Citation
The contribution of role playing techniques to self-concept enhancement and weight loss in overweight college women

Material Information

Title:
The contribution of role playing techniques to self-concept enhancement and weight loss in overweight college women
Creator:
Schreiber, Fred M ( Fred Mark ), 1953-
Publication Date:
Language:
English
Physical Description:
v, 73 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Food ( jstor )
Obesity ( jstor )
Overweight ( jstor )
Pretreatment ( jstor )
Self concept ( jstor )
Self esteem ( jstor )
Weight control ( jstor )
Weight loss ( jstor )
Weight reduction ( jstor )
Women ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Obesity -- Psychological aspects ( lcsh )
Psychology thesis Ph. D ( lcsh )
Role playing ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1980.
Bibliography:
Includes bibliographical references (leaves 69-72).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Fred M. Schreiber.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
07093151 ( OCLC )

Downloads

This item has the following downloads:


Full Text










THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN
OVERWEIGHT COLLEGE WOMEN







By

Fred M. Schreiber




















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1980















ACKNOWLEDGEMENTS


I would like to thank Dr. Harry Grater for the time he has contributed and the interest he has shown in seeing this study through to its completion. His guidance and support throughout this entire study has been greatly appreciated.

I would like to thank the following people for their interest in and contributions to this study:

Dr. Afesa Bell-Nathaniel: for her suggestion and encouragement to study the self-concepts of overweight people.

Dr. Sig Fagerberg: for his support in teaching weight control classes in the Department of Health Education and Safety.

Dr. Linda Moody: for the opportunity to develop the weight control manuals used in this study.

Dr. Paul Schauble: for his help in research methods and design, and his encouragement to examine various facets of the overweight person.

Dr. Ted Landsman: for his personal support not only during this study, but throughout my entire graduate program.

Special thanks to Dr. Al Kahn and Dr. Mike Omizo for contributions

beyond the call of duty. They made the statistical portion of this study more manageable and understandable.

Thanks to Dr. Bob Ziller for his stimulating ideas on the selfconcept and theory of personal change.




ii
















TABLE OF CONTENTS




ACKNOWLEDGEMENTS ii ABSTRACT iv

INTRODUCTION 1

Problem 1 Behavioral Approaches to Weight Control 1 The Self-Concept 5 METHOD 27

Subjects 27 Measures 29 Procedures 31 RESULTS 43

Hypotheses 43 Pretreatment to Follow-up Change Scores 49 Correlations Between Variables 50 DISCUSSION 52 APPENDIX A INSTRUMENTS AND FORMS 62 APPENDIX B COURSE OUTLINE 66 APPENDIX C LETTER SENT TO SUBJECTS AT THE FOLLOW-UP 68 REFERENCES 69 BIOGRAPHICAL SKETCH 73









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

THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN OVERWEIGHT COLLEGE WOMEN

By

Fred M. Schreiber

August 1980

Chairman: Harry A. Grater
Department: Psychology

Research in psychology has indicated that the self-concept may be a moderator variable in the maintenance of behavioral changes. In addition, previous research has suggested that role playing techniques may facilitate a change in the self-concept. This study investigated the effectiveness of role playing techniques in a behavior modification weight reduction group in order to help clients enhance their self-concepts and to determine the effect this has on clients' abilities to lose weight and maintain their losses over a follow-up period.

Two eight-week weight reduction groups were conducted. The
goal of each group was gradual weight loss through improving eating habits. In addition, one group included an emphasis on using role playing techniques to facilitate clients' abilities to change their self-concepts. The other group was presented with theories and information on weight control.

General self-concept, physical self-concept, and complexity of self were measured at pretreatment and after a four month follow-up


iv









period. The scales used to measure these variables were the Tennessee Self-Concept Scale and the Ziller Complexity of Self Scale.

Both groups had significant weight losses at posttreatment and these losses were maintained after the four month follow-up period. There were no significant differences in weight loss between the two groups at posttreatment. However, at the follow-up, subjects in the "information" treatment group tended to maintain a greater weight loss than subjects in the "role playing" treatment group.

At the follow-up, there were no significant differences in selfconcept, complexity, or physical self between the two groups. In addition, there were no significant differences in self-concept, complexity, or physical self between "successful" and "unsuccessful" weight reducers at the follow-up. However, "successful" weight reducers tended to have higher scores on the Ziller Complexity of Self Scale at pretreatment than "unsuccessful" weight reducers. These results are discussed with regard to Ziller's Helical Theory of Personal Change and future work in weight control.





Chairman














v














INTRODUCTION

Problem

The problem of obesity has reached such a high magnitude that the U.S. Public Health Service has classified it as "one of the most prevalent health problems in the United States today" (.p.. 547, Abramson, 1973). Stuart and Davis (1972) estimate that there are currently between 40 and 80 million obese individuals in this country alone. Traditional treatments for the problem have included medication, psychotherapy, and therapeutic starvation. However, these treatments have generally been unsuccessful, and Stunkard (1958) has concluded that

Most obese persons will not remain in treatment.
Of those that remain in treatment, most will not
lose weight, and of those who do lose weight,
most will regain it.(p. 79)

Behavioral Approaches to Weight Control

Recently, some successes have been reported using a behavioral approach to weight reduction. This approach views eating habits as learned behaviors and sees the overweight person as someone who has learned inappropriate patterns of eating. Behavioral approaches to weight reduction focus on helping the overweight person become aware of his eating patterns and helping him change inappropriate or problematic behaviors. It is presumed that by improving one's eating habits it will be easier to reduce one's caloric intake, and consequently, lose weight. In addition, by changing one's habits permanently, weight loss should also be permanent.
1




2



Techniques used in behavioral approaches to weight reduction are derived from both operant and classical conditioning. Operant conditioning focusses on the antecedents and consequences which control a behavior. When applied to weight control, the overweight person is taught to bring his eating under appropriate situational cues and to reinforce improvements in his eating patterns. Classical conditioning involves the pairing of two stimuli so that eventually they bring about the same response in an organism. The overweight person has paired eating with a number of different external cues so that eventually the external cues alone elicit a response of "feeling hungry." The overweight person needs to disassociate eating from other activities in order to extinguish eating as a conditioned response to other activities.

Stuart (1967) utilized operant and respondent conditioning techniques to help clients gain control over their eating behavior. These techniques included controlling the antecedent and consequent conditions of eating as well as record keeping and exercise. In addition, reinforcement was provided in the following three ways: (1) through the clients' experience of success in self-control; (2) through the reduction of the aversive consequences caused by a lack of self-control; and (3) through considerable reassurance by the therapist. Stuart treated eight women on an individual basis. Weight loss over a twelve month period for the eight women ranged from twenty-six to forty-seven pounds.

Wollersheim (1970) compared the effectiveness of three group
treatments (behavior self-control, positive expectation--social pressure, and non-specific therapy) with a no-treatment control group. The behavioral self-control group utilized operant conditioning techniques




3



to learn appropriate eating behaviors in order to lose weight. In addition, they were trained in relaxation in order to counter tension in situations which would ordinarily result in eating. The positive expectation--social pressure group utilized group pressure to help subjects lose weight. The non-specific therapy group helped subjects understand their underlying motives for being overweight. Wollersheim reported that all three treatment groups lost significantly more weight than the control group after a twelve week treatment program. Furthermore, after an eight week follow-up subjects in the behavioral self-control group showed a significantly greater weight loss (pretreatment to follow-up) than either of the other two treatment groups.

Stuart (1971) treated six overweight women on an individual basis in a fifteen week program which stressed environmental control of overeating, nutritional planning, and regulated increase in energy expenditure. Half the women started treatment immediately while the other group started treatment fifteen weeks later. Approximately six months following termination of treatment for the first group and three months following termination of treatment for the second group, followup data were collected. Results indicated that subjects in the first group lost an average of thirty-five pounds while those in the second group lost an average of twenty-one pounds.

Research in the treatment of obesity has indicated that behavioral approaches are generally effective in helping people lose weight. However, outcomes vary across programs; that is, not all clients are equally successful at losing weight during treatment or maintaining weight losses over a follow-up period. This point is brought out in the following studies.





4

Jeffrey, Christensen, and Katz (1975) reported the results of four subjects whom they had treated for obesity using a behavioral approach. After six months of treatment weight losses ranged from twenty to thirtytwo pounds. However, after a six month follow-up period weight losses (pretreatment to follow-up) ranged from three to thirty-two pounds; two subjects had maintained their weight loss while the other two subjects gained back part or most of their weight.

Harris and Bruner (1971) reported two studies examining the effectiveness of a self-control behavioral program and a contract procedure for weight reduction. In the first study self-control procedures included positive reinforcement, stimulus control, and breaking or lengthening the chain of eating while the contract program consisted of subjects receiving a monetary reward for each pound lost. Both groups lost a significant portion of body weight after two months of treatment. Furthermore, the contract group had lost a larger proportion of initial weight than the self-control group. However, after a ten month follow-up period there were no significant weight losses for either group. In a second study, Harris and Bruner (1971) compared a self-control weight group with a no-treatment control group and found no significant weight losses for either group after sixteen weeks.

Penick, Filion, Fox, and Stunkard (1971) compared a behavior

modification weight control gorup with a control group which received supportive psychotherapy as well as instruction about dieting and nutrition. After three months of treatment the median weight loss for the behavior modification group was greater than that of the control group, but this difference did not reach significance. Penick et al. attribute this result to the significantly larger variability of outcome in the






5

behavior modification group. The five best reducers belonged to this group as did the single least successful one, the only subject who gained weight during treatment.


The Self-Concept


In order to explain the large variability of outcome in behavior

therapy,Ince (1972) suggests that the self-concept may be a moderator variable affecting outcome, but that it is largely ignored by behavioral practitioners. The self-concept, according to Rogers (1951) consists of an organized conceptual pattern of the "I" or "me" together with the values attached to those concepts.

Ince (1972) cited two clients whom he treated using behavioral

techniques. In both cases, treatment of the overt symptoms by behavior modification proved insufficient to effect change. Underlying each client's difficulties was a poor self-concept which needed to be enhanced in order for the treatment to be effective. Hence, the therapist began verbally reinforcing each client for positive self-references with the goal of increasing such verbalizations which would then become internalized and thus modify the self-concept. This approach met with success, and suggested that the modification of an individual's self-concept can be an important variable and one upon which success of therapy might depend.

,Ziller's (1973) helical theory of personal change supports

Ince's (1972) proposition that a change in the self-concept is a necessary and desired outcome of personal change procedures. This theory suggests a hierarchy of potentially changeable characteristics which include attitudes, values, behaviors, roles, and self-concepts. It





6

is assumed that these characteristics are ordered according to their ease of change, with attitudes being the least difficult to change while self-concepts are the most resistant to change.

Ziller's helical theory proposes that the self-concept is the

anchoring characteristic of the system, or the ultimate level of the personal change hierarchy.

Those who would effect change in attitudes, values, behavior, or roles are necessarily
concerned with a change in the self-concept, for it is assumed here that unless a change
in the self-concept is achieved which is congruent with changes at lower levels in the
hierarchy, the lower level change is likely to be reversed and the organism returned to
the initial state of equilibrium.(p. 174)

Self-Concept and Body-Cathexis

Secord and Jourard (1953) investigated the relationship between an individual's self-concept and body-cathexis. By body-cathexis is meant the degree of feeling or satisfaction with the various parts or processes of the body. In order to appraise body-cathexis subjects were presented with a list of forty-six items, each describing a different part or function of the body, and were asked to indicate on a five point scale their degree of satisfaction with each item. In order to measure selfcathexis, individuals were asked to rate in a similar fashion fifty-five items believed to represent a sampling of the various conceptual aspects of the self. Subjects were seventy college males and fifty-six college females.

Intercorrelations between body-cathexis and self-cathexis scores were .58 for men and .66 for women. Hence, individuals had a moderate tendency to cathect their body to the same degree and in the same






7

direction that they cathected their self. In addition, it was found that females cathected their bodies, irrespective of direction, more highly than did males, in that they did not assign as many threes (have no particular feeling one way or the other) to body items. Secord and Jourard suggested that women would be more likely than men to develop anxiety concerning their bodies because of the social importance of the female body.

Rosen and Ross (1968) noted that in the correlations obtained by Secord and Jourard (1953) between body-cathexis and self-concept, they did not take into account that certain parts or processes may be more important to an individual than other parts or processes in evaluating his body- and self-concepts. Hence, Rosen and Ross (1968) investigated the relationship between body image and self-concept, taking into account the relative subjective importance of the aspects being rated.

Eighty-two undergraduates were presented with a list of twentyfour body parts and seventeen adjectives from the Adjective Check List. They were asked to indicate on a five-point scale for each body part or adjective how satisfied they were with that aspect and how important that aspect was to them. Correlations of subjects' mean satisfaction scores between body image and self-concept were; r = .52 for all items; r = .62 for all items above mean importance; and r = .28 for items below mean importance. Hence, these findings support those of Secord and Jourard (1953), and in addition, they indicate that the relationship between body-cathexis and self-concept can be refined if the subjective importance of each component is considered.
Lerner, Karabenick, and Stuart (1973) asked 118 male and 190 female college students to rate twenty-four body characteristics in terms of





8


how satisfied they were with each characteristic of their own body and how important each part was in determining their own physical attractiveness. In addition, subjects responded to a short self-concept scale consisting of sixteen bipolar dimensions derived from the Adjective Check List. Results indicated that the degree of positive self-concept increased with the degree of positive attitude toward one's body characteristics (males: r = .33; females: r = .43). However, weighting satisfaction ratings by corresponding importance ratings did not significantly increase the satisfaction/self-concept relation (males: r = .33; females: r = .44). Hence, while this study supports the general relationship between self-concept and body-cathexis put forth by Secord and Jourard (1953), it failed to support the results of Rosen and Ross (1968) that importance ratings strengthened this relationship.

Mahoney (1974) replicated the Lerner et al. findings also using college students as subjects and self-report attitude measures. He found the correlation between mean unweighted (by importance) body-cathexis and self-esteem to be .45 for males and .37 for females. For weighted body-cathexis the respective correlations for males and females were .37 and .41.

Further support for the relationship between body attitude and self-esteem has been provided by Weinberg (1960), Zion (1965), and Mahoney and Finch (1976). In each study, moderate correlations were found using college students and self-report instruments. The use of an identical scaling procedure to measure self-esteem and body attitude raises the question that the high correlation found between bodycathexis and self-esteem may represent method variance and not an intertrait relationship. The following study addresses this issue.






9

Kurtz (1971) tested the hypothesis that body attitudes are related to feelings of self-esteem using two instruments which share very little method similarity. He used a self-report body attitude scale (methodologically similar to that used by Secord and Jourard, 1953) and the Ziller (1969) Self-Esteem Scale, which is a more indirect or projective technique for measuring a person's feelings toward himself. For the group of college students sampled, the hypothesis that a positive evaluative body attitude and a positive sense of self-esteem are related was confirmed. Of importance was the strength of this relationship considering that the scales used shared little methodological similarities.

The relationship between body-cathexis and self-concept sheds some light on the self-concept of overweight individuals. It is assumed that people who volunteer to participate in weight reduction programs are dissatisfied with their bodies. Since people tend to cathect their bodies to the same degree that they cathect themselves, overweight individuals in weight control groups are probably somewhat dissatisfied with themselves. Hence, the preceding research on the relationship between body-cathexis and self-concept implies that obese individuals in weight reduction groups may be partly characterized by feelings of low self-esteem.


Obese Self-Concept and Body-Cathexis


,Werkman and Greenberg (1967) compared eighty-eight obese adolescent girls at a medically oriented camp for overweight girls with forty-two normal-weight girls at an ordinary summer camp on a number of personality and interest measures (MMPI, Strong Vocational Interest Blank, Semantic Differential, and Sentence Completion Blank). One finding which cut





10


across all the test results was the presence of a response set. The obese were defensive about revealing any psychological problems and, in short, presented themselves as "hypernormal." In addition, there was a great difference between the obese and the control group on the social introversion scale of the MMPI, suggesting that the obese girls were more uncomfortable and anxious in social situations.

Vocational interest patterns of the two groups also differed.

While the control group identified with professions in which imagination, ambitiousness, creativity, and intellectual strivings were paramount, the obese girls' interests were more consistent with persons in nurturant professions. The obese showed a kind of maturityy; that is, their interests were similar to those people whose occupations are stable and "realistic."

These results suggest that the obese tend to restrict themselves both socially and vocationally. They appear to live within a conventional life pattern in which one does not attempt situations which might provoke anxiety. In addition, through their efforts to appear normal they may be sacrificing spontaneity and flexibility in many spheres of psychological functioning. This pattern may hamper them by preventing the full development of character through conflict and acceptance of challenges.

Held and Snow (1972) studied twenty-three obese adolescent girls and twenty-three non-obese adolescent girls who were randomly selected from patients being seen for medical reasons at an out-patient clinic. Subjects were individually administered the MMPI, Mooney Problem Check List, and the Rotter Internal-External Locus of Control Scale. The obese group scored significantly higher on five of the ten clinical scales of





11

the MMPI than the non-obese group. These scales were D (depression), Pd (psychopathic deviate), Pa (paranoia), Pt (psychasthenia), and Sc (schizophrenia). The generally elevated profile of the obese group, particularly the Pd and Sc scales, suggested that many had serious psychological disturbances characterized by feelings of depression, alienation, and low self-worth. In addition, they tended to be non-conforming, had problems in impulse control, and were distrustful of others.

Held and Snow failed to find the obese group defensive about revealing psychological problems in that they admitted to significantly more difficulties on the Mooney Problem Check List than the non-obese group. These findings suggested that obese adolescent girls may need help reconciling some of the feelings they have toward themselves and other people in addition to programs designed to help them lose weight.

Wunderlich, Johnson, and Ball (1973) administered the Adjective

Check List (ACL) and the Edwards Personal Preference Schedule (EPPS) to sixteen subjects ranging from 64.6% to 214.7% overweight. The obese group endorsed a significantly fewer number of favorable adjectives on the ACL than did the normative sample. In addition, the obese were characterized as having a low need for achievement, poor personal adjustment, and few preferences for close attachments to other people. They did not describe themselves as being dependable, cautious, stable, or inhibited. Instead, they described themselves as hasty, pleasure-seeking, headstrong, and rebellious. Order for them was low, as they described themselves as being careless, changeable, and disorderly, as well as having poor self-control.

The obese scored low on endurance, suggesting that they were less

willing to expend energy toward a goal or persist at a task. Hence, the





12

obese tended to prefer sedentary activities or an inactive, passive lifestyle. The obese also scored low on dominance, indicating that they are seldom ambitious, determined, or assertive. They also characterized themselves as being manipulative, hostile, and aggressive. It seemed, therefore, that they would be likely to react to unexpected demands with sudden withdrawal and underlying hostility. Finally, the obese were very self-conscious about their size and expressed strong sexual conflicts and frustrations. This seemed to be produced by the desire to enjoy heterosexual contacts but the inability to engage in these behaviors due to the attitudes of themselves and others concerning their bodies.

Quereshi (1972) studied 180 female members of TOPS (Take Off Pounds Sensibly) who had considerable difficulty with weight reduction, and on the average weighed over 200 pounds. He compared them with ninety-eight females who had been through the TOPS program and were successful over a six month period at staying within 5% of their ideal weight. These people were know as KOPS (Keep Off Pounds Sensibly). All subjects were administered the Michil Adjective Rating Scale (MARS) which consisted of forty-eight adjectives such as "nervous," "talkative," and "ambitious," and was accompanied by a five-point scale ranging from "very atypical" to "very typical." This scale yielded four personality factors, which were labeled as unhappiness (Factor 1), extraversion (Factor 2), selfassertiveness (Factor 3), and productive-persistence (Factor 4).

The mean for TOPS' self-ratings on Factor 1, unhappiness, was significantly larger than that of KOPS, which indicated that TOPS perceived themselves as generally unhappy, nervous, tense, and dissatisfied. TOPS also had a significantly larger mean than KOPS on Factor 2, extra-




13




version, which suggested that the obese considered themselves to be more outgoing and friendly than KOPS. However, Quereshi noted that obese persons' perceptions of themselves as extraverts did not necessarily mean that they were since previous research (Mayer & Thomas, 1967, cited in Quereshi, 1972) utilizing projective techniques with obese females evidenced traits such as passivity and withdrawal accompanied by feelings of rejection.

A more realistic explanation may be gleaned from considering the

high self-ratings on Factors 1 and 2 together. This suggests that obese females, despite their attempts to gain approval from others by means of friendliness and congeniality, perceive themselves as lonely and rejected. This explanation would also support the findings of Werkman and Greenberg (1967) and Held and Snow (1972).

Gottesfeld (1962) compared self-drawings of thirty super-obese

subjects with thirty neurotics' self-drawings in terms of each group's body-cathexis, or degree of satisfaction with the parts and processes of their bodies. The super-obese subjects showed a more negative bodycathexis than the neurotic group on the following three criteria; (1) the super-obese were judged as having more negative body-cathexis by a group of clinicians; (2) their drawings had more major parts of the body missing; and (3) their drawings were less differentiated.

Gottesfeld (1962) also gave a list of twenty-eight personal traits to the same group of super-obese individuals and neurotics and asked them to rate themselves first as how they are and second as how they would like to be. The discrepancy between self and ideal self served as a measure of the degree of satisfaction with the self.




14



Results indicated that the super-obese reported to be more satisfied with themselves than the neurotics. Later, however, while the superobese subjects were hospitalized for a two week evaluation, three independent judges observed their interactions with other patients and staff, and rated them on the same list of twenty-eight personal traits. The differences between self-ratings and observer (objective) ratings were significantly greater than the differences between self-ratings and ideal ratings. Gottesfeld concluded that the super-obese patients seemed to present a facade of satisfaction. They denied that they were dissatisfied with themselves on a self-report trait list, but they could not as easily guard against a negative self-picture in their drawings (projective test) or in observers' ratings.

Stunkard and Mendelson (1967) have found very low body-concepts in some overweight people. These people felt that their bodies were grotesque and loathesome and that others viewed them with hostility and contempt. This feeling was associated with self-consciousness and impaired social functioning. In addition, the person took a very narrow view of himself, expecially during times of misfortune or unhappiness. All the unpleasant aspects of his life became focussed on his obesity; that is, his body became the explanation and the symbol of his unhappiness. Buchanan (1973) stated that the body was the receptacle for selfhate for obese individual; when they felt self-hate they complained they felt fat.

Negative body-concepts have been found to be most prevalent among persons who became obese during childhood or adolescence (Stunkard and Mendelson, 1967; Stunkard and Burt, 1967). In addition, this condition seemed to persist despite weight reduction and prolonged maintenance of normal body weight (Stunkard and Burt, 1967).






15

The research review thus far suggests that there are many psychological and social problems associated with obesity. The obese have problems with impulse control and tend to become easily angered, irritated, or resentful. They also describe themselves as disorderly, pleasure-seeking, and rebellious. In addition, they are seldom active or competitive and hardly expend a great deal of effort to accomplish a difficult task. They sometimes take a very unrealistic view of themselves and tend to blame many of their difficulties on their obesity.

Obese adolescents and young adults seem to have a poor self-concept which is reflected in their dissatisfaction with their bodies as well as with themselves. They harbor deep feelings of insecurity and have exaggerated needs for attention and social approval. The world is seen as a threatening and rejecting place and their response is to withdraw as a defense against being hurt. As a result, they often feel lonely and alienated, distrust other people, and avoid close interpersonal relationships.

These obese persons are self-conscious about their size and seem to feel uncomfortable and anxious in a variety of situations. Consequently, they tend to restrict their interests both socially and vocationally and they may be deficient in basic social skills. In addition, these individuals seem to have serious concerns about their masculinity or feminity, feeling afraid that they cannot perform adequately in sexual situations. Weight Reduction and Self-Concept


Glucksman and Hirsch (1969) compared obese subjects' performance on a body size estimation task before, during, and after weight loss with that of a normal control group. Six obese subjects with a mean initial





16

weight of 334 pounds were hospitalized for eight months and lost an average of 86.7 pounds. Using an adjustable distorting image apparatus, subjects were requested to make a distorted screen image of themselves correspond to their body size as they perceived it at that moment. Obese subjects increasingly overestimated their own body size from pretreatment, during fifteen weeks of weight loss, and after a six week maintenance period. In effect, despite their weight loss, they perceived themselves as if they had lost almost no weight. In addition, three of the obese subjects in this study were retested after an additional year of weight loss, and they continued to overestimate their actual body size.

This "phantom body size" phenomenon was accompanied by supportive clinical and figure-drawing data. For example, these same subjects drew progressively larger figure-drawings during weight loss and at the end of the final weight maintenance period. In addition, they reported that they continued to feel obese despite weight loss (Glucksman, Hirsch, McCully, Barron, and Knittle, 1968).

Suczek (1955) tried to delineate psychological aspects of weight reduction. He administered the Interpersonal Dimension of Personality System (IDPS) to 100 obese women before and after a sixteen week treatment program. This instrument analyzes behavior in terms of five discrete levels: I, the level of public communication; II, the level of conscious description; III, the level of private symbolization, IV, the level of unexpressed unconscious; and V, the level of ego ideal. In this study, only data from levels I and II (i.e., overt, facade behavior) were described.

There was little variability or inconsistency between level I and level II data at pretreatment. In other words, obese women saw little




17



difference between their views of themselves as they impress others through overt behavior and their conscious views of themselves (i.e., between how they are and how they claim to be). In essence, they presented themselves as conflict-free.

Subjects were assigned to groups of ten to fifteen individuals, which met once a week for sixteen weeks. Half the groups were led by psychiatric social workers whose aim was to promote change toward self and others while the other groups were led by dieticians and nutritionists whose aim was to guide spontaneous discussion and provide dietary information. Subjects in both types of groups were relatively successful in losing weight. However, there was little or no change at posttreatment on the IDPS. Thus, at the level of overt behavior and their conceptions of themselves, the obese women resisted change through their experiences in the group--their personalities did not change.

