measures of assertive­ ness, although differences were less pronounced, pre- to post-treatment scores indicated that the Assertive Training group improved the most on both measures. A related hypothesis, that, for low-anxiety women, Asser­ tive Training would be more effective in increasing assertive responses than any other condition, was supported. In addition, within the Asser­ tive Training treatment itself, greater generalization of treatment ef­ fects to untreated situations occurred for low-anxiety women than for high-anxiety women. These results suggest that the level of general anx­ iety

does influence the effect of assertion training; the lower the

anxiety level, the greater the amount of improvement that can be expect­ ed to generalize across situations.

PROJECT FOUR; Assertive training and discussion for elderty In the first three projects, clients were from a broad social back­ ground, ranging in age from 16 to 56 with the median age being 30. The question was raised whether this same or a similar program could be uti­ lized with an elderly population. In Project Four and Five the focus was on the elderly.

72

Project Four was concerned with the elderly in an

institution (Berger

and Rose, 1977)- Of the 160 patients in the institution only 90 were considered as neither too ill nor too disoriented to participate. Of these 90, ^0 who agreed to participate in the study were from a wide variety of socio-economic "backgrounds. Most of them had serious medical problems. Of the ho,

27 were randomly assigned to either the Assertive

Training condition or one of the two control conditions. The others had participated in the development of the role play situations.

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

Because of special problems of illness and irregular group attendance, an individualized assertiveness program was developed to be followed in a later project by a group program. A new audio-taped role play test was also developed. In this project, as in the previous, the format suggested by Goldfried and D'Zurilla (1969) was carefully followed. First, problem situa­ tions were developed which could be used in both the training program and in the role play test used to evaluate the program. Program and Test Development In order to elicit problem situations the investigator interviewed 22 patients. The experimenter asked the patient to recall recent inter­ personal situations, particularly those in which the patient felt she or he could have responded more successfully or felt better than he actually did. Specific descriptions were elicited by asking for each situation: l) What led up to the situation?; 2) What was the purpose of the interaction?; 3) What were the characteristics of the other per­ son?; k) What was the physical setting?; and 5) What was the flow of interaction? i.e. who said what to whom? From these interviews a set of 3^ highly specific problematic situa­ tions experienced by residents in a home for the elderly were construct­ ed. These situations involved making a reasonable request, refusing an inappropriate request, or responding to an unfair action or statement, and initiating and maintaining a conversation. Most of the situations involved interactions with peers, a few involved nursing home staff and relatives. In order to find the most relevant situations all 3h were tape re-

73

corded and presented to seven female and four male patients, all of whom had participated in the initial interview, who rated the items on perceived comfort and competence, as well as whether it was likely to arise at all.

In this way 16 items were obtained for which fewer than

half the subjects rated their response to the situation as "comfortable and able to handle it well" and fewer than one-third rated the situa­ tion as one that "would never arise". In order to obtain responses to these situations eight staff persons

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

including the nursing home supervisor and other professional and non­ professional staff listened to a tape recording of the 16 most relevant situations. To each situation they were asked to respond out loud as if the patient was actually experiencing that situation. Their role plays provided a list of eight possible responses to each of the situations. These responses were rated by other experts from very ineffective to very effective on a five point scale and interviewees were asked to ex­ plain what it was that made it so. These 16 items constituted the beha­ viour role play test, eight of which were employed in training. The coding manual consisted of the items plus responses. The Assertive Training Program The experimenter read the situation to the patient. The experimenter read coaching material which had been prepared from the response evalua­ tions of significant others in the program development phase. This mate­ rial explained the general nature of the situation, suggested one way in which the patient could respond, and often indicated

probable

consequences of the suggested response. As an example of the coaching material, the experimenter played the tape recorded response of a model to the situation. All modeling responses were recorded by an 86-yearold female. The experimenter read additional coaching material. This material commented on the modeled response and reviewed the previous coaching material. In a brief discussion the experimenter asked the patient what he or she thought of the modeled response and if the respon'se would be effec­ tive for him or her. The patients who disagreed with the response or felt it would not be appropriate for themselves were asked to rehearse Ik

for the "sake of practice". The experimenter re-read the situation which was abbreviated to avoid unnecessary repetition. The patient who then roleplayed a response to the situation was ver­ bally reinforced for any roleplayed response. Additional reinforcement was provided for aspects of the response that satisfied the coaching material. In addition, the experimenter clearly indicated to the patient the ways in which his response did and/or did not satisfy the coaching

