A VIDEO-TAPE PRESENTATION OF DESIRABLE WORK BEHAVIORS TO THE‘ - CLIENTS OF A SHELTERED WORK EVALUATION AND TRAINING PROGRAM Thesis for the Degree of Ph. D: MICHIGAN STATE UNIVERSITY DON LEE VAN OSTENBERG 1972 I.) LIBRARY Michigan State University THESIS This is to certify that the thesis entitled A VIDEO-TAPE PRESENTATION OF DESIRABLE WORK BEHAVIORS TO THE CLIENTS OF A SHELTERED WORK EVALUATION AND TRAINING PROGRAM presented by Don Lee Van Ostenberg has been accepted towards fulfillment of the requirements for Ph.D. degreeinDeEartment of Coun- seling, Personnel Se:- vices, and Educational Psychology 0-7639 =3 am‘gmo av ‘1‘ IIOAS 8: SIINS' I“. 4 800K BINIIEBY IIIII. I LIBRARY muons Grog", MICHIGAN :4 A“ I SPRIN ABSTRACT A VIDEO-TAPE PRESENTATION OF DESIRABLE WORK BEHAVIORS TO THE CLIENTS OF A SHELTERED WORK EVALUATION AND TRAINING PROGRAM BY Don Lee Van Ostenberg This study was designed to identify and opera- tionally define the client behaviors desired by the staff of a particular work evaluation and training facility. These desired behaviors were then modeled on video tape and administered to an experimental sample of clients in an effort to more effectively orient and educate them to the behavioral demands and goals of the facility program for the purpose of shortening or improving their training period. Clients who have viewed the video tape of model behaviors for the program should demonstrate better work performance, enhanced interpersonal relationships, greater self-esteem, and general feelings of psychological well- being when compared to those clients who viewed a control video tape. Clients exposed to the model behaviors at the beginning of their program should show greater Don Lee Van Ostenberg improvement in overall functioning than those clients exposed during the middle or at the end of their program. Three behavioral rating scales and two personality tests were used to collect data in a repeated measures design for two of the behavioral scales, and a pre-post test design was used for the remaining behavioral scale and the personality tests. Separate multi-variate analyses of variance were carried out for each of the behavioral ratings, except the monthly evaluation report, and for each of the personality measures, using the sub~ scales as dependent variables and groups and conditions as the independent variables. A two-way analysis of co- variance was carried out on the monthly evaluation report. A cluster analysis using a cOmponent of the official version of the BC TRY System (Tryon—Bailey Associates, Inc.) was carried out using the 81 variables on which data were collected in the present research project. The major finding of this study was that the treatment had no measured effect on the experimental groups when compared to controls over a four-week period of time. A VIDEO-TAPE PRESENTATION OF DESIRABLE WORK BEHAVIORS TO THE CLIENTS OF A SHELTERED WORK EVALUATION AND TRAINING PROGRAM BY Don Lee Van Ostenberg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services, and Educational Psychology 1972 PREFAC E WOrk is essentially a social activity, carried through because of a great array of social expec- tations and demands, both conscious and uncon- scious, overt and covert. The world of work is a kind of subculture, complete with traditions, customs, laws, rituals, compulsions, rewards, and sanctions. We work at certain times, in certain places, and in certain prescribed ways. Training in certain work habits is one of the kinds of training which every organized society, no matter how primitive and underdeveloped, imposes upon its children. --Walter S. Neff, 1968 ii ACKNOWLEDGMENTS This long and arduous task was made into a uniquely gratifying learning experience through the support and cooperation of the Sheltered WOrkshop staff at Pine Rest and the expert guidance of my doctoral com- mittee. Mr. Herbert Start, Director of the Workshop, and Mr. Robert Kieth, Director of Client Services, deserve special mention for their unfailing support of this project. Mr. David Vandergoot made a significant con- tribution to the data collection which was most helpful. The members of my committee, Doctors James Engelkes, chairman, George W. Fairweather, Richard Johnson, and Maryellen McSweeny, gave freely of their time and demonstrated genuine concern for my welfare and the excellence of my endeavors. Dr. Engelkes has served as academic advisor and dissertation committee chairman throughout my doctoral studies. His warm support, his expert guidance, and his friendship are deeply appreciated. iii The loving support of my wife, Diane, and the tolerance of a missing father by my daughter, Jane, are difficult to capture on paper. Their contribution to my effort has been essential and is acknowledged with the utmost gratitude. iv TABLE OF CONTENTS Chapter Page I. THE PROBLEM . . . . . . . . . . . . 1 Need 0 O O O O O O O O O I O O 3 Purpose 0 O O O O O O 0 O O O C 4 Theory. . . . . . . . . . . . . 4 II. REVIEW OF THE LITERATURE. . . . . . . . 8 Wbrk as a Social Role. . . . . . . . 9 Social Model Learning. . . . . 10 Factors Pertinent to Social Modeling. . . 12 Symbolic Models . . . . . . . . . 14 Summary . . . . . . . . . . . . 15 III. METHODOLOGY . . . . . . . . . . . . l7 San‘ple I O O O O I O O O O O C 18 Measures . . . . . . . . 18 Social Modeling Video Tape . . . . . . 25 Procedures . . . . . . . . . . . 27 Hypotheses . . . . . . . . . . 31 Design and Analysis . . . . . . . . 31 IV. ANALYSIS OF RESULTS . . . . . . . . . 34 Cluster Analysis . . . . . . . . . 52 Summary . . . . . . . . . . . . 54 V. SUMMARY AND CONCLUSIONS . . . . . . . . 55 Limitations of the Study. . . . . . . 57 Implications for Future Research . . . . 63 BIBLIWMPHY O O O O O I O O I O O O O 65 APPENDICES Appendix A. Evaluation Report . . . . . . . . . 69 Appendix B. Structured Interview Behavioral Tally . weekly Job Behavior Scale Daily Behavior Check List Functional Self-Attitude Scale. Cluster Analysis . Correlational Matrix: FSAS. vi Page 70 71 75 78 79 85 87 LIST OF TABLES Experimental Design . . . . . . . . Characteristics of the Sample. . . . . Experimental Design for Daily Behavior Checklist. . . . . . . . . . . Summary of Analysis of Variance for Daily Behavior Checklist. . . . . . . . Experimental Design for Weekly Behavior Rating Scale . . . . . . . . . Summary of Analysis of Variance for Weekly Behavior Rating Scale. . . . . . . Experimental Design for Evaluation Report. Summary of Analysis of Co-variance for Evaluation Report . . . . . . . . Experimental Design for 16 Personality Factor Test . . . . . . . . . . Summary of Analysis of Variance for 16 Personality Factor Test . . . . . . Experimental Design for Functional Self- Attitude Scale . . . . . . . . . Summary of Analysis of Variance for Func- tional Self-Attitude Scale . . . . . Correlational Matrix Pre-Post FSAS Scores. vii Page 32 35 38 39 41 43 44 45 47 48 50 51 87 Appendix Structured Interview Behavioral Tally . Weekly Job Behavior Scale Daily Behavior Check List Functional Self-Attitude Scale. Cluster Analysis . Correlational Matrix: FSAS. Page 70 71 75 78 79 85 87 CHAPTER I THE PROBLEM Facilities offering sheltered work evaluation and training are a rapidly growing treatment modality in the field of vocational rehabilitation. The use of such facilities is commonly promoted as good rehabilitation practice by federal and state officials, as well as numerous authorities in the field. The Rehabilitation Services Administration (RSA), United States Department of Health, Education, and Welfare, recently identified and studied 2,656 existing rehabilitation facilities. They conclude that 1,130 of these facilities have defi— nite improvement needs requiring an estimated federal expenditure of $41 million dollars. This RSA survey also indicates that 1,839 new rehabilitation facilities are immediately needed nationally at an estimated cost of $282 million in federal funds (Rehabilitation Facility Needs in the '70's, Monograph #1). Extensive use of facilities by state rehabili— tation agencies is illustrated by the following figures. In 1965, 46,000 handicapped persons were referred to facilities at a case service cost of $19 million. During 1970, referrals increased to 142,000 and expenditures to $77 million. Currently, 16% of state agency clients are referred to rehabilitation facilities and 25% of all case service funds are expended in facilities (Massie, 1971). Rehabilitation facilities may be broadly defined as those facilities which offer services of a rehabilita— tive nature to handicapped persons. Such facilities vary from large metropolitan hospital sheltered workshop (SWS) complexes offering comprehensive medical, social, psy- chological, and vocational services to small rural sheltered workshOps offering work training to only a handful of clients. The facility staff may vary from a director and assistant to a full complement of voca- tional, social, psychological, and medical personnel. In accordance with physical capacity and staffing, a facility may offer a single program or many sophisticated programs and serve a few disability types or all disabled persons. Thus, average or typical characteristics of rehabilitation facilities are not meaningful unless broken down into complex sub-categories. In addition to the above difficulties, there is considerable dis- agreement as to the actual number of facilities existing today. Counts range from 1,200 to 7,000, with the Inter- national Association of Rehabilitation Facilities list- ing a membership of 650. In general, rehabilitation facilities do not pro- vide their associations or other interested organizations with sufficient information to hold them accountable for their functions. When information is provided, compari- sons among facilities are most difficult due to a lack of common accounting procedures. Need As the demand for facility programs increases, the range of disabilities they serve broadens, and their costs continue to inflate, it becomes imperative that the con- tent and goals of these programs be carefully specified. Few facility programs are stated in well-defined behavioral objectives which would enhance measurement of goal attainment and fewer yet make a conscious and controlled use of the learning theory principles upon which most of their programs are based. Rehabilitation facilities apparently suffer from the ills of most social welfare programs, i.e., they possess programs based upon opinion rather than empirical data, insufficient or a total lack of meaningful evalu- ation of goal attainment, and little commonality of accounting procedures. The need for field research in facilities is clear and the increasing demand for an accounting by those who pay the costs makes it imperative. Purpose This study was designed to identify and oper- ationally define the client behaviors desired by the staff of a particular work evaluation and training facility. These desired behaviors were then modeled on video tape and administered to an experimental sample of clients in an effort to more effectively orient and educate them to the behavioral demands and goals of the facility program for the purpose of shortening or improv- ing their training period. Clients who have viewed the video tape of model behaviors for the program should demonstrate better work performance, enhanced interpersonal relationships, greater self-esteem, and general feelings of psychologi- cal well-being when compared to those clients who viewed a control video tape. Clients exposed to the model behaviors at the beginning of their program should Show greater improvement in overall functioning than those clients exposed during the middle or at the end of their program. Theory The basic assumptions of this study are pre- dicted upon social model learning theory. Social imi- tation, or modeling, is a broadly based concept in learning theory which has been addressed by many writers. Stugart (1970) provides an extensive review of this literature dating back to the earliest formulations of the concept. Bandura (Bandura & Walters, 1963; Bandura, 1965, 1969) and his collaborators are most closely identified with theories of imitation or social modeling at this time, and have produced the major research con- tributions to this area of learning. To Bandura (1969) it seems likely that autonomic and instrumental responses can be learned, and conditioned emotional responses extin— guished, as a result of social imitation. He theorizes that no-trial learning occurs because the observer does not engage in any gzggt responding trials, although there may be many observational trials. He indicates that two representational systems are necessary for this type of learning, an imaginal and a verbal one. These representations function as mediators for the response retrievals and reproductions. Bandura distinguishes between acquisition of responses and performance of responses. The consequences of the model's behavior seem to affect the latter, with positive reinforcement greatly