Suczek later compared the obese group with a sample of neurotics undergoing group therapy and found that neurotics did undergo personality changes. Hence, changes in behavior and in self-conception were achievable by means of group methods and measurable by the IDPS. However, the obese women were not readily amenable to change and did not change their attitudes about themselves to any appreciable degree in the weight reduction groups. Unfortunately, no long term follow-up results were reported to determine if weight losses were maintained.

,Collingwood and Willett (1971) investigated the effects of physical training on self-attitude changes of five male teenagers enrolled in a special YMCA obese physical training program. The program consisted of ten hours of exercise per week for three weeks, and a total of three hours of group counseling-discussion. Two attitude scales were




18


administered at pretreatment and again at posttreatment. A Body Attitude Scale, containing fifteen body concepts, measured subjects' attitudes toward their bodies along the following three dimensions: (1) Evaluative dimension with good-bad, awkward-graceful, and beautiful-ugly bipolar adjectives; (2) Potency dimension with weak-strong, hard-soft, and thin-thick bipolar adjectives; and (3) Activity dimension with activepassive, cold-hot, and fast-slow bipolar adjectives. In addition, Bills' Index of Adjustment and Values (IAV) containing twenty-four adjectives, measured subjects' attitudes toward themselves along the following three dimensions: (1) Self-Concept ("seldom like me" to "most of the time like me"); (2) Self-Acceptance ("I very much dislike being as I am in this respect" to "I very.much like being as I am in this respect"); and (3) Ideal Self ("seldom would I like to be that way" to "most of the time I would like to be that way").

Results at posttreatment indicated a significant weight decrease

and a significant increase on both the evaluative and potency dimensions of the Body Attitude Scale. In addition, on the IAV there were significant increases on the Self-Concept and Self-Acceptance dimensions, and a significant decrease on the discrepancy between Self-Concept and Ideal Self. These results indicated that physical training experiences can help teenagers lose weight and enhance their attitudes toward themselves. However, long-term follow-ups are needed to determine the stability or permanency of these changes.

Rohrbacher (1973) studied 204 overweight boys between the ages of eight and eighteen in an eight week weight reduction camp program. Body image and self-concept assessments (Secord and Jourard, 1953, scales) were made before and after the camp program, and sixteen weeks





19

after the camp program had ended and subjects had returned home. Significant weight losses were recorded at posttreatment and at the followup, although subjects did regain some weight during the follow-up period. In addition, body image showed a significant positive change as a result of the camp program, but self-concept remained unchanged. Rohrbacher suggested that changing body image may be an important factor related to the program's success, but that longer follow-ups are needed to accurately assess the program's effectiveness.

Weller, Arad, and Levit (1977) compared the self-concepts of

twenty-five women who had successfully reduced their weight with those of twenty-five women who were unsuccessful at weight reduction. On the Tennessee Self-Concept Scale, successful dieters revealed a better self-image (general self-concept score) and a better physical self-image (physical self score) than did the unsuccessful dieters. These results suggested that those individuals who were most successful at weight reduction had also modified their self-concepts in the direction of feeling better about themselves and their bodies.

McCall (1973) studied MMPI profiles of two groups of women belonging to TOPS (Take Off Pounds Sensibly). One group had considerable difficulty losing weight and/or maintaining whatever weight loss they had achieved. Such refractorily or irremediably obese were designiated R-TOPS. The other group had successfully lost weight and had maintained their ideal body weight for at least six months. These women were referred to as KOPS (Keep Off Pounds Sensibly).

On nine of the ten MMPI clinical scales, R-TOPS women were significantly more deviant than KOPS women. On six of the nine scales, the differences were significant at the .01 level; on three scales the level




20



of significance was .05. From the six scales on which the relative differences were greatest R-TOPS women exhibited more body concern (hypochondriasis), psychic "hurting" (depression), somatization (hysteria), rebelliousness (psychopathic deviate), compulsive and ruminative tendencies (psychasthenia), and bizarre or confused thinking (schizophrenia).

McCall (1974) studied nineteen women TOPS club members with a mean weight of 204 pounds. He broke them down into three experimental subgroups as follows: One group was chosen because their MMPI profiles closely resembled those of successful weight reducers previously studied (KOPS); another group consisted of women whose MMPI profiles closely resembled those of the resistively obese (R-TOPS); and a third group fell in-between these two extreme groups. The question asked was whether the MMPI profiles that distinguished between these three subgroups would have any bearing on success in group therapy and weight reduction.

Subjects were randomly assigned to one of three groups which met for sixteen weeks and were oriented toward the development of self-control. Pre- and posttherapy data showed that only the R-TOPS-like subgroup showed significant changes after therapy. They changed on the following six clinical scales: Hs (hypochondriasis), D (depression), Hy (hysteria), Pd (psychopathic deviate), Pt (psychasthenia), and Sc (schizophrenia). These six scales had previously most differentiated the refractorily obese (R-TOPS) from the remediated obese (KOPS). In addition, only the R-TOPS-like subgroup showed a significant weight loss at posttreatment. Hence, obese women who had the "worst" MMPI profiles tended to benefit most from group therapy as indicated both by weight loss and profile improvement.




21


The research in weight reduction and concomitant changes in the

self-concept suggests that the self-concept does not change as quickly as a person loses weight. Weight control programs which solely emphasize restricting your caloric intake seem to have little immediate effects on the self-concept. These programs tend to ignore many of the psychological or interpersonal problems associated with being overweight. Since they do not address themselves to all facets of the problem of obesity they appear to be incomplete. While clients may lose weight, they do not always change their attitudes toward themselves and will often manifest a "phantom body size." According to Ziller (1973):

A change in the self-concept is the desired
outcome of personal change procedures but the
change processes must involve changes in attitudes, values, behaviors, and roles.
. If the desired changes in the selfother orientations have not been achieved, the
client is likely to revert to earlier attitudes,
behaviors, and roles. (p. 176)
Hence, the relative stability of the obese person's self-concept may account for the poor long-term success of many weight reduction programs.

The optimal strategy for personal change seems to be to change the self-concept. Physical training programs and some experiences in group therapy seemed to have some effect on the obese person's self-concept (Collingwood & Willett, 1971; Rohrbacher, 1973; McCall, 1974), while other group experiences yielded little or no self-concept change (Suczek, 1955).

Role Playing and Self-Concept

Ziller's (1973) helical theory of personal change indicates that changing a person's role may exert a strong influence on a person's




22


self-concept; therefore, focussing on taking on new roles may be a valuable dimension in the treatment of obesity. In review, Ziller's theory suggests a hierarchy of potentially changeable characteristics which include attitudes, values, behaviors, roles, and self-concepts. It is assumed that these characteristics are ordered according to their ease of change, with attitudes being the least difficult to change while selfconcepts are the most resistant to change.

If one of the components in the system is changed a state of

disequilibrium within the system ensues. It is assumed that in this imbalanced state there is a tendency for the components lower in the hierarchy to change in a way which will render the components congruent. In addition, a change in one component will exert some press toward change in components higher in the system. A change in behavior, for example, will exert some press toward changes in attitudes, values, roles, and self-concepts, but with diminishing force with regard to the higher level components.

Behavioral approaches to weight control are concerned with changing behaviors, and this will cause the system to be in a state of disequilibrium. However, a change in behavior will not exert a very strong press on the individual's self-concept. Hence, in order for the system to regain equilibrium the person is likely to revert back to old ways of behaving (i.e., old eating habits) which are congruent with the self-concept, rather than change the self-concept. Consequently, weight maintenance is unlikely.

A change in a person's role will exert a stronger press on a person's self-concept than a change in behavior since roles are closer than behaviors to self-concepts on the personal change hierarchy. This




23


suggests that weight control clients should be more likely to change their self-concepts, and therefore, be more likely to maintain weight losses if treatment foccusses on changing their roles in addition to changing their behaviors.

Role playing techniques or behavioral rehearsal may facilitate role changes. This, in turn, may have a strong effect on a person's self-concept. Horrocks and Jackson (1972) explain this process. First, they distinguish between role taking and role playing behavior. Role taking is a concrete manifestation or implementation of a hypothesized identity and presents an observable product of the self-process. Role playing, however, represents performance by an individual of prescribed, demand behavior determined by a situational context. The role player behaves according to situational expectations, but the behavior does not represent anything the person believes himself to be in that context. Role playing behavior, therefore, is the manifestation by an individual of anti-identities, or those roles not conceptualized by an individual of himself at that moment.

When an individual role plays anti-identities, locus of control of his behavior is external to him, imposed upon him by situational or behavioral demands. When the person receives feedback on his performance, he assimilates into his cognitive structures the effectiveness and appropriateness of these actions. In this manner, cognitive dissonance may alter previous cognitions and change a conception of the self to include aspects of the anti-identity. As a result, new meanings of the self may be conceptualized, and role taking behavior replaces role playing.

Horrocks and Jackson suggest that the greater the array of identities an individual incorporates into his conceptualization of himself






24

the greater is his potential for flexible adaptation. Ziller (1973) supports this idea and states that

individuals with complex self-concepts may be
aware of or consider a greater number of
stimuli as being potentially associated with
the self. In terms of interpersonal perception,
the complex person has a higher probability
of matching some facet of the self with a
facet of the other person, since there are a
larger number of possible matches.(p. 79)

Hence, complex persons are less likely to be disturbed by new experiences or situations which appear to be incongruent with their selfconcepts.

The theories of Ziller (1973) and Horrocks and Jackson (1972),

taken together, suggest that role playing experiences facilitate role taking behavior, and effect the complexity of the self-concept. Hence, the individual perceives and incorporates new facets of the self; therefore, he becomes more adaptive or less disturbed by change.

The value of role playing techniques in effecting role change has been demonstrated by Lazarus (1966). He compared the effectiveness of behavior rehearsal with two other techniques, direct advice and nondirective reflection-interpretation, in the management of specific interpersonal problems. Seventy-five clients were randomly assigned to one of the three procedures (twenty-five clients per subgroup) and a maximum of four thirty-minute sessions was devoted to each treatment condition. If there was no evidence of change or learning within one month the treatment was regarded as having failed. The criterion of change or learning was objective evidence that the client was behaving adaptively in the area which had previously constituted a problem; e.g., the socially awkward girl was going out on dates, the company executive






25

was effecting a promising merger, or the considerate husband had persuaded his wife to move out of her parents' house into a home of their own.

Results indicated evidence of learning in eight (32%) of the clients treated by reflection-interpretation, eleven (44%) of the clients by advice, and twenty-three (92%) of the clinets treated by behavior rehearsal. In addition, of the thirty-one clients who did not benefit from reflection-interpretation or from advice, twenty-seven were then treated by behavior rehearsal and there was evidence of learning in twenty-two (81%) of them. Thus, the overall effectiveness of behavior rehearsal in fiftytwo cases was 86.5%. Hence, behavior rehearsal appeared to be significantly more effective in changing behavior outside the therapy session than direct advice or non-directive therapy.

The literature review revealed that the self-concept of obese

adolescents and young adults tends to be characterized by feelings of low self-esteem, a restricted view of themselves focussed mainly on their bodies or their size, and social isolation or social anxiety. Moreover, a change in this self-concept may be a factor facilitating long-term maintenance of weight loss.

It appears that adapting new roles allows a person to see himself in a variety of different ways, or expands a person's view of himself, and makes him more adaptive to change. In addition, role playing techniques may facilitate role taking behavior, and may exert a strong press on changing a person's self-concept. Based on these premises, the present study was devised to test the efficacy of role playing techniques in a behavior modification weight reduction group in order to help clients modify their self-concepts, and to determine the effects this






26

on clients' abilities to lose weight and maintain their losses over a follow-up period.

In order to carry out this study, two weight reduction groups
were conducted. Both groups emphasized behavioral techniques in order to lose weight. In addition, one group included an emphasis on using role playing techniques to facilitate clients' abilities to take on new roles and change their self-concepts. The other group was presented with theories and information on weight control.

The following served as the hypotheses:

1. Both groups will show significant reductions in weight at posttreatment.

2. The "role playing" treatment group will show a significantly greater weight loss than the "information" treatment group at the follow-up.

3. The "role playing" treatment group will have a higher self-concept than the "information" treatment group at the follow-up.

4. The "role playing" treatment group will have a higher complexity of the self than the "information" treatment group at the follow-up.

5. The "role playing" treatment group will have a higher physical selfconcept than the "information" treatment group at the follow-up.

6. At the follow-up, "successful" weight reducers will have a higher self-concept than "unsuccessful" weight reducers.














METHOD


Subjects

The subjects in this study were twenty-nine female college students enrolled in a weight reduction course taught through the Department of Health Education at the University of Florida. Two sections of the course were offered--one met Monday afternoon ("information" group) and one met Thursday afternoon ("role playing" group). Sixteen women enrolled in the Monday afternoon section ("information" group) and thirteen women enrolled in the Thursday afternoon section ("role playing" group). Subjects signed up for the section they preferred, mostly depending on their schedule of classes. Subjects were unaware of any difference between the two treatment groups at the time of registration.

Pretreatment measurements for the two groups are summarized in

Table 1. In order to determine if there were any differences between the two groups at pretreatment a two-group discriminant analysis was performed using all pretreatment variables. This analysis indicated that the "family self" subscale of the Tennessee Self-Concept Scale had the most discriminating power of all the variables entered (F = 4.79', df = 1,27, p <.05). The analysis produced no additional sigificant discriminating variables.

Since twenty-two variables were put into the analysis, it is likely that some differences may appear by chance and may not reflect actual differences between the two groups. In addition, "family self" appears

27




28



Table 1

Pretreatment Measurements of "Role Playing" Group and "Information" Group



Role Playing Information (n = 13) (n = 16)

M SD M SD


Initial Weight 176.23 44.49 166.94 25.20 Amount Overweight 45.62 37.41 36.19 23.20 Percent Overweight 34.15 26.04 27.75 18.07 Age 19.77 2.32 20.25 2.50 G.P.A. 2.74 0.42 2.77 0.50 Years Overweight 10.08 7.08 10.44 7.43 Class (1=freshman,
2=sophomore, 3=jr.,
4=sr.) 2.08 0.86 2.44 0.81

Ziller Complexity of
Self Scale 41.15 14.15 44.81 15.78

Tennessee Self-Concept
Scale (raw scores
converted to Std. T scores: M = 50; SD
= 10)

Self-Criticism 48.23 10.17 52,69 12,48 Total Positive 43.46 11.46 46.69 7.74 Identity 45.85 13.09 50.25 6.93 Self-Satisfaction 44.69 8.53 46.25 7.71 Behavior 42.15 11.78 45.44 8.41 Physical Self 34.69 7.22 36.00 11.58 Moral-Ethical Self 45.31 12.05 48.69 6.73 Personal Self 44.77 11.74 47.81 8.73 Family Self 46.54 10.63 53.31 5.78 Social Self 51.31 14.03 50.50 10.15 Total Variability 54.08 7.65 55.25 9.23 Row Variability 53.92 8.64 54.69 8.22 Column Variability 52.62 6.70 53.81 10.69 Distribution 44.77 11.03 48.88 8.42




29


to be unrelated to the focus of this study, and therefore, should have little effect on the outcome. Hence, it was assumed that there were no differences between the two groups at pretreatment.

Students in this class were graded on a pass/fail basis depending solely on their record of attendance; those who attended eight out of nine sessions received a passing grade. Only female subjects were used because weight reduction programs attract mostly female clients. Subjects met a number of criteria for acceptance into the program: (1) a need to lose at least twenty pounds; (2) not currently enrolled in another weight reduction program; (3) not taking medication for weight; and (4) having no history of a metabloic or hormonal imbalance that could affect weight.


Measures


Tennessee Self-Concept Scale (TSCS)--Counseling Form


The TSCS (Fitts, 1965) was administered to all subjects at pretreatment and at the follow-up. The Scale is a self-administered paper and pencil test consisting of 100 self-descriptive statements which the subject uses to portray a picture of herself. For each item the respondent chooses one of five response options labeled from "completely false" to "completely true." Fourteen scores are derived from these items: self-criticism; nine self-esteem scores (identity, self-satisfaction, behavior, physical self, moral-ethical self, personal self, family self, social self, total); three variability of response scores (variation across the first three self-esteem scores, variation across the next five self-esteem scores, total variation); and a distribution score. Raw scores on the TSCS have been converted to Standard T-Scores; i.e., M = 50, SD = 10.






30



The TSCS was normed on a sample of 626 persons of varying age, sex, race, socioeconomic status, intellectual level, and educational level. The standardization group was overrepresented in number of college students, white subjects, and persons in the twelve to thirty year age bracket. Test-retest reliability with sixty college students over a two week period ranged from .80 to .92 for nine self-esteem scales and ranged from .60 to .89 for the other five scales.


Complexity of Self Scale (see Appendix A)


The Ziller (1973) Complexity of the Self Scale was administered to all subjects at pretreatment and at the follow-up. This scale measures the degree of differentiation of the self-concept. A person with high complexity of the self-concept requires more words to describe herself.

The Ziller Complexity of the Self Scale consists of a list of

109 adjectives and the task is to check the works that describe yourself. The score is the number of words checked. This scale has a range of zero to 109 and is scored in the direction of high complexity.

In a study involving 100 randomly selected students from grades

seven through twelve, the split-half reliability was .92. Test-retest reliability after one month for college sophomores was .72.


Other Instruments


A questionnaire asking for specific demographic information was given to all subjects at the initial meeting (see Appendix A). In addition, "role playing" subjects were asked to fill out a second questionnaire regarding how they thought weight reduction would affect






31

their interpersonal relationships (see Appendix A). The information gathered from the second questionnaire was used to develop appropriate role training situations.

Dependent Variable

The dependent variable used was a standardized index called the weight reduction index (RI), which is equal to the percent of excess weight lost multiplied by the relative initial obesity. In addition to weight, this index takes into account height, amount overweight, and weight goal, and is expressed by the following formula:

RI = (WI/Ws) X (Wi/Wt) X 100

where W1 = weight lost, Ws = surplus weight, Wi = initial weight and Wt = target weight.


Procedures

The experimenter served as the moderator or therapist for each group. The emphasis in both groups was on gradual weight loss using self-control techniques designed to improve eating habits. Each group met for two hours, once a week for nine weeks. In addition, there was a four month follow-up.

Week 1

At the initial meeting subjects were given a course outline (see

Appendix B) and the experimenter explained the principles of the program. Subjects filled out the TSCS, Complexity of Self Scale, and Questionnaire #1. In addition, "role playing" subjects filled out Questionnaire #2. At the end of the initial meeting, subjects were weighed, were given a






32

weight control manual, and were asked to take a baseline of their eating and exercise behavior for the first week. Week 2

Each subject weighed-in before the class period began. The first forty-five minutes of this session was an interaction period in which subjects discussed their reactions to taking a baseline (i.e., "In what situations were you most likely to eat?"). The experimenter facilitated discussion by reflecting the content and feelings being expressed.

During the next forty-five minutes of this session the experimenter explained the food program to be used by subjects. An exchange system was used where subjects selected from lists of six food groups (meat, cereal, milk, fruits, vegetables, and miscellaneous). In addition, each subject figured out her daily caloric intake limit in order to lose one pound per week. Subjects were also asked to keep a graph of their daily caloric intake, their daily exercise level, and their weekly weight.

During the final part of the session subjects participated in an exercise period. Exercise consisted of walking or running around the Florida Field for six minutes. During the third week this time was increased to nine minutes, and for weeks four through nine subjects spent twelve minutes exercising.

Weeks 3 9

Standard Procedures

At the beginning of every class session each subject was privately






33

weighed on a balance scale. The first forty-five minutes of each class session in both groups was an interaction period. Subjects spent the time discussing difficulties and successes they had during the past week while the experimenter facilitated the discussion by reflecting the content and feelings being expressed by the subjects.

At the end of the interaction period each week the experimenter explained a few of the behavioral techniques for subjects to practice. Lesson plans followed the program developed by Moody and Schreiber (1979a, 1979b). Techniques focussed on gradual habit improvement through stimulus control, breaking or extending the chain of eating, and positive reinforcement. Some of these techniques included:

1. Eat slower: Chew thoroughly and swallow before you pick up another forkful. Take small bites and set a minimum time of twenty minutes for which your meal must last. Enjoy your food; do not gobble it.

2. Use small dishes: Use smaller dishes and smaller food containers. This will make small portions appear larger.

3. Use positive reinforcement: Silently praise yourself each time you adhere to one of the guidelines. In addition, reward yourself points each time you obey one of the principles, and when you accrue a certain amount of points, administer a pre-selected reward.

4. Develop a routine schedule of eating: Eat three meals each day and eat at the same time each day. Do not eat at other times! Do not eat on impulse.

5. Never engage in any other activities while eating: Pay attention to eating; do not watch T. V., read the newspaper, listen to the radio, or engage in similar activities. Do not eat unless you are hungry.

6. Prepare only one portion of each course at a time: Take only one





34


helping, one course at a time. Take small portions. Do not feel that you must complete a meal, especially if you planned the meal when you were hungry. If you are not hungry, leave the remaining food on your plate. Be guided by your feelings of hunger, not by the amount of food on the table.

7. Do not buy prepared foods: Limit your diet to foods which must be prepared before eating. Eat low calorie foods or foods that are difficult to eat (e.g., eat five carrots or crackers rather than one piece of candy). Do not tempt yourself.

8. Eat only in one place in one room: Only eat while sitting at the table. Do not take snacks into other rooms; do not eat at your desk; don't eat standing up. Do not eat casually.

9. Eat before a party: If you plan to attend a social event at which there will be a great temptation to eat high calorie foods, eat a small, low calorie meal before you go.

10. Do not prepare food while you are hungry: If you tend to nibble while preparing meals, prepare the meal at a time when you are not hungry (e.g., prepare dinner soon after lunch). If you go grocery shopping, do that right after eating. 11. Develop a pre-potent repertory: Think of times you have a hard time resisting food. Plan other activities you can be involved in during these times so that you won't be around food.

,The last part of each class session in both groups was spent
exercising. Subjects were also encouraged to exercise during the week between classes and to note improvements in the number of laps they completed during the twelve minute walk/run period.






35

Variable Procedures


Role Playing Group


During the second forty-five minute period of the third week "role playing" subjects were introduced to role playing. The experimenter explained to the class that when people lose weight other people may begin reacting to them differently. Hence, as a thin person they may be faced with new situations which will require new roles. As a result, part of this class was used to help students practice these new roles. Week 3


Exercise 1: Getting Acquainted

Purpose
a. To help group members get acquainted in a relatively non-threatening manner.

b. To explore feelings generated by "becoming another person." c. To explore the dimensions of a brief encounter. d. To emphasize the need for careful, active listening as well as self-disclosure during conversation. Instructions

1. Group members are paired in dyads, and the facilitator instructs participants to "get to know your partner" for the next few minutes (5 to 10 minutes). Participants are instructed to listen to the "free information" or clues' their partners give about themselves and to follow-up on this free information. In addition, participants are instructed to "self-disclose" or give information about themselves to their partners.






36

2. After the interviewing phase, group members reassemble in the larger group. The facilitator indicates that they now have the responsibility of introducing their partner to the group. Each group member, in turn, is to introduce her partner by standing behind her and speaking in the first person, as if she were that partner. There should be no rechecking between partners during this phase. The individual who is being introduced should hold her comments for the discussion period.

3. After all the introductions have been made, the facilitator leads a discussion of the exercise, focussing on feelings generated and/or the issues inherent in the goals of the exercise.


Exercise 2: On Being Fat and Thin

Purpose

a. To examine stereotypical differences between fat and thin people. b. To get subjects thinking about their feelings concerning being fat and being thin.

Instructions

1. Have the group generate a list of generalizations about fat and thin people. Brainstorm in this manner for five to ten minutes, creating an arsenal of stereotypes.

s. Discuss this exercise, focussing on subjects' feelings about the

list.


Homework: Think about new situations that you may encounter as a result of losing weight. What new roles will you become involved in as a thin person?

Purpose

a. To encourage students to expand their views of themselves, i.e.,





37


to increase their complexity of themselves. b. To have students think about situations they could role play. Week 4


Exercise 1: Giving Feedback Purpose: To teach students how to give helpful, appropriate feedback. Instructions

1. The facilitator defines "feedback" to the group as:

A way of helping another person to consider
changing her behavior. It is communication to
a person (or group) which gives that person
information about how she affects others.
Feedback helps an individual keep her behavior
on target and thus better achieve her goals.

2. Group members discuss various criteria for useful feedback. Exercise 2: Role Playing Fat and Thin People Purpose

a. To compare how it feels to be a fat person with how it feels to be a thin person.

b. To rehearse how to act in different situations. c. To develop an awareness of oneself in relation to other people, d. To see oneself in a variety of different ways. e. To learn how to give appropriate, helpful feedback. Role Playing Instructions

1. Make up a situation that you would like to role play. Discuss the incident carefully with the group and how the different characters might feel in the situation. Discuss the main character's goal and the best way to achieve this goal considering the way in.which other people are feeling. Next, have group members volunteer to act in the different





38

roles. Now dramatize the situation for five to ten minutes, or until it comes to a natural conclusion. Do the same situation twice: first, as you would act as an overweight person, and second, as you would act as a person of normal weight.

2. After you have enacted the situation both ways (i.e., overweight and normal weight roles) discuss the situation and give each other feedback in the following order: (1) main character: hwo did you feel about acting overweight and about acting normal weight? (2) other person(s) in the exercise: how did you feel about yourself in relation to the main character? (3) non-participant observers: what did you observe was going on? During the feedback session other group members may role play alternate ways of responding to the situation.

3. Think of various ways of responding no matter how strange they seem. Contrast responses and see what "fits."

4. There are no right or wrong answers in this exercise, so don't be afraid of making mistakes. We are practicing and comparing ways of acting as overweight and normal weight people so feel free to try out any role you like.

Weeks 5 to 9

During the second forty-five minute period of each week "role playing" subjects continued the role playing exercise. The following is a list,of various situations that were role played.