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

material. A brief discussion focused on whether the patient understood and agreed with the feedback. Rehearsal and feedback continued until the experimenter determined that the patient adequately satisfied the coach­ ing content in two rehearsals. If the patient had not met this crite­ rion within three minutes the experimenter went on to the next coachingmodeling segment of the training sequence. In order to facilitate learning, the training sequence was structured in such a way that only one principle of effective behaviour was presented at a time, in one coaching-modeling segment (steps 2 to h above). When the patient's response to the situation evidenced learning of that seg­ ment, the experimenter proceeded to the next segment. Patients in the control condition were told that discussing social situations makes one better able to handle these situations. The expe­ rimenter read out loud the same situations as used in the Assertive Training sessions, in the same order. Each situation was followed by a 15 to 20-minute discussion which focused on similar experiences which occurred to the patient, the patient's feelings in those situations, and a discussion of probable causes for those feelings. Specific behaviou­ ral solutions to these situations were not presented. Another control consisted of patients who participated only in the assessment condition. Results and Discussion On the behavioural role play test, patients who participated in Asser­ tive Training were significantly (p < .05) more effective than Control patients in responding to social situations to which effective response had been explicitly taught. Assertive Training patients retained this learning even two months after training. The results of this study how75

ever, provided only limited evidence for the efficacy of Assertive Train­ ing with, elderly patients. On the role play test it was clear that learn­ ing did not generalize to novel situations, i.e., Assertive Training patients did not learn how to apply general rules of effective behaviour to new situations. In addition, although these patients did increase their effectiveness, this was not reflected in the self-report measures. That is, Assertive Training patients were not more satisfied with their responses and did not rate social situations as less problematic or dif­

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

ficult. Also, Assertive Training had no effect on the patient's perfor­ mance in a simulated interaction with a social worker.

PROJECT FIVE: Assertive training, diseussion and problem solving for elderly In contrast to Project Four which did not use small groups and which worked solely with institutionalized elderly, Toseland (1977) in Pro­ ject Five sought to work exclusively in small groups with the elderly who still lived in the community. Moreover, he sought to compare asser­ tive training in groups with problem-solving groups as well as discus­ sion groups. Treatment Methods The introductions were standardized across the three group methods. All participants received the same explanation for money payments, con­ fidentiality, and an introduction to the topic of social skills in dif­ ficult interpersonal situations. Passive, aggressive and assertive be­ haviour was differentiated. All groups, meeting for six 1 1/2 hour meetings, covered the same six topic areas: session 1 , initiating interactions;session 2, confronting others; session 3, handling service situations and making requests; session k, giving negative feedback; session 5, responding to criticism and turning down requests; and ses­ sion 6, expressing opinions. There were fifteen groups, five groups in each of three treatment methods (l) Assertive Training, (2) Problem Solving, and (3) general discussion. Each treatment method contained a package of treatment components.

76

The Assertive Training condition included role playing, behaviour rehear­ sal, modeling, feedback and coaching. The Problem Solving condition in­ cluded a general orientation, defining the situation, an orientation to problem solving (cognitive restructuring, generating alternative solu­ tion, decision making criteria and implementation discussion). The Dis­ cussion Method which is similar to traditional social group work ap­ proaches included a time limited focus, a discussion of the person-insituation configuration with the leader acting as the facilitator of

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

the group's definition and discussion of the situation. In this project the discussion condition served as a placebo control. Design Project Five used a pre-, post-, and three month follow-up assess­ ment design with two experimental conditions and a placebo control. Fifty-three subjects above the age of 55 participated in the project. Data was analyzed in terms of group means which provided a sample size of fifteen. These groups were made up of three to five participants. Each group was randomly assigned to treatments with the constraint that once a treatment contained five groups no more groups would be assigned to it. Subjects

came from three community agencies. All subjects were

volunteers who were paid two dollars for each group session attended. Twelve leaders including six community agency staff personnel, three students and three older people were trained in the use of the treat­ ment methods and randomly assigned to groups. Three leaders led two groups. All other leaders led one group. Three assessment measures were used. A role play test of eight situa­ tions was developed by interviewing sixty older persons over a one year period. Four of the situations were used as training situations by all groups. Four of the situations were used as untrained generalization situations. Two coders rated the tapes based on a coder rating manual developed in conjunction with the role play test. The Assertion Inven­ tory (Gambrill and Richey, 1975) was used to assess change in anxiety and response probability as a result of training. A Group Evaluation Inventory was used to assess the subjective responses of participants in terms of satisfaction with the group method, perceived change in 77

social skill and number of situations in which social skills learned in the group were used in everyday situations. Results and Discussion On both the trained and untrained test situations, subjects in both the Assertive Training and Problem Solving groups were significantly (p<.Ol) more assertive at

posttest than those in the Discussion groups.