improving the possibility that a subject will actually perform a model's acts. Bandura concludes that vicarious modeling and reinforcement of the model's behavior are probably as effective as live modeling and subsequent reinforcement of the subject's behavior. A secondary assumption upon which this study rests is Bandura's (1969) assertion that social learning procedures produce correlated changes in attitudes. He states that . . . unlike the inconsistent effects of persuasive communications, desensitization and modeling approaches are capable of producing correlated changes in refractory behavior and attitudes. These positive relationships range for r = .39 when changes in attitudes are measured by seman- tic differential technique to r = .59 and r = .72 for change scores based on a variety of attitude scales (Bandura, Blanchard, & Ritter, 1968; Blanchard, 1969, p. 597). Bandura further states that these correlated changes in attitudes and behavior can be interpreted according to most contemporary attitude theories by positing the existence of a strong drive for consistency among beliefs, feelings, and actions. Bandura, Blanchard, and Ritter (1969) suggest that . . . an alternative interpretation is that social influences have similar but independent effects on attitudes, behavior, and emotional arousal. In this view, attitude-behavior consistencies represent correlated coeffects rather than outcomes of a process in which modification of type of behavior forces changes in other forms of responding to eliminate cognitive disequilibrium [p. 196]. They further suggest that these alternative theoretical formulations should be regarded as complementary rather than as competing. In summary, Bandura (1969) concludes that . . . one of the fundamental means by which new modes of behavior are acquired and existing patterns are modified entails modeling and vicarious processes. Indeed, research conducted within the framework of social learning theory demonstrates that virtually all learning phenomena resulting from direct exper- iences can occur on a vicarious basis through observation of other person's behavior and its con- sequences for them. . . . Modeling procedures are, therefore, ideally suited for effecting diverse out- comes including elimination of behavioral deficits, reduction of excessive fears and inhibitions, trans- mission of self-regulating systems, and social facili- tation of behavioral patterns on a group-wide scale [p. 118]. CHAPTER II REVIEW OF THE LITERATURE This review will begin with Neff's (1968) state- ment of work as a social role. This statement has explan- atory value for the program and population under study and provides a logical rationale for the use of a learn- ing theory approach, and particularly social modeling techniques, to produce change in people needing sheltered work evaluation and training programs. Studies important in the recent development of social model learning theory will then be reviewed with particular attention being paid to the range of behaviors and types of clients to whom the techniques have been applied. Following this section, three aspects of social modeling will be treated briefly: attentional factors or cues; vicarious reinforcement; and characteristics of the model. From this review a rationale will flow for the use of video tape in the present treatment process. Studies which use video tapes of social models in teach- ing roles will be briefly reviewed. Finally, a summary of the above research will be offered and the unique characteristics of the present research will be noted. Work as a Social Role Neff (1968) states that The relatively new professional field of vocational rehabilitation has had to recognize that it is often not sufficient to provide a disabled person with a prosthetic device or to train him in a new job-skill. To a greater extent than most industrial psycholo- gists, the rehabilitation eXpert has been made aware that the ability to work is not only the function of a given set of cognitive and motor aptitudes, but also a wide range of other factors, some of which arise because most kinds of work today are carried on in quite complex social settings and require the ability to adapt to a great variety of interpersonal demands [p. 26]. In summarizing the components of the work personality, Neff further states that The general source of what we are calling the work personality lies in the precepts of society that an individual should play a productive role. . . . All children are faced with a demand to achieve, since the process of growth to adulthood involves the mastery of a great many biologically and culturally defined tasks. . . . People may work poorly or may not be able to work at all for varied reasons. Some may not have internalized the precepts of society to play a productive role; they may not have become enculturated to the work subculture. In other instances, the general demand to be productive may have internalized, but may be in conflict with other demands of the work situation-~the requirement to relate appropriately to peers and authority figures on the job, the requirement to moderate certain needs for intimacy and privacy, the requirement to meet standards imposed by others, and so on [p. 164]. Neff concludes that work behavior is largely the function of a semi-autonomous area of personality. He further believes that adapting to work is, in essence, 10 a process of assimilating and internalizing a set of cultural norms and practices which can be defined as the subculture of work. It seems logically to follow from these formulations that the provision of effective social models of desirable behaviors for the subculture of work would provide an efficacious treatment program for those who cannot compete in the world of work for other than physical reasons, and are in need of sheltered employment and training. Social Model Learning The theoretical development of social modeling or imitation in learning theory, including the research it has spawned, has recently been extensively reviewed by several authors (Bandura, 1969; Dahm, 1970; Krieger, 1969; LaFleur, 1970; Stugart, 1970). Dahm, LaFleur, and Stugart unanimously conclude, in reviews which include Bandura, that social modeling has been shown to be an effective means for changing many behaviors. Recent research on social modeling has generally fallen into three categories of application. The first involves the use of various modeling procedures with children, and this literature has been reviewed exten- sively by Bandura and Walters (1963) and Bandura (1965, 1969). The studies reviewed are typically of attempts to extinguish some unacceptable behavior such as ll aggression, or to teach a new skill. The results of these studies have generally been positive. A second category of research in social modeling is comprised of studies dealing directly with the treat- ment of pathological behavior, usually in adults. These studies have also been extensively reviewed by Bandura (1969). The recent study of Bandura, Blanchard, and Ritter (1969) is an excellent example of this type of research. These investigators compared desensitization and modeling approaches in reducing the fear arousal of snake phobic subjects. Subjects were administered either live modeling combined with contact desensiti— zation, symbolic modeling, symbolic desensitization, or no treatment. Bandura and his colleagues concluded that while all methods used produced fear reduction, modeling with guided participation produced the most positive results. A third category of research in social modeling is that of using models to increase the effectiveness of counseling. For example, Krumboltz and his associates (Krumboltz & Schroeder, 1965; Krumboltz & Thoresen, 1964; Krumboltz, Varenhorst, & Thoresen, 1967; Thoresen, Hosford, & Krumboltz, 1970) have studied the use of models to increase information seeking behavior in young adults. The criterion measures were typically taken two to three weeks after the experimental 12 counseling session and consisted of subject self-reports on the frequency and variety of their information-seeking behaviors. Myrick (1969) used video and audio tapes of models making frequent self—references to increase that type of behavior in the initial interview of eighth—grade stu- dents. He obtained positive results testing his hypothe- sis at the .25 level of significance. Whalen (1969), in an often-quoted study, found that a film model of interpersonal openness accompanied by detailed exhortative and descriptive instructions sig- nificantly increased these behaviors in groups of four male college students. Heller (1969), after reviewing the effects of modeling in helping relationships, concludes that such techniques can be important facilitators of change. He feels, however, that the limited number of studies using modeling techniques in clinical settings impedes the use- fulness of modeling in clinically helpful relationships. Heller cites the need for research designed to clearly Specify the uses and limitations of modeling techniques. Factors Pertinent to Social Modeling Reviews of social modeling literature by Bandura (1969), Dahm (1969), LaFleur (1970), and Stugart (1970) all indicate quite clearly that subjects exposed to 13 social modeling must perceive a high probability that their imitation of the model's behavior will be rewarded, to insure their performance of it. These literature reviews further indicate that vicarious reinforcement, i.e. the model being reinforced, is sufficient reinforce- ment for the subject. Some attention has been paid to the characteris- tics of a good model for various subjects. Recent studies of model characteristics include that of Thoresen, Hos- ford, and Krumboltz(l970), who studied the effects of models of varying competencies upon subjects of varying competencies and found the effectiveness of a model for their high school sample was moderated by differences between students from particular schools, in addition to home-community environments. Stugart (1970) studied the effect of adult versus peer models and of the model's race upon subjects. Bandura (1969) has reviewed other investigators who have studied the effects of the model's prestige, his perceived intelligence, and his social status upon subjects, all with generally inconclusive results. Bandura (1969) and Kagan (1967) suggest theo- retically that the characteristics of a model may only serve as attention-getting devices. Cues which draw attention to modeled behaviors have been shown to facilitate learning by aiding the observer in the process of mental coding (Bandura, 1969). l4 LaFleur (1970) found that adding "pre-organizers" to the modeled behaviors, prior to their presentation, greatly enhanced his subject's ability to learn the desired modeled behavior and increased their generalizability. Both "pre-organizers" and "post-summaries" were used in the present study. Symbolic Models Few investigators have found it practical to use live models in social modeling experimentation. There seems to be general agreement that symbolic models are as effective as live models (Bandura, 1969; Dahm, 1970; LaFleur, 1970; Stugart, 1970). Video tape was selected for the present study because its audio and visual proper- ties were thought necessary to model the complexity and breadth of behaviors selected for the experiment. Recently, portable video-tape equipment has become widely available and it has been used extensively in education and training (Public Health Service Publication No. 1862, Part IV, 1969) and to achieve psychotherapeutic goals (Donk, 1970). Although it has been used primarily by psychotherapists to produce self-confrontation in clients or trainees, some investigators have made effec- tive use of video-tape models with others' behaviors as a teaching device. Higgins, Ivey, and Uhlemann (1970) used the video method of "media therapy" to train clients in "direct, mutual communication." Their video models 15 emphasized listening and sharing in a "here and now" con- text. The experimental subjects of this study were rated significantly better on two interaction scales than the controls. Eisenberg and Delaney (1970) found that modeling procedures were effective in teaching counselor trainees response leads which would evoke client responses stated in behavioral terms. Summary Neff's (1968) conceptualization of the work personality, and of work as a subculture, provides a theoretical rationale for facilities offering sheltered evaluation and training and especially to those clients who have little or no work experience. It is for this population that social model learning theory seems par- ticularly appropriate. Numerous recent reviews of social model research have been cited and the research reviewed here has been generally divided into three basic cate- gories. The variables of model characteristics, attentional cues, and vicarious reinforcement were briefly reviewed. It was concluded that symbolic model- ing may be