1. You see a friend whom you haven't seen in a long time. He remarks, "You look different. Have you been sick?"

2. You're home on vacation and your mother keeps "pushing" food on you. You'd love to taste some of the food, but you are already full.





39

Your mother feels very hurt because you don't want to eat her food.

3. You're at a party with some of your girlfriends. A guy you've been attracted to for awhile notices you and strikes up a conversation with you.

4. You've lost some weight but haven't yet reached your goal. You are sitting on a bus next to a guy whom you've just met. After talking with him for awhile he remarks, "You have a beautuiful face. Do you realize that if you lost some weight you'd really be a knock-out!"

5. You're with a girlfriend on campus and meet a guy that she knows. He ignores you and speaks only to your girlfriend. Even after your girlfriend has introduced you and has made several references to you during the conversation he continues to address his attention solely to your girlfriend.

6. You are at a bar and strike up a conversation with a guy. He seems friendly but as time goes by he becomes overly aggressive.

7. A guy calls you up and wants to take you out to eat. You want to go out with him but have already eaten.

8. You are with a friend and you are both anxious about an exam. Your friend says, "Let's go get something to eat."

9. You've lost some weight and you're walking home across campus. You meet a guy that you know and he tells you now great you look and insists on taking you to the Arrendondo Room for lunch. 10. ,You've lost weight and your roommate, who is a little overweight, begins making snide remarks about your body. You suspect she may be feeling a bit jealous of you.






40

Information Group


During the second forty-five minute period of each week the "information" treatment group was presented with various weight control theories. The following is an outline of the topics discussed each week.


Week 3 Hypnotherapy: Dr. Sig Fagerberg A. Conscious vs. subconscious B. The power of suggestion to the subconscious C. Hypnotic state D. Suggestibility and self-hypnosis

1. Daily use of suggestion

2. Simple, repetitive, positive imagery pertaining to the future E. Books

1. LeCron, Lesley. Self-Hypnosis

2. Peale. Power of Positive Thinking Week 4 External Cue Sensitivity A. Perceived time B. Food cues

C. Response cost D. Taste

E. Emotional state F. Feedback and competing cues Week 5 Nutrition and Dieting: Stephanie Fredette, Graduate Student in Nutrition

A. Basic four food groups






41

B. Definition of obesity C. Adipose cell theory

D. Basal metabolic rate E. Additives and preservatives F. Pinch test


Week 6 Surgical Methods of Weight Reduction: Dr. Joun Kuldeau, J. Hillis Miller Health Center A. Interstinal bypass surgery

1. Diagram and description of operation

2. Adverse physical side effects

3. Benefits of operation B. Stapling the stomack shut

1. Diagram and description of operation

2. Advantages of this procedure over intestinal bypass operation C. Case reports of super-obese clients D. Mood changes during rapid weight reduction Week 7 Hereditary and environmental factors in obesity A. Family research B. Twin studies

1. Variability in weights in identical twins

2. Identical twins reared in same environment vs. identical twins reared in different environments

3. Identical twins vs. fraternal twins vs. non-twin siblings C. Adopted children research Week 8 TOPS (Take Off Pounds Sensibly): Marilyn poss, Gainesville





42

Chapter of TOPS A. History and background of TOPS B. TOPS basic program

1. No prescribed diet

2. Changing eating habits 3. Group therapy approach

4. Competition--honors and rewards

C. KOPS (Keep Off Pounds Sensibly) Maintenance of weight loss D. Case reports of successful clients Week 9 Other Approaches to Weight Control A. Aversive conditioning B. Counting mouthfuls C. Fad diets and crash diets D. Scarsdale diet


Follow-up


During the week of July 9, 1979, all subjects were contacted by mail (see Appendix C). They were asked to report their current weight and to fill out the Tennessee Self-Concept Scale and the Ziller Complexity of Self Scale.













RESULTS

The sample at posttreatment consisted of 29 subjects; 13 were in the "role playing" treatment group and 16 were in the "information" treatment group. The attrition rate during the program was 0%. The follow-up sample consisted of 28 subjects; 12 were in the "role playing" group and 16 were in the "information" group. One subject from the "role playing" group had left the country and could not be contacted.


Hypotheses

Hypothesis 1: Both groups will show significant reductions in weight at posttreatment.
The mean reduction index at posttreatment was 20.00 for subjects in the "role playing" group and 14.06 for subjects in the "information" group. Additional analysis revealed that the mean reduction index at the follow-up was 18.58 for subjects in the "role playing" group and 38.19 for subjects in the "information" group. One-way t tests indicated that all four of these indices were significant at the .05 level or greater (Table 2). Hypothesis 1 was accepted.











43






44


Table 2

Mean Reduction Indices for "Information" and "Role Playing"
Groups



Role Playing Information M SD df t M SD df t


Posttreatment 20.00 24.91 11 -2.78** 14.06 17.51 15 -3.21** Follow-up 18.58 32.34 11 -1.99* 38.19 29.37 15 -5.20***


*E< .05
** 2. .01
*** < .001

Hypothesis 2: The "role playing" group will show a significantly greater weight loss than the "information" group at the follow-up. Hypothesis 3: The "role playing" group will have a higher self-concept than the "information" group at the follow-up. Hypothesis 4: The "role playing" group will have a higher complexity of the self than the "information"group at the follow-up. Hypothesis 5: The "role playing" group will have a higher physical selfconcept than the "information" group at the follow-up.

The following measures were used in order to test the above hypotheses: (1) the Total Positive subscale of the Tennessee Self-Concept Scale (TSCS) was used to assess self-concept; (2) the Ziller Complexity of Self Scale was used to assess complexity of the self; and (3) the Physical Self subscale of the TSCS was used to assess physical self-concept. Follow-up measurements for the two treatment groups are summarized in Table 3.





45


Table 3

Means and Standard Deviations for "Role Playing" and "Information"
Treatment Groups on all Measures at the Follow-up



Role Playing Information (n = 12) (n = 16) M SD M SD


Self-Concept 45.50 11.39 47.25 8.18 Complexity 39.42 15.40 42.69 17.53 Physical Self 37.67 8.60 38.81 9.92 Reduction Index 18.58 32.34 38.19 29.37


In order to test Hypotheses 2, 3, 4 and 5 Multivariate Analysis of Variance (MANOVA) and Discriminant Analysis techniques as discussed by Kerlinger and Pedhazur (1973) were used. MANOVA procedures to determine significant differences between the "role playing" and "information" groups on the four follow-up measures as a group (self-concept, complexity, physical self-concept, and reduction index) revealed an F ratio of .95, df = 4,24, p > .05.

Univariate F values, standardized discriminant coefficients (SDC), and F values for SDC for the "role playing" and "information" treatment groups on all follow-up measures are presented in Table 4. Results from the discriminant analysis revealed that there were no significant differences between the treatment groups relative to self-concept, complexity, and physical self-concept as noted by the respective F values of 1.23, 0.89, and 0.88. The reduction index approached




46


significance in the opposite direction than predicted as indicated by an F value of 3.46 ( (<.10; df = 4,24).

When each variable was analyzed controlling for the effects of all other variables, none of the follow-up measures proved to be significant discriminators. In other words, none of the follow-up measures were significantly different for the two groups (Table 4, F values for SDC). Hypotheses 2, 3, 4, and 5 were rejected.


Table 4

Univariate F Values, Standardized Discriminant Coefficients (SDC),
and F Values for SDC for the "Role Playing" and "Information"
Treatment Groups on all Follow-up Measures



Standardized
Univariate Discriminant F Values F Values Coefficients (SDC) Tfor SDC


Self-Concept 1.23 0.29 0.05 Complexity 0.89 0.19 0.05 Physical Self 0.88 0.07 0.01 Reduction Index 3.46* 0.80 0.30


* ( .10

Hypothesis 6: At the follow-up, those subjects who have maintained the greatest weight loss will have a higher self-concept than unsuccessful weight reducers.

Subjects in both treatment groups were combined at the follow-up and divided into "successful" and "unsuccessful" weight reducers. A median split of the reduction indices was used to divide the two groups.






47

A subject was labeled "successful" if her reduction index was greater than 25.00 and "unsuccessful" if her reduction index was less than 25.00. Pretreatment and follow-up measures of these two groups are summarized in Table 5.


Table 5

Means and Standard Deviations of "Successful" and
"Unsuccessful" Weight Reducers on Pretreatment and Follow-up Measures



Successful Unsuccessful (n = 14) (n 14) M SD M SD


Pretreatment

Self-Concept 46.43 9.99 44.14 9.67 Complexity 47.47 15.54 37.43 12.41 Physical Self 37.00 12.41 33.71 6.62 Follow-up

Self-Concept 47.93 7.20 45.07 11.50 Complexity 46.93 16.93 35.64 14.33 Physical Self 40.07 8.00 36.57 10.30 Reduction Index 54.71 21.60 4.86 16.67



A two-group discriminant analysis of the "successful" and "unsuccessful" weight reducers revealed a canonical correlation of 0.42, corresponding to a chi-square of 4.80 (df = 2, p ( .10). When each of the variables was analyzed controlling for the effects of all other




48


variables pretreatment complexity and treatment group accounted for the most variance between the two groups, as indicated by the Standardized Discriminant Weights in Table 6. The variable self-concept failed to be a significant discriminator between the two groups (hypothesis 6 was rejected). When the type of treatment and each of the pretest and follow-up measures were analyzied independently pretreatment and posttreatment complexity scores approached a significant differency between the two groups ( (<.10; df = 1,26) as indicated by the univariate F values in Table 6.


Table 6

Univariate F Values and Standardized Discriminant
Weights for "Successful" and "Unsuccessful"
Weight Reducers on Treatment Group and
all Pretest and Follow-up Measures



Standardized
Discriminant Univariate Weights F Values


Pretreatment

Self-Concept 0.03 0.38 Physical Self 0.04 0.76 Complexity 0.74 3.64* Treatment Group 0.55 2.36 Follow-up

Self-Concept 0.50 0.62 Physical Self -- 1.01

Complexity 0.32 3.62*


Standardized Discriminant Weights greater than 0.50 are underlined
* p < .10






49

Pretreatment to Follow-up Change Scores


Paired t tests on mean pretreatment and follow-up measures for

each group indicated no significant differences on any of the following variables: self-concept, complexity, and physical self (Table 7).


Table 7

Comparison of Pretreatment and Follow-up Scores
for "Role Playing" and "Information" Treatment Groups on all Measures



Pretreatment Follow-up M SD M SD df t p


"Role Playing" Group

Self-Concept 43.46 11.46 45.50 11.39 11 -1.15 n.s.

Complexity 41.15 14.15 39.42 15.40 11 0.58 n.s.

Physical Self 34.69 7.22 37.67 8.60 11 -1.85 n.s. "Information Group"

Self-Concept 46.69 7.74 47.25 8.18 15 -0.27 n.s.

Complexity 44.81 15.78 42.69 17.53 15 1.00 n.s.

Physical Self 36.00 11.58 38.81 9.92 15 -1.15 n.s.





50


Correlations Between Variables


Correlation coefficients were calculated between each of the following variables: initial weight, percent overweight, G.P.A., years overweight, pretreatment self-concept, pretreatment complexity, pretreatment physical self, treatment group, follow-up self-concept, follow-up complexity, follow-up physical self, and follow-up reduction index (Table 8).








51





"o

r N LOOO m C Oom

L. C C) I I

4


C r- 0 -- N 1 -- O 0 C0 0 0 L 00 000 00 0 0 0
I I I I

LO C) 1 Dc c 0 r- 0
C. r- 0 r- c CO 4t r- 0 0 O 0 0000 00 00
LLn
L D J N L O O 0 0

SLv ddddddd a 0
I I













0 0
u 0 0 0 0 0 0 0)










SI I C C) C) C)
4



L C~ D C O O
r4-.. .. .
0 >0 O 0 0 0r0 to C I I
- ..C





0J
WE 0 C00 0 0 '
LO I




C >



O,- c,
+3 WC 1 1







00 0
* W 0 0 0 0
-L 0. O O






















; 0 L ..) O O O-r I 0 O
o .0 0 0 0

I 0










*- W





2 4) L. 0. Un \ L0 2 0




A C.D i- 0.. 0. 0.. C0 L.1.. Li.. IA.. LL.. 4<'














DISCUSSION


The findings of the present study indicate that a behavioral

self-control approach to overeating produces significant weight losses during the treatment period. In addition, this study supports Jeffrey's (1974) findings that a self-control approach produces significant long-term weight losses. In the present study mean reduction indices for both treatment groups were significant after a four month follow-up period.

At the follow-up, a comparison of the mean reduction indices between the two treatment groups approached significance in the opposite direction than predicted; that is, there was a trend toward greater weight loss for subjects in the "information" treatment group. A possible reason for this outcome may be related to the issue of client/ treatment compatibility. Several studies have indicated that outcome in psychotherapy is a function of the interaction between treatment parameters and client variables (Abramowitz et al., 1974; Friedman & Dies, 1974; Kilman, Albert, & Sotile, 1973). In other words, clients differ in the type of treatment to which they best respond.

Subjects in the "information" treatment group were exposed to a broader range of theories and treatment concerning obesity than subjects in the "role playing" treatment group. As a result, they may have had a greater opportunity to select an appropriate or compatible treatment approach.


52




53


Weight control programs can be greatly improved if they could

specify the type of client with whom they are more likely to be successful. In addition, treatment procedures may be enhanced if client characteristics which hinder effective weight reduction efforts could be delineated and then modified.

In the present study, pretreatment self-concept measurements

indicated that subjects presented a fairly positive view of themselves (general self-concept); they maintained a poor opinion only of their physical beings (physical self). Hence, they denied any dissatisfaction with themselves except with their physical bodies.

Past research in obesity has revealed that obese individuals often manifest other emotional and social problems (Held & Snow, 1972; Querehi, 1972; Werkman & Greenberg, 1967; Wunderlich, Johnson, & Ball, 1973). This seemed to hold true of the subjects in the present study. During the weekly meetings subjects disclosed conflicts regarding their identities, social relationships, families, and other personal concerns. However, research also indicates that obese persons may deny their problems on a self-report questionnaire (Gottesfeld, 1962; Suczek, 1955; Werkman & Greenbert, 1967). This may be the reason why subjects' scores fell within the normal range on the total positive scale of the Tennessee Self-Concept Scale. The only difficulty these subjects admitted to was the obvious one; hence, their scores were below average on the physical self subscale of the Tennessee Self-Concept Scale, but not on the total positive scale. A projective instrument may have proven to be a more valid indicator of subjects' general self-concepts.

At the follow-up no significant differences in self-concept were found between the two treatment groups. Although total self-concept scores may not have been accurate depictions of subjects' general level





54

of self-esteem, the failure of role playing techniques to affect selfconcepts is probably a valid result. This is supported by other results which indicated that role playing techniques had no significant affect on those parts of the self-concept concerning subjects' complexity or physical selves. The physical self subscale of the Tennessee SelfConcept Scale dealt with concerns subjects were more open about and the Ziller Complexity of Self Scale is more indirect in measuring complexity. Hence, these scores are more likely to be valid, yet they too remained unaffected by role playing techniques.

A possible explanation for the failure of role playing techniques to effectuate self-concept changes involves the relative commitment an individual has to each of the five components of Ziller's (1973) helical theory of personal change (attitudes, values, behaviors, roles, and self-concept). According to Roby (1960; cited in Ziller, 1973), "commitment refers to a reduction in further alternatives associated with a particular choice." (p. 151)

The amount of commitment an individual has to each of the components of Ziller's system increases as one goes up the hierarchy. For example, having reported certain attitudes is less restricting than having accepted a certain role, such as group leader. Violation of the expectations associated with certain attitudes is more acceptable to the self and others than violation of role expectancies. The greatest commitment is associated with the self-concept; these constructs are associated with the stability, regularity, and consistency of personal behavior.

Since attitudes require the least amount of commitment, they are subject to the most rapid change. A change in attitudes permits some experimentation within the personal system; little commitment is






55

involved and the personal system may return to its previous state with relatively little difficulty.

Increasing commitment is made as changes in
the other components proceed. The hierarchy suggests an orderly progression up the scale.
Attitude changes are tried first, then values, behaviors, roles, and finally the self-concept, in turn, under favorable circumstances. (p. 154,
Ziller, 1973)

In the present study, the "role playing" treatment group was a

social learning approach with an emphasis on practicing or rehearsing new social roles. However, subjects may not have been ready to commit themselves to role changes; more likely, they were at the stage of experimenting with new attitudes, values, and behaviors. There was evidence of this during the role playing sessions. During the discussions subjects often expressed their feelings on such issues as jealousy, intimacy, respect, and assertiveness. Hence, it seems that subjects were concerned with the lower components of Ziller's hierarchy.

Learning new social roles may not become a need for clients until after they have changed their behavior. At this point they would have lost a significant amount of weight, would be confronted with new situations, and may feel a social role deficit.

In the present study, the intervention of role playing may have been pre-mature and may have even served to arouse some anxiety about the prospect of losing weight. Future work in weight control might postpone the intervention of role playing until after clients are more committed to new attitudes, values, and behaviors. Then, it is more likely that role playing techniques would facilitate role changes and exert a strong press toward a change in the self-concept. A longer term treatment intervention would be required.





56

Although role playing techniques did not have an immediate impact on subjects' reported self-concepts, these techniques may have facilitated subjects' own awareness of themselves. This additional selfawareness may eventually prove helpful in the process of losing weight.

Previous research indicates that young overweight people are selfconscious about their size and seem to feel uncomfortable in interpersonal or social situations. Consequently, they may become socially withdrawn. Thus, while overweight people remain overweight because of poor eating habits, social problems may develop as well. However, these problems are often denied; hence, many overweight people may remain unaware of their feelings and needs. In the present study, role playing scenarios partly served to help subjects become aware of their feelings in interpersonal situations. In addition, the feedback sessions following role playing gave subjects an arena in which to share their feelings.

Remaining overweight may become a defense against having to deal with personal concerns. Losing weight removes this defense. Some people may not want to deal with this issue and may resist losing weight. In this respect, losing weight is similar to any other type of personal change; i.e., it is often resisted.

People may not be aware of their own defenses, and may not be aware of the consequences of losing them. In the present study, role playing techniques served to make subjects aware of personal issues. Role playing experiences brought out many anxiety provoking situations. As a result, these subjects became more aware of the challenges they might face while losing weight.

These speculations would only hold true for overweight people with






57

interpersonal anxiety. It is possible that only subjects who were not socially anxious were successful at losing weight. It was found that they tended to be more complex people. Since complexity is defined as the ability to match an aspect of oneself with an aspect of the environment, this may also imply more ease in social and interpersonal situations. It would be difficult to measure the relationship between social anxiety and success at weight reduction since social anxiety is difficult to measure; i.e., overweight people tend to deny personal problems.

This line of reasoning implies that weight reduction groups may

only work to the extent that feelings can be expressed and an awareness of personal problems can be developed. Groups that are restrictive in these areas will not help clients whose personal problems are entwined with their weight. If an awareness of personal concerns is fostered then success at weight control may depend upon the person's ability to deal with these concerns. The leader's responsibility would be to help a person mobilize or find appropriate support systems.

An analysis of the self-concepts of "successful" and "unsuccessful" weight reducers revealed no significant differences between the two groups. Both "successful" and "unsuccessful" weight reducers presented a fairly positive total self-concept while their physical selfpicture remained poor. These results support previous research indicating that the physical self-concepts of obese persons do not change commensurate with weight loss (Glucksman & Hirsch, 1969; Glucksman et al., 1968; Rohrbacher, 1973; Suczek, 1955). In addition, these results are compatible with Ziller's theory regarding change. According to this theory, people are more likely to change the self






58

system starting with the lower components and gradually progressing to the higher components. The ultimate level of change is a change in the self-concept, and this needs to be achieved if lower level changes are to be maintained.

Weight reduction reflects a change in behavior, the third

component in Ziller's hierarchy. According to Ziller's theory, if weight loss (behavior change) is to be maintained it must be followed by changes in roles and self-concept. Since these additional changes take time, it follows that self-concept changes would lag behind weight loss. Hence, in the present study, follow-up measures were probably taken some time after "successful" subjects had changed their behavior, but before they changed their self-concepts. As a result, no selfconcept differences were found between "successful" and "unsuccessful" weight reducers at the time of the follow-up. A longer term follow-up might reveal one of the following results: (1) "Successful" weight reducers would eventually change their self-concepts; therefore, behavior changes would be retained and weight loss would be maintained, or

(2) The self-concept of "successful" weight reducers would remain unchanged; therefore, behavior would return to its previous state and weight would be regained.

A further analysis of "successful" and "unsuccessful" weight

reducers revealed that those people who lost the most weight tended to be more complex people at pretreatment. This supports the theories of Horrocks and Jackson (1972) and Ziller (1973) who state that more complex people incorporate a greater array of facets or stimuli into their conceptualization of themself. Complex people are less likely to be disturbed by new experiences because there is a higher probability





59


of matching some aspect of themself with an aspect of the situation. Hence, they have a greater potential for flexible adaptation.

Ziller (1973) suggests that

strategies of personal change directed toward the
self-concept may be accomplished with greater
facility if they begin with concern for complexity
of the self-concept, in order to render the self
system more adaptive.(pp. 155-156)

Future work in weight control might investigate the usefulness of various techniques to help people expand their view of themselves and consequently have a greater potential for change.

The present study suggests the need for longer term treatment of overweight clients. It is proposed that unless positive changes in the self-concept accompany changes in behavior following behavior modification procedures, the treatment should be continued until such changes are observed.



























APPENDICES



























APPENDIX A
INSTRUMENTS AND FORMS








questionnaire #1


Name: Height: Address: Size of frame: small medium large Phone: (circle one)


Where will you be during the week of July 9, 1979? Address:

Phone:

Are you taking medication for weight? Are you enrolled in another weight program now? Do you have a metabolic or hormonal imbalance which could affect your weight?

What is your class? freshman, sophomore, junior, senior, grad. student (circle one)
What is your age? What is your major area of study? What is your grade point average? For how may years have you been overweight?



















62






63

Questionnaire #2


1. When you lose weight, your physical appearance will change. Who do you think will notice?




2. How do you think people will react to you when you lose weight?







3. Do you think you will act differently as a thin person? If yes, please specify how you will act differently.







4. Do you expect to encounter new situations when you lose weight? If yes, please specify.






64

Ziller Complexity of Self Scale

INSTRUCTIONS: Here is a list of words. You are to read the words quickly and check each one that you think describes YOU. You may check as many or as few words as you like--but be HONEST. Don't check words that tell what kind of a person you should be. Check words that tell what kind of a person you really are.

1. able 29. delicate 57. large 85. serious
2. active 30. delightful 58. lazy 86. sharp 3. afraid 31. different 59. little 87. silly 4. alone 32. difficult 60. lively 88. slow 5. angry 33. dirty 61. lonely 89. small 6. anxious 34. dull 62. loud 90. smart
7. ashamed 35. dumb 63. lucky 91. soft
8. attractive 36. eager 64. mild 92. special 9. bad 37. fair 65. miserable 93. strange 10. beautiful 38. faithful 66. modest 94. stupid 11. big 39. false 67. neat 95. strong 12. bitter 40. fine 68. old 96. sweet 13. bold 41. fierce 69.___patient 97. terrible 14. brave 42. foolish 70. peaceful 98. ugly 15. bright 43. friendly 71._perfect 99. unhappy 16. busy 44. funny 72.__ pleasant 100. unusual 17. calm 45. generous 73.__polite 101. useful 18. capable 46. gentle 74. poor 102. valuable 19. careful 47._glad 75._ popular 103. warm 20. careless 48. good 76. proud 104. weak 21. charming 49.___ great 77. quiet 105. wild 22. cheerful 50. happy 78. quick 106. wise 23. clean 51. humble 79. responsible 107. wonderful 24. clever 52. idle 80. rough 108. wrong 25. comfortable 53. important 81. rude 109. ____young 26. content 54. independent 82. sad 27. cruel 55. jealous 83. selfish 28. curious 56. kind 84. sensible




























APPENDIX B COURSE OUTLINE








Course Outline


Title: Hes 4905 Variable Topics in Health Education Weight Control: Self-Managed Behavior Change

Instructors: Fred Schreiber 377-7604
Sig Fagerberg

Thursday Monday

January 4 8 11 15
18 22
25 29
February 1 5
8 12
15 19
22 26
March 1 5

July 9, 1979: 4 month follow-up

Required Reading: Toward Permanent Weight Balance: Student's Manual

Course Description and Requirements:

This is a nine week weight control program. You will be graded on an S/U basis. Grades will be given strictly on the basis of your attendance (not on how much weight you lose). In order to pass all you need to do is attend class. In cases of emergency, or other unforseen circumstances, you will be allowed to miss one class, but you must notify the instructor as to your reason for missing class.

In addition, you are asked to participate in a 4 month follow-up.
This will be held during the week of July 9, 1979, and will involve filling out a few questionnaires and weighing in. If you are going to be out of town at that time, please note where I may contact you.

This weight control program is part of a research project designed to determine if certain techniques which are used to help individuals lose weight are more effective with one type of personality or another. Therefore, we are asking you to fill out some information forms, and your responses will be strictly confidential. Both groups will employ techniques that have been shown to be effective in helping people lose weight.

The groups will meet once a week for nine weeks with a trained
leader. The techniques utilized in treatment will include various selfcontrol procedures which will be discussed during the sessions and practiced at home. No techniques which will cause pain or discomfort will be used during the program.


66




























APPENDIX C
LETTER SENT TO SUBJECTS AT THE FOLLOW-UP








1700 S.W. 16 Ct. #A-1
Gainesville, Fla, 32608
July 9, 1979
Dear

It has been approximately four months since the end of our weight control class and now I am doing the follow-up. Please record your current weight at the bottom of this page, fill out both of the enclosed questionnaires, and return all the materials (this letter, 1 test booklet, 2 answer sheets) in the enclosed envelope. I realize that not all of you may have continued losing weight, and that some of you may have gained weight. However, please be honest in reporting your present weight and in answering the enclosed questionnaires so that my results will be valid.