The Assertive Training method subjects were consistently more assertive

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

on the role play test than Problem Solving subjects but the differences were not significant. Though scores on the role play test continued to increase for all three conditions from posttest

to follow-up, because

scores increased more rapidly for the discussion group, differences at follow-up were not quite significant (p<.07). No differences were found between any of the treatment methods on the Assertion Inventory anxiety and response probability scales. Since there was a great deal of random variation of scores within conditions, the Assertion Inventory may not have been

appropriate for use with

older persons. The Group Evaluation Inventory showed that participants in all group method approaches were very satisfied with the group in which they were participants. All participants indicated moderate increases in social skill learned by participation in a group. There was very little differ­ ence

on perceived increase in social skill between treatment methods.

In terms of the number of responses where social skills were used in everyday situations the Assertive Training condition participants re­ ported a greater number of situations where skills were used (3.10) as compared to the Problem Solving (1.50) or the Discussion (2.23) group participants. These results indicate that although no effects are noted on an as­ sertion inventory in terms of behavioural outcomes the Assertive Training group method is slightly more effective than the Problem Solving group method and both are more effective than the Discussion Method in in­ creasing the social skill of older adults.

78

PROJECT SIX: Assertive training and discussion for social workers In all the previous projects, clinical population were trained in assertive responses. In Project Six the effectiveness of an interperso­ nal skill training program for professional social workers (Rose and Edleson, 1977) is evaluated. In order to examine this population's response to assertive training in groups, 62 senior and graduate social work students were assigned to experimental and control conditions. The

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

assignment was not random, however, but evidence suggested that the grou" were equivalent on major variables. Subjects in the experimental condi­ tion were taught to deal with critical situations commonly encountered by professional social workers in their clinical practice. As in asser­ tive training, modeling, overt and covert rehearsal, videotaped and group feedback, coaching, buddy contacts, selected readings, behavioural assignments, contingency contracts, and delegation of group leadership were utilized in the training program. Experimental subjects met in small groups of four to six members for two hour sessions each week over a period of six weeks. In addition, data on group interaction gathered by pairs of trained observers at each session were presented to group members by the leader as feedback on their interaction. Role Play Test Development The control subjects received no training. Subjects in both condi­ tions participated in behavior role play test prior to and following the training program for the experimental group. In order to develop items for the test professional social workers were asked to list as many situations with clients, peers, or supervi­ sors which required some kind of assertive response from the social worker. These items were edited and submitted to judges who selected the best six items in terms of relevancy to professional practice, varia­ tion, clarity of the item, and amenability to multiple responses. At this point the test development differed from the previous pro­ jects. Rather than eliciting a single response, multiple responses from the subject were desired. In order to obtain these responses a set of general statements which could apply to any response to a given item

79

were interspersed between each client response. (Similar to a procedure developed by Galassi and Galassi, 197*0. Prior to the training program, expert judges were used to develop the criteria for evaluating effectiveness of most possible sets of re­ sponses. These criteria consisted of overall competency, expression of feelings and opinions, persistence, giving and seeking clarification, appropriate affect, latency, fluency, volume, and the timing within which critical statements were made. Only those criteria relevant to

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

each specific role play situation were used in evaluating the subject's response. (See Rose, Caynor, and Edleson, 1 9 7 7 , for a detailed descrip­ tion of this test). Results and Discussion The results of the study indicated that following treatment the ex­ perimental subjects were significantly (p<.0l) more effective than the controls in dealing with both trained and untrained items on the beha­ viour role play test.Data about interaction among members also indica­ ted that leadership could be successfully delegated to group members with an increase in member satisfaction and group productivity.

PROJECT SEVEN: Multiple-baseline effects of contingency contracting In contrast with the earlier projects, which studied the effective­ ness of assertive training, the purpose of Project Seven was to analyze the effectiveness of contingency contracting with adults as a means of increasing the percentage of behavioural assignments completed in asser­ tive training groups (Rose, 1 9 7 7 b ) . Since contingency contracting had thus far rarely been used with adults especially in assertive training groups, this project provided a test of its efficacy under clinical con­ ditions . Research Design In the fall of 1976, six assertive training groups were organized in three different agencies: a Family Service Agency, the Department of Adult Medicine of the Medical School, and the Interpersonal Skill Train-

80

ing and Research Project of the School of Social Work of the University of Wisconsin. Each of the groups consisted of 3 to 8 members and met weekly for two hour sessions. Clients, recruited primarily by adverti­ sing in local media or through posters, were chiefly university staff and students, and professional and business people. In all groups and for all meetings except the last one, leaders were instructed at the end of each session to develop specific home assign­