as effective as live models, and support for using video tape as a medium was cited. The present study differs from the previously cited research in its methodology for selecting behaviors 16 to be modeled; the breadth of behaviors modeled; the extensiveness of outcome measurement; the breaking down into phases of both treatment and measurement; and the population studied. These factors will be further examined in Chapter III. CHAPTER III METHODOLOGY The work evaluation and training clients of the Pine Rest Christian Hospital Sheltered WOrkshop, Grand Rapids, Michigan, were chosen as the population for this study. This facility has a total enrollment of some 250 clients, 40 to 70 of whom are usually in rehabili- tation status. The rehabilitation program is staffed by two master's level rehabilitation counselors, a work evaluator, a rehabilitation aide, and a job placement coordinator. The program is directed by a master's level Director of Client Services who reports to the Director of the Sheltered Workshop. The work evaluation and training clients are enrolled in the work sample evaluation center, an on- the-job evaluation program, or a work training program on one or more of 15 different work stations. On the average, six out of ten clients enrolled in this program are housed in the hospital's halfway houses located near the sheltered workshop. Group counseling is the primary modality used for 17 18 personal-social and vocational counseling. Clients may also be assigned to classes which-teach social skills. Recreation and meals are provided by the program. Sample The sample for this study was comprised of all work evaluation and training clients in rehabilitation status who were enrolled in the program during the months of August, September, and October, 1971, with two exclusions: clients who were blind or deaf. Although the sample was drawn in time, an examination of the records indicated no reason to assume that this sample is atypical of the population served by this workshop. Measures Bozarth (1970) in an article on applied research, made a plea for "more than one measure of outcome pref- erably diverse measures [p. 23]. . . ." Caro (1971) writing on the evaluation of social programs, also advocated comprehensive measurement. As behavioral measures are commonly used in research on social modeling (Bandura, 1969) three behavioral rating scales of varying objectivity are included in the present project to provide a reality based objective measure of change. 19 Although Fairweather and his colleagues (Fair- weather gt_al., 1960, 1964, 1969) have repeatedly demonstrated that little relationship exists between measures of personality and behavior, Bandura (1969) in a study quoted earlier cites contrary evidence. At least one investigator (Eber, 1967) has demonstrated that personality traits are related to outcome in rehabili- tation research. Lane and Barry (1970) in their review of recent research on client motivation for rehabili- tation substantiate the need to measure personality variables in rehabilitation research projects. For these reasons it was thought that the use of behavioral measures alone would be an unnecessary limitation on the present study. In addition, con- ceptualization of the present treatment as the social modeling of desirable behaviors in a subculture of work, seems to logically dictate a number and variety of out- come measures. The three behavioral rating scales and the two personality tests, selected to gather data for this study, are described below. Evaluation Report.--This rating form was filled out by a client's supervisor at the end of each working month and/or at the end of a client's training on a particular work station. The form was used by the Pine Rest Sheltered WOrkshop prior to the present study, and was designed to measure a client's performance compared 20 to community standards in nine general areas and on numerous specific work skills peculiar to a work station. A five-point rating scale was used for the present study and anchor points were established. The scale score used in this study was expressed as a percentage of top performance achieved (see Appendix A). The work super- visors were retrained on rating with this scale. Because each Evaluation Report is highly specific to a work station, no inter-rater reliability could be computed across trainers. Pre- and post-treatment data were collected on all clients using this scale. Weekly Job Behavior RatingyScale.--This scale was filled out at the end of each working week on each client by his work trainer. As a behavioral rating scale which adequately measures the desirable behaviors taught in a particular facility was crucial to this study, the development of this scale will be described in detail. The work trainers, vocational counselors, and supervisory personnel were interviewed and asked to list and describe the behaviors they were attempting to teach the Sheltered Workshop clients. These behaviors were then presented by the vocational counselors to the clients for their reaction in the form of a structured interview (see Appendix B). Seventy of these structured interviews were tape recorded and a behavioral tally was made of the clients' responses (see Appendix C). 21 For example, counselors and trainers thought that the client's relationships with his co-workers was an impor- tant behavior in their program. In the structured inter- views, the clients were asked "What do you like about the people you work with?" and "What do you like most about your job?" and finally, "What did you learn about people this week?" "How did you learn it?" Clients responding to the structured interview were placed in three groups on the basis of their length of enrollment in the facility's program. No appreciable differences in behavior responses were found among the three groups. The behavioral responses were then ordered for each question in terms of frequency of occurrence. The behaviors occurring more than once in the tally were then used as the criteria for selecting a rating scale (see Appendix C). The behavioral tally was also used to script the modeling tape used as the treatment in the present study. As a result of the above process, behaviors were modeled and clients' performance on them rated, which were thought to be important by both staff and clients. Sensitivity of the criterion measure to the treatment method was insured by using the same pool of behaviors for the modeling tape and the Weekly Behavior Rating Scale. Bozarth (1970) states that outcome measures in applied research "should be standard measures which have been used in other studies. Too often the measures 22 are esoteric and few in numbers." In agreement with Bozarth's plea for enhancing the comparability of applied research, existing job behavior scales were compared to the present behavioral tally. A scale which had been used previously in extensive research by Fairweather (Fairweather gt_§1,, 1960) and his colleagues was selected for use in the present study because its items incorporated most of the tallied behaviors and its simple format encouraged high inter-rater reliability. This scale was adapted for use in the present study by changing the word "patient" to "client" in all items, and changing the wording or verb tense of each item so that the items read in the past tense. Four items, representing heavily tallied behaviors, written in similar format, were added to the Fairweather scale and it was given the title "Weekly Job Behavior Scale" (see Appendix D). Subjective evaluation of the scale quite clearly indicates that items 1 through 18 measure behaviors related to accomplishing a task. Items 19 through 35 appear to measure interpersonal relation- ships on the job. These two groups of items were thus scored separately for analysis. Reliability for the instrument was computed using the intra-class method (Stueland, 1971). The reliability of all raters participating in the last two training sessions was .672. When two grossly deviant raters were 23 excluded, the reliability was .917. The two deviant, raters were given further training, before the present study was begun, resulting in their ratings becoming almost identical to the other raters. To check the pos- sibility that the raters as a group began to rate more stringently as they gained familiarity with the instru- ment, a pre-treatment and post-treatment average rating on the training video tape was computed. The pre- treatment means were 14.07 and 12.61 for the two scales. The post-treatment means were 14.14 and 13.57 respectively. The Weekly Job Behavior Scale was incorporated into the ongoing assessment program of the Pine Rest Sheltered Workshop. Client ratings taken during the four weeks of each client's treatment period were utilized in the present study. Dailnyehavior Checklist.--This behavioral check- list was created to provide a purely objective behavioral count, in contrast to the more subjective measures above. Ratings were made on all clients daily by their trainer which were averaged into a weekly score and included as part of the research data (see Appendix E). Average scores were obtained on four subscales: Attendance, Appearance, Promptness, and Perseverance. These sub- scales were created subjectively to gain more specificity in the data. Data were collected on each client during the four weeks of his treatment period for the present study. 24 16 P.F., Form E.--This 1967 experimental form of the 16 P.F. is intended for "use with persons of limited educational and cultural background" (Interim Manual Supplement for Form B, 16 Personality Factor Question- naire, 1968). Sufficient data have not been gathered on this form to permit the calculation of validity and reliability coefficients. All forms of the 16 P.F. are assumed by their authors to measure the same personality factors and possess the same statistical properties, therefore assumptions concerning the utility of this form can be based in part upon extensive research on the other forms (Handbook for the 16 P.F., 1970). This test was included as a criterion measure for purposes of general personality assessment. The low— literate Form E is intended for use with persons who read at the sixth-grade level or below and who may suffer from cultural deprivation. The test questions can be tape recorded for use with clients whose reading skills fall below the third-grade reading level. This tape recorded administration was used with the current sample. The 16 P.F., Form E was administered at the beginning and at the end of the four-week treatment period. The norms for males and females found in the Interim Manual Supplement for Form E were used to convert the raw scores into standard ten-point scores (stens). 25 The sten scores were used to calculate personality change in the present sample. Functional Self-Attitude Scale.--This experimental measure was developed by Holbert (1971) to measure the attitudes rehabilitation clients have about themselves. The test was initially developed for use with a client population quite similar to the present one. This 60- item scale uses a Likert-type five-point response scale and is divided into eight sub-scales, including a Lie Scale, and a total score. Its reading level has been determined not to exceed grade 3.9. The scale is cur- rently being used experimentally at the University of Kentucky Counseling Center and in several rehabilitation projects (Holbert, 1971). No norms, validity, or reliability data are available as yet (see Appendix F). This test was administered before and after the four-week treatment period of the present study and raw sub-scale scores were recorded for use in analysis of the data. Social Modeling Video Tape A rough outline of the script, based upon the behavioral tally described above, was written for pro- duction of the experimental video tape. A model was selected from the active treatment unit of Pine Rest Christian Hospital to avoid the possible confounding 26 variable of using a current Sheltered workshop client as a model for the other clients. The model selected was an l8-year—old male Caucasian with average intelligence. His style of dress and mannerisms were those which seemed typical of a young man in his age range. His performance on the tape was voluntary and he seemed genuinely inter- ested in playing the roles assigned to him. The tape was scripted to tell the story of a successful facility client and was edited to 30 minutes as a reasonable length of time in consideration of client attention span and in practical consideration of staff and clients' time away from their work. The story began with a scene of the model arriving at the job on which he was placed after successful completion of his program. Reinforcers such as a desirable job, a car, and a nice place to live were built into the early scenes. The film then flashed back in time to the model's first day at the facility and showed him moving through intake procedures and being assigned to a job in the