Please read all directions carefully. When you fill out the
Tennessee Self-Concept Scale remember that the answer sheet is arranged so that you respond to every other item on it. Please do not omit any item. If you want to change an answer, do not erase it; rather, mark an X through the incorrect response and circle the response you want.

Please send these materials back to me AS SOON AS POSSIBLE as I am very anxious to finish my paper before the end of the summer. If you are interested in discussing the results of my research with me please call me in mid-August. My phone number is 377-7604.

Sincerely,


Fred M. Schreiber




















68













REFERENCES


Abramowitz, C. V., Abramowitz, S. I., Roback, H. B. & Jackson, C.
Differential effectiveness of directive and nondirective group
therapies as a function of client internal-external control.
Journal of Consulting and Clinical Psychology, 1974, 42, 849-853.

Abramson, E. E. A review of behavioral approaches to weight control.
Behavior Research and Therapy, 1973, 11, 547-556.

Buchanan, J. R. Five year psychoanalytic study of obesity. American
Journal of Psychoanalysis, 1973, 33, 30-38.

Collingwood, T. R. & Willett, L. The effects of physical training upon
self-concept and body attitude. Journal of Clinical Psychology,
1971, 27, 411-412.

Fitts, W. H. Tennessee Self-Concept Scale Manual. Nashville, Tenn.:
Counselor Recordings and Tests, 1965.

Friedman, M. L. & Dies, R. R. Reactions of internal and external testanxious students to counseling and behavior therapies. Journal of
Consulting and Clinical Psychology, 1974, 42, 921.

Glucksman, M. L. & Hirsch, J. The response of obese patients to weight
reduction: III. The perception of body size. Psychosomatic Medicine,
1969, 31, 1-7.

Glucksman, M. L., Hirsch, J., McCully, R. S., Barron, B. A. & Knittle,
J. L. The response of obese patients to weight reduction. II. A quantitative evaluation of behavior. Psychosomatic Medicine, 1968,
30, 359-373.

Gottesfeld,. H. Body and self-cathexis of super-obese patients.
Journal of Psychosomatic Research, 1962, 6, 177-183.

Harris, M. B. & Bruner, C. G. A comparison of a self-control and a
contract procedure for weight loss. Behavioral Research and Therapy,
1971, 9, 347-354.

Held, M. L. & Snow, D. L. MMPI, Internal-external control, and problem
check list scores of obese adolescent females. Journal of Clinical
Psychology, 1972, 28, 523-525.





69






70

Horrocks, J. E. & Jackson, D. W. Self and Role: A Theory of SelfProcess and Role Behavior. Boston: Houghton Mifflin Company, 1972.

Ince, L. P. The self-concept variable in behavior therapy. Psychotherapy: Theory, Research, and Practice, 1972, 9, 223-225.

Jeffrey, D. B. A comparison of the effects of external control and selfcontrol on the modification and maintenance of weight. Journal of
Abnormal Psychology, 1974, 83, 404-410.

Jeffrey, D. B., Christensen, E. R. & Katz, R. C. Behavior therapy
weight reduction programs: Some preliminary findings on the need
for follow-ups. Psychotherapy: Theory, Research, and Practice,
1975, 12, 311-313.

Kerlinger, F. N. & Pedhazur, E. J. Multiple Regression in Behavior
Research. New York: Holt & Winston, Inc., 1973.

Kilman, P. R., Albert, B. M. & Sotile, W. M. Relationship between locus
of control, structure of therapy, and outcome. Journal of Consulting
and Clinical Psychology, 1975, 43, 588.

Kurtz, R. M. Body attitude and self-esteem. Proceedings, 79th Annual
Convention, APA, 1971, 6, 467-468.

Lazarus, A. A. Behaviour rehearsal vs. non-directive therapy vs. advice
in effecting behaviour change. Behaviour Research and Therapy, 1966,
4, 209-212.

Lerner, R. M., Karabenick, S. A. & Stuart, J. L. Relations among
physical attractiveness, body attitudes, and self-concept in male
and female college students. Journal of Psychology, 1973, 85, 119-129.

Mahoney, E. R. Body-cathexis and self-esteem: The importance of subjective importance. Journal of Psychology, 1974, 88, 27-30.

Mahoney, E. R. & Finch, M. D. Body-cathexis and self-esteem: A
reanalysis of the differential contribution of specific body aspects.
Journal of Social Psychology, 1976, 99, 251-258.

McCall, R. J. MMPI factors that differentiate remediably from
irremediably obese women. Journal of Community Psychology, 1973, 1,
34-36.
McCall, R. J. Group therapy with obese women of varying MMPI profiles.
Journal of Clinical Psychology, 1974, 30, 466-470.

Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Instructor's Manual. Gainesville, Fla.: Florida Cooperative
Extension Service, 1979(a).

Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Student's Manual. Gainesville, Fla.: Florida Cooperative
Externsion Service, 1979(b).





71

Penick, S. B., Filion, R., Fox, S. & Stunkard, A. Behavior modification
in the treatment of obesity. Psychosomatic Medicine, 1971, 33, 49-55,

Quereshi, M. Y. Some psychological factors that distinguish between
the remediably and irremediably obese. Journal of Clinical Psychology,
1972, 28, 17-22.

Rogers, C. R. Client-Centered Therapy. Boston: Houghton Mifflin
Company, 1951.

Rohrbacher, R. Influence of a special camp program for obese boys on
weight loss, self-concept, and body image. Research Quarterly,
1973, 44, 150-157.

Rosen, G. M. & Ross, A. 0. Relationship of body image to self-concept.
Journal of Consulting and Clinical Psychology, 1968, 32, 100.

Secord, P. F. & Jourard, S. M. The appraisal of body-cathexis: Bodycathexis and the self. Journal of Consulting Psychology, 1953,
17, 343-347.

Stuart, R. B. Behavioral control of overeating. Behaviour Research and
Therapy, 1967, 5, 357-365.

Stuart, R. B. A three-dimensional program for the treatment of obesity.
Behaviour Research and Therapy, 1971, 9, 177-186.

Stuart, R. B. & Davis, B. Slim Chance in a Fat World: Behavioral Control of Obesity. Research Press: Champaign, Ill., 1972.

Stunkard, A. J. The management of obesity. New York State Journal of
Medicine, 1958, 58, 79-87.

Stunkard, A. & Burt, V. Obesity and body image: II. Age at onset of
disturbances in the body image. American Journal of Psychiatry,
1967, 123, 1443-1447.

Stunkard, A. & Mendelson, M. Obesity and the body image: I. Characteristics of disturbance in the body image of some obese persons.
American Journal of Psychiatry, 1967, 123, 1296-1300.

Suczek, R. F. Psychological aspects of weight reduction. In Eppright,
E. S., Swanson, P. & Iverson, C. A. (Committee in charge). Weight
Control: A collection of papers presented at the weight control
colloquium. Ames, Iowa: Iowa State College Press, 1955.

Weinberg, J. R. A further investigation of body-cathexis and the self.
Journal of Consulting Psychology, 1960, 24, 277.

Weller, L., Arad, T. & Levit, R. Self-concept, delayed gratification,
and field dependence of successful and unsuccessful dieters. Israel
Annals of Psychiatry and Related Disciplines, 1977, 15, 41-46.






72


Werkman, S. L. & Greenberg, E. S. Personality and interest patterns in
obese adolescent girls. Psychosomatic Medicine, 1967, 29, 72-80.

Wollersheim, J. P. Effectiveness of group therapy based upon learning
principles in the treatment of overweight women. Journal of
Abnormal Psychology, 1970, 76, 462-474.

Wunderlich, R. A., Johnson, W. G. & Ball, M. F. Some personality
correlates of obese persons. Psychological Reports, 1973, 32,
1267-1277.

Ziller, R. C. The Social Self. New York: Pergamon Press Inc., 1973.

Zion, L. Body concept as it relates to self-concept. Research Quarterly,
1965, 36, 490-495.














BIOGRAPHICAL SKETCH


Fred M. Schreiber was born in New York City on May 2, 1953.

He graduated from Baldwin Senior High School in Baldwin, New York, in June, 1971. He attended the University of Rochester in Rochester, New York, from which he received a B. A. degree in psychology in May, 1975. In September, 1975, he enrolled at the University of Florida for graduate study in psychology, and received an M. A. degree in psychology in June, 1977. Presently he is a Counseling Intern at the Counseling and Testing Service at the University of Houston, and expects to receive a Ph.D. in psychology from the University of Florida upon completion of his internship in August, 1980.
























73








I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.



)arfy A. Gra er, Chairman Professor of Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.



Ted Land man
Professor of Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.



Paul G. Schauble Professor of Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.



Afesa Bell-Nathaniel Assistant Professor of Psychology








I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.



Seigf d Fagerberg
Associate Professor of
Health Education and Safety


This dissertation was submitted to the Graduate Faculty of the Department of Psychology in the College of Liberal Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.

August 1980


Dean, Graduate School




Full Text
23
suggests that weight control clients should be more likely to change
their self-concepts, and therefore, be more likely to maintain weight
losses if treatment foccusses on changing their roles in addition to
changing their behaviors.
Role playing techniques or behavioral rehearsal may facilitate
role changes. This, in turn, may have a strong effect on a person's
self-concept. Horrocks and Jackson (1972) explain this process. First,
they distinguish between role taking and role playing behavior. Role
taking is a concrete manifestation or implementation of a hypothe
sized identity and presents an observable product of the self-process.
Role playing, however, represents performance by an individual of pre
scribed, demand behavior determined by a situational context. The role
player behaves according to situational expectations, but the behavior
does not represent anything the person believes himself to be in that
context. Role playing behavior, therefore, is the manifestation by
an individual of anti-identities, or those roles not conceptualized by
an individual of himself at that moment.
When an individual role plays anti-identities, locus of control of
his behavior is external to him, imposed upon him by situational or be
havioral demands. When the person receives feedback on his performance,
he assimilates into his cognitive structures the effectiveness and appro
priateness of these actions. In this manner, cognitive dissonance may
alter previous cognitions and change a conception of the self to include
aspects of the anti-identity. As a result, new meanings of the self may
be conceptualized, and role taking behavior replaces role playing.
Horrocks and Jackson suggest that the greater the array of identi
ties an individual incorporates into his conceptualization of himself


Table 8
Correlations Between Variables
Initial
Wt
Percent
Overwt G.P.A.
Years
Overwt
Pretreat Pretrt
Self-con Comp
Pretreat Trtmt
Phys Self Group
F. up
S.C.
F. up
Comp
F.up F.up
Phys Self Red Ind
Init'l Wt 1
.00
0.90 -0.06
0.00
-0.05
0.22
-0.04
-0.16
-0.14
0.25
-0.15
-0.27
1 Overwt
1.00 0.00
0.00
0.05
0.08
-0.04
-0.16
0.05
0.19
-0.09
-0.22
G.P.A.
1.00
0.11
0.19
0.06
-0.04
0.03
0.12
0.07
-0.13
0.15
Yrs Overwt
1.00
-0.09
-0.24
-0.22
0.08
-0.15
-0.15
-0.23
0.07
Pre S.C.
1.00
0.43
0.77*
0.17
0.70
0.39
0.50
-0.03
Pre Comp
1.00
0.57
0.50
0.00
0.83
0.12
0.20
Pre Phys Self
1.00
0.08
0.36
0.43
0.63
0.00
Grp (l=role
play, 2=info)
1.00
0.09
0.10
0.06
0.31
F.up S.C.
1.00
0.01
0.60*
0.15
F.up Comp
1.00
0.07
0.19
F.up Phys Self
1.00
0.09
F.up Red Ind
1.00
* Spuriously high because of overlapping test items


45
Table 3
Means and Standard Deviations for "Role Playing" and "Information"
Treatment Groups on all Measures at the Follow-up
Role Playing Information
(n = 12) (n = 16)
M
SD
M
SD
Self-Concept
45.50
11.39
47.25
8.18
Complexity
39.42
15.40
42.69
17.53
Physical Self
37.67
8.60
38.81
9.92
Reduction Index
18.58
32.34
38.19
29.37
In order to test Hypotheses 2, 3, 4 and 5 Multivariate Analysis of
Variance (MANOVA) and Discriminant Analysis techniques as discussed by
Kerlinger and Pedhazur (1973) were used. MANOVA procedures to determine
significant differences between the "role playing" and "information"
groups on the four follow-up measures as a group (self-concept,
complexity, physical self-concept, and reduction index) revealed an
£ ratio of .95, df = 4,24, jd > .05.
Univariate £ values, standardized discriminant coefficients (SDC),
and £ values for SDC for the "role playing" and "information" treatment
groups on all follow-up measures are presented in Table 4. Results from
the discriminant analysis revealed that there were no significant
differences between the treatment groups relative to self-concept,
complexity, and physical self-concept as noted by the respective £
values of 1.23, 0.89, and 0.88. The reduction index approached


ACKNOWLEDGEMENTS
I would like to thank Dr. Harry Grater for the time he has con
tributed and the interest he has shown in seeing this study through to
its completion. His guidance and support throughout this entire study
has been greatly appreciated.
I would like to thank the following people for their interest in and
contributions to this study:
Dr. Afesa Bell-Nathaniel: for her suggestion and encouragement to
study the self-concepts of overweight people.
Dr. Sig Fagerberg: for his support in teaching weight control
classes in the Department of Health Education and Safety.
Dr. Linda Moody: for the opportunity to develop the weight control
manuals used in this study.
Dr. Paul Schauble: for his help in research methods and design, and
his encouragement to examine various facets of the overweight person.
Dr. Ted Landsman: for his personal support not only during this
study, but throughout my entire graduate program.
Special thanks to Dr. A1 Kahn and Dr. Mike Omizo for contributions
beyond the call of duty. They made the statistical portion of this study
more manageable and understandable.
Thanks to Dr. Bob Ziller for his stimulating ideas on the self-
concept and theory of personal change.


57
interpersonal anxiety. It is possible that only subjects who were not
socially anxious were successful at losing weight. It was found that
they tended to be more complex people. Since complexity is defined as
the ability to match an aspect of oneself with an aspect of the environ
ment, this may also imply more ease in social and interpersonal
situations. It would be difficult to measure the relationship between
social anxiety and success at weight reduction since social anxiety
is difficult to measure; i.e., overweight people tend to deny personal
problems.
This line of reasoning implies that weight reduction groups may
only work to the extent that feelings can be expressed and an awareness
of personal problems can be developed. Groups that are restrictive in
these areas will not help clients whose personal problems are entwined
with their weight. If an awareness of personal concerns is fostered
then success at weight control may depend upon the person's ability to
deal with these concerns. The leader's responsibility would be to
help a person mobilize or find appropriate support systems.
An analysis of the self-concepts of "successful" and "unsuccessful
weight reducers revealed no significant differences between the two
groups. Both "successful" and "unsuccessful" weight reducers presented
a fairly positive total self-concept while their physical self
picture remained poor. These results support previous research
indicating that the physical self-concepts of obese persons do not
change commensurate with weight loss (Glucksman & Hirsch, 1969;
Glucksman et al., 1968; Rohrbacher, 1973; Suczek, 1955). In addition,
these results are compatible with Ziller's theory regarding change.
According to this theory, people are more likely to change the self


39
Your mother feels very hurt because you don't want to eat her food.
3. You're at a party with some of your girlfriends. A guy you've been
attracted to for awhile notices you and strikes up a conversation with
you.
4. You've lost some weight but haven't yet reached your goal. You
are sitting on a bus next to a guy whom you've just met. After talking
with him for awhile he remarks, "You have a beautuiful face! Do you
realize that if you lost some weight you'd really be a knock-out!"
5. You're with a girlfriend on campus and meet a guy that she knows.
He ignores you and speaks only to your girlfriend. Even after your girl
friend has introduced you and has made several references to you during
the conversation he continues to address his attention solely to your
girlfriend.
6. You are at a bar and strike up a conversation with a guy. He
seems friendly but as time goes by he becomes overly aggressive.
7. A guy calls you up and wants to take you out to eat. You want to go
out with him but have already eaten.
8. You are with a friend and you are both anxious about an exam. Your
friend says, "Let's go get something to eat."
9. You've lost some weight and you're walking home across campus. You
meet a guy that you know and he tells you now great you look and insists
on taking you to the Arrendondo Room for lunch.
10. ,You've lost weight and your roommate, who is a little overweight,
begins making snide remarks about your body. You suspect she may be
feeling a bit jealous of you.


22
self-concept; therefore, focussing on taking on new roles may be a valu
able dimension in the treatment of obesity. In review, Ziller's theory
suggests a hierarchy of potentially changeable characteristics which in
clude attitudes, values, behaviors, roles, and self-concepts. It is as
sumed that these characteristics are ordered according to their ease of
change, with attitudes being the least difficult to change while self-
concepts are the most resistant to change.
If one of the components in the system is changed a state of
disequilibrium within the system ensues. It is assumed that in this
imbalanced state there is a tendency for the components lower in the
hierarchy to change in a way which will render the components congruent.
In addition, a change in one component will exert some press toward
change in components higher in the system. A change in behavior, for
example, will exert some press toward changes in attitudes, values, roles,
and self-concepts, but with diminishing force with regard to the higher
level components.
Behavioral approaches to weight control are concerned with chang
ing behaviors, and this will cause the system to be in a state of
disequilibrium. However, a change in behavior will not exert a very
strong press on the individual's self-concept. Hence, in order for the
system to regain equilibrium the person is likely to revert back to
old ways of behaving (i.e., old eating habits) which are congruent with
the self-concept, rather than change the self-concept. Consequently,
weight maintenance is unlikely.
A change in a person's role will exert a stronger press on a per
son's self-concept than a change in behavior since roles are closer than
behaviors to self-concepts on the personal change hierarchy. This


19
after the camp program had ended and subjects had returned home. Sig
nificant weight losses were recorded at posttreatment and at the follow
up, although subjects did regain some weight during the follow-up period.
In addition, body image showed a significant positive change as a re
sult of the camp program, but self-concept remained unchanged. Rohr-
bacher suggested that changing body image may be an important factor
related to the program's success, but that longer follow-ups are needed
to accurately assess the program's effectiveness.
Weller, Arad, and Levit (1977) compared the self-concepts of
twenty-five women who had successfully reduced their weight with those
of twenty-five women who were unsuccessful at weight reduction. On
the Tennessee Self-Concept Scale, successful dieters revealed a better
self-image (general self-concept score) and a better physical self-image
(physical self score) than did the unsuccessful dieters. These results
suggested that those individuals who were most successful at weight re
duction had also modified their self-concepts in the direction of feeling
better about themselves and their bodies.
McCall (1973) studied MMPI profiles of two groups of women belong
ing to TOPS (Take Off Pounds Sensibly). One group had considerable
difficulty losing weight and/or maintaining whatever weight loss they
had achieved. Such refractorily or irremediably obese were designiated
R-TOPS. The other group had successfully lost weight and had maintained
their ideal body weight for at least six months. These women were re
ferred to as KOPS (Keep Off Pounds Sensibly).
On nine of the ten MMPI clinical scales, R-TOPS women were sig
nificantly more deviant than KOPS women. On six of the nine scales, the
differences were significant at the .01 level; on three scales the level


50
Correlations Between Variables
Correlation coefficients were calculated between each of the
following variables: initial weight, percent overweight, G.P.A.,
years overweight, pretreatment self-concept, pretreatment complexity,
pretreatment physical self, treatment group, follow-up self-concept,
follow-up complexity, follow-up physical self, and follow-up reduction
index (Table 8).


Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN
OVERWEIGHT COLLEGE WOMEN
By
Fred M. Schreiber
August 1980
Chairman: Harry A. Grater
Department: Psychology
Research in psychology has indicated that the self-concept may
be a moderator variable in the maintenance of behavioral changes.
In addition, previous research has suggested that role playing
techniques may facilitate a change in the self-concept. This study
investigated the effectiveness of role playing techniques in a
behavior modification weight reduction group in order to help clients
enhance their self-concepts and to determine the effect this has on
clients' abilities to lose weight and maintain their losses over a
follow-up period.
Two eight-week weight reduction groups were conducted. The
goal of each group was gradual weight loss through improving eating
habits. In addition, one group included an emphasis on using role
playing techniques to facilitate clients' abilities to change their
self-concepts. The other group was presented with theories and informa
tion on weight control.
General self-concept, physical self-concept, and complexity of
self were measured at pretreatment and after a four month follow-up
TV


THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN
OVERWEIGHT COLLEGE WOMEN
By
Fred M. Schreiber
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1930

ACKNOWLEDGEMENTS
I would like to thank Dr. Harry Grater for the time he has con
tributed and the interest he has shown in seeing this study through to
its completion. His guidance and support throughout this entire study
has been greatly appreciated.
I would like to thank the following people for their interest in and
contributions to this study:
Dr. Afesa Bell-Nathaniel: for her suggestion and encouragement to
study the self-concepts of overweight people.
Dr. Sig Fagerberg: for his support in teaching weight control
classes in the Department of Health Education and Safety.
Dr. Linda Moody: for the opportunity to develop the weight control
manuals used in this study.
Dr. Paul Schauble: for his help in research methods and design, and
his encouragement to examine various facets of the overweight person.
Dr. Ted Landsman: for his personal support not only during this
study, but throughout my entire graduate program.
Special thanks to Dr. A1 Kahn and Dr. Mike Omizo for contributions
beyond the call of duty. They made the statistical portion of this study
more manageable and understandable.
Thanks to Dr. Bob Ziller for his stimulating ideas on the self-
concept and theory of personal change.

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT iv
INTRODUCTION I
Problem I
Behavioral Approaches to Weight Control 1
The Self-Concept 5
METHOD 27
Subjects 27
Measures 29
Procedures 31
RESULTS 43
Hypotheses 43
Pretreatment to Follow-up Change Scores 49
Correlations Between Variables 50
DISCUSSION 52
APPENDIX A INSTRUMENTS AND FORMS 62
APPENDIX B COURSE OUTLINE 66
APPENDIX C LETTER SENT TO SUBJECTS AT THE FOLLOW-UP 68
REFERENCES 69
BIOGRAPHICAL SKETCH 73
i i i

Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN
OVERWEIGHT COLLEGE WOMEN
By
Fred M. Schreiber
August 1980
Chairman: Harry A. Grater
Department: Psychology
Research in psychology has indicated that the self-concept may
be a moderator variable in the maintenance of behavioral changes.
In addition, previous research has suggested that role playing
techniques may facilitate a change in the self-concept. This study
investigated the effectiveness of role playing techniques in a
behavior modification weight reduction group in order to help clients
enhance their self-concepts and to determine the effect this has on
clients' abilities to lose weight and maintain their losses over a
follow-up period.
Two eight-week weight reduction groups were conducted. The
goal of each group was gradual weight loss through improving eating
habits. In addition, one group included an emphasis on using role
playing techniques to facilitate clients' abilities to change their
self-concepts. The other group was presented with theories and informa
tion on weight control.
General self-concept, physical self-concept, and complexity of
self were measured at pretreatment and after a four month follow-up
TV

period. The scales used to measure these variables were the Tennessee
Self-Concept Scale and the Ziller Complexity of Self Scale.
Both groups had significant weight losses at posttreatment and
these losses were maintained after the four month follow-up period.
There were no significant differences in weight loss between the two
groups at posttreatment. However, at the follow-up, subjects in the
information" treatment group tended to maintain a greater weight loss
than subjects in the "role playing" treatment group.
At the follow-up, there were no significant differences in self-
concept, complexity, or physical self between the two groups. In
addition, there were no significant differences in self-concept,
complexity, or physical self between "successful" and "unsuccessful"
weight reducers at the follow-up. However, "successful" weight
reducers tended to have higher scores on the Ziller Complexity of Self
Scale at pretreatment than "unsuccessful" weight reducers. These
results are discussed with regard to ZiTier's Helical Theory of
Personal Change and future work in weight control.
v

INTRODUCTION
Problem
The problem of obesity has reached such a high magnitude that the
U.S. Public Health Service has classified it as "one of the most preva
lent health problems in the United States today" (p. 547, Abramson, 1973).
Stuart and Davis (1972) estimate that there are currently between 40
and 80 million obese individuals in this country alone. Traditional
treatments for the problem have included medication, psychotherapy, and
therapeutic starvation. However, these treatments have generally been
unsuccessful, and Stunkard (1958) has concluded that
Most obese persons will not remain in treatment.
Of those that remain in treatment, most will not
lose weight, and of those who do lose weight,
most will regain it. (p. 79)
Behavioral Approaches to Weight Control
Recently, some successes have been reported using a behavioral ap
proach to weight reduction. This approach views eating habits as
learned behaviors and sees the overweight person as someone who has
learned inappropriate patterns of eating. Behavioral approaches to
weight reduction focus on helping the overweight person become aware
of his eating patterns and helping him change inappropriate or problem
atic behaviors. It is presumed that by improving one's eating habits it
will be easier to reduce one's caloric intake, and consequently, lose
weight. In addition, by changing one's habits permanently, weight loss
should also be permanent.
1

2
Techniques used in behavioral approaches to weight reduction are
derived from both operant and classical conditioning. Operant condi
tioning focusses on the antecedents and consequences which control a
behavior. When applied to weight control, the overweight person is
taught to bring his eating under appropriate situational cues and to
reinforce improvements in his eating patterns. Classical conditioning
involves the pairing of two stimuli so that eventually they bring about
the same response in an organism. The overweight person has paired eat
ing with a number of different external cues so that eventually the
external cues alone elicit a response of feeling hungry." The over
weight person needs to disassociate eating from other activities in
order to extinguish eating as a conditioned response to other activities.
Stuart (1967) utilized operant and respondent conditioning tech
niques to help clients gain control over their eating behavior. These
techniques included controlling the antecedent and consequent conditions
of eating as well as record keeping and exercise. In addition, rein
forcement was provided in the following three ways: (1) through the
clients' experience of success in self-control; (2) through the reduction
of the aversive consequences caused by a lack of self-control; and (3)
through considerable reassurance by the therapist. Stuart treated
eight women on an individual basis. Weight loss over a twelve month
period for the eight women ranged from twenty-six to forty-seven pounds.
Wollersheim (1970) compared the effectiveness of three group
treatments (behavior self-control, positive expectation--social pressure,
and non-specific therapy) with a no-treatment control group. The
behavioral self-control group utilized operant conditioning techniques

3
to learn appropriate eating behaviors in order to lose weight. In addi
tion, they were trained in relaxation in order to counter tension in
situations which would ordinarily result in eating. The positive expecta-
tion--social pressure group utilized group pressure to help subjects lose
weight. The non-specific therapy group helped subjects understand their
underlying motives for being overweight. Wollersheim reported that all
three treatment groups lost significantly more weight than the control
group after a twelve week treatment program. Furthermore, after an
eight week follow-up subjects in the behavioral self-control group showed
a significantly greater weight loss (pretreatment to follow-up) than
either of the other two treatment groups.
Stuart (1971) treated six overweight women on an individual basis
in a fifteen week program which stressed environmental control of
overeating, nutritional planning, and regulated increase in energy
expenditure. Half the women started treatment immediately while the
other group started treatment fifteen weeks later. Approximately six
months following termination of treatment for the first group and three
months following termination of treatment for the second group, follow
up data were collected. Results indicated that subjects in the first
group lost an average of thirty-five pounds while those in the second
group lost an average of twenty-one pounds.
Research in the treatment of obesity has indicated that behavioral
approaches are generally effective in helping people lose weight.
However, outcomes vary across programs; that is, not all clients are
equally successful at losing weight during treatment or maintaining
weight losses over a follow-up period. This point is brought out in
the following studies.