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

ments with all clients. At the beginning of the subsequent session, as­ signments were monitored by asking each member to review the assign­ ments he or she had agreed to for the previous week and to provide evi­ dence of their completion. In groups one and two contingency contracts were introduced at the end of the second session; in groups three and four at the end of
Conolusion In summary, the Interpersonal Skill Training and Research Program has provided a variety of evidence to support the contention that assertive training in groups is an efficient and beneficial approach to the treat­ ment of clinical populations, women, young adults, and the elderly. It

81

can also provide effective training for professional social workers in assertiveness-related skills. Another finding, at least with women, suggests that the level of anx­ iety plays a major role in determining the degree of generalization. High-anxiety subjects tend to show less generalization of improvement across situations than low-anxiety subjects. The outcome results with the elderly, though positive, are the most tentative; the results with women and young adults in a clinical situa­

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

tion appear to be the strongest. Follow-up data were obtained in three of the studies; again the young adult population seemed to be somewhat more effective in maintaining changes obtained in the group than the elderly. However, differences may be due to methodological considera­ tions. It appears that with the elderly, greater use of memory aids, more frequent sessions, and periodic booster session may be required to achieve the same level of change and maintenance of change that younger populations seem to obtain. Looking at the process findings in several projects, we find that contingency contracts appear to be a useful addition to assertive train­ ing in groups as a means of increasing the completion rates of behaviour assignments. It is also possible to increase participation of members through simple reinforcement procedures and to effectively delegate leadership activities to members of the group. It is clear that a great deal more research in this area is needed. Two kinds of training programs were used in this study: semi-auto­ mated and partially leader directed and structured, and completely lea­ der directed. In the latter condition the leader tried to fade his or her activities as the group approached termination. Both conditions tended to yield the same levels of change and the same degree of satis­ faction. Though both are highly structured experiences, these structures is objectionable to very few persons. In all of the projects evaluated outcome, a behaviour role play test was developed which was uniquely tailored to the population being served, In two projects item development adhered closely to the complex approach suggested by Goldfried and D'Zurilla (1969). In these projects items were developed which were used in training and others which were not.

82

This made it possible to test the degree to which the training genera­ lized to new items and even new categories of items. This appears to be a promising use of the role play test. In the other projects a sim­ pler procedure was used in which most of the items to be used on the test were developed by the subjects in the study or by similar subjects in previous studies. In all but the professional training group, a single response was demanded to each item on the role play test. In the latter group, multiple responses interspersed with responses from the

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

antagonist were used. Although somewhat less reliable than the other tests because of the complexity in coding, it appeared to represent the real life situation more nearly and as such may prove to be a more valid measure. In all cases however the role play test was a sensitive measure regardless of how it was developed. Except in the initial noncontrol group project and in the DeLange (1976) project, the Rathus Assertiveness Schedule and the GambrillRichey Assertion Inventory fared poorly as discriminating instruments. Almost all conditions showed improvement on these tests, but usually only small differences between conditions

were

discernable.

One might

conclude that on such inventories subjects tend to show how much they liked the training rather than how much they actually changed. Alterna­ tively, one might ascertain that any treatment focusing on assertive behavior is sufficient to bring about self-reported change, though, only Assertive Training group approaches are likely to bring about motoricverbal demonstrations of change. Although in nearly every study some methodological limitations restric­ ted confidence in the findings,the persistence of the role-play test in demonstrating the effectiveness of Assertiveness Training in groups in all projects can not be dismissed lightly. Of course, there are still many unanswered questions. To what degree must this approach be restricted to behaviors commonly described as assertive? What other populations might profit from this approach, e.g., depressives, psychotherapists, public service workers, receptionists? How important is the group condition for maintenance of changes achiev­ ed in the group? How important is our concern for group process in achieving individual treatment gains? What is the relationship of the

83

role play test to results to real world "behavior? These are some of the questions with which the Interpersonal Skill Training and Research Pro­ gram will be dealing with in the coming months and years.

References Berger R M & Rose S D: Interpersonal skill training with institutiona­ lized elderly patients. (In press).

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

DeLange J M: Relative effectiveness of assertive skill training and desensitization for high and low anxiety women. (Doctoral disser­ tation, University of Wisconsin, Madison, 1 9 7 6 ) . Eisler R M, Hersen M & Miller P M: Effects of modeling on components of assertive behavior. Journal of Behavior Therapy and Experimen­ tal Psychiatry, 4, 1 - 6 ,

1973.