Sheltered Workshop. His first day on the job was then portrayed with undesirable work behaviors and contrasted with a second day on the job some time later in which he modeled most of the desirable behaviors appearing on the behavioral tally (see Appen- dix C). The video tape ended with a reinforcing inter- view of the model by his work supervisor, underscoring his good work behaviors. The model client was then 27 referred to the job placement specialist and placed on the job seen at the beginning of the tape. Summations of pertinent behaviors and cues to forthcoming behaviors were edited into the tape during transition between scenes. As dialogue was not scripted, several tapings were required to gain the best possible performance of model behaviors. This procedure, although more time- consuming, allowed the actors to be spontaneous and natural, which seemed to enhance the credibility of their amateur performance. A second video tape was made for viewing by the control groups. This tape was the same length as the experimental tape (30 minutes) and showed scenes of the same buildings, however, without people being present. An off-camera narration consisted of a description and history of the buildings and grounds which were filmed. Procedures All subjects in the work evaluation and training programs at the beginning of the experimental period were divided into three groups based upon their length of enrollment at the facility. The groups were established in consultation with the facility staff who felt that clients in the program 13 to 49 weeks comprised a group which they considered long-term rehabilitative cases or clients who were in a holding status for placement, and 28 therefore should be minimally affected by the treatment. These clients were labeled Group I. Group II was com- prised of clients enrolled in the program from 2 weeks to 12 weeks. The above two groups of clients had begun their program prior to the experimental period and were therefore exposed to the treatment immediately. All clients admitted to the facility during the experimental period were labeled Group III. The experimental period (as contrasted with the treatment period) was defined as that period of time which was necessary to accumulate an adequate number of Group III clients to allow appropriate statistical analysis of the data. Thirty clients were admitted to the facility's program during this period and remained for a time long enough to include them in the four-week treatment program. Randomization of subjects was accomplished by listing all clients in the program immediately prior to the experimental period in chronological order according to number of weeks in their program. Subjects were then randomly assigned to treatment or control groups within each week and again randomly assigned to treatment days Monday or Tuesday. Subjects within treatment and control groups were again randomly assigned to viewing groups of five or six subjects and treatment and control groups were alternately shown their tape during the two treat- ment mornings. These last several procedures were 29 necessary because of the size of the initial group, the size of the video tape viewing room, and a desire to hold the variables of time and fatigue constant. A table of random numbers was used for randomization procedures. All new clients who entered the facility's pro- gram during the experimental period were randomly assigned to treatment or control groups each week and again ran- domly assigned to a viewing morning. After the initial large group of subjects finished the four-week treatment period, all subjects viewed the video tape on Monday mornings for economy in procedure. All subjects in Groups I and II were administered all measures at the beginning of the experimental period. For the following three weeks they were shown the experi- mental or control film at the beginning of each week and measured at the end of each week on the Weekly Behavior Rating Scale and at the end of each day on the Daily Behavior Checklist. At the end of the treatment period (Week 4) all measures were again administered to the sub- jects. Personality measures were administered by the staff vocational evaluator. Subjects were reminded of appointments for testing and times for viewing their video tape by the work trainers on the basis of a weekly schedule which was made available to them. Both experimental and control video tapes were shown in the Group Room of the Client Services office. 3O Clients were greeted by the experimenter and told that they would be viewing a video tape of the program so that they could learn more about it. They were then seated in the Group Room, the tape was turned on, and the experi- menter left the room. At the end of 30 minutes the experimenter re-entered the room, turned off the video tape, and thanked the subjects for their attention. They were then dismissed and returned to their work stations. All subjects were told that they would be seeing this video tape three times so that they could benefit from it fully. Assignment of subjects to treatment and control groups was known only to the experimenter. The work trainers who served as raters were not informed of the content of the film and as the subjects of any one view- ing group were spread throughout several work stations, raters did not learn whether their clients were experi- mental or control subjects. All subjects were rated by their trainers each Friday and the forms were returned to the Client Services office by 5:00 P.M. The forms were scored and the data entered on data sheets by the following Monday noon so that a check could be.made for missing data. Any data missing were then immediately collected and entered on the data sheets, thus assuring maximum objectivity of the behavioral ratings. 31 Hypotheses The following hypotheses were tested: 1. Experimental subjects in all treatment groups will receive higher behavioral ratings than the con- trol subjects on the post-treatment measures. 2. There will be a differential effect on treatment as a function of groups and of time on the behavioral rating scales. 3. Experimental subjects in all treatment groups will receive more positive scores than the control sub- jects on the post-treatment personality measures. 4. There will be a differential effect on treatment as a function of groups and of time on the personality measures. Design and Analysis The design of the study is shown in Table 3.1. The basic pre-test/post-test control group design was used with the addition of repeated treatments and repeated measures with the behavioral ratings. Separate multi-variate analyses of variance were carried out for each of the behavioral ratings, except the monthly evaluation report, and for each of the personality measures, using the subscales as dependent variables and groups and conditions as the independent variables. A two-way analysis of co-variance was car- ried out on the monthly evaluation report. A cluster analysis using a component of the official version of the BC TRY System (Tryon-Bailey 32 mcflumm Hmuofl>mnmm aaxooz can unwaxoocu now>moom maflmo sues mcwumm .mamom mcwumm uoa>mcmm maxomz one umHonmoO How>mcmm mawmo ouH3 mcflumm .mucoasnumcfl o>onm on» mo Ham nufl3 mcwumm one mcwumoa II D .mamom II U as m a. .unomom coflpmsam>m cam ..m.m ma may .mamom mosufluufilmaom Hmcofiuocsm mo coflumuumflcflfio< u d i ox ox ox oxm mu H xmmz um mcwumucm mucoflau Ham ox ox ox oxm mm HHH macaw ox ox ox oxm vo maum mxomz ca mucmflau Ham ox ox ox oxm mm HH msouo ox ox ox oxm mo manna mxomz ca mucmaao Haa ox ox ox oxm Hm H ozone an s0 «m «4 e m m a mxmmz unosumoua cmflmoo Hmucoaflummxm H.m mqm¢9 33 Associates, Inc.) was carried out using the 81 variables on which data were collected in the present research project. CHAPTER IV ANALYSIS OF RESULTS Table 4.1 presents the characteristics of the sample used in this study. The clients were predominately white teenagers and young adults who were schooled in special education programs. Most frequently they had not been previously employed and had little or no experience in what Neff (1968) calls the "subculture of work." The sample was almost equally male and female. The majority of clients had mental retardation or a psy~ chiatric diagnosis listed as their primary disability. This classification of clients was strongly influenced by the fact that 51 of the 69 subjects were Vocational Rehabilitation Services (VRS) referrals and thus had been fit into the VRS classification system. No attempt was made by the Pine Rest staff to reclassify the clients as the treatment program was focused on the individual client's behavior regardless of his disability label. To further understand the characteristics of this sample, it is worth noting that the Pine Rest Work 34 35 TABLE 4.1 Characteristics of the Sample Characteristic Mean Mode Standard DeV1ation Age in years 24.5 19 8.87 Education in years includ- ing special education 10.2 12 3.41 Years of previous employment 1.9 0 4.97 Sex Males - 33 Females - 36 Disability Mental Retardation 25 Mental Retardation and others 8 Psychiatric l9 Psychiatric and others 2 Physical 13 Sociogenic 2 Referral Source Vocational Reha- bilitation Services 51 Private 7 Coldwater State Home 4 G.R. Consultation Center 2 Pine Rest Christian Hospital 2 Kent Occupational High School 1 Disability Determi- nation Service 1 Veterans Adminis- tration 1 Housing During the Program Pine Rest Halfway Housing 44 Private Home 25 36 Evaluation and Training Program is frequently seen as a "dumping place" or a program of last resort by community agencies and thus the clients may be more severely dis- abled than the tabled disability classification indicates. No objective measurement has been made of this assertion. The analysis of the five criterion measures used in this study will be presented by first restating Hypotheses l and 2. The analysis of the three behavioral measures which test these two hypotheses will then be presented along with a table portraying the experimental design used with each measure. Following this presen- tation Hypotheses 3 and 4 will be restated and the analysis of the two personality tests used as criterion measures for these hypotheses will be presented with their respective experimental designs. Hypothesis 1: Experimental subjects in all treatment groups will receive higher behavioral ratings than their control subjects on the post-treatment measures. Hypothesis 2: There will be a differential effect on treatment as a function of groups and of time on the behavioral rating scales. Three behavioral rating scales were used as cri- terion measures to test the above hypotheses--the Daily Behavior Checklist, the Weekly Behavior Rating Scale, and the (monthly) Evaluation Report. The experimental 37 design and analysis of variance table for each measure is presented below. Each summary of analysis table presents the analysis of variance F value, its proba- bility of occurrence, its source in the analysis, and the degrees of freedom used in its computation. The .05 level of significance was used in all analyses. Table 4.2 displays the design used for the Daily Behavior Checklist. The Checklist consisted of four behavioral ratings which were averaged for each subject for each of the four weeks of the experimental period, providing measures one (Ml) through four (M4). The groups of subjects denote length of enrollment in the SWS program, and the number of subjects in each con- dition is shown. No subjects were lost from any con- dition on this measure. Table 4.3 presents the summary data from the multivariate analysis of variance carried out on the Daily Behavior Checklist using the Finn program with a repeated measures design. The first three entries on Table 4.3 indicate that this instrument did not measure any significant differences between treatment and control subjects due to time in the program (groups); the treatment admin- istered (conditions); or an interaction of these two factors. The first hypothesis is therefore not sup- ported by analysis of the data produced by the Daily Behavior Checklist. 38 TABLE 4.2 Experimental Design for Daily Behavior Checklist Group 1 U) o o o o H Week 1 Week 2 Week 3 M1 M2 M3 M4 M1 M2 M3 M4 M1 M2 M3 M 4 M M M M 1 Week 4 2 3 4 Group 2 Group 3 7—.- __ 39 TABLE 4.3 Summary of Analysis of Variance for Daily Behavior Checklist Degrees of Source Freedom F P Groups 2,63 0.60 0.5532 Conditions 1,63 0.86 0.3558 Group X Conditions 2,63 0.20 0.8164 Measures Across Time 15,49 2.23 0.0179* Times 3:61 0.24 0.8691* Measures 3:61 4.04 0.0110 Time X Measures Interaction 9,55 0.32 0.9659 Measures Across Time X Groups Interaction 30,98 1.14 0.3125 Times X Group Interaction 6,122 0.84 0.5394 Measures X Group Interaction 6,122 2.23 0.0445* Times X Measure X Groups ’ Interaction 18,110 0.69 0.8171 Measures Across Time X Conditions Interactions 15.49 0.77 0.6969 Times X Conditions Interactions 3,61 1.05 0.3952 Measures X Conditions Interactions 3.61 0.28 0.8356 Times X Measures X Conditions Interaction 9,55 1.05 0.4143 Measures Across Time X Groups X Conditions Interaction 39,98 0.77 0.7923 Times X Groups X Conditions Interaction 6,122 1.78 0.1087 Measures X Groups X Conditions Interaction 6,122 0.21 0.9741 Times X Measures X Groups X Conditions Interaction 18,110 0.54 0.9338 * .05 level of significance used for this study. 