4
Jeffrey, Christensen, and Katz (1975) reported the results of four
subjects whom they had treated for obesity using a behavioral approach.
After six months of treatment weight losses ranged from twenty to thirty-
two pounds. However, after a six month follow-up period weight losses
(pretreatment to follow-up) ranged from three to thirty-two pounds; two
subjects had maintained their weight loss while the other two subjects
gained back part or most of their weight.
Harris and Bruner (1971) reported two studies examining the
effectiveness of a self-control behavioral program and a contract
procedure for weight reduction. In the first study self-control pro
cedures included positive reinforcement, stimulus control, and breaking
or lengthening the chain of eating while the contract program consisted
of subjects receiving a monetary reward for each pound lost. Both
groups lost a significant portion of body weight after two months of
treatment. Furthermore, the contract group had lost a larger proportion
of initial weight than the self-control group. However, after a ten
month follow-up period there were no significant weight losses for
either group. In a second study, Harris and Bruner (1971) compared a
self-control weight group with a no-treatment control group and found no
significant weight losses for either group after sixteen weeks.
Penick, Filion, Fox, and Stunkard (1971) compared a behavior
modification weight control gorup with a control group which received
supportive psychotherapy as well as instruction about dieting and nutri
tion. After three months of treatment the median weight loss for the
behavior modification group was greater than that of the control group,
but this difference did not reach significance. Penick et al. attribute
this result to the significantly larger variability of outcome in the

5
behavior modification group. The five best reducers belonged to this
group as did the single least successful one, the only subject who
gained weight during treatment.
The Self-Concept
In order to explain the large variability of outcome in behavior
therapyInce (1972) suggests that the self-concept may be a moderator var
iable affecting outcome, but that it is largely ignored by behavioral
practitioners. The self-concept, according to Rogers (1951) consists of
an organized conceptual pattern of the "I" or "me" together with the
values attached to those concepts.
Ince (1972) cited two clients whom he treated using behavioral
techniques. In both cases, treatment of the overt symptoms by behavior
modification proved insufficient to effect change. Underlying each
client's difficulties was a poor self-concept which needed to be enhanced
in order for the treatment to be effective. Hence, the therapist began
verbally reinforcing each client for positive self-references with the
goal of increasing such verbalizations which would then become internal
ized and thus modify the self-concept. This approach met with success,
and suggested that the modification of an individual's self-concept can
be an important variable and one upon which success of therapy might
depend.
filler's (1973) helical theory of personal change supports
Ince's (1972) proposition that a change in the self-concept is a neces
sary and desired outcome of personal change procedures. This theory
suggests a hierarchy of potentially changeable characteristics which
include attitudes, values, behaviors, roles, and self-concepts. It

6
is assumed that these characteristics are ordered according to their
ease of change, with attitudes being the least difficult to change while
self-concepts are the most resistant to change.
Ziller's helical theory proposes that the self-concept is the
anchoring characteristic of the system, or the ultimate level of the
personal change hierarchy.
Those who would effect change in attitudes,
values, behavior, or roles are necessarily
concerned with a change in the self-concept,
for it is assumed here that unless a change
in the self-concept is achieved which is con
gruent with changes at lower levels in the
hierarchy, the lower level change is likely
to be reversed and the organism returned to
the initial state of equilibrium, (p. 174)
Self-Concept and Body-Cathexis
Secord and Jourard (1953) investigated the relationship between an
individual's self-concept and body-cathexis. By body-cathexis is meant
the degree of feeling or satisfaction with the various parts or processes
of the body. In order to appraise body-cathexis subjects were presented
with a list of forty-six items, each describing a different part or
function of the body, and were asked to indicate on a five point scale
their degree of satisfaction with each item. In order to measure self-
cathexis, individuals were asked to rate in a similar fashion fifty-five
items believed to represent a sampling of the various conceptual aspects
of the self. Subjects were seventy college males and fifty-six college
females.
Intercorrelations between body-cathexis and self-cathexis scores
were .58 for men and .66 for women. Hence, individuals had a moderate
tendency to cathect their body to the same degree and in the same

7
direction that they cathected their self. In addition, it was found that
females cathected their bodies, irrespective of direction, more highly
than did males, in that they did not assign as many threes (have no
particular feeling one way or the other) to body items. Secord and
Jourard suggested that women would be more likely than men to develop
anxiety concerning their bodies because of the social importance of the
female body.
Rosen and Ross (1968) noted that in the correlations obtained by
Secord and Jourard (1953) between body-cathexis and self-concept, they
did not take into account that certain parts or processes may be more
important to an individual than other parts or processes in evaluating
his body- and self-concepts. Hence, Rosen and Ross (1968) investigated
the relationship between body image and self-concept, taking into account
the relative subjective importance of the aspects being rated.
Eighty-two undergraduates were presented with a list of twenty-
four body parts and seventeen adjectives from the Adjective Check List.
They were asked to indicate on a five-point scale for each body part or
adjective how satisfied they were with that aspect and how important that
aspect was to them. Correlations of subjects' mean satisfaction scores
between body image and self-concept were; r = .52 for all items; _r = .62
for all items above mean importance; and = .28 for items below mean
importance. Hence, these findings support those of Secord and Jourard
(1953), and in addition, they indicate that the relationship between
body-cathexis and self-concept can be refined if the subjective impor
tance of each component is considered.
Lerner, Karabenick, and Stuart (1973) asked 118 male and 190 female
college students to rate twenty-four body characteristics in terms of

8
how satisfied they were with each characteristic of their own body and
how important each part was in determining their own physical attractive
ness. In addition, subjects responded to a short self-concept scale
consisting of sixteen bipolar dimensions derived from the Adjective Check
List. Results indicated that the degree of positive self-concept in
creased with the degree of positive attitude toward one's body character
istics (males: r. = -33; females: r. = *43). However, weighting satis
faction ratings by corresponding importance ratings did not significantly
increase the satisfaction/self-concept relation (males: r. = .33; females
21 = .44). Hence, while this study supports the general relationship
between self-concept and body-cathexis put forth by Secord and Jourard
(1953), it failed to support the results of Rosen and Ross (1968)
that importance ratings strengthened this relationship.
Mahoney (1974) replicated the Lerner et al. findings also using col
lege students as subjects and self-report attitude measures. He found
the correlation between mean unweighted (by importance) body-cathexis
and self-esteem to be .45 for males and .37 for females. For weighted
body-cathexis the respective correlations for males and females were
.37 and .41.
Further support for the relationship between body attitude and
self-esteem has been provided by Weinberg (1960), Zion (1965), and
Mahoney and Finch (1976). In each study, moderate correlations were
found using college students and self-report instruments. The use of
an identical scaling procedure to measure self-esteem and body attitude
raises the question that the high correlation found between body-
cathexis and self-esteem may represent method variance and not an
intertrait relationship. The following study addresses this issue.

9
Kurtz (1971) tested the hypothesis that body attitudes are related
to feelings of self-esteem using two instruments which share very
little method similarity. He used a self-report body attitude scale
(methodologically similar to that used by Secord and Jourard, 1953) and
the Ziller (1969) Self-Esteem Scale, which is a more indirect or pro
jective technique for measuring a person's feelings toward himself. For
the group of college students sampled, the hypothesis that a positive
evaluative body attitude and a positive sense of self-esteem are related
was confirmed. Of importance was the strength of this relationship con
sidering that the scales used shared little methodological similarities.
The relationship between body-cathexis and self-concept sheds some
light on the self-concept of overweight individuals. It is assumed that
people who volunteer to participate in weight reduction programs are
dissatisfied with their bodies. Since people tend to cathect their
bodies to the same degree that they cathect themselves, overweight in
dividuals in weight control groups are probably somewhat dissatisfied
with themselves. Hence, the preceding research on the relationship
between body-cathexis and self-concept implies that obese individuals in
weight reduction groups may be partly characterized by feelings of low
self-esteem.
Obese Self-Concept and Body-Cathexis
Werkman and Greenberg (1967) compared eighty-eight obese adolescent
girls at a medically oriented camp for overweight girls with forty-two
normal-weight girls at an ordinary summer camp on a number of personality
and interest measures (MMPI, Strong Vocational Interest Blank, Semantic
Differential, and Sentence Completion Blank). One finding which cut

10
across all the test results was the presence of a response set. The
obese were defensive about revealing any psychological problems and,
in short, presented themselves as "hypernormal." In addition, there
was a great difference between the obese and the control group on the
social introversion scale of the MMPI, suggesting that the obese girls
were more uncomfortable and anxious in social situations.
Vocational interest patterns of the two groups also differed.
While the control group identified with professions in which imagination,
ambitiousness, creativity, and intellectual strivings were paramount, the
obese girls' interests were more consistent with persons in nurturant
professions. The obese showed a kind of "maturity"; that is, their
interests were similar to those people whose occupations are stable and
"real is tic."
These results suggest that the obese tend to restrict themselves
both socially and vocationally. They appear to live within a con
ventional life pattern in which one does not attempt situations which
might provoke anxiety. In addition, through their efforts to appear nor
mal they may be sacrificing spontaneity and flexibility in many spheres
of psychological functioning. This pattern may hamper them by preventing
the full development of character through conflict and acceptance of
challenges.
Held and Snow (1972) studied twenty-three obese adolescent girls
and twenty-three non-obese adolescent girls who were randomly selected
from patients being seen for medical reasons at an out-patient clinic.
Subjects were individually administered the MMPI, Mooney Problem Check
List, and the Rotter Internal-External Locus of Control Scale. The obese
group scored significantly higher on five of the ten clinical scales of

n
the MMPI than the non-obese group. These scales were D (depression),
Pd (psychopathic deviate), Pa (paranoia), Pt (psychasthenia), and Sc
(schizophrenia). The generally elevated profile of the obese group,
particularly the Pd and Sc scales, suggested that many had serious psy
chological disturbances characterized by feelings of depression,
alienation, and low self-worth. In addition, they tended to be non-con-
forming, had problems in impulse control, and were distrustful of others.
Held and Snow failed to find the obese group defensive about reveal
ing psychological problems in that they admitted to significantly more
difficulties on the Mooney Problem Check List than the non-obese group.
These findings suggested that obese adolescent girls may need help recon
ciling some of the feelings they have toward themselves and other people
in addition to programs designed to help them lose weight.
Wunderlich, Johnson, and Ball (1973) administered the Adjective
Check List (ACL) and the Edwards Personal Preference Schedule (EPPS) to
sixteen subjects ranging from 64.6% to 214.7% overweight. The obese
group endorsed a significantly fewer number of favorable adjectives on
the ACL than did the normative sample. In addition, the obese were
characterized as having a low need for achievement, poor personal adjust
ment, and few preferences for close attachments to other people. They
did not describe themselves as being dependable, cautious, stable, or
inhibited. Instead, they described themselves as hasty, pleasure-seeking,
headstrong, and rebellious. Order for them was low, as they described
themselves as being careless, changeable, and disorderly, as well as
having poor self-control.
The obese scored low on endurance, suggesting that they were less
willing to expend energy toward a goal or persist at a task. Hence, the

12
obese tended to prefer sedentary activities or an inactive, passive
lifestyle. The obese also scored low on dominance, indicating that
they are seldom ambitious, determined, or assertive. They also character
ized themselves as being manipulative, hostile, and aggressive. It
seemed, therefore, that they would be likely to react to unexpected
demands with sudden withdrawal and underlying hostility. Finally, the
obese were very self-conscious about their size and expressed strong
sexual conflicts and frustrations. This seemed to be produced by the
desire to enjoy heterosexual contacts but the inability to engage in
these behaviors due to the attitudes of themselves and others concerning
their bodies.
Quereshi (1972) studied 180 female members of TOPS (Take Off Pounds
Sensibly) who had considerable difficulty with weight reduction, and on
the average weighed over 200 pounds. He compared them with ninety-eight
females who had been through the TOPS program and were successful over a
six month period at staying within 5% of their ideal weight. These peo
ple were know as KOPS (Keep Off Pounds Sensibly). All subjects were ad
ministered the Michil Adjective Rating Scale (MARS) which consisted of
forty-eight adjectives such as "nervous," "talkative," and "ambitious,"
and was accompanied by a five-point scale ranging from "very atypical"
to "very typical." This scale yielded four personality factors, which
were labeled as unhappiness (Factor 1), extraversin (Factor 2), self-
assertiveness (Factor 3), and productive-persistence (Factor 4).
The mean for TOPS' self-ratings on Factor 1, unhappiness, was sig
nificantly larger than that of KOPS, which indicated that TOPS per
ceived themselves as generally unhappy, nervous, tense, and dissatisfied.
TOPS also had a significantly larger mean than KOPS on Factor 2, extra-

13
version, which suggested that the obese considered themselves to be more
outgoing and friendly than KOPS. However, Quereshi noted that obese
persons' perceptions of themselves as extraverts did not necessarily mean
that they were since previous research (Mayer & Thomas, 1967, cited in
Quereshi, 1972) utilizing projective techniques with obese females evi
denced traits such as passivity and withdrawal accompanied by feelings
of rejection.
A more realistic explanation may be gleaned from considering the
high self-ratings on Factors 1 and 2 together. This suggests that obese
females, despite their attempts to gain approval from others by means of
friendliness and congeniality, perceive themselves as lonely and rejected.
This explanation would also support the findings of Werkman and Greenberg
(1967) and Held and Snow (1972).
Gottesfeld (1962) compared self-drawings of thirty super-obese
subjects with thirty neurotics' self-drawings in terms of each group's
body-cathexis, or degree of satisfaction with the parts and processes of
their bodies. The super-obese subjects showed a more negative body-
cathexis than the neurotic group on the following three criteria; (1)
the super-obese were judged as having more negative body-cathexis by a
group of clinicians; (2) their drawings had more major parts of the body
missing; and (3) their drawings were less differentiated.
Gottesfeld (1962) also gave a list of twenty-eight personal traits
to the same group of super-obese individuals and neurotics and asked
them to rate themselves first as how they are and second as how they
would like to be. The discrepancy between self and ideal self served as
a measure of the degree of satisfaction with the self.

14
Results indicated that the super-obese reported to be more satisfied
with themselves than the neurotics. Later, however, while the super-
obese subjects were hospitalized for a two week evaluation, three inde
pendent judges observed their interactions with other patients and staff,
and rated them on the same list of twenty-eight personal traits. The
differences between self-ratings and observer (objective) ratings were
significantly greater than the differences between self-ratings and ideal
ratings. Gottesfeld concluded that the super-obese patients seemed to
present a facade of satisfaction. They denied that they were dissatis
fied with themselves on a self-report trait list, but they could not
as easily guard against a negative self-picture in their drawings (pro
jective test) or in observers' ratings.
Stunkard and Mendelson (1967) have found very low body-concepts in
some overweight people. These people felt that their bodies were gro
tesque and loathesome and that others viewed them with hostility and
contempt. This feeling was associated with self-consciousness and im
paired social functioning. In addition, the person took a very narrow
view of himself, expecially during times of misfortune or unhappiness.
All the unpleasant aspects of his life became focussed on his obesity;
that is, his body became the explanation and the symbol of his unhappi
ness. Buchanan (1973) stated that the body was the receptacle for self-
hate for obese individual; when they felt self-hate they complained
they felt fat.
Negative body-concepts have been found to be most prevalent among
persons who became obese during childhood or adolescence (Stunkard and
Mendelson, 1967; Stunkard and Burt, 1967). In addition, this condition
seemed to persist despite weight reduction and prolonged maintenance of
normal body weight (Stunkard and Burt, 1967).

15
The research review thus far suggests that there are many psycho
logical and social problems associated with obesity. The obese have
problems with impulse control and tend to become easily angered, irri
tated, or resentful. They also describe themselves as disorderly,
pleasure-seeking, and rebellious. In addition, they are seldom active
or competitive and hardly expend a great deal of effort to accomplish a
difficult task. They sometimes take a very unrealistic view of them
selves and tend to blame many of their difficulties on their obesity.
Obese adolescents and young adults seem to have a poor self-concept
which is reflected in their dissatisfaction with their bodies as well as
with themselves. They harbor deep feelings of insecurity and have
exaggerated needs for attention and social approval. The world is seen
as a threatening and rejecting place and their response is to withdraw
as a defense against being hurt. As a result, they often feel lonely and
alienated, distrust other people, and avoid close interpersonal relation
ships .
These obese persons are self-conscious about their size and seem to
feel uncomfortable and anxious in a variety of situations. Consequently,
they tend to restrict their interests both socially and vocationally and
they may be deficient in basic social skills. In addition, these indi
viduals seem to have serious concerns about their masculinity or feminity,
feeling afraid that they cannot perform adequately in sexual situations.
Weight Reduction and Self-Concept
Glucksman and Hirsch (1969) compared obese subjects' performance on
a body size estimation task before, during, and after weight loss with
that of a normal control group. Six obese subjects with a mean initial

16
weight of 334 pounds were hospitalized for eight months and lost an
average of 86.7 pounds. Using an adjustable distorting image apparatus,
subjects were requested to make a distorted screen image of themselves
correspond to their body size as they perceived it at that moment. Obese
subjects increasingly overestimated their own body size from pretreatment,
during fifteen weeks of weight loss, and after a six week maintenance
period. In effect, despite their weight loss, they perceived themselves
as if they had lost almost no weight. In addition, three of the obese
subjects in this study were retested after an additional year of weight
loss, and they continued to overestimate their actual body size.
This "phantom body size" phenomenon was accompanied by supportive
clinical and figure-drawing data. For example, these same subjects
drew progressively larger figure-drawings during weight loss and at the
end of the final weight maintenance period. In addition, they reported
that they continued to feel obese despite weight loss (Glucksman, Hirsch,
McCully, Barron, and Knittle, 1968).
Suczek (1955) tried to delineate psychological aspects of weight
reduction. He administered the Interpersonal Dimension of Personality
System (IDPS) to 100 obese women before and after a sixteen week treat
ment program. This instrument analyzes behavior in terms of five
discrete levels: I, the level of public communication; II, the level of
conscious description; III, the level of private symbolization, IV, the
level of unexpressed unconscious; and V, the level of ego ideal. In this
study, only data from levels'I and II (i.e., overt, facade behavior) were
described.
There was little variability or inconsistency between level I and
level II data at pretreatment. In other words, obese women saw little

17
difference between their views of themselves as they impress others
through overt behavior and their conscious views of themselves (i.e., be
tween how they are and how they claim to be). In essence, they pre
sented themselves as conflict-free.
Subjects were assigned to groups of ten to fifteen individuals,
which met once a week for sixteen weeks. Half the groups were led by
psychiatric social workers whose aim was to promote change toward self
and others while the other groups were led by dieticians and nutritionists
whose aim was to guide spontaneous discussion and provide dietary infor
mation. Subjects in both types of groups were relatively successful in
losing weight. However, there was little or no change at posttreatment
on the IDPS. Thus, at the level of overt behavior and their conceptions
of themselves, the obese women resisted change through their experiences
in the group--their personalities did not change.
Suczek later compared the obese group with a sample of neurotics
undergoing group therapy and found that neurotics did undergo person
ality changes. Hence, changes in behavior and in self-conception were
achievable by means of group methods and measurable by the IDPS. However,
the obese women were not readily amenable to change and did not change
their attitudes about themselves to any appreciable degree in the weight
reduction groups. Unfortunately, no long term follow-up results were
reported to determine if weight losses were maintained.
Collingwood and Willett (1971) investigated the effects of physical
training on self-attitude changes of five male teenagers enrolled in a
special YMCA obese physical training program. The program consisted of
ten hours of exercise per week for three weeks, and a total of three
hours of group counseling-discussion. Two attitude scales were

18
administered at pretreatment and again at posttreatment. A Body Attitude
Scale, containing fifteen body concepts, measured subjects' attitudes
toward their bodies along the following three dimensions: (1) Evalua
tive dimension with good-bad, awkward-graceful, and beautiful-ugly
bipolar adjectives; (2) Potency dimension with weak-strong, hard-soft,
and thin-thick bipolar adjectives; and (3) Activity dimension with active-
passive, cold-hot, and fast-slow bipolar adjectives. In addition,
Bills' Index of Adjustment and Values (IAV) containing twenty-four
adjectives, measured subjects' attitudes toward themselves along the
following three dimensions: (1) Self-Concept ("seldom like me" to "most
of the time like me"); (2) Self-Acceptance ("I very much dislike being
as I am in this respect" to "I"very-much like being as I am in this
respect"); and (3) Ideal Self ("seldom would I like to be that way" to
"most of the time I would like to be that way").
Results at posttreatment indicated a significant weight decrease
and a significant increase on both the evaluative and potency dimensions
of the Body Attitude Scale. In addition, on the IAV there were sig
nificant increases on the Self-Concept and Self-Acceptance dimensions,
and a significant decrease on the discrepancy between Self-Concept and
Ideal Self. These results indicated that physical training experiences
can help teenagers lose weight and enhance their attitudes toward them
selves. However, long-term follow-ups are needed to determine the
stability or permanency of these changes.
Rohrbacher (1973) studied 204 overweight boys between the ages
of eight and eighteen in an eight week weight reduction camp program.
Body image and self-concept assessments (Secord and Jourard, 1953,
scales) were made before and after the camp program, and sixteen weeks

19
after the camp program had ended and subjects had returned home. Sig
nificant weight losses were recorded at posttreatment and at the follow
up, although subjects did regain some weight during the follow-up period.
In addition, body image showed a significant positive change as a re
sult of the camp program, but self-concept remained unchanged. Rohr-
bacher suggested that changing body image may be an important factor
related to the program's success, but that longer follow-ups are needed
to accurately assess the program's effectiveness.
Weller, Arad, and Levit (1977) compared the self-concepts of
twenty-five women who had successfully reduced their weight with those
of twenty-five women who were unsuccessful at weight reduction. On
the Tennessee Self-Concept Scale, successful dieters revealed a better
self-image (general self-concept score) and a better physical self-image
(physical self score) than did the unsuccessful dieters. These results
suggested that those individuals who were most successful at weight re
duction had also modified their self-concepts in the direction of feeling
better about themselves and their bodies.
McCall (1973) studied MMPI profiles of two groups of women belong
ing to TOPS (Take Off Pounds Sensibly). One group had considerable
difficulty losing weight and/or maintaining whatever weight loss they
had achieved. Such refractorily or irremediably obese were designiated
R-TOPS. The other group had successfully lost weight and had maintained
their ideal body weight for at least six months. These women were re
ferred to as KOPS (Keep Off Pounds Sensibly).
On nine of the ten MMPI clinical scales, R-TOPS women were sig
nificantly more deviant than KOPS women. On six of the nine scales, the
differences were significant at the .01 level; on three scales the level

20
of significance was .05. From the six scales on which the relative dif
ferences were greatest R-TOPS women exhibited more body concern (hypo
chondriasis), psychic "hurting" (depression), somatization (hysteria),
rebelliousness (psychopathic deviate), compulsive and ruminative tenden
cies (psychasthenia), and bizarre or confused thinking (schizophrenia).
McCall (1974) studied nineteen women TOPS club members with a mean
weight of 204 pounds. He broke them down into three experimental sub
groups as follows: One group was chosen because their MMPI profiles
closely resembled those of successful weight reducers previously studied
(KOPS); another group consisted of women whose MMPI profiles closely re
sembled those of the resistively obese (R-TOPS); and a third group fell
in-between these two extreme groups. The question asked was whether the
MMPI profiles that distinguished between these three subgroups would
have any bearing on success in group therapy and weight reduction.
Subjects were randomly assigned to one of three groups which met
for sixteen weeks and were oriented toward the development of self-con
trol. Pre- and posttherapy data showed that only the R-TOPS-like sub
group showed significant changes after therapy. They changed on the
following six clinical scales: Hs (hypochondriasis), D (depression), Hy
(hysteria), Pd (psychopathic deviate), Pt (psychasthenia), and Sc
(schizophrenia). These six scales had previously most differentiated
the refractorily obese (R-TOPS) from the remediated obese (KOPS). In
addition, only the R-TOPS-like subgroup showed a significant weight loss
at posttreatment. Hence, obese women who had the "worst" MMPI profiles
tended to benefit most from group therapy as indicated both by weight
loss and profile improvement.