Endler N S & Okada M: A multidimensional measure of trait anxiety: The S-R Inventory of General Trait Anxiousness. Journal of Consulting and Clinical Psychology, 43, 319-329, 1975. Galassi J P & Galassi M D: Validity of a measure of assertiveness. Journal of Counseling Psychology, 21, 21+8-250, 197*+. Galassi K P, Kostka M P & Galassi M D: Assertive training: A one year follow-up. Journal of Counseling Psychology, 22, U51-U52, 1975. Gambrill B D & Richey C A: An assertion inventory for use in assess­ ment and research. Behavior Therapy, 6, 550-561, 1975Goldfried M R & D'Zurilla T J: A behavioral-analytic model for assess­ ing competence. In C D Spielberger (Ed): Current topics in clini­ cal and community psychology* Vol 1. New York: Academic Press, 1969Gough H G: California psychological inventory. Palo Alto, California: Consulting Psychologists Press, Inc., 1957. Hedquist F J & Weinhold B K: Behavioral group counseling with socially anxious and unassertive college students. Journal of Counseling Psychology, 17, 237-21+2, 1970. Hersen M, Eisler R M & Miller P M: An experimental analysis of general­ ization

in assertive training. Behaviour Research and Therapy, 12, 1

295-310, 197 *.

81+

Hersen M, Eisler R M, Miller P M, Johnson M B & Pinkston S G: Effects of practice, instructions, and modeling on components of assertive behavior. Behaviour Research and Therapy, 11, UU3-U51, 1973. Lomont J F, Gilner F H, Spector N J & Skinner K K: Group assertive training and group insight therapies. Psychological Reports, 25,

1+63-Vro, 1969MacDonald M L, Lindquist C U, Kramer J A, McGrath R A & Rhyne L L: Social skills training: The effects of behavior rehearsal in groups

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

on dating skills. Journal of Counseling Psychology, 22, 2214-239,

1975. McFall R M & Lillesand D B: Behavioral rehearsal with modeling and coaching in assertion training. Journal of Abnormal Psychology, 77, 313-323, 1 9 7 1 . McFall R M & MarstonA R: An experimental investigation of behavior re­ hearsal in assertive training. Journal of Abnormal Psychology, 76, 295-303, 1970. McFall R M & Twentymen C T: Four experiments on the relative contribu­ tions of rehearsal, modeling, and coaching to assertion training. Journal of Abnormal Psychology, 81, 1 9 9 - 2 1 8 , 1 9 7 3 . Rathus S A: An experimental investigation of assertive training in a group setting. Journal of Behavior Therapy and Experimental Psychi­ atry, 3, 81-86, 1972. Rathus S A: A 30-item schedule for assessing assertive behavior. Behavior Therapy, 4,

397-^06, 1973.

Rimm D C, Hill G A, Brown K N & Stuart J E: Group-assertive training in treatment of expression of inappropriate anger. Psychological Reports, 34,

791-798,

19lh.

Rimm D C, Keyson M & Hunziker J: Group assertion training in the treat­ ment of antisocial behavior. Unpublished manuscript, Arizona State University, 1 9 7 1 . Rose S D: In pursuit of social competence. Social Work, 20, 33-39, 1975. Rose S D: Group Therapy: A behavioral approach. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1977a. Rose S D: The effect of contingency contracting on the completion rate of behavioral assignments in assertive training groups. Unpublished

manuscript, University of Wisconsin, Madison, 197Tb. Rose S'D, Caynor J & Edleson J: Measuring interpersonal skills. Social Work, 22, 1 2 5 - 1 2 9 , 1977Rose S D & Edleson J: Interpersonal skill training for social workers. Unpublished manuscript, University of Wisconsin, Madison, 1976. Schinke S P & Rose S D: Interpersonal skill training in groups. Journal of Counseling Psychology, 1976.

Downloaded by [Carnegie Mellon University] at 22:32 02 February 2015

Shoemaker M E & Paulson T L: Group assertive training for mothers: A family intervention strategy. In E J Mash (Ed): Parenting: The change, maintenance and direction of healthy family behaviors. New York: Brunner/Mazel, 1976. Spielberger C D, Gorsuch R L & Luchene R E: STAI manual for the statetrait inventory. Palo Alto, California: Consulting Psychological Press, 1970. Toseland R;

Assertive training and problem solving groups:

A comparative study. Unpublished doctoral dissertation, University of Wisconsin, Madison, 1977. Wolpe J: Practice of behavior therapy, 2nd ed. New York: Pergamon, 1973.

86