40 The grand mean of measures across time is sig- nificant (P = .0179). This significance seems due pri- marily to the measures themselves (P = .0110) which are of different traits measured on four subscales across for four weeks and thus the comparison is of no real value. The significance of a measures by group inter- action (P = .0445) indicates that the four measures of different traits on this instrument (attendance, appearance, promptness, and perseverance) interact in a different manner with each of the six groups of sub- jects. This significance seems primarily an artifact of the measures themselves as stated above and should be qualified by noting that the overall effect of measures across time by groups was not significant. The lack of significant differences found in all the analyses of interaction effects other than measures clearly indicates a lack of support for the second hypotheses positing a differential effect on treatment as a function of groups and of time. Table 4.4 portrays the experimental design for the Weekly Behavior Rating Scale used as a criterion measure for Hypotheses l and 2. As in the previous design the groups refer to time of the subjects in the SWS program. The weeks (1-4) represent the experimental period for these subjects and the two scores derived 41 TABLE 4.4 Experimental Design for Weekly Behavior Rating Scale Group 1 week 1 Week 2 Week 3 M1 M2 M M M M l 2 l 2 Week 4 M M Group 2 Group 3 42 1 and M2 in the design. No subjects were lost from this design during from the scale are represented by M the experimental period. Table 4.5 presents the summary data from the multivariate analysis of variance repeated measures design, using the Finn program, for the Weekly Behavior Rating Scale. There are no significant differences due to length of time in the SWS program; experimental con- ditions; time of the treatment period; or an interaction effect in any combination of the above three factors. Measurements with the Weekly Behavior Rating Scale do not therefore support either of the two hypotheses stated above. Table 4.6 presents the experimental design used for the (monthly) Evaluation Report. The groups (1-3) refer to length of time of the subjects in the SW8 pro- gram. The number of subjects in each cell is shown on the design. The Evaluation Report was administered pre- and post-treatment period with the pre—rating being used as a covariate in the analysis. Table 4.7 presents the summary data of the analysis of covariance for the Evaluation Report. No significant differences were found between groups, between experimental and control subjects, or due to an interaction of these two factors. 43 TABLE 4.5 Summary of Analysis of Variance for Weekly Behavior Rating Scale Degrees of Source Freedom F P Groups 2,63 1.14 0.3249 Conditions 1,63 1.47 0.2301 Group X Conditions Interaction 2,63 1.17 0.3159 Measures Across Time 7,57 1.43 0.2121 Times 3,61 1.59 0.2010 Measures 1,63 1.50 0.2260 Time X Measures Interaction 3,61 0.81 0.4935 Measures Across Time X Groups Interaction 14,114 0.70 0.7693 Times X Groups Interaction 6,122 0.60 0.7326 Measures X Groups Interaction 2,63 1.38 0.2601 Times X Measures x Groups Interaction 6,122 .74 0.6160 Measures Across Time X Conditions Interaction 7,57 1.38 0.2324 Times X Conditions Interaction 3,61 1.23 0.3054 Measures X Conditions Interaction 1,63 0.07 0.7985 Times X Measures X Conditions Interaction 3,61 2.31 0.0848 Measures Across Time X Groups X Conditions Interaction 14,114 0.39 0.9754 Times X Groups X Conditions Interaction 6,122 0.26 0.9559 Measures X Groups X Conditions Interaction 2,63 1.54 0.2234 Times X Measures X Groups x Conditions Interaction 6,122 0.97 0.2244 * .05 level of significance used in this study. 44 TABLE 4.6 Experimental Design for Evaluation Report Groupil Group 2 Group 3 l S 1’ S 1 Experimental ' ' ' Conditions ' ' ' 811 S 9 814 S 1 S 1 S I Control ' ' ' Conditions ° ' ' S 9 S11 815 45 TABLE 4 . 7 Summary of Analysis of Co-variance for Evaluation Report Sources Degrees of Freedom F ‘P Groups 2,62 2.29 0.1095 Conditions 1,62 1.73 0.1931 Group X Conditions Interaction 2,62 1.17 0.3160 .05 level of significance used in this study. 46 Hypothesis 3: Experimental subjects in all treatment groups will receive more positive scores than their control subjects on the post-treatment personality measures. Hypothesis 4: There will be a differential effect on treatment as a function of groups and of time on the personality measures. Two personality tests, the Functional Self- Attitude Scale and the 16 P.F., were used as criterion measures for the above two hypotheses. Table 4.8 presents the experimental design for the 16 P.F. The groups are the same as those used in the above designs and refer to the length of time in the SWS program for the subjects. Subjects are shown separ- ately for pre- and post-testing as a few were lost from groups two and three due to intellectual inability to take the test or a failure to administer post-testing before the subject left the program. Ml . . . . . M16 denotes the 16 scores repre- senting the 16 personality factors measured by this test. Table 4.9 presents a summary of the multivariate analysis of variance, repeated measures design, for the 16 P.F. 'The Finn program was also used for this analysis. Primary analysis of the data indicates that dif- ferences between groups and between experimental and Experimental Design for 16 Factor Test 47 TABLE 4.8 Personality Time 1 (Pre) Time 2 (Post) M1 000...... M16 M1 0000...... M16 S 1 S 1 E O O 811 811 Group 1 $12 812 C ' ' $20 820 821 821 E ‘ ' 829 528 Group 2 S30 829 C O 0 S40 S37 S41 S38 E ' ' 555 545 Group 3 $54 850 C ' ' S66 S61 48 TABLE 4.9 Summary of Analysis of Variance for 16 Personality Factor Test Degrees of Source Freedom F P Groups 2,55 0.06 0.9364 Conditions 1:55 0.45 0.5039 Group X Conditions Interaction 2:55 1.68 0.1953 Measures Across Time 31.25 7.63 0.0001* Times 1:55 0.62 0.4344 Measures 15.41 10.46 0.0001: Time X Measures Interaction 15:41 2.17 0.0252 Measures Across Time X Group Interaction 62,50 1.12 0.3374 Times X Group Interaction 2,55 0.17 0.8450 Measures X Groups Interaction 30,82 1.37 0.1311 Times X Measures X Groups Interaction 30,82 0.89 0.6320 Measures Across Time X Conditions Interaction 31,25 1.11 0.3996 Times X Conditions Interaction 1,55 2.85 0.0968 Measures X Conditions Interaction 15,41 0.73 0.7446 Times X Measures X Conditions Interaction 15,41 1.07 0.4083 Measures Across Time X Groups X Conditions Interaction 62,50 0.95 0.5723 Times x Groups X Conditions Interaction 2,55 1.43 0.2490 Measures X Groups X Conditions Interaction 30,82 1.23 0.2262 Times X Measures X Groups X Conditions Interaction 30,82 1.20 0.2542 .05 level of significance used in this study. 49 control subjects were not significant. The interaction effects of groups and conditions were also not signifi- cant. Further analysis of the data indicates that the grand mean of measures across time was significant (P = .0001) and that most of this significance was due to the measures themselves (P = .0001) being of dif- ferent traits. The interaction of time and measures when analyz- ing the effect of measures across time was significant (P = .0252). This interaction may be a real significance but in the practical sense does not have any meaning. All other interaction effects of measures across time with groups, conditions, or groups and conditions were non- significant. Table 4.10 presents the experimental design for the Functional Self-Attitude Scale. It is the same design used for the 16 P.F. and some subjects were lost for the same reasons. Table 4.11 presents a summary of the multivariate analysis of variance repeated measures design used for the FSAS. The Finn program was used for this analysis. Significant differences were not found between groups on this measure, nor between experimental and control subjects. No significant interaction was found between groups and conditions. Further analysis indi- cates that the grand mean of measures across time has 50 TABLE 4.10 Experimental Design for Functional Self—Attitude Scale Time 1 (Pre) Time 2 (Post) Measures M1 ...... M8 M1 ...... M8 5 1 s 1 E O; O Grou 811 $11 1 P 312 517' c . I I; I 520L 519 $21 { i§§o E .1 . :1 ' Grou 529: 327 2 9 s30? sze I c .1 . s40! 536 541 537 E . . Grou 553 €38 3 p 3549 549 c .3 . 866 $60 51 TABLE 4.11 Summary of Analysis of Variance for Functional Self-Attitude Scale Degrees of Source Freedom F P Groups 2,54 1.95 0.1522 Conditions 1,54 0.29 0.8645 Group X Conditions Interaction 2,54 0.11 0.8952 Measures Across Time 15,40 183.99 0.0000* Times 1,54 1.62 0.2089 Measures 7,48 404.25 0.0000* Time X Measures Interaction 7,48 4.32 0.0009* Measures Across Time X Groups Interaction 30,80 1.54 0.0651 Times X Groups Interaction 2,54 0.42 0.6593 Measures X Groups Interaction 14.96 1.56 0.1062 Times X Measures X Groups * Interaction 14,96 2.06 0.0210 Measures Across Time X Conditions Interactibn 15,40 1.10 0.3846 Times X Conditions Interaction 1,52 0.24 0.6238 Measures X Conditions Interaction 7,48 0.77 0.6146 Times X Measures X Conditions Interaction 7,48 1.45 0.2081 Measures Across Time X Groups X Conditions Interactibn 30,80 1.19 0.3372 Times X Groups X Conditions Interaction 2,54 0.66 0.5201 Measures X Groups X Conditions Interaction 14,96 0.95 0.5058 Times X Measures X Groups Interaction 14,96 1.25 0.2511 * .05 level of significance used for this study. 52 a significant F value (P = .0000). This significance seems mostly accounted for by the measures themselves (P = .0000) being of different traits and thus not com- parable with each other across time. There is a time by measures interaction effect (P = .0009) when looking at measures across time. Other interaction effects of measures across time when combined with groups, conditions, or groups and conditions are not significant. Cluster Analysis A cluster analysis of 81 variables measured in this study was executed using a component of the official version of the BC TRY System (TryoneBailey Associates, Inc.). Nine clusters were formed in the analysis accounting for the total variance of the variables (see Appendix G). Cluster 1 is defined by the weekly behavior ratings of work and interpersonal behavior. This cluster also included the monthly ratings of the Evaluation Report and accounted for .2053 of the total variance of the'measures. Cluster 2 is made up entirely of the attendance ratings of the daily scale and accounts for .0745 of the total variance. Cluster 3 is defined by the tension- stability and dependence scales of the 16 P.F. and accounts for .2143 of the total variance of the measures. 53 Cluster 4 contains the intelligence measures of the 16 P.F. and I.Q. testing. It accounts for .0838 of the total variance. Cluster 5 is several measures of the single variable appearance. This daily rating accounts for .0874 of the total variance. Cluster 6 contains the 16 P.F. factor reserved- outgoing and reflects .0725 of the total variance. Cluster 7 is also primarily 16 P.F. factors and is defined by measures of mood and the submissive-aggressive continuum. This cluster accounts for .0990 of the total variance. Cluster 8 is made up of the hostility, depression, delinquency, and mental alertness scales of the FSAS. It accounts for .0960 of the total Variance. Cluster 9 contains the variables of promptness and perseverances as.measured by the Daily Behavior Checklist and accounts for .0672 of the total variance. The cluster analysis is meaningful to this study in several ways. It is immediately evident that no measured relationship was found between behavior and personality. The 16 P.F. Test which proports to measure 16 different personality factors collapsed into four clusters of related factors when combined with the other measures. Two of the behavioral rating scales, the weekly and monthly, fell into the same cluster and are 54 evidently measuring the same behaviors. The more objec- tive Daily scale, which was scored for four behaviors, formed three unrelated clusters. In this study attendance, appearance, and promptness-perseverance were not only unrelated to each other but were unrelated to the other two behavioral rating scales which rated more behaviors on a more subjective basis. The cluster analysis also provides some infor- mation on the test-retest reliability of the FSAS (Appendix H). The scale score correlations ranged from .175 to .358. As there were no significant dif- ferences found in the groups between pre- and post- testings the low correlations found would indicate low reliability of this instrument. Summary Primary analysis of the results of this study indicate that the treatment had no effect on the three groups of subjects over the four-week treatment period. The grand mean of measures over time was significant for the Daily Behavior Rating Scale, the 16 P.F., and the FSAS. There was also a times by measures inter- action on the 16 P.F. and the FSAS. Each of these instruments contains several measures of different behavior traits which make the differences between them over time of little interest. CHAPTER V SUMMARY AND CONCLUSIONS Analysis of data derived from three behavioral rating scales and two personality tests used as criterion measures in this study of social modeling indicate most clearly that these instruments did not record any sig- nificant treatment effects over the four—week treatment period. No significant F values emerged from the analysis of variance for the Evaluation Report or the Weekly Behavior Rating Scale, both behavioral measures. These two scales were thought to be the strongest measuring devices in the study. The Evaluation Report has been the basic objective measurement of progress in the Pine Rest Sheltered Workshop for several years and was con- sidered a valid and reliable instrument by the staff. The weekly Job Behavior Rating Scale was adapted, with minor modifications, from a scale used successfully by Fairweather and his associates (Fairweather gE_21., 1960) with similar populations. 