21
The research in weight reduction and concomitant changes in the
self-concept suggests that the self-concept does not change as quickly
as a person loses weight. Weight control programs which solely empha
size restricting your caloric intake seem to have little immediate ef
fects on the self-concept. These programs tend to ignore many of the
psychological or interpersonal problems associated with being overweight
Since they do not address themselves to all facets of the problem of
obesity they appear to be incomplete. While clients may lose weight,
they do not always change their attitudes toward themselves and will
often manifest a "phantom body size." According to Ziller (1973):
A change in the self-concept is the desired
outcome of personal change procedures but the
change processes must involve changes in
attitudes, values, behaviors, and roles.
... If the desired changes in the self-
other orientations have not been achieved, the
client is likely to revert to earlier attitudes,
behaviors, and roles, (p. 176)
Hence, the relative stability of the obese person's self-concept may
account for the poor long-term success of many weight reduction programs
The optimal strategy for personal change seems to be to change the
self-concept. Physical training programs and some experiences in group
therapy seemed to have some effect on the obese person's self-concept
(Collingwood & Willett, 1971; Rohrbacher, 1973; McCall, 1974), while
other group experiences yielded little or no self-concept change
(Suczek, 1955).
Role Playing and Self-Concept
Ziller's (1973) helical theory of personal change indicates that
changing a person's role may exert a strong influence on a person's

22
self-concept; therefore, focussing on taking on new roles may be a valu
able dimension in the treatment of obesity. In review, Ziller's theory
suggests a hierarchy of potentially changeable characteristics which in
clude attitudes, values, behaviors, roles, and self-concepts. It is as
sumed that these characteristics are ordered according to their ease of
change, with attitudes being the least difficult to change while self-
concepts are the most resistant to change.
If one of the components in the system is changed a state of
disequilibrium within the system ensues. It is assumed that in this
imbalanced state there is a tendency for the components lower in the
hierarchy to change in a way which will render the components congruent.
In addition, a change in one component will exert some press toward
change in components higher in the system. A change in behavior, for
example, will exert some press toward changes in attitudes, values, roles,
and self-concepts, but with diminishing force with regard to the higher
level components.
Behavioral approaches to weight control are concerned with chang
ing behaviors, and this will cause the system to be in a state of
disequilibrium. However, a change in behavior will not exert a very
strong press on the individual's self-concept. Hence, in order for the
system to regain equilibrium the person is likely to revert back to
old ways of behaving (i.e., old eating habits) which are congruent with
the self-concept, rather than change the self-concept. Consequently,
weight maintenance is unlikely.
A change in a person's role will exert a stronger press on a per
son's self-concept than a change in behavior since roles are closer than
behaviors to self-concepts on the personal change hierarchy. This

23
suggests that weight control clients should be more likely to change
their self-concepts, and therefore, be more likely to maintain weight
losses if treatment foccusses on changing their roles in addition to
changing their behaviors.
Role playing techniques or behavioral rehearsal may facilitate
role changes. This, in turn, may have a strong effect on a person's
self-concept. Horrocks and Jackson (1972) explain this process. First,
they distinguish between role taking and role playing behavior. Role
taking is a concrete manifestation or implementation of a hypothe
sized identity and presents an observable product of the self-process.
Role playing, however, represents performance by an individual of pre
scribed, demand behavior determined by a situational context. The role
player behaves according to situational expectations, but the behavior
does not represent anything the person believes himself to be in that
context. Role playing behavior, therefore, is the manifestation by
an individual of anti-identities, or those roles not conceptualized by
an individual of himself at that moment.
When an individual role plays anti-identities, locus of control of
his behavior is external to him, imposed upon him by situational or be
havioral demands. When the person receives feedback on his performance,
he assimilates into his cognitive structures the effectiveness and appro
priateness of these actions. In this manner, cognitive dissonance may
alter previous cognitions and change a conception of the self to include
aspects of the anti-identity. As a result, new meanings of the self may
be conceptualized, and role taking behavior replaces role playing.
Horrocks and Jackson suggest that the greater the array of identi
ties an individual incorporates into his conceptualization of himself

24
the greater is his potential for flexible adaptation. Ziller (1973)
supports this idea and states that
individuals with complex self-concepts may be
aware of or consider a greater number of
stimuli as being potentially associated with
the self. In terms of interpersonal perception,
the complex person has a higher probability
of matching some facet of the self with a
facet of the other person, since there are a
larger number of possible matches, (p. 79)
Hence, complex persons are less likely to be disturbed by new experi
ences or situations which appear to be incongruent with their self-
concepts.
The theories of Ziller (1973) and Horrocks and Jackson (1972),
taken together, suggest that role playing experiences facilitate role
taking behavior, and effect the complexity of the self-concept. Hence,
the individual perceives and incorporates new facets of the self; there
fore, he becomes more adaptive or less disturbed by change.
The value of role playing techniques in effecting role change has
been demonstrated by Lazarus (1966). He compared the effectiveness of
behavior rehearsal with two other techniques, direct advice and non
directive reflection-interpretation, in the management of specific
interpersonal problems. Seventy-five clients were randomly assigned to
one of the three procedures (twenty-five clients per subgroup) and a
maximum of four thirty-minute sessions was devoted to each treatment
condition. If there was no evidence of change or learning within one
month the treatment was regarded as having failed. The criterion of
change or learning was objective evidence that the client was behaving
adaptively in the area which had previously constituted a problem; e.g.
the socially awkward girl was going out on dates, the company executive

25
was effecting a promising merger, or the considerate husband had per
suaded his wife to move out of her parents' house into a home of their
own.
Results indicated evidence of learning in eight (32%) of the clients
treated by reflection-interpretation, eleven (44%) of the clients by ad
vice, and twenty-three (92%) of the clinets treated by behavior rehearsal.
In addition, of the thirty-one clients who did not benefit from reflec
tion-interpretation or from advice, twenty-seven were then treated by
behavior rehearsal and there was evidence of learning in twenty-two (81%)
of them. Thus, the overall effectiveness of behavior rehearsal in fifty-
two cases was 86.5%. Hence, behavior rehearsal appeared to be signifi
cantly more effective in changing behavior outside the therapy session
than direct advice or non-directive therapy.
The literature review revealed that the self-concept of obese
adolescents and young adults tends to be characterized by feelings of low
self-esteem, a restricted view of themselves focussed mainly on their
bodies or their size, and social isolation or social anxiety. Moreover,
a change in this self-concept may be a factor facilitating long-term
maintenance of weight loss.
It appears that adapting new roles allows a person to see himself
in a variety of different ways, or expands a person's view of himself,
and makes him more adaptive to change. In addition, role playing tech
niques may facilitate role taking behavior, and may exert a strong press
on changing a person's self-concept. Based on these premises, the
present study was devised to test the efficacy of role playing techniques
in a behavior modification weight reduction group in order to help
clients modify their self-concepts, and to determine the effects this

26
on clients' abilities to lose weight and maintain their losses over a
follow-up period.
In order to carry out this study, two weight reduction groups
were conducted. Both groups emphasized behavioral techniques in order
to lose weight. In addition, one group included an emphasis on using
role playing techniques to facilitate clients' abilities to take on new
roles and change their self-concepts. The other group was presented
with theories and information on weight control.
The following served as the hypotheses:
1. Both groups will show significant reductions in weight at post
treatment.
2. The "role playing" treatment group will show a significantly greater
weight loss than the "information" treatment group at the follow-up.
3. The "role playing" treatment group will have a higher self-concept
than the "information" treatment group at the follow-up.
4. The "role playing" treatment group will have a higher complexity
of the self than the "information" treatment group at the follow-up.
5. The "role playing" treatment group will have a higher physical self-
concept than the "information" treatment group at the follow-up.
6. At the follow-up, "successful" weight reducers will have a higher
self-concept than "unsuccessful" weight reducers.

METHOD
Subjects
The subjects in this study were twenty-nine female college students
enrolled in a weight reduction course taught through the Department of
Health Education at the University of Florida. Two sections of the course
were offered--one met Monday afternoon ("information" group) and one met
Thursday afternoon ("role playing" group). Sixteen women enrolled in
the Monday afternoon section ("information" group) and thirteen women
enrolled in the Thursday afternoon section ("role playing" group).
Subjects siqned up for the section they preferred, mostly depending on
their schedule of classes. Subjects were unaware of any difference
between the two treatment groups at the time of registration.
Pretreatment measurements for the two groups are summarized in
Table 1. In order to determine if there were any differences between
the two groups at pretreatment a two-group discriminant analysis was
performed using all pretreatment variables. This analysis indicated
that the "family self" subscale of the Tennessee Self-Concept Scale
had the most discriminating power of all the variables entered (F_ =
4.79', df = 1,27, £ <.05). The analysis produced no additional sig
nificant discriminating variables.
Since twenty-two variables were put into the analysis, it is likely
that some differences may appear by chance and may not reflect actual
differences between the two groups. In addition, "family self" appears
27

28
Table 1
Pretreatment Measurements of "Role Playing" Group
and "Information" Group
Role Playing Information
(n = 13) (n = 16)
M
SD
M
SD
Initial Weight
176.23
44.49
166.94
25.20
Amount Overweight
45.62
37.41
36.19
23.20
Percent Overweight
34.15
26.04
27.75
18.07
Age
19.77
2.32
20.25
2.50
G.P.A.
2.74
0.42
2.77
0.50
Years Overweight
10.08
7.08
10.44
7.43
Class (l=freshman,
2=sophomore, 3=jr.,
4=sr.)
2.08
0.86
2.44
0.81
Ziller Complexity of
Self Scale
41.15
14.15
44.81
15.78
Tennessee Self-Concept
Scale (raw scores
converted to Std. T
scores: M = 50; SD
= 10)
Self-Criticism
48.23
10.17
52,69
12,48
Total Positive
43.46
11.46
46.69
7.74
Identity
45.85
13.09
50.25
6.93
Self-Satisfaction
44.69
8.53
46.25
7.71
Behavior
42.15
11.78
45.44
8.41
Physical Self
34.69
7.22
36.00
11.58
Moral-Ethical Self
45.31
12.05
48.69
6.73
Personal Self
44.77
11.74
47.81
8.73
Family Self
46.54
10.63
53.31
5.78
Social Self
51.31
14.03
50.50
10.15
Total Variability
54.08
7.65
55.25
9.23
Row Variability
53.92
8.64
54.69
8.22
Column Variability
52.62
6.70
53.81
10.69
Distribution
44.77
11.03
48.88
8.42

29
to be unrelated to the focus of this study, and therefore, should have
little effect on the outcome. Hence, it was assumed that there were
no differences between the two groups at pretreatment.
Students in this class were graded on a pass/fail basis depending
solely on their record of attendance; those who attended eight out
of nine sessions received a passing grade. Only female subjects
were used because weight reduction programs attract mostly female
clients. Subjects met a number of criteria for acceptance into the
program: (1) a need to lose at least twenty pounds; (2) not currently
enrolled in another weight reduction program; (3) not taking medication
for weight; and (4) having no history of a metabloic or hormonal
imbalance that could affect weight.
Measures
Tennessee Self-Concept Scale (TSCS)--Counseling Form
The TSCS (Fitts, 1965) was administered to all subjects at pretreat
ment and at the follow-up. The Scale is a self-administered paper and
pencil test consisting of 100 self-descriptive statements which the
subject uses to portray a picture of herself. For each item the
respondent chooses one of five response options labeled from "completely
false" to "completely true." Fourteen scores are derived from these
items: self-criticism; nine self-esteem scores (identity, self-satis
faction, behavior, physical self, moral-ethical self, personal self, fam
ily self, social self, total); three variability of response scores (var
iation across the first three self-esteem scores, variation across the
next five self-esteem scores, total variation); and a distribution score.
Raw scores on the TSCS have been converted to Standard T-Scores; i.e.,
M = 50, SD = 10.

30
The TSCS was normed on a sample of 626 persons of varying age, sex,
race, socioeconomic status, intellectual level, and educational level.
The standardization group was overrepresented in number of college
students, white subjects, and persons in the twelve to thirty year age
bracket. Test-retest reliability with sixty college students over a
two week period ranged from .80 to .92 for nine self-esteem scales and
ranged from .60 to .89 for the other five scales.
Complexity of Self Scale (see Appendix A)
The Ziller (1973) Complexity of the Self Scale was administered to
all subjects at pretreatment and at the follow-up. This scale measures
the degree of differentiation of the self-concept. A person with high
complexity of the self-concept requires more words to describe herself.
The Ziller Complexity of the Self Scale consists of a list of
109 adjectives and the task is to check the works that describe your
self. The score is the number of words checked. This scale has a range
of zero to 109 and is scored in the direction of high complexity.
In a study involving 100 randomly selected students from grades
seven through twelve, the split-half reliability was .92. Test-retest
reliability after one month for college sophomores was .72.
Other Instruments
A questionnaire asking for specific demographic information was
given to all subjects at the initial meeting (see Appendix A). In
addition, "role playing" subjects were asked to fill out a second
questionnaire regarding how they thought weight reduction would affect

31
their interpersonal relationships (see Appendix A). The information
gathered from the second questionnaire was used to develop appropriate
role training situations.
Dependent Variable
The dependent variable used was a standardized index called the
weight reduction index (RI), which is equal to the percent of excess
weight lost multiplied by the relative initial obesity. In addition
to weight, this index takes into account height, amount overweight,
and weight goal, and is expressed by the following formula:
RI = (Wl/Ws) X (Wi/Wt) X 100
where W1 = weight lost, Ws = surplus weight, Wi = initial weight and
Wt = target weight.
Procedures
The experimenter served as the moderator or therapist for each
group. The emphasis in both groups was on gradual weight loss using
self-control techniques designed to improve eating habits. Each group
met for two hours, once a week for nine weeks. In addition, there was
a four month follow-up.
Week 1
At the initial meeting subjects were given a course outline (see
Appendix B) and the experimenter explained the principles of the program.
Subjects filled out the TSCS, Complexity of Self Scale, and Questionnaire
#1. In addition, "role playing" subjects filled out Questionnaire #2.
At the end of the initial meeting, subjects were weighed, were given a

32
weight control manual, and were asked to take a baseline of their eating
and exercise behavior for the first week.
Week 2
Each subject weighed-in before the class period began. The first
forty-five minutes of this session was an interaction period in which
subjects discussed their reactions to taking a baseline (i.e., "In what
situations were you most likely to eat?"). The experimenter facilitated
discussion by reflecting the content and feelings being expressed.
During the next forty-five minutes of this session the experimenter
explained the food program to be used by subjects. An exchange system
was used where subjects selected from lists of six food groups (meat,
cereal, milk, fruits, vegetables, and miscellaneous). In addition, each
subject figured out her daily caloric intake limit in order to lose
one pound per week. Subjects were also asked to keep a graph of their
daily caloric intake, their daily exercise level, and their weekly
weight.
During the final part of the session subjects participated in an
exercise period. Exercise consisted of walking or running around the
Florida Field for six minutes. During the third week this time was
increased to nine minutes, and for weeks four through nine subjects
spent twelve minutes exercising.
Weeks 3-9
Standard Procedures
At the beginning of every class session each subject was privately

33
weighed on a balance scale. The first forty-five minutes of each class
session in both groups was an interaction period. Subjects spent the
time discussing difficulties and successes they had during the past
week while the experimenter facilitated the discussion by reflecting the
content and feelings being expressed by the subjects.
At the end of the interaction period each week the experimenter
explained a few of the behavioral techniques for subjects to practice.
Lesson plans followed the program developed by Moody and Schreiber
(1979a, 1979b). Techniques focussed on gradual habit improvement
through stimulus control, breaking or extending the chain of eating, and
positive reinforcement. Some of these techniques included:
1. Eat slower: Chew thoroughly and swallow before you pick up another
forkful. Take small bites and set a minimum time of twenty minutes for
which your meal must last. Enjoy your food; do not gobble it.
2. Use small dishes: Use smaller dishes and smaller food containers.
This will make small portions appear larger.
3. Use positive reinforcement: Silently praise yourself each time you
adhere to one of the guidelines. In addition, reward yourself points
each time you obey one of the principles, and when you accrue a certain
amount of points, administer a pre-selected reward.
4. Develop a routine schedule of eating: Eat three meals each day and
eat at the same time each day. Do not eat at other times! Do not eat
on impulse.
5. Never engage in any other activities while eating: Pay attention to
eating; do not watch T. V., read the newspaper, listen to the radio, or
engage in similar activities. Do not eat unless you are hungry.
6. Prepare only one portion of each course at a time: Take only one

34
helping, one course at a time. Take small portions. Do not feel that
you must complete a meal, especially if you planned the meal when you
were hungry. If you are not hungry, leave the remaining food on your
plate. Be guided by your feelings of hunger, not by the amount of
food on the table.
7. Do not buy prepared foods: Limit your diet to foods which must be
prepared before eating. Eat low calorie foods or foods that are dif
ficult to eat (e.g., eat five carrots or crackers rather than one piece
of candy). Do not tempt yourself.
8. Eat only in one place in one room: Only eat while sitting at the
table. Do not take snacks into other rooms; do not eat at your dsk;
dont eat standing up. Do not eat casually.
9. Eat before a party: If you plan to attend a social event at which
there will be a great temptation to eat high calorie foods, eat a small,
low calorie meal before you go.
10. Do not prepare food while you are hungry: If you tend to nibble
while preparing meals, prepare the meal at a time when you are not
hungry (e.g., prepare dinner soon after lunch). If you go grocery
shopping, do that right after eating.
11. Develop a pre-potent repertory: Think of times you have a hard time
resisting food. Plan other activities you can be involved in during
these times so that you won't be around food.
,The last part of each class session in both groups was spent
exercising. Subjects were also encouraged to exercise during the week
between classes and to note improvements in the number of laps they
completed during the twelve minute walk/run period.

35
Variable Procedures
Role Playing Group
During the second forty-five minute period of the third week "role
playing" subjects were introduced to role playing. The experimenter
explained to the class that when people lose weight other people may
begin reacting to them differently. Hence, as a thin person they may be
faced with new situations which will require new roles. As a result,
part of this class was used to help students practice these new roles.
Week 3
Exercise 1: Getting Acquainted
Purpose
a. To help group members get acquainted in a relatively non-threatening
manner.
b. To explore feelings generated by "becoming another person."
c. To explore the dimensions of a brief encounter.
d. To emphasize the need for careful, active listening as well as
self-disclosure during conversation.
Instructions
1. Group members are paired in dyads, and the facilitator instructs
participants to "get to know your partner" for the next few minutes
(5 to 10 minutes). Participants are instructed to listen to the "free
information" or clues their partners give about themselves and to
follow-up on this free information. In addition, participants are in
structed to "self-disel ose" or give information about themselves to
their partners.

36
2. After the interviewing phase, group members reassemble in the
larger group. The facilitator indicates that they now have the responsi
bility of introducing their partner to the group. Each group member,
in turn, is to introduce her partner by standing behind her and speaking
in the first person, as if she were that partner. There should be no
rechecking between partners during this phase. The individual who is
being introduced should hold her comments for the discussion period.
3. After all the introductions have been made, the facilitator leads a
discussion of the exercise, focussing on feelings generated and/or the
issues inherent in the goals of the exercise.
Exercise 2: On Being Fat and Thin
Purpose
a. To examine stereotypical differences between fat and thin people.
b. To get subjects thinking about their feelings concerning being fat
and being thin.
Instructions
1. Have the group generate a list of generalizations about fat and thin
people. Brainstorm in this manner for five to ten minutes, creating
an arsenal of stereotypes.
s. Discuss this exercise, focussing on subjects' feelings about the
1 ist.
Homework: Think about new situations that you may encounter as a result
of losing weight. What new roles will you become involved in as a thin
person?
Purpose
a. To encourage students to expand their views of themselves, i.e.,

37
to increase their complexity of themselves.
b.To have students think about situations they could role play.
Week 4
Exercise 1: Giving Feedback
Purpose: To teach students how to give helpful, appropriate feedback.
Instructions
1. The facilitator defines "feedback" to the group as:
A way of helping another person to consider
changing her behavior. It is communication to
a person (or group) which gives that person
information about how she affects others.
Feedback helps an individual keep her behavior
on target and thus better achieve her goals.
2. Group members discuss various criteria for useful feedback.
Exercise 2: Role Playing Fat and Thin People
Purpose
a. To compare how it feels to be a fat person with how it feels to be
a thin person.
b. To rehearse how to act in different situations.
c. To develop an awareness of oneself in relation to other people,
d. To see oneself in a variety of different ways.
e. To learn how to give appropriate, helpful feedback.
Role Playing Instructions
1. Make up a situation that you would like to role play. Discuss the
incident carefully with the group and how the different characters
might feel in the situation. Discuss the main character's goal and the
best way to achieve this goal considering the way in which other people
are feeling. Next, have group members volunteer to act in the different

38
roles. Now dramatize the situation for five to ten minutes, or until it
comes to a natural conclusion. Do the same situation twice: first, as
you would act as an overweight person, and second, as you would act as
a person of normal weight.
2. After you have enacted the situation both ways (i.e., overweight and
normal weight roles) discuss the situation and give each other feedback
in the following order: (1) main character: hwo did you feel about
acting overweight and about acting normal weight? (2) other person(s)
in the exercise: how did you feel about yourself in relation to the
main character? (3) non-participant observers: what did you observe
was going on? During the feedback session other group members may
role play alternate ways of responding to the situation.
3. Think of various ways of responding no matter how strange they seem.
Contrast responses and see what "fits."
4. There are no right or wrong answers in this exercise, so don't be
afraid of making mistakes. We are practicing and comparing ways of act
ing as overweight and normal weight people so feel free to try out any
role you like.
Weeks 5 to 9
During the second forty-five minute period of each week "role play
ing" subjects continued the role playing exercise. The following is a
list,of various situations that were role played.
1. You see a friend whom you haven't seen in a long time. He remarks,
"You look different. Have you been sick?"
2. You're home on vacation and your mother keeps "pushing" food on
you. You'd love to taste some of the food, but you are already full.

39
Your mother feels very hurt because you don't want to eat her food.
3. You're at a party with some of your girlfriends. A guy you've been
attracted to for awhile notices you and strikes up a conversation with
you.
4. You've lost some weight but haven't yet reached your goal. You
are sitting on a bus next to a guy whom you've just met. After talking
with him for awhile he remarks, "You have a beautuiful face! Do you
realize that if you lost some weight you'd really be a knock-out!"
5. You're with a girlfriend on campus and meet a guy that she knows.
He ignores you and speaks only to your girlfriend. Even after your girl
friend has introduced you and has made several references to you during
the conversation he continues to address his attention solely to your
girlfriend.
6. You are at a bar and strike up a conversation with a guy. He
seems friendly but as time goes by he becomes overly aggressive.
7. A guy calls you up and wants to take you out to eat. You want to go
out with him but have already eaten.
8. You are with a friend and you are both anxious about an exam. Your
friend says, "Let's go get something to eat."
9. You've lost some weight and you're walking home across campus. You
meet a guy that you know and he tells you now great you look and insists
on taking you to the Arrendondo Room for lunch.
10. ,You've lost weight and your roommate, who is a little overweight,
begins making snide remarks about your body. You suspect she may be
feeling a bit jealous of you.

40
Information Group
During the second forty-five minute period of each week the "infor
mation" treatment group was presented with various weight control
theories. The following is an outline of the topics discussed each
week.
Week 3 Hypnotherapy: Dr. Sig Fagerberg
A. Conscious vs. subconscious
B. The power of suggestion to the subconscious
C. Hypnotic state
D. Suggestibility and self-hypnosis
1. Daily use of suggestion
2. Simple, repetitive, positive imagery pertaining to the future
E. Books
1. LeCron, Lesley. Self-Hypnosis
2. Peale. Power of Positive Thinking
Week 4 External Cue Sensitivity
A. Perceived time
B. Food cues
C. Response cost
D. Taste
E. Emotional state
F. Feedback and competing cues
Week 5 Nutrition and Dieting: Stephanie Fredette, Graduate Student
in Nutrition
A. Basic four food groups

41
B. Definition of obesity
C. Adipose cell theory
D. Basal metabolic rate
E. Additives and preservatives
F. Pinch test
Week 6 Surgical Methods of Weight Reduction: Dr. Joun Kuldeau,
J. Hillis Miller Health Center
A. Interstinal bypass surgery
1. Diagram and description of operation
2. Adverse physical side effects
3. Benefits of operation
B. Stapling the stomack shut
1. Diagram and description of operation
2. Advantages of this procedure over intestinal bypass operation
C. Case reports of super-obese clients
D. Mood changes during rapid weight reduction
Week 7 Hereditary and environmental factors in obesity
A. Family research
B. Twin studies
1. Variability in weights in identical twins
2. Identical twins reared in same environment vs. identical twins
reared in different environments
3. Identical twins vs. fraternal twins vs. non-twin siblings
C. Adopted children research
Week 8 TOPS (Take Off Pounds Sensibly): Marilyn poss, Gainesville

42
Chapter of TOPS
A. History and background of TOPS
B. TOPS basic program
1. No prescribed diet
2. Changing eating habits
3. Group therapy approach
4. Competitionhonors and rewards
C. KOPS (Keep Off Pounds Sensibly) Maintenance of weight loss
D. Case reports of successful clients
Week 9 Other Approaches to Weight Control
A. Aversive conditioning
B. Counting mouthfuls
C. Fad diets and crash diets
D. Scarsdale diet
Follow-up
During the week of July 9, 1979, all subjects were contacted by
mail (see Appendix C). They were asked to report their current weight
and to fill out the Tennessee Self-Concept Scale and the Ziller
Complexity of Self Scale.

RESULTS
The sample at posttreatment consisted of 29 subjects; 13 were in
the "role playing" treatment group and 16 were in the "information"
treatment group. The attrition rate during the program was 0%. The
follow-up sample consisted of 28 subjects; 12 were in the "role play
ing" group and 16 were in the "information" group. One subject from
the "role playing" group had left the country and could not be contacted.
Hypotheses
Hypothesis 1: Both groups will show significant reductions in weight at
posttreatment.
The mean reduction index at posttreatment was 20.00 for subjects in
the "role playing" group and 14.06 for subjects in the "information"
group. Additional analysis revealed that the mean reduction index at
the follow-up was 18.58 for subjects in the "role playing" group and
38.19 for subjects in the "information" group. One-way t tests indica
ted that all four of these indices were significant at the .05 level
or greater {Table 2). Hypothesis 1 was accepted.
43

44
Table 2
Mean Reduction Indices for "Information" and "Role Playing"
Groups
Role
Playing
Information
M
SD
df t
M
SD
df
t
Posttreatment
20.00
24.91
11 -2.78**
14.06
17.51
15
-3.21**
Follow-up
18.58
32.34
11 -1.99*
38.19
29.37
15
-5.20***
* £ < .05
** £ < .01
*** £ < .001
Hypothesis 2: The "role playing" group will show a significantly
greater weight loss than the "information" group at the follow-up.
Hypothesis 3: The "role playing" group will have a higher self-concept
than the "information" group at the follow-up.
Hypothesis 4: The "role playing" group will have a higher complexity of
the self than the "information"group at the follow-up.
Hypothesis 5: The "role playing" group will have a higher physical self-
concept than the "information" group at the follow-up.
The following measures were used in order to test the above hypothe
ses: (1) the Total Positive subscale of the Tennessee Self-Concept Scale
(TSCS) was used to assess self-concept; (2) the Ziller Complexity of
Self Scale was used to assess complexity of the self; and (3) the Physi
cal Self subscale of the TSCS was used to assess physical self-concept.
Follow-up measurements for the two treatment groups are summarized in
Table 3.

45
Table 3
Means and Standard Deviations for "Role Playing" and "Information"
Treatment Groups on all Measures at the Follow-up
Role Playing Information
(n = 12) (n = 16)
M
SD
M
SD
Self-Concept
45.50
11.39
47.25
8.18
Complexity
39.42
15.40
42.69
17.53
Physical Self
37.67
8.60
38.81
9.92
Reduction Index
18.58
32.34
38.19
29.37
In order to test Hypotheses 2, 3, 4 and 5 Multivariate Analysis of
Variance (MANOVA) and Discriminant Analysis techniques as discussed by
Kerlinger and Pedhazur (1973) were used. MANOVA procedures to determine
significant differences between the "role playing" and "information"
groups on the four follow-up measures as a group (self-concept,
complexity, physical self-concept, and reduction index) revealed an
£ ratio of .95, df = 4,24, jd > .05.
Univariate £ values, standardized discriminant coefficients (SDC),
and £ values for SDC for the "role playing" and "information" treatment
groups on all follow-up measures are presented in Table 4. Results from
the discriminant analysis revealed that there were no significant
differences between the treatment groups relative to self-concept,
complexity, and physical self-concept as noted by the respective £
values of 1.23, 0.89, and 0.88. The reduction index approached

46
significance in the opposite direction than predicted as indicated by
an £ value of 3.46 (jd < .10; df = 4,24).
When each variable was analyzed controlling for the effects of all
other variables, none of the follow-up measures proved to be significant
discriminators. In other words, none of the follow-up measures were
significantly different for the two groups (Table 4, £ values for
SDC). Hypotheses 2, 3, 4, and 5 were rejected.
Table 4
Univariate £ Values, Standardized Discriminant Coefficients (SDC),
and £ Values for SDC for the "Role Playing" and "Information"
Treatment Groups on all Follow-up Measures
Univariate
£ Values
Standardized
Discriminant
Coefficients (SDC)
F Values
for SDC
Self-Concept
1.23
0.29
0.05
Complexity
0.89
0.19
0.05
Physical Self
0.88
0.07
0.01
Reduction Index
3.46*
0.80
0.30
*£ < .10
Hypothesis 6: At the follow-up, those subjects who have maintained the
greatest weight loss will have a higher self-concept than unsuccessful
weight reducers.
Subjects in both treatment groups were combined at the follow-up
and divided into "successful" and "unsuccessful" weight reducers. A
median split of the reduction indices was used to divide the two groups.

47
A subject was labeled "successful" if her reduction index was greater
than 25.00 and "unsuccessful" if her reduction index was less than
25.00. Pretreatment and follow-up measures of these two groups are
summarized in Table 5.
Table 5
Means and Standard Deviations of "Successful" and
"Unsuccessful" Weight Reducers on Pretreatment
and Follow-up Measures
Successful Unsuccessful
(n
= 14)
(n =
14)
M
SD
M
SD
Pretreatment
Self-Concept
46.43
9.99
44.14
9.67
Complexity
47.47
15.54
37.43
12.41
Physical Self
37.00
12.41
33.71
6.62
Follow-up
Self-Concept
47.93
7.20
45.07
11.50
Complexity
46.93
16.93
35.64
14.33
Physical Self
40.07
8.00
36.57
10.30
Reduction Index
54.71
21.60
4.86
16.67
A two-group discriminant
analysis
of the "successful"
and "unsuc-
cessful" weight reducers revealed a canonical correlation
of 0.42, cor-
responding to a chi-
square of
4.80 (df
= 2, £ variables was analyzed controlling for
the effects of all
other

48
variables pretreatment complexity and treatment group accounted for the
most variance between the two groups, as indicated by the Standardized
Discriminant Weights in Table 6. The variable self-concept failed to
be a significant discriminator between the two groups (hypothesis 6
was rejected). When the type of treatment and each of the pretest
and follow-up measures were analyzied independently pretreatment and
posttreatment complexity scores approached a significant differency
between the two groups (]3 variate F values in Table 6.
Table 6
Univariate £ Values and Standardized Discriminant
Weights for "Successful" and "Unsuccessful"
Weight Reducers on Treatment Group and
all Pretest and Follow-up Measures
Standardized
Discriminant
Weights
Univariate
£ Values
Pretreatment
Self-Concept
0.03
0.38
Physical Self
0.04
0.76
Complexity
0.74
3.64*
Treatment Group
0.55
2.36
Follow-up
Self-Concept
0.50
0.62
Physical Self
--
1.01
Complexity
0.32
3.62*
Standardized Discriminant Weights greater than 0.50 are underlined
* £ < .10

49
Pretreatment to Follow-up Change Scores
Paired jt tests on mean pretreatment and follow-up measures for
each group indicated no significant differences on any of the following
variables: self-concept, complexity, and physical self (Table 7).
Table 7
Comparison of Pretreatment and Follow-up Scores
for "Role Playing" and "Information" Treat
ment Groups on all Measures
Pretreatment Follow-up
M SD M SD df t p
Role Playing" Group
Self-Concept
43.46
11.46
Complexity
41.15
14.15
Physical Self
34.69
7.22
Information Group"
Self-Concept
46.69
7.74
Complexity
44.81
15.78
Physical Self
36.00
11.58
45.50
11.39
11
-1.15
n.s.
39.42
15.40
11
0.58
n.s.
37.67
8.60
11
-1.85
n.s.
47.25
8.18
15
-0.27
n.s.
42.69
17.53
15
1.00
n.s.
38.81
9.92
15
-1.15
n.s.

50
Correlations Between Variables
Correlation coefficients were calculated between each of the
following variables: initial weight, percent overweight, G.P.A.,
years overweight, pretreatment self-concept, pretreatment complexity,
pretreatment physical self, treatment group, follow-up self-concept,
follow-up complexity, follow-up physical self, and follow-up reduction
index (Table 8).

Table 8
Correlations Between Variables
Initial
Wt
Percent
Overwt G.P.A.
Years
Overwt
Pretreat Pretrt
Self-con Comp
Pretreat Trtmt
Phys Self Group
F. up
S.C.
F. up
Comp
F.up F.up
Phys Self Red Ind
Init'l Wt 1
.00
0.90 -0.06
0.00
-0.05
0.22
-0.04
-0.16
-0.14
0.25
-0.15
-0.27
1 Overwt
1.00 0.00
0.00
0.05
0.08
-0.04
-0.16
0.05
0.19
-0.09
-0.22
G.P.A.
1.00
0.11
0.19
0.06
-0.04
0.03
0.12
0.07
-0.13
0.15
Yrs Overwt
1.00
-0.09
-0.24
-0.22
0.08
-0.15
-0.15
-0.23
0.07
Pre S.C.
1.00
0.43
0.77*
0.17
0.70
0.39
0.50
-0.03
Pre Comp
1.00
0.57
0.50
0.00
0.83
0.12
0.20
Pre Phys Self
1.00
0.08
0.36
0.43
0.63
0.00
Grp (l=role
play, 2=info)
1.00
0.09
0.10
0.06
0.31
F.up S.C.
1.00
0.01
0.60*
0.15
F.up Comp
1.00
0.07
0.19
F.up Phys Self
1.00
0.09
F.up Red Ind
1.00
* Spuriously high because of overlapping test items

DISCUSSION
The findings of the present study indicate that a behavioral
self-control approach to overeating produces significant weight losses
during the treatment period. In addition, this study supports Jef
frey's (1974) findings that a self-control approach produces significant
long-term weight losses. In the present study mean reduction indices
for both treatment groups were significant after a four month follow-up
period.
At the follow-up, a comparison of the mean reduction indices be
tween the two treatment groups approached significance in the opposite
direction than predicted; that is, there was a trend toward greater
weight loss for subjects in the "information" treatment group. A pos
sible reason for this outcome may be related to the issue of client/
treatment compatibility. Several studies have indicated that outcome
in psychotherapy is a function of the interaction between treatment
parameters and client variables (Abramowitz et al., 1974; Friedman & Dies,
1974; Kilman, Albert, & Sotile, 1973). In other words, clients differ
in the type of treatment to which they best respond.
Subjects in the "information" treatment group were exposed to a
broader range of theories and treatment concerning obesity than sub
jects in the "role playing" treatment group. As a result, they may
have had a greater opportunity to select an appropriate or compatible
treatment approach.
52

53
Weight control programs can be greatly improved if they could
specify the type of client with whom they are more likely to be success
ful. In addition, treatment procedures may be enhanced if client
characteristics which hinder effective weight reduction efforts could be
delineated and then modified.
In the present study, pretreatment self-concept measurements
indicated that subjects presented a fairly positive view of themselves
(general self-concept); they maintained a poor opinion only of their
physical beings (physical self). Hence, they denied any dissatisfaction
with themselves except with their physical bodies.
Past research in obesity has revealed that obese individuals often
manifest other emotional and social problems (Held & Snow, 1972;
Querehi, 1972; Werkman & Greenberg, 1967; Wunderlich, Johnson, & Ball,
1973). This seemed to hold true of the subjects in the present study.
During the weekly meetings subjects disclosed conflicts regarding their
identities, social relationships, families, and other personal concerns.
However, research also indicates that obese persons may deny their
problems on a self-report questionnaire (Gottesfeld, 1962; Suczek, 1955;
Werkman & Greenbert, 1967). This may be the reason why subjects'
scores fell within the normal range on the total positive scale of the
Tennessee Self-Concept Scale. The only difficulty these subjects ad
mitted to was the obvious one; hence, their scores were below average on
the physical self subscale of the Tennessee Self-Concept Scale, but not
on the total positive scale. A projective instrument may have proven
to be a more valid indicator of subjects' general self-concepts.
At the follow-up no significant differences in self-concept were
found between the two treatment groups. Although total self-concept
scores may not have been accurate depictions of subjects' general level

54
of self-esteem, the failure of role playing techniques to affect self-
concepts is probably a valid result. This is supported by other results
which indicated that role playing techniques had no significant affect
on those parts of the self-concept concerning subjects' complexity or
physical selves. The physical self subscale of the Tennessee Self-
Concept Scale dealt with concerns subjects were more open about and the
Ziller Complexity of Self Scale is more indirect in measuring complexity.
Hence, these scores are more likely to be valid, yet they too remained
unaffected by role playing techniques.
A possible explanation for the failure of role playing techniques
to effectuate self-concept changes involves the relative commitment an
individual has to each of the five components of ZiTier's (1973) helical
theory of personal change (attitudes, values, behaviors, roles, and
self-concept). According to Roby (1960; cited in Ziller, 1973), "com
mitment refers to a reduction in further alternatives associated with a
particular choice." (p. 151)
The amount of commitment an individual has to each of the components
of Ziller's system increases as one goes up the hierarchy. For example,
having reported certain attitudes is less restricting than having accepted
a certain role, such as group leader. Violation of the expectations
associated with certain attitudes is more acceptable to the self and
others than violation of role expectancies. The greatest commitment
is associated with the self-concept; these constructs are associated
with the stability, regularity, and consistency of personal behavior.
Since attitudes require the least amount of commitment, they are
subject to the most rapid change. A change in attitudes permits some
experimentation within the personal system; little commitment is

55
involved and the personal system may return to its previous state with
relatively little difficulty.
Increasing commitment is made as changes in
the other components proceed. The hierarchy
suggests an orderly progression up the scale.
Attitude changes are tried first, then values,
behaviors, roles, and finally the self-concept,
in turn, under favorable circumstances (p. 154,
Ziller, 1973)
In the present study, the "role playing" treatment group was a
social learning approach with an emphasis on practicing or rehearsing
new social roles. However, subjects may not have been ready to commit
themselves to role changes; more likely, they were at the stage of
experimenting with new attitudes, values, and behaviors. There was
evidence of this during the role playing sessions. During the dis
cussions subjects often expressed their feelings on such issues as
jealousy, intimacy, respect, and assertiveness. Hence, it seems that
subjects were concerned with the lower components of Ziller's hierarchy.
Learning new social roles may not become a need for clients until
after they have changed their behavior. At this point they would have
lost a significant amount of weight, would be confronted with new
situations, and may feel a social role deficit.
In the present study, the intervention of role playing may have
been pre-mature and may have even served to arouse some anxiety about
the prospect of losing weight. Future work in weight control might
postpone the intervention of role playing until after clients are more
committed to new attitudes, values, and behaviors. Then, it is more
likely that role playing techniques would facilitate role changes and
exert a strong press toward a change in the self-concept. A longer
term treatment intervention would be required.

56
Although role playing techniques did not have an immediate impact
on subjects' reported self-concepts, these techniques may have facili
tated subjects' own awareness of themselves. This additional self-
awareness may eventually prove helpful in the process of losing weight.
Previous research indicates that young overweight people are self-
conscious about their size and seem to feel uncomfortable in interper
sonal or social situations. Consequently, they may become socially with
drawn. Thus, while overweight people remain overweight because of
poor eating habits, social problems may develop as well. However,
these problems are often denied; hence, many overweight people may
remain unaware of their feelings and needs. In the present study, role
playing scenarios partly served to help subjects become aware of their
feelings in interpersonal situations. In addition, the feedback ses
sions following role playing gave subjects an arena in which to share
their feelings.
Remaining overweight may become a defense against having to deal
with personal concerns. Losing weight removes this defense. Some
people may not want to deal with this issue and may resist losing weight.
In this respect, losing weight is similar to any other type of personal
change; i.e., it is often resisted.
People may not be aware of their own defenses, and may not be aware
of the consequences of losing them. In the present study, role playing
techniques served to make subjects aware of personal issues. Role
playing experiences brought out many anxiety provoking situations. As.
a result, these subjects became more aware of the challenges they might
face while losing weight.
These speculations would only hold true for overweight people with

57
interpersonal anxiety. It is possible that only subjects who were not
socially anxious were successful at losing weight. It was found that
they tended to be more complex people. Since complexity is defined as
the ability to match an aspect of oneself with an aspect of the environ
ment, this may also imply more ease in social and interpersonal
situations. It would be difficult to measure the relationship between
social anxiety and success at weight reduction since social anxiety
is difficult to measure; i.e., overweight people tend to deny personal
problems.
This line of reasoning implies that weight reduction groups may
only work to the extent that feelings can be expressed and an awareness
of personal problems can be developed. Groups that are restrictive in
these areas will not help clients whose personal problems are entwined
with their weight. If an awareness of personal concerns is fostered
then success at weight control may depend upon the person's ability to
deal with these concerns. The leader's responsibility would be to
help a person mobilize or find appropriate support systems.
An analysis of the self-concepts of "successful" and "unsuccessful
weight reducers revealed no significant differences between the two
groups. Both "successful" and "unsuccessful" weight reducers presented
a fairly positive total self-concept while their physical self
picture remained poor. These results support previous research
indicating that the physical self-concepts of obese persons do not
change commensurate with weight loss (Glucksman & Hirsch, 1969;
Glucksman et al., 1968; Rohrbacher, 1973; Suczek, 1955). In addition,
these results are compatible with Ziller's theory regarding change.
According to this theory, people are more likely to change the self

58
system starting with the lower components and gradually progressing
to the higher components. The ultimate level of change is a change in
the self-concept, and this needs to be achieved if lower level changes
are to be maintained.
Weight reduction reflects a change in behavior, the third
component in Ziller's hierarchy. According to Ziller's theory, if
weight loss (behavior change) is to be maintained it must be followed by
changes in roles and self-concept. Since these additional changes take
time, it follows that self-concept changes would lag behind weight
loss. Hence, in the present study, follow-up measures were probably
taken some time after "successful" subjects had changed their behavior,
but before they changed their self-concepts. As a result, no self-
concept differences were found between "successful" and "unsuccessful"
weight reducers at the time of the follow-up. A longer term follow-up
might reveal one of the following results: (1) "Successful" weight
reducers would eventually change their self-concepts; therefore, beha
vior changes would be retained and weight loss would be maintained, or
(2) The self-concept of "successful" weight reducers would remain
unchanged; therefore, behavior would return to its previous state and
weight would be regained.
A further analysis of "successful" and "unsuccessful" weight
reducers revealed that those people who lost the most weight tended to
be more complex people at pretreatment. This supports the theories
of Horrocks and Jackson (1972) and Ziller (1973) who state that more
complex people incorporate a greater array of facets or stimuli into
their conceptualization of themself. Complex people are less likely
to be disturbed by new experiences because there is a higher probability

59
of matching some aspect of themself with an aspect of the situation.
Hence, they have a greater potential for flexible adaptation.
Ziller (1973) suggests that
strategies of personal change directed toward the
self-concept may be accomplished with greater
facility if they begin with concern for complexity
of the self-concept, in order to render the self
system more adaptive, (pp. 155-156)
Future work in weight control might investigate the usefulness of
various techniques to help people expand their view of themselves and
consequently have a greater potential for change.
The present study suggests the need for longer term treatment of
overweight clients. It is proposed that unless positive changes in
the self-concept accompany changes in behavior following behavior
modification procedures, the treatment should be continued until such
changes are observed.

APPENDICES

APPENDIX A
INSTRUMENTS AND FORMS

Questionnaire #1
Name: Height:
Address: Size of frame:
small medium large
Phone: (circle one)
Where will you be during the week of July 9, 1979?
Address:
Phone:
Are you taking medication for weight?
Are you enrolled in another weight program now?
Do you have a metabolic or hormonal imbalance which could affect
your weight?
What is your class? freshman, sophomore, junior, senior, grad, student
(circle one)
What is your age?
What is your major area of study?
What is your grade point average?
For how may years have you been overweight?
62

63
Questionnaire #2
1. When you lose weight, your physical appearance will change. Who do
you think will notice?
2. How do you think people will react to you when you lose weight?
3. Do you think you will act differently as a thin person?
If yes, please specify how you will act differently.
4.Do you expect to encounter new situations when you lose weight?
If yes, please specify.

64
Ziller Complexity of Self Scale
INSTRUCTIONS: Here is a list of words. You are to read the words quick
ly and check each one that you think describes YOU. You may check as
many or as few words as you like--but be HONEST. Don't check words that
tell what kind of a person you should be. Check words that tell what
kind of a person you really are.
1.
able
29.
delicate 57.
large
85.
serious
2.
active
30.
delightful 58.
lazy
86.
sharp
3.
afraid
31.
different 59.
1ittle
87.
silly
4.
alone
32.
difficult 60.
lively
88.
slow
5.
angry
33.
di rty 61.
lonely
89.
smal 1
6.
anxious
34.
dull 62.
loud
90.
smart
7.
ashamed
35.
dumb 63.
lucky
91.
soft
8.
attractive
36.
eager 64.
mild
92.
special
9.
bad
37.
fair 65.
miserable
93.
strange
10.
beautiful
38.
faithful 66.
modest
94.
stupid
11.
big
39.
false 67.
neat
95.
strong
12.
bitter
40.
fine 68.
old
96.
sweet
13.
bold
41.
fierce 69.
patient
97.
terrible
14.
brave
42.
foolish 70.
peaceful
98.
ugly
15.
bright
43.
friendly 71.
perfect
99.
unhappy
16.
busy
44.
funny 72.
pleasant
100.
unusual
17.
calm
45.
generous 73.
_pol ite
101.
useful
18.
capable
46.
gentle 74.
poor
102.
valuable
19.
careful
47.
glad 75.
popular
103.
warm
20.
careless
48.
good 76.
proud
104.
weak
21.
charming
49.
great 77.
quiet
105.
wild
22.
cheerful
50.
happy 78.
quick
106.
wise
23.
clean
51.
humble 79.
responsible
107.
wonderful
24.
clever
52.
idle 80.
rough
108.
wrong
25.
comfortable
53.
important 81.
rude
109.
young
26.
content
54.
independent 82.
sad
27.
cruel
55.
jealous 83.
sel fish
28.
curious
56.
kind 84.
sensible

APPENDIX B
COURSE OUTLINE

Course Outline
Title: Hes 4905 Variable Topics in Health Education Weight Control:
Self-Managed Behavior Change
Instructors: Fred Schreiber 377-7604
Sig Fagerberg
Thursday Monday
January 4 8
11 15
18 22
25 29
February 1 5
8 12
15 19
22 26
March 1 5
July 9, 1979: 4 month follow-up
Required Reading: Toward Permanent Weight Balance: Student's Manual
Course Description and Requirements:
This is a nine week weight control program. You will be graded on
an S/U basis. Grades will be given strictly on the basis of your
attendance (not on how much weight you lose). In order to pass all you
need to do is attend class. In cases of emergency, or other unforseen
circumstances, you will be allowed to miss one class, but you must
notify the instructor as to your reason for missing class.
In addition, you are asked to participate in a 4 month follow-up.
This will be held during the week of July 9, 1979, and will involve fill
ing out a few questionnaires and weighing in. If you are going to be
out of town at that time, please note where I may contact you.
This weight control program is part of a research project designed
to determine if certain techniques which are used to help individuals
lose weight are more effective with one type of personality or another.
Therefore, we are asking you to fill out some information forms, and
your responses will be strictly confidential. Both groups will employ
techniques that have been shown to be effective in helping people lose
weight.
The groups will meet once a week for nine weeks with a trained
leader. The techniques utilized in treatment will include various self-
control procedures which will be discussed during the sessions and
practiced at home. No techniques which will cause pain or discomfort
will be used during the program.
66

APPENDIX C
LETTER SENT TO SUBJECTS AT THE FOLLOW-UP

1700 S.W. 16 Ct. #A-1
Gainesville,, Fla, 32608
July 9 1979
Dear
It has been approximately four months since the end of our weight
control class and now I am doing the follow-up. Please record your
current weight at the bottom of this page, fill out both of the enclosed
questionnaires, and return all the materials (this letter, 1 test
booklet, 2 answer sheets) in the enclosed envelope. I realize that
not all of you may have continued losing weight, and that some of you
may have gained weight. However, please be honest in reporting your
present weight and in answering the enclosed questionnaires so that
my results will be valid.
Please read all directions carefully. When you fill out the
Tennessee Self-Concept Scale remember that the answer sheet is arranged
so that you respond to every other item on it. Please do not omit
any item. If you want to change an answer, do not erase it; rather,
mark an X through the incorrect response and circle the response you
want.
Please send these materials back to me AS SOON AS POSSIBLE as I am
very anxious to finish my paper before the end of the summer. If you
are interested in discussing the results of my research with me please
call me in mid-August. My phone number is 377-7604.
Sincerely,
Fred M. Schreiber

REFERENCES
Abramowitz, C. V., Abramowitz, S. I., Roback, H. B. & Jackson, C.
Differential effectiveness of directive and nondirective group
therapies as a function of client internal-external control.
Journal of Consulting and Clinical Psychology, 1974, 42, 849-853.
Abramson, E. E. A review of behavioral approaches to weight control.
Behavior Research and Therapy, 1973, Vl_, 547-556.
Buchanan, J. R. Five year psychoanalytic study of obesity. American
Journal of Psychoanalysis, 1973, ^33, 30-38.
Collingwood, T. R. & Willett, L. The effects of physical training upon
self-concept and body attitude. Journal of Clinical Psychology,
1971, 27, 411-412.
Fitts, W. H. Tennessee Self-Concept Scale Manual. Nashville, Tenn.:
Counselor Recordings and Tests, 1965.
Friedman, M. L. & Dies, R. R. Reactions of internal and external test-
anxious students to counseling and behavior therapies. Journal of
Consulting and Clinical Psychology, 1974, 42_, 921.
Glucksman, M. L. & Hirsch, J. The response of obese patients to weight
reduction: III. The perception of body size. Psychosomatic Medicine,
1969, 31_, 1-7.
Glucksman, M. L., Hirsch, J., McCully, R. S., Barron, B. A. & Knittle,
J. L. The response of obese patients to weight reduction. II. A
quantitative evaluation of behavior. Psychosomatic Medicine, 1968,
30, 359-373.
Gottesfeld, H. Body and self-cathexis of super-obese patients.
Journal of Psychosomatic Research, 1962, 6_, 177-183.
Harris, M. B. & Bruner, C. G. A comparison of a self-control and a
contract procedure for weight loss. Behavioral Research and Therapy,
1971,9,347-354.
Held, M. L. & Snow, D. L. MMPI, Internal-external control, and problem
check list scores of obese adolescent females. Journal of Clinical
Psychology, 1972, 28, 523-525.
69

70
Horrocks, J. E. & Jackson, D. W. Self and Role: A Theory of Self-
Process and Role Behavior. Boston: Houghton Mifflin Company, 1972.
Ince, L. P. The self-concept variable in behavior therapy. Psycho
therapy: Theory, Research, and Practice, 1972, _9, 223-225,
Jeffrey, D. B. A comparison of the effects of external control and self-
control on the modification and maintenance of weight. Journal of
Abnormal Psychology, 1974, 83, 404-410.
Jeffrey, D. B., Christensen, E. R. & Katz, R. C. Behavior therapy
weight reduction programs: Some preliminary findings on the need
for follow-ups. Psychotherapy: Theory, Research, and Practice,
1975, 12, 311-313.
Kerlinger, F. N. & Pedhazur, E. J. Multiple Regression in Behavior
Research. New York: Holt & Winston, Inc., 1973.
Kilman, P. R., Albert, B. M. & Sotile, W. M. Relationship between locus
of control, structure of therapy, and outcome. Journal of Consulting
and Clinical Psychology, 1975, 43, 588.
Kurtz, R. M. Body attitude and self-esteem. Proceedings, 79th Annual
Convention, APA, 1971, 6_, 467-468.
Lazarus, A. A. Behaviour rehearsal vs. non-directive therapy vs. advice
in effecting behaviour change. Behaviour Research and Therapy, 1966,
4, 209-212.
Lerner, R. M., Karabenick, S. A. & Stuart, J. L. Relations among
physical attractiveness, body attitudes, and self-concept in male
and female college students. Journal of Psychology, 1973, 85, 119-129.
Mahoney, E. R. Body-cathexis and self-esteem: The importance of sub
jective importance. Journal of Psychology, 1974, 88, 27-30.
Mahoney, E. R. & Finch, M. D. Body-cathexis and self-esteem: A
reanalysis of the differential contribution of specific body aspects.
Journal of Social Psychology, 1976, 99, 251-258.
McCall, R. J. MMPI factors that differentiate remediably from
irremediably obese women. Journal of Community Psychology, 1973, 1,
34-36.
McCall, R. J. Group therapy with obese women of varying MMPI profiles.
Journal of Clinical Psychology, 1974, 30, 466-470.
Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Instructor's Manual. Gainesville, Fla.: Florida Cooperative
Extension Service, 1979(a).
Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Student1s Manual. Gainesville, Fla.: Florida Cooperative
Externsion Service, 1979(b).