55 56 Analysis of the Daily Behavior Checklist and the two personality measures did yield significant F values in the secondary analysis of the grand means of measures over time, in the analysis of variance. Inspection of the summary data tables for these three instruments and a visual comparison of their mean scores suggests that the significance of these grand means is due to charac— teristics of the measures. Non-significant individual mean score differences on these multi-score tests yield a significant grand mean difference which is not directly related to groups of subjects or the treatment and con- trol conditions. In addition, as Sakoda and his colleagues (Sakoda, Cohen, & Beall, 1954) have pointed out, the probability of finding a significant F value in a series of F tests increases with the number of tests made. In this study the high number of such tests made on the same data could account for the several significant F values. As the hypothesized significant differences were not found in this study, it could be concluded that Bandura's theory does not hold in this type of a learn- ing situation. Something not now evident may be limit— ing the generalizability of his theory. It seems more likely, however, that the hypothesized differences were not found due to the limitations of this study discussed below. 57 Limitations of the Study The attempt to model two broad a range of behaviors may have contributed generally to the non- significance of the findings of this study. Certainly, the broad range of behaviors modeled in this study goes beyond the attempts of previously reviewed research to model simple discrete behaviors in limited numbers. It is quite possible that when such demands are placed upon social modeling as a medium of learning it lacks suf- ficient strength to produce a measurable change in behavior. Bandura (1969) has pointed out that acquisition of behaviors may occur without the overt manifestation of the behavior unless the latter is at least vicariously reinforced. Although the desired behaviors were strongly reinforced on the video tape itself, perhaps vicarious reinforcement alone is not sufficient when more than one behavior is being modeled and no other reinforcers are utilized. As the trainers also served as raters in this experiment, they could not be exposed directly to the modeled behaviors and thus could not reinforce the clients for practicing them on the job. It also became apparent as the experiment pro- gressed that the Pine Rest Sheltered Workshop is exper- iencing a rather common workshOp conflict between pro- duction and rehabilitation. The trainers serve primarily 58 in a quality control role and receive little reinforcement themselves for the direct training of clients and the reinforcement of desired behaviors leading to rehabili- tation. Although the video tape of modeled behaviors is seen as a strong factor in this study due to the method of its production as described in the preceding section on Procedures, it may well have been shown an insuf- ficient number of times to the clients. As it was not possible to analyze the measures of client behavior as the experiment progressed, the experimenter had no objective feedback on the success of his modeling pro- cedures. The number of times the video tape was shown (three) to each client was chosen primarily on the basis of practical consideration which may have contributed to the weakness of the treatment. It is also possible, as Myrick (1969) suggests, that the "video tape might introduce too great a number of stimuli to which to attend." In this respect perhaps a shorter video tape, more simplistic scripting, or an audio tape might have been more effective. A judgment based on this study faulting the effectiveness of social modeling is further complicated by the lack of specificity with which some of the instru- ments used measured the behaviors modeled on the video tape. The daily behavior checklist consisted entirely 59 of behaviors modeled directly on the tape.‘ The other instruments, however, measured modeled behaviors more generally and in the case of the Evaluation Report measured more behaviors than were modeled on the video tape. The personality measures, of course, constituted an indirect measurement of the effects of the treatment procedure. As the Daily Behavior Checklist was simple and entirely objective, no rater training was attempted with this instrument. On the Weekly Job Behavior Rating Scale, a high inter-rater reliability was achieved, however, it became evident in scoring this scale that the raters were rating uniformly near the top of the scale early in the experiment, which left little room for improvement of the rated behaviors. Total mean scores on Measure I of this scale ranged from 14.1 the first week to a high of 14.5 the fourth week out of a possible 18 points. On Measure 2 of this scale total mean scores for all groups ranged from 13.5 the first week to 14.5 the fourth week out of a total possible 17 points. This initial high rating was also evident on the Daily Behavior Checklist where most clients received nearly perfect scores during all four weeks of the treatment period. This could be, however, attributable to the fact that they were rated on attendance, 60 promptness, appearance, and perseverance. These behaviors, although modeled directly in the video tape, are evidently not behaviors which need to be taught directly to the clients in Pine Rest Sheltered Workshop. The trainers' high ratings of new clients was not evident in the Evaluation Report where the total mean score for all clients was 64.5% out of a possible 100% score. In spite of the room for improvement left the clients on this scale, no significant differences were nOted in the analysisdue to length of time in the program or treatment and control conditions. Most of the subjects fell in the "average" or "normal" range for the test norms of the personality measures. This was especially true of the 16 P.F. and true of all scale means on the Functional Self Attitude Scale, with the exception of the hostility scale, where the Pine Rest Population received a total mean score of 25.1 on the pre-testings and 27.4 on the post-testing, with a score of 18 to 24 being considered "within normal limits." Mean scores on all other subtests of this scale were within normal limits and did not vary more than a few tenths of a point from pre— to post-testing. The import of the difference in the hostility scale score is not known due to the lack of reliability and validity data on this scale. 61 The absence of validity and reliability data on the Functional Self Attitude Scale is a further limitation of the study. The scale was selected for inclusion in the study as an additional measurement in response to considerable literature pointing to the importance of a change in self-concept for the successful rehabilitation of sheltered workshop clients (Holbert, 1971). A few other factors should be mentioned as pos- sible limitations of this study. As can be seen from Table 4.1, the population of this study was limited by many factors such as low-level intelligence, a lack of education, and a lack of work experience. It is quite possible that social modeling procedures that have worked successfully with brighter, more advantaged clients might need considerable strengthening to be effective with this sheltered workshop population. An additional possible limitation on this study might be that the clients were measured over only a four- week span. Thislength of time for measurement was chosen, as was the time for treatment, on the basis of practical considerations. No previous research was uncovered which could be directly applied in either situation and the translation from theory was difficult to make. The four-week measurement time was chosen primarily because of the high attrition rate from the program beginning the fifth week of the client's enroll- ment. 62 A final limitation on this study is a grand one and difficult to assess. The Pine Rest Sheltered WOrk- shop is part of a private mental health complex with an exceptionally strong service orientation. The current research project was the first large research project to be attempted in the sheltered workshop and one of very few to be attempted in the hospital complex at large. This strong social service orientation of the staff places research in a less valued position than service and had far-reaching consequences for all aspects of the project. For example, it was consis- tently difficult to get all clients to appear at the proper time to view the treatment or control video tapes. It was also consistently difficult throughout the project to get the trainers to turn in their rating scales on time and filled correctly. For example, it became neces- sary after the first week to place a research assistant in charge of immediately recording all weekly ratings so that he could uncover missing ratings and personally request them from the trainers on the next working day. Rather frequently the counselors would question the validity of the trainers' ratings, explaining that if their client was rated so near the top of the scale, then he should be ready for employment, and they certainly did not see him that way. In general, the entire research project was seen as burdensome to the line 63 personnel in the sheltered workshop and their general lack of support may have effected the outcome of the study in some unspecificable manner. Implications for Future Research It is not the author's conclusion, based upon this study, that social modeling is a totally ineffective procedure for teaching desired behaviors to sheltered workshop clients. But rather, social modeling as utilized in this study is seen as a generally weak method of teaching these behaviors. Future research should attend to strengthening procedures used in this study and to improving the criterion measures. Specifically, it would seem desirable to use social modeling in conjunction with such procedures as guided practice or group discussion and with direct reinforcement of the modeled behaviors by the trainers and counselors. Fewer behaviors should be modeled at one time and comparative research should be carried out on the mediums available for social modeling, such as video tape, audio tape, video slides, etc. No research comparing these mediums was discovered in a review of the literature. Future researchers should also attend more closely to the raters' attitudes toward the behaviors they are rating and their attitudes toward the research project in general. It would also seem wise on the basis of this study to avoid measuring 64 behaviors other than those modeled, even though common sense consideration indicates that some generalizing of the modeled behaviors should occur and indeed there is some research evidence for this (Bandura, 1969). Future researchers in this area would also do well to devise a means of acquiring more immediate feedback as to the success of their treatments so that they can make adjust- ments in the time of exposure to the modeled behaviors. A test of acquisition of the knowledge transmitted by the modeling treatment would also be of value in this respect. In summary, the social modeling of desired work- shop behaviors, using vicarious reinforcement on video tape, was demonstrated to be a generally weak procedure for accomplishing rather grand goals. More limited goals, with a stronger treatment program, would probably accomplish the original intent of this project, i.e. to teach sheltered workshop clients how to more effectively utilize a complex program which is attempting to teach them the complex subculture of work. BIBL IOGRAPHY BIBLIOGRAPHY Bandura, A. Vicarious processes: A case of no-trial learning. In L. Berkowitz (Ed.), Advances in experimental social psychology. Vail II. New York: Academic Press, I965. Bandura, A. Principles of behavior modification. New York: HOIt, Rifiehart and Winston, Inc., 1969. Bandura, A., Blanchard, E. B., & Ritter, B. Relative efficacy of desensitization and modeling approaches for inducing behavioral, affective, and attitudi- nal changes. Journal of Personalipy and Social ngcholggy, 1969, 13, 173-199. Bandura, A. & Huston, C. Identification as a process of incidental learning. Journal of Abnormal and Social Psychology, 1961, 63, 3114318. Bandura, A., Ross, D., & Ross, S. A comparative test of the status envy, social power, and secondary reinforcement theories of identificatory learn- ing. Journal of Abnormal and Social Psychology, 1963, El, 527, 534. Bandura, A. & Walters, R. Social imitation and person- ality_development. New York: Holt, Rinehart and’Winston,ii963. Bozarth, J. D. Some problems in applied research. Rehabilitation Research and Practice Review, 1970, 2721-25. Caro, F. G. Issues in the evaluation of social programs. Review of Educational Research, 1971, 41, 87-114} Cattell, R. B., Eber, H. W., & Tatsuoka, Maurice M. Handbook for the 16 P.F. Champaign, Ill.: Institute for Personality and Ability Testing, 1970. 65 66 Dahm, J. W. The effects of modeling and information--Only as counseling self—referral among middle school students. Unpublished doctoral dissertation, Michigan State University, 1970. Donk, L. J. The dimensions of videotape self—cognition in psychology and psychiatry. Unpublished manu- script, 1971. Eber, H. W. Multivariate analysis of a rehabilitation system: Cross validation and extension. Multi- variate Behavioral Research, 1967, 2, 477-484. Eisenberg, S. & Delaney, D. J. Using video simulation of counseling for training counselors. Journal of Counselipg ngchology, 1970, 11, 15—19} Fairweather, G. W. (Ed.) Social psychology in treating mental illness. New Yofk: John Wiley & Sons, Inc., 1964f Fairweather, G. W. Methods for experimental social inno- vation. New YorK} John Wiley & Sons, Inc., 1968. Fairweather, G. W. et a1. Relative effectiveness of psychotherapeutic programs: A multiculture com- parison of four programs for three different patient groups. Ps cholo ical Mono ra hs: General and Applie ,‘__ 5, ole No. ), 1960. Fairweather, G. W. et a1. Community life for the mentally ill: An alternative to institutiOnal care. ChiCago: Aldine PubliShing Co., 1969. Heller, K. Effects of modeling procedures in helping relationships. Journal of Consultipg and Clinical Psychology, 1969, 33, 5224526. Higgins, W. H., Ivey, A. E., & Uhlemann, M. R. Media therapy: A programmed approach to teaching behavioral skills. Journal of Counselipg Psy- chology, 1970, 11, 20-26. Holbert, W. M. Personal communication, June 11, 1971. Interim Manual Supplement for Form E, 16 P.F. Champaign, Ill.: Institute for Personality and Ability Testing, 1968. 67 Krieger, G. W. An exploratory study of the effect of model-reinforcement counseling on the vocational behavior of a group of male retarded adolescents. Unpublished doctoral dissertation, Michigan State University, 1969. Krumboltz, J. D. & Schroeder, W. W. Promoting career planning through reinforcement. Personnel and Guidance Journal, 1965, 44, 19-26. Krumboltz, J. D. & Thoresen, C. E. The effect of behavioral counseling in group and individual settings on information seeking behavior. Journal of Counseling Psychology, 1964, 11, 324-333. '— Krumboltz, J. D., Varenhorst, B. B., & Thoresen, C. E. Nonverbal factors in the effectiveness of models in counseling. Journal of Counseling_P§ychology, 1967, 14, 412-418. LaFleur, N. K. The separate and combined effects of models, reinforcements, and attentional variables on adolescent vocational behaviors. Unpublished doctoral dissertation, Michigan State University, 1970. Lane, J. M., Jr. & Barry, J. R. Recent research on client motivation. Rehabilitation Research and Practice Review, 1, 5-25. Massie, W. A. Partners in providing services to the handicapped. Newsletter, International Associ- ation of Rehabiliiation Facilities, 1971. Myrick, R. D. Effect of a model in verbal behavior in counseling. Journal of Counseling Psychology, 1969, 16, 185-I90. Neff, W. S. Work and human behavior. New York: Atherton Press, I968. Rehabilitation Facility Needs in the '70's, Monograph #1. U.S. Department of Health, Education, and Welfare. Social and Rehabilitation Service, Rehabilitation Services Administration, Washington, D.C., 20201. Sakoda, J. M., Cohen, B. H., & Beall, T. Test of significance for a series of statistical tests. Psychological Bulletin, 1954, 51, 172-175. 68 Stueland, J. R. Training and estimation of reliability of trainer's ratings. Unpublished manuscript, Michigan State University, 1971. Stugart, D. B. An experimental study investigating the effects of model race and model age--referrent group upon the vocational information-seeking behaviors of male black eleventh graders. Unpublished doctoral dissertation, Michigan State University, 1970. Thoresen, C. E., Hosford, R. E., & Krumboltz, J. D. Determining effective models for counseling clients of varying competencies. Journal of Counselipg Psychology, 1970, 11, 369-375. Training Methodology: IV Audiovisual Theory, Aids, and Equipment: An Annotated Bibliography. Public Health Service Publications, 1969, No. 1862-IV, 80p. Tryon-Bailey Associates, Inc. 728 10th Street, Boulder, Colorado, 80302. Whalen, C. Effects of a model and instructions on group verbal behaviors. Journal of consulting and clinical psychology, I969,‘33, 509-521. APPENDICES APPENDIX A EVALUATION REPORT EVALUATION REPORT SWS CLIENT NAME PERIOD COVERED TRAINER NAME DAYS IN ATTENDANCE THERAPIST NAME REASONS FOR ABSENCE COPIES TO: DAYS LATE 0-5 minutes 6-15 minutes l6+ minutes WORK RATING 1 2 3 4 ntit GENERAL AREAS: arance ene n ave 3 ca us en 0 e a ons p Su vision e a ns p o Co-workers e war erseverance SPECIFIC AREAS: nve o e u o a Trainer Date 69 APPENDIX B STRUCTURED INTERVIEW INSTRUCTIONS TO COUNSELORS Please ask each of your clients these questions at the beginning of each weekly interview for the next few weeks. Tape record the responses and write the client's name and number of weeks in thepprogram on the tape. We are looking for behavioral responses. Push your client to translate feelings into behaviors and to be as specific as possible. Please tell your client, "We are interested in your evaluation of our program, especially what is good about it. Your answers are being taped so that they may be studied and our program improved." Please keep the tape as brief as possible. If your client expresses feelings you wish to pursue, ask him if you can talk further about them later in the inter- view. The interest you demonstrate in his welfare by ask- ing these questions should enhance discussion of feelings even though it is postponed. STRUCTURED INTERVIEW QUESTIONNAIRE First Administration Only: What did you do the day you started your program? Did you want to come to this program? Why? What did you expect when you arrived at Pine Rest? What do you remember most about your first day at work? What happened your first night in the Homestead/Lodge? U'IhUJNH o o o .0 All Administrations: 1. What happened at work last/this week? What did it teach you? 2. What do you like about your trainer? What do you dis- like? . What do you like about the peOple you work with? 4. What do you like most about your job? What do you dislike most? 5. What happened at the Homestead/Lodge last/this week? 6. Do you like living there? What's good about it? What's bad about it? 7. Did you do anything interesting in evening activities last/this week? 8. What did you learn about peOple? (Getting along with people last/this week?) (H 3 did you learn it?) 9. What did you learn about yourself last/this week? (How did you learn it?) Describe yourself. 10. DS—you like your counseling appointments with me? Why? 11. What do you like best about being in this program? (Why or how?) 12. How do you think this program is going to help you? (Goals) 70 APPENDIX C BEHAVIORAL TALLY BEHAVIORAL TALLY First Administration Only: Number of Responses 1. What did you do the day you started your program? (21) nervous, scared (6) shown around (6) went to work (4) went to room, unpacked, met roommates (3) testing, physical (2) 2. Did you want to come to this program? Why? (22) yes - 19 no - 2 training or to get a job (8) nice place (2) instead of something else (5) 3. What did you expect when you arrived at Pine Rest? (23) don't know (8) dorms (2) get on my own (2) 4. What do you remember most about your first day at work? (22) described tasks (7) nervous, didn't know what to expect (3) nothing much (4) met people (4) 5. What happened your first night in the Homestead/Lodge? (21) nervous, shy, frightened (6) unpacked (4) met roommate, other kids (9) watched TV, played pool (3) A11 Administrations: 1. What happened at work last/this week? What did it teach you? (55) described tasks (13) just worked, nothing new, not much (14) worked hard, busy (3) 71 72 Continued new job instructor helped me, complimented me, jumped me helped or helped by co-worker What do you like about your trainer? What do you dislike? nice, like, good guy OK, just people, fair great, cool, friendly patient, understanding, calm got angry encouraging ' helpful, showed me the work doesn't holler at me good personality What do you like about the people you work with? friendly, nice, swell guys OK, alright, fine hardworking, do work right goldbrickers don't act funny easy to get along with help me, teach me, help each other are goofy, have a lot of problems talk to you What do you like most about your job? What do you dislike most? don't know keep busy it's easy described tasks the people, new people like change, variety something to do working with patients being rewarded by trainer work by myself What happened at the Homestead/Lodge last/ this week? nothing watched TV, played records, read talk, play, tease, goof off, shopping put on restriction dull, routine, average quiet fighting, drinking, getting high (4) (5) (5) (55) (15) (18) (3) (8) (4) (2) (7) (2) (3) (55) (14) (19) (3) (6) (2) (2) (9) (3) (2) (54) (9) (2) (3) (9) (7) (5) (2) (2) (2) (2) (53) (14) (8) (6) (4) (3) (2) (3) 10. 73 Do you like living there? What's good about it? What's bad about it? yes - 34 no - 10 alright, OK people are friendly, like the kids, meet new kids private room can cook talking to houseparents too noisy, can't cook don't like houseparents don't like houses, rules some guys steal, bug others freedom to come and go likes food, girls, sports Did you do anything interesting in evening activities last/this week? shopping movie bowling bike riding played ball MTC ride, walk TV, records, pool out to eat What did you learn about people? (Getting along with people last/this week?) (How did you learn it?) nothing -—_- how to get along peOple are different people are all alike if I'm friendly, others are others have problems too learn about people by watching What did you learn about yourself last/ this week? (How did you learn it?) Describe yourEEIf nothing much growing up, handle self better, indepen- dent Do you like your counseling appointments with me? Why? yes - 47 no - 1 OK - 6 (55) (9) (13) (9) (2) (3) (5) (3) (14) (3) (7) (4) (56) (18) (7) (6) (2) (5) (5) (5) (8) (2) (49) (9) (8) (8) (2) (6) (3) (2) (51) (5) (ll) (54) 10. 11. 12. 74 Continued like you, your're nice, good person helps to talk understanding, you listen What do you like best about being in this program? (Why or how?) the training meeting new peOple, friends get away from home, freedom don't know having a job, making money helping others doing different things like the place group meetings counseling, self-improvement everytning finding me a job How do you think this program is going to help you? (Goals) get a job get a better job don't know get along with people get on my own see what I can do straighten me out get me started train me (10) (16) (9) (7) (10) (5) (5) (6) (2) (2) (2) (3) (5) (3) (2) (52) (19) (8) (2) (5) (11) (3) (9) (2) (4) APPENDIX D WEEKLY JOB BEHAVIOR SCALE WEEKLY JOB BEHAVIOR SCALE Client's Name Rater's Name Date DIRECTIONS: This scale measures weekly changes in behavior. Please fill this scale out every Friday and turn in the completed forms to the Client Service office by 5:00 P.M. You may use a v’or X to mark your choice. Do not skip any statements. l. The client made few errors. The client made many errors. 2. The client seldom needed prodding to get job done. The client usually needed prodding to get job done. 3. The client rarely needed help. The client often needed help. 4. The client followed directions on the job. The client did not follow directions on the job. 5. The client worked for extended periods of time. The client failed to work on a task for any reasonable period of time. 6. The client's work was usually well-planned. The client's work was usually poorly planned. 7. The client abided by the rules and regulations. The client violated the job rules and regulations. 8. The client usually profited from constructive criticism. The client usually did not profit from con- structive criticism. 9. The client seldom became upset by failure. The client usually became upset by failure. 10. The client usually did not complain about tasks that were given him. The client usually complained about tasks that were given him. 75 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 76 The client usually applied himself to the task. The client usually did not apply himself to the task. The client worked with average accuracy. The client's work was often inaccurate. The client worked constantly on task. The client worked rarely on task. The client reported regularly for his assignment. The client often failed to show up for his assignment. The client usually profited by his mistakes. The client repeated his mistakes over and over again. The client often helped his co-workers. The client seldom helped his co-workers. The client was usually dressed appropriately for his job. The client was seldom dressed apprOpriately for his job. The client often expressed a desire for a job in the community. The client seldom expressed a desire for a job in the community. The client was likable as a person. The client was not likable as a person. The client did not do or say anything that was disturbing to others. The client sometimes said or did things that disturbed others. The client rarely or never said his work or skill was of poor quality. The client often said his work or skill was of poor quality. The client was courteous and considerate, showing a normal concern for the feelings of others. The client was discourteous, and often showed no concern for the feelings for others. The client was reasonably friendly and agreeable. The client was usually unfriendly and disagreeable. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 77 The client never commented in an uncomplimentary manner on the work or activity of his associates. The client frequently commented in an uncompli- mentary manner on the work or activity of his associates. The client often spoke to others. The client rarely or never spoke to others. The client's ideas were never odd. The client's ideas were often odd. The client was usually happy. The client was usually sad. The client did not tell others that he had exceptional abilities that he did not have. The client told others that he had exceptional abilities that he did not have. The client took pride in the quality of his work; i.e., showed work or talked about work to others and/or supervisors. The client did not take pride in the quality of hiw work; i.e., did not show work or talk about work to others and/or supervisors. The client seldom found fault with others and/or their work. The client often found fault with others and/or their work. The client paid attention to the work and activi- ties of others. ' The client did not pay attention to the work or activities of others--was in a world of his own. The client made some worthwhile suggestions about tasks. The client made no worthwhile suggestions about tasks. The client usually accepted constructive sug- gestions of the supervisor. The client seldom accepted constructive sug- gestions of the supervisor. The client was not usually disturbed by con- structive criticism. The client was usually disturbed by constructive criticism. The client seldom found fault with his work. The client often found fault with his work. APPENDIX E DAILY BEHAVIOR CHECK LIST 78 31': 2:9:354m 2|: EIBEZE4m m 219:354m m 2:95:94m m 2:8th 2!: 293894 m 2:85:54m It: 239:354m If: ZESE—‘fiq €13 BEBE-iii: m EBZBM Cl: EBSBfi-c m Eli-438th n 2643994 3'1 2936454 m SENSE-ch :13 219::e+m :r: 26436494 :1: 2:95:54m m 2:9:254m a: 2:95:54m :13 2835'an 2!: 2543th :11 28389-0 a: SBSE—fh m ::e::e+m m E185364m m Sit-038m m 2:85:94m LT: 2:9:294m :13 293834 53 536436-054 m 283E494 I11 2:9:354m m 25'3th :13 EIBEIE4M m SB3BEH :13 21BE£E+W m Elf-£39k: 33 EIBEIE4h m 2643=BEH II: 22936-094 a: SE—lzfilu m 2:95:64m a: SPOKE-H34 m 283E011: m 5.283913. 33 2292364m :1: 283894 {I} 2936430 :1 293E4h l1: Elf-43894 m 2:95:54m :1: 2:9:354m :13 2:9:354m 2L1 Eli-13994 m 2:9:ze+m =13 283E494 :13 2385364h =3 EIBEIEdh m 53938:!4 :11 EIBIZE+m :13 SENSE-Hi4 El: 2:95:54m a: Ell-4394b :13 293994 :r: 21936494 x SBSBEH 213 Eifilifidh m 293E454 m 25'3th :1: 2836-194 2!: SIS-138k: m . EIBEZE+h :1: 25'3th a 28393: 2!: 2:95:54m :12 SPIKE-GIN a: EIBEZE+m :1: 5.2643894 :13 SIS-139% m SBSE-‘fi-a m 2:95:54m 11': ZE-le-Hh m 2:9:254m :13 EBZBFH a: SB3E-IFH :13 2936-”:24 2B SBSBI’A m Elli-435454 =1: 2938B: a: SB3E-ifia :13 2183894 2: 2285364h M T W T F :1: 2382364m m 2:95:94m :1: ZESBfi-a m 2:8:354m I: 2936494 :12 283BFH W TH F I: 2939!:4 53 29385:: 21': 2836494 DAILY BEHAVIOR CHECK LIST DATES Client came to work today. I. CIient came to work on time. 23 Client's personaI appearance and hygiene are adequate. 3. morning coffee break T Cliént returned from M on time. I. from lunch on 5. Client returned time. Client returned from afternoon coffee break on time. 6. the end of the 7. Client stayed till day. APPENDIX F FUNCTIONAL SELF-ATTITUDE SCALE FUNCT IONAL SELF-ATTITUDE SCALE 1. I tell the truth. 2. I'll never be able to make enough money to live on. 3. I obey the law. 4. I'd rather work at a job I like, even if it pays less. 5. I like having something to do. 6. People lie to me. 7. I can't make enough money. 8. I don't need any help from anybody. 9. I have trouble with the law. 10. I feel better when I am busy. 11. It takes "pull" to get what you want. 12. Working is the only way to get ahead. 13. When I was in school, I went along with my teachers. 14. I get treated like dirt. 15. I talk things over with my family. 16. Life is lonely. 17. Most things work out for the best. 18. I am satisfied with my looks. 19. Things that happen in life scare me. 20. I am healthy. 21. I can make a go of it in life. 22. I need to be alone. 23. My family understands my problems. 24. I'll never get anywhere. 25. It doesn't do any good to talk about my problems to anybody. 26. I like for friends to invite me over. 27. I'll never get out of debt. 28. I can keep a good job. 29. I didn't like school. 30. I forget what I am doing. 31. When I think about a new job, I really get nervous. 32. I like being by myself. 33. I like to learn new things. 34. The jobs I get don't work out for me. 35. I made good grades in school. 36. I can't talk to my bosses. 37. My folks used to whip us kids. 38. My appearance is against me. 39. I freeze up in a new setting, and can't do anything. 40. I can learn. 41. When I try something, I stick to it. 42. Kids need to have the.meanness taken out of them. 43. I have determination. 44. I have quite a few accidents. 45. It helps to talk things over. 79 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 80 I feel good. I ask for help when I need it. It's not what you can do, it's who you know that matters. I'm not good enough for some people. I have poor health. I'm smarter than most people. Teachers disliked me. I feel "down." Bosses don't pay attention to me. The most important thing about a job is how.much it pays. I feel alone. I like to have someone come and visit. I lie when it is necessary. PeOple don't care about me. People don't tell the truth. Pine Rest Sheltered Workshop Attitude Scale Scores 1. Name: 2. Center No. __ ‘__ __ __ -7cc 1:6) 3. Card No. 4. Length of Stay (in weeks): (cc 7) (cc 8-9) 5. Pre-program Scores: Sub test W. (cc 10-11) Sub test H. (cc 12-13) Sub test N. (cc 14-15) Sub test D. (cc 16-17) Sub test I. (cc 18-19) Sub test A. (cc 20-21) Sub test P. (cc 22_23) 7. Post-program Scores: Sub test W. Sub test L. (cc 24 25) (cc 48-49) ———~——— Total Pre. (cc 26-28) Sub test H. 6. Mid-program Scores: (cc 50-51) Sub test N. Sub test W. (cc 29 30) (cc 52_53) ___.___ Sub test H. (cc 31-32) Sub test D. Sub test N. (cc 33-34) (cc 54-55) Sub test D. (co 35-36) Sub test I. (cc 56-57) Sub test 1. (cc 37-38) Sub test A. Sub test A. (cc 39 40) (cc 58-59) ——— ——— Sub test P. (cc 41-42) Sub test P. Sub test L. (cc 43-44) (cc 60-61) Total Mid. (cc 45-47) Sub test L. (cc 62-63) Total Post (64-66) 81 FUNCTIONAL SELF-ATTITUDE SCALE KEY Work and Achievement--Sub-scale W 12. 27. I'll never be able to make enough money to live on. I can't make enough money. I'd rather work at a job I like even if it pays less. The most important thing about a job is how much it pays. I like having something to do. I feel better when I am busy. Working is the only way to get ahead. I'll never get out of debt. Hostility Feelings--Sub-sca1e H 36. 54. 13. 52. 14. 59. 37. 42. I can't talk to my bosses. Bosses don't pay attention to me. When I was in school I got along with my teachers. Teachers disliked me. I get treated like dirt. People don't care about me. My folks used to whip us kids. Kids need to have the meanness taken out of them. Neurotic Tendencies-Sub-scale N 56. 16. 15. 23. 17. 19. 18. 38. I feel alone. Life is lonely. I talk things over with my family. My family understands my problems. Most things work out for the best. Things that happen in life scare me. I am satisfied with my looks. My looks are against me. Depressive Tendencies--Subscale D 46. 53. I feel good. I feel "down." 82 83 Depressive Tendencies--Continued 20. 50. 21. 24. 22. 32. I am healthy. I have poor health. I can make a go of it in life. I'll never get anywhere. I need to be alone. I like being by myself. Inadequagy Feelings-Sub-scale I 45. 25. 26. 57. 28. 34. 39. 31. It helps to talk things over. It doesn't do any good to talk about my problems to anyone. I like for friends to invite me over. I like to have someone come and visit. I can keep a good job. The jobs I get don't work out for me. I freeze up in a new setting and can't do anything. When I think about a new job I really get nervous. Mental Alertness--Sub-scale A 44. 30. 40. 33. 41. 43. 35. 29. I have quite a few accidents. I forget what I am doing. I can learn. I like to learn new things. When I try something I stick to it. I have determination. I made good grades in school. I didn't like school. Delinquency Tendencies--Sub-scale P 3. 9. 47. 8. 48. 11. I obey the law. I have trouble with the law. I ask for help when I need it. I don't need any help from anybody. It's not what you can do, it's who you know that matters. It takes "pull" to get what you want. 84 Delinqgency Tendencies--Continued 51. I'm smarter than most people. 49. I'm not good enough for some people. Lie Scale-Sub-scale L l. I tell the truth. 60. People tell me the truth. 58. I lie when it is necessary. 6. People lie to me. APPENDIX G CLUSTER ANALYSIS THE NINE CLUSTERS Cluster 1--w0rk and Interpersonal Behavior 1. weekly Behavior Rating Scale a. b. c. d. e. f. g. h. i. j. week 3 - WOrk Behavior week 4 - " " Week 3 - Interpersonal Behavior Week 1 - Wbrk Behavior Week 2 - " " Post-monthly Evaluation Report Week 4 - Interpersonal Behavior Pre-monthly Evaluation Report Week 1 - Interpersonal Behavior Week 2 - " " Cluster 2-Attendance 1. Daily Behavior Rating Scale a. Week 3 b. Week 4 Cluster 3--Tension, Stability and Dependency 1. 16 P.F. a. Pre-tension b. Pre-stability c. Post-tension d. Pre-confidence e. Post-dependency f. 'Post-submissive-aggressive g. Post-emotional Responsiveness h. Pre-group Dependency i. Post-confident-apprehensive j. Post-Timid-Venturesome k. Pre-Impulsive-Controlled Cluster 4--Intelligence 1. l6 P.F. a. Post-measure b. Pre-measure 2. I.Q. Test Cluster 5--Appearance 1. Daily Ratings a. b. c. Week 3 Week 2 Week 1 85 .9532 .8067 .7881 .7815 .7544 .6423 .6391 .6074 .6010 .5791 .9675 .7612 .9098 .7442 .6999 .6718 .6680 .6630 .6529 .6526 .5668 .5464 .5281 .9172 .7928 .5762 .9104 .8700 .5907 86 Cluster 6--Reserved-Outgoing l. 16 P.F. a. Pre-measure b. Post-measure Cluster 7-Mood, Submissive-Aggressive 1. 16 P.F. a. Post-Mood b. Post-Suhmissive-Aggressive c. Pre-Mood d. Pre-Submissive-Aggressive Cluster 8--Hostility, Depression, Delinquency, and Mental Alertness 1. Functional Self-Attitude Scale a. Post-Depression b. Post-Hostility c. Post-Mental Alertness d. Post-Delinquency Cluster 9--Promptness-Perseverance 1. Daily Behavior Check List a. week 2 - Promptness b. Week 2 - Perseverance c. week 3 - Promptness .8583 .8515 .7468 .7178 .6485 .5901 .7319 .7118 .5566 .5400 .9085 .7004 .6518 APPENDIX H CORRELATIONAL MATRIX: FSAS 87 mamom can u a mmwocmocma mocooocaamo u m mmmcuuoafi Hmucmz u 4 mmcflaowm accommoocH u H moflocoocoa o>wmmmummo u o moflocmocoa oauousmz u z mmcfiammm huflHflumom u m ucosw>mfisofl one Muoz u 3 mmm. oma. med. vqa. mma. mHH. med. mum. A mmo. mmm. «Hm. mma. mmo.. moa.l mma. eve. m mmo.| moo.| med. mmo. H¢N.I mao.n mmH.I «mm. 4 ooo.l one. meo.| 0mm. amH.I mmo. emo.l who. H mmo. Hma. How. mmo.| Hem. mmo.| mum. cow. 0 mma. ema. mmo. voo.l mmH. mma. «ma. med. 2 mma. mma. ham. oma. mmm. mmo.l mmm. mma. m maa. mad. evo.l emu. vma.l mmo. «mo. nmm. 3 m