71
Penick, S. B., Filion, R., Fox, S. & Stunkard, A. Behavior modification
in the treatment of obesity. Psychosomatic Medicine, 1971 213, 49-55,
Quereshi, M. Y. Some psychological factors that distinguish between
the remediably and irremediably obese. Journal of Clinical Psychology,
1972, 28, 17-22.
Rogers, C. R. Client-Centered Therapy. Boston: Houghton Mifflin
Company, 1951.
Rohrbacher, R. Influence of a special camp program for obese boys on
weight loss, self-concept, and body image. Research Quarterly,
1973, 44, 150-157.
Rosen, G. M. & Ross, A. 0. Relationship of body image to self-concept.
Journal of Consulting and Clinical Psychology, 1968, 32, 100.
Secord, P. F. & Jourard, S. M. The appraisal of body-cathexts: Body-
cathexis and the self. Journal of Consulting Psychology, 1953,
17, 343-347.
Stuart, R. B. Behavioral control of overeating. Behaviour Research and
Therapy, 1967, 1, 357-365.
Stuart, R. B. A three-dimensional program for the treatment of obesity.
Behaviour Research and Therapy, 1971, 1, 177-186.
Stuart, R. B. & Davis, B. Slim Chance in a Fat World: Behavioral Con
trol of Obesity. Research Press: Champaign, Ill., 1972.
Stunkard, A. J. The management of obesity. New York State Journal of
Medicine, 1958, 58, 79-87.
Stunkard, A. & Burt, V. Obesity and body image: II. Age at onset of
disturbances in the body image. American Journal of Psychiatry,
1967, 123, 1443-1447.
Stunkard, A. & Mendelson, M. Obesity and the body image: I. Char
acteristics of disturbance in the body image of some obese persons.
American Journal of Psychiatry, 1967, J_23, 1296-1300.
Suczek, R. F. Psychological aspects of weight reduction. In Eppright,
E. S., Swanson, P. & Iverson, C. A. (Committee in charge). Weight
Control: A collection of papers presented at the weight control
colloquiurn. Ames, Iowa: Iowa State College Press, 1955.
Weinberg, J. R. A further investigation of body-cathexts and the self.
Journal of Consulting Psychology, 1960, 24, 277.
Weller, L., Arad, T. & Levit, R. Self-concept, delayed gratification,
and field dependence of successful and unsuccessful dieters. Israel
Annals of Psychiatry and Related Disciplines, 1977, J_5, 41-46.

72
Werkman, S. L. & Greenberg, E. S. Personality and interest patterns in
obese adolescent girls. Psychosomatic Medicine, 1967, 29_, 72-80.
Wollersheim, J. P. Effectiveness of group therapy based upon learning
principles in the treatment of overweight women. Journal of
Abnormal Psychology, 1970, 715, 462-474.
Wunderlich, R. A., Johnson, W. G. & Ball, M. F. Some personality
correlates of obese persons. Psychological Reports, 1973, 32,
1267-1277.
Ziller, R. C. The Social Self. New York: Pergamon Press Inc., 1973.
Zion, L. Body concept as it relates to self-concept. Research Quarterly,
1965, 36, 490-495.

BIOGRAPHICAL SKETCH
Fred M. Schreiber was born in New York City on May 2, 1953.
He graduated from Baldwin Senior High School in Baldwin, New York,
in June, 1971. He attended the University of Rochester in Rochester,
New York, from which he received a B, A. degree in psychology in
May, 1975. In September, 1975, he enrolled at the University of
Florida for graduate study in psychology, and received an M. A. degree
in psychology in June, 1977. Presently he is a Counseling Intern
at the Counseling and Testing Service at the University of Houston,
and expects to receive a Ph.D. in psychology from the University of
Florida upon completion of his internship in August, 1980.
73

I certify that I have read this study and that in my opinion it
conforms, to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
farry A. Grader, Chairman
'Professor or Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Ted Landsman
Professor of Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
/.eld
Paul G. Schauble
Professor of Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Afesa Bel 1-Nathaniel
Assistant Professor of
Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Seigfrd Fagerbergy
Associate Professor of
Health Education and Safety
This dissertation was submitted to the Graduate Faculty of the
Department of Psychology in the College of Liberal Arts and Sciences
and to the Graduate Council, and was accepted as partial fulfillment
of the requirements for the degree of Doctor of Philosophy.
August 1980
Dean, Graduate School



9
Kurtz (1971) tested the hypothesis that body attitudes are related
to feelings of self-esteem using two instruments which share very
little method similarity. He used a self-report body attitude scale
(methodologically similar to that used by Secord and Jourard, 1953) and
the Ziller (1969) Self-Esteem Scale, which is a more indirect or pro
jective technique for measuring a person's feelings toward himself. For
the group of college students sampled, the hypothesis that a positive
evaluative body attitude and a positive sense of self-esteem are related
was confirmed. Of importance was the strength of this relationship con
sidering that the scales used shared little methodological similarities.
The relationship between body-cathexis and self-concept sheds some
light on the self-concept of overweight individuals. It is assumed that
people who volunteer to participate in weight reduction programs are
dissatisfied with their bodies. Since people tend to cathect their
bodies to the same degree that they cathect themselves, overweight in
dividuals in weight control groups are probably somewhat dissatisfied
with themselves. Hence, the preceding research on the relationship
between body-cathexis and self-concept implies that obese individuals in
weight reduction groups may be partly characterized by feelings of low
self-esteem.
Obese Self-Concept and Body-Cathexis
Werkman and Greenberg (1967) compared eighty-eight obese adolescent
girls at a medically oriented camp for overweight girls with forty-two
normal-weight girls at an ordinary summer camp on a number of personality
and interest measures (MMPI, Strong Vocational Interest Blank, Semantic
Differential, and Sentence Completion Blank). One finding which cut


59
of matching some aspect of themself with an aspect of the situation.
Hence, they have a greater potential for flexible adaptation.
Ziller (1973) suggests that
strategies of personal change directed toward the
self-concept may be accomplished with greater
facility if they begin with concern for complexity
of the self-concept, in order to render the self
system more adaptive, (pp. 155-156)
Future work in weight control might investigate the usefulness of
various techniques to help people expand their view of themselves and
consequently have a greater potential for change.
The present study suggests the need for longer term treatment of
overweight clients. It is proposed that unless positive changes in
the self-concept accompany changes in behavior following behavior
modification procedures, the treatment should be continued until such
changes are observed.


49
Pretreatment to Follow-up Change Scores
Paired jt tests on mean pretreatment and follow-up measures for
each group indicated no significant differences on any of the following
variables: self-concept, complexity, and physical self (Table 7).
Table 7
Comparison of Pretreatment and Follow-up Scores
for "Role Playing" and "Information" Treat
ment Groups on all Measures
Pretreatment Follow-up
M SD M SD df t p
Role Playing" Group
Self-Concept
43.46
11.46
Complexity
41.15
14.15
Physical Self
34.69
7.22
Information Group"
Self-Concept
46.69
7.74
Complexity
44.81
15.78
Physical Self
36.00
11.58
45.50
11.39
11
-1.15
n.s.
39.42
15.40
11
0.58
n.s.
37.67
8.60
11
-1.85
n.s.
47.25
8.18
15
-0.27
n.s.
42.69
17.53
15
1.00
n.s.
38.81
9.92
15
-1.15
n.s.


34
helping, one course at a time. Take small portions. Do not feel that
you must complete a meal, especially if you planned the meal when you
were hungry. If you are not hungry, leave the remaining food on your
plate. Be guided by your feelings of hunger, not by the amount of
food on the table.
7. Do not buy prepared foods: Limit your diet to foods which must be
prepared before eating. Eat low calorie foods or foods that are dif
ficult to eat (e.g., eat five carrots or crackers rather than one piece
of candy). Do not tempt yourself.
8. Eat only in one place in one room: Only eat while sitting at the
table. Do not take snacks into other rooms; do not eat at your dsk;
dont eat standing up. Do not eat casually.
9. Eat before a party: If you plan to attend a social event at which
there will be a great temptation to eat high calorie foods, eat a small,
low calorie meal before you go.
10. Do not prepare food while you are hungry: If you tend to nibble
while preparing meals, prepare the meal at a time when you are not
hungry (e.g., prepare dinner soon after lunch). If you go grocery
shopping, do that right after eating.
11. Develop a pre-potent repertory: Think of times you have a hard time
resisting food. Plan other activities you can be involved in during
these times so that you won't be around food.
,The last part of each class session in both groups was spent
exercising. Subjects were also encouraged to exercise during the week
between classes and to note improvements in the number of laps they
completed during the twelve minute walk/run period.


1700 S.W. 16 Ct. #A-1
Gainesville,, Fla, 32608
July 9 1979
Dear
It has been approximately four months since the end of our weight
control class and now I am doing the follow-up. Please record your
current weight at the bottom of this page, fill out both of the enclosed
questionnaires, and return all the materials (this letter, 1 test
booklet, 2 answer sheets) in the enclosed envelope. I realize that
not all of you may have continued losing weight, and that some of you
may have gained weight. However, please be honest in reporting your
present weight and in answering the enclosed questionnaires so that
my results will be valid.
Please read all directions carefully. When you fill out the
Tennessee Self-Concept Scale remember that the answer sheet is arranged
so that you respond to every other item on it. Please do not omit
any item. If you want to change an answer, do not erase it; rather,
mark an X through the incorrect response and circle the response you
want.
Please send these materials back to me AS SOON AS POSSIBLE as I am
very anxious to finish my paper before the end of the summer. If you
are interested in discussing the results of my research with me please
call me in mid-August. My phone number is 377-7604.
Sincerely,
Fred M. Schreiber


31
their interpersonal relationships (see Appendix A). The information
gathered from the second questionnaire was used to develop appropriate
role training situations.
Dependent Variable
The dependent variable used was a standardized index called the
weight reduction index (RI), which is equal to the percent of excess
weight lost multiplied by the relative initial obesity. In addition
to weight, this index takes into account height, amount overweight,
and weight goal, and is expressed by the following formula:
RI = (Wl/Ws) X (Wi/Wt) X 100
where W1 = weight lost, Ws = surplus weight, Wi = initial weight and
Wt = target weight.
Procedures
The experimenter served as the moderator or therapist for each
group. The emphasis in both groups was on gradual weight loss using
self-control techniques designed to improve eating habits. Each group
met for two hours, once a week for nine weeks. In addition, there was
a four month follow-up.
Week 1
At the initial meeting subjects were given a course outline (see
Appendix B) and the experimenter explained the principles of the program.
Subjects filled out the TSCS, Complexity of Self Scale, and Questionnaire
#1. In addition, "role playing" subjects filled out Questionnaire #2.
At the end of the initial meeting, subjects were weighed, were given a


29
to be unrelated to the focus of this study, and therefore, should have
little effect on the outcome. Hence, it was assumed that there were
no differences between the two groups at pretreatment.
Students in this class were graded on a pass/fail basis depending
solely on their record of attendance; those who attended eight out
of nine sessions received a passing grade. Only female subjects
were used because weight reduction programs attract mostly female
clients. Subjects met a number of criteria for acceptance into the
program: (1) a need to lose at least twenty pounds; (2) not currently
enrolled in another weight reduction program; (3) not taking medication
for weight; and (4) having no history of a metabloic or hormonal
imbalance that could affect weight.
Measures
Tennessee Self-Concept Scale (TSCS)--Counseling Form
The TSCS (Fitts, 1965) was administered to all subjects at pretreat
ment and at the follow-up. The Scale is a self-administered paper and
pencil test consisting of 100 self-descriptive statements which the
subject uses to portray a picture of herself. For each item the
respondent chooses one of five response options labeled from "completely
false" to "completely true." Fourteen scores are derived from these
items: self-criticism; nine self-esteem scores (identity, self-satis
faction, behavior, physical self, moral-ethical self, personal self, fam
ily self, social self, total); three variability of response scores (var
iation across the first three self-esteem scores, variation across the
next five self-esteem scores, total variation); and a distribution score.
Raw scores on the TSCS have been converted to Standard T-Scores; i.e.,
M = 50, SD = 10.


Course Outline
Title: Hes 4905 Variable Topics in Health Education Weight Control:
Self-Managed Behavior Change
Instructors: Fred Schreiber 377-7604
Sig Fagerberg
Thursday Monday
January 4 8
11 15
18 22
25 29
February 1 5
8 12
15 19
22 26
March 1 5
July 9, 1979: 4 month follow-up
Required Reading: Toward Permanent Weight Balance: Student's Manual
Course Description and Requirements:
This is a nine week weight control program. You will be graded on
an S/U basis. Grades will be given strictly on the basis of your
attendance (not on how much weight you lose). In order to pass all you
need to do is attend class. In cases of emergency, or other unforseen
circumstances, you will be allowed to miss one class, but you must
notify the instructor as to your reason for missing class.
In addition, you are asked to participate in a 4 month follow-up.
This will be held during the week of July 9, 1979, and will involve fill
ing out a few questionnaires and weighing in. If you are going to be
out of town at that time, please note where I may contact you.
This weight control program is part of a research project designed
to determine if certain techniques which are used to help individuals
lose weight are more effective with one type of personality or another.
Therefore, we are asking you to fill out some information forms, and
your responses will be strictly confidential. Both groups will employ
techniques that have been shown to be effective in helping people lose
weight.
The groups will meet once a week for nine weeks with a trained
leader. The techniques utilized in treatment will include various self-
control procedures which will be discussed during the sessions and
practiced at home. No techniques which will cause pain or discomfort
will be used during the program.
66


n
the MMPI than the non-obese group. These scales were D (depression),
Pd (psychopathic deviate), Pa (paranoia), Pt (psychasthenia), and Sc
(schizophrenia). The generally elevated profile of the obese group,
particularly the Pd and Sc scales, suggested that many had serious psy
chological disturbances characterized by feelings of depression,
alienation, and low self-worth. In addition, they tended to be non-con-
forming, had problems in impulse control, and were distrustful of others.
Held and Snow failed to find the obese group defensive about reveal
ing psychological problems in that they admitted to significantly more
difficulties on the Mooney Problem Check List than the non-obese group.
These findings suggested that obese adolescent girls may need help recon
ciling some of the feelings they have toward themselves and other people
in addition to programs designed to help them lose weight.
Wunderlich, Johnson, and Ball (1973) administered the Adjective
Check List (ACL) and the Edwards Personal Preference Schedule (EPPS) to
sixteen subjects ranging from 64.6% to 214.7% overweight. The obese
group endorsed a significantly fewer number of favorable adjectives on
the ACL than did the normative sample. In addition, the obese were
characterized as having a low need for achievement, poor personal adjust
ment, and few preferences for close attachments to other people. They
did not describe themselves as being dependable, cautious, stable, or
inhibited. Instead, they described themselves as hasty, pleasure-seeking,
headstrong, and rebellious. Order for them was low, as they described
themselves as being careless, changeable, and disorderly, as well as
having poor self-control.
The obese scored low on endurance, suggesting that they were less
willing to expend energy toward a goal or persist at a task. Hence, the


26
on clients' abilities to lose weight and maintain their losses over a
follow-up period.
In order to carry out this study, two weight reduction groups
were conducted. Both groups emphasized behavioral techniques in order
to lose weight. In addition, one group included an emphasis on using
role playing techniques to facilitate clients' abilities to take on new
roles and change their self-concepts. The other group was presented
with theories and information on weight control.
The following served as the hypotheses:
1. Both groups will show significant reductions in weight at post
treatment.
2. The "role playing" treatment group will show a significantly greater
weight loss than the "information" treatment group at the follow-up.
3. The "role playing" treatment group will have a higher self-concept
than the "information" treatment group at the follow-up.
4. The "role playing" treatment group will have a higher complexity
of the self than the "information" treatment group at the follow-up.
5. The "role playing" treatment group will have a higher physical self-
concept than the "information" treatment group at the follow-up.
6. At the follow-up, "successful" weight reducers will have a higher
self-concept than "unsuccessful" weight reducers.


30
The TSCS was normed on a sample of 626 persons of varying age, sex,
race, socioeconomic status, intellectual level, and educational level.
The standardization group was overrepresented in number of college
students, white subjects, and persons in the twelve to thirty year age
bracket. Test-retest reliability with sixty college students over a
two week period ranged from .80 to .92 for nine self-esteem scales and
ranged from .60 to .89 for the other five scales.
Complexity of Self Scale (see Appendix A)
The Ziller (1973) Complexity of the Self Scale was administered to
all subjects at pretreatment and at the follow-up. This scale measures
the degree of differentiation of the self-concept. A person with high
complexity of the self-concept requires more words to describe herself.
The Ziller Complexity of the Self Scale consists of a list of
109 adjectives and the task is to check the works that describe your
self. The score is the number of words checked. This scale has a range
of zero to 109 and is scored in the direction of high complexity.
In a study involving 100 randomly selected students from grades
seven through twelve, the split-half reliability was .92. Test-retest
reliability after one month for college sophomores was .72.
Other Instruments
A questionnaire asking for specific demographic information was
given to all subjects at the initial meeting (see Appendix A). In
addition, "role playing" subjects were asked to fill out a second
questionnaire regarding how they thought weight reduction would affect


4
Jeffrey, Christensen, and Katz (1975) reported the results of four
subjects whom they had treated for obesity using a behavioral approach.
After six months of treatment weight losses ranged from twenty to thirty-
two pounds. However, after a six month follow-up period weight losses
(pretreatment to follow-up) ranged from three to thirty-two pounds; two
subjects had maintained their weight loss while the other two subjects
gained back part or most of their weight.
Harris and Bruner (1971) reported two studies examining the
effectiveness of a self-control behavioral program and a contract
procedure for weight reduction. In the first study self-control pro
cedures included positive reinforcement, stimulus control, and breaking
or lengthening the chain of eating while the contract program consisted
of subjects receiving a monetary reward for each pound lost. Both
groups lost a significant portion of body weight after two months of
treatment. Furthermore, the contract group had lost a larger proportion
of initial weight than the self-control group. However, after a ten
month follow-up period there were no significant weight losses for
either group. In a second study, Harris and Bruner (1971) compared a
self-control weight group with a no-treatment control group and found no
significant weight losses for either group after sixteen weeks.
Penick, Filion, Fox, and Stunkard (1971) compared a behavior
modification weight control gorup with a control group which received
supportive psychotherapy as well as instruction about dieting and nutri
tion. After three months of treatment the median weight loss for the
behavior modification group was greater than that of the control group,
but this difference did not reach significance. Penick et al. attribute
this result to the significantly larger variability of outcome in the


55
involved and the personal system may return to its previous state with
relatively little difficulty.
Increasing commitment is made as changes in
the other components proceed. The hierarchy
suggests an orderly progression up the scale.
Attitude changes are tried first, then values,
behaviors, roles, and finally the self-concept,
in turn, under favorable circumstances (p. 154,
Ziller, 1973)
In the present study, the "role playing" treatment group was a
social learning approach with an emphasis on practicing or rehearsing
new social roles. However, subjects may not have been ready to commit
themselves to role changes; more likely, they were at the stage of
experimenting with new attitudes, values, and behaviors. There was
evidence of this during the role playing sessions. During the dis
cussions subjects often expressed their feelings on such issues as
jealousy, intimacy, respect, and assertiveness. Hence, it seems that
subjects were concerned with the lower components of Ziller's hierarchy.
Learning new social roles may not become a need for clients until
after they have changed their behavior. At this point they would have
lost a significant amount of weight, would be confronted with new
situations, and may feel a social role deficit.
In the present study, the intervention of role playing may have
been pre-mature and may have even served to arouse some anxiety about
the prospect of losing weight. Future work in weight control might
postpone the intervention of role playing until after clients are more
committed to new attitudes, values, and behaviors. Then, it is more
likely that role playing techniques would facilitate role changes and
exert a strong press toward a change in the self-concept. A longer
term treatment intervention would be required.


46
significance in the opposite direction than predicted as indicated by
an £ value of 3.46 (jd < .10; df = 4,24).
When each variable was analyzed controlling for the effects of all
other variables, none of the follow-up measures proved to be significant
discriminators. In other words, none of the follow-up measures were
significantly different for the two groups (Table 4, £ values for
SDC). Hypotheses 2, 3, 4, and 5 were rejected.
Table 4
Univariate £ Values, Standardized Discriminant Coefficients (SDC),
and £ Values for SDC for the "Role Playing" and "Information"
Treatment Groups on all Follow-up Measures
Univariate
£ Values
Standardized
Discriminant
Coefficients (SDC)
F Values
for SDC
Self-Concept
1.23
0.29
0.05
Complexity
0.89
0.19
0.05
Physical Self
0.88
0.07
0.01
Reduction Index
3.46*
0.80
0.30
*£ < .10
Hypothesis 6: At the follow-up, those subjects who have maintained the
greatest weight loss will have a higher self-concept than unsuccessful
weight reducers.
Subjects in both treatment groups were combined at the follow-up
and divided into "successful" and "unsuccessful" weight reducers. A
median split of the reduction indices was used to divide the two groups.


58
system starting with the lower components and gradually progressing
to the higher components. The ultimate level of change is a change in
the self-concept, and this needs to be achieved if lower level changes
are to be maintained.
Weight reduction reflects a change in behavior, the third
component in Ziller's hierarchy. According to Ziller's theory, if
weight loss (behavior change) is to be maintained it must be followed by
changes in roles and self-concept. Since these additional changes take
time, it follows that self-concept changes would lag behind weight
loss. Hence, in the present study, follow-up measures were probably
taken some time after "successful" subjects had changed their behavior,
but before they changed their self-concepts. As a result, no self-
concept differences were found between "successful" and "unsuccessful"
weight reducers at the time of the follow-up. A longer term follow-up
might reveal one of the following results: (1) "Successful" weight
reducers would eventually change their self-concepts; therefore, beha
vior changes would be retained and weight loss would be maintained, or
(2) The self-concept of "successful" weight reducers would remain
unchanged; therefore, behavior would return to its previous state and
weight would be regained.
A further analysis of "successful" and "unsuccessful" weight
reducers revealed that those people who lost the most weight tended to
be more complex people at pretreatment. This supports the theories
of Horrocks and Jackson (1972) and Ziller (1973) who state that more
complex people incorporate a greater array of facets or stimuli into
their conceptualization of themself. Complex people are less likely
to be disturbed by new experiences because there is a higher probability


8
how satisfied they were with each characteristic of their own body and
how important each part was in determining their own physical attractive
ness. In addition, subjects responded to a short self-concept scale
consisting of sixteen bipolar dimensions derived from the Adjective Check
List. Results indicated that the degree of positive self-concept in
creased with the degree of positive attitude toward one's body character
istics (males: r. = -33; females: r. = *43). However, weighting satis
faction ratings by corresponding importance ratings did not significantly
increase the satisfaction/self-concept relation (males: r. = .33; females
21 = .44). Hence, while this study supports the general relationship
between self-concept and body-cathexis put forth by Secord and Jourard
(1953), it failed to support the results of Rosen and Ross (1968)
that importance ratings strengthened this relationship.
Mahoney (1974) replicated the Lerner et al. findings also using col
lege students as subjects and self-report attitude measures. He found
the correlation between mean unweighted (by importance) body-cathexis
and self-esteem to be .45 for males and .37 for females. For weighted
body-cathexis the respective correlations for males and females were
.37 and .41.
Further support for the relationship between body attitude and
self-esteem has been provided by Weinberg (1960), Zion (1965), and
Mahoney and Finch (1976). In each study, moderate correlations were
found using college students and self-report instruments. The use of
an identical scaling procedure to measure self-esteem and body attitude
raises the question that the high correlation found between body-
cathexis and self-esteem may represent method variance and not an
intertrait relationship. The following study addresses this issue.


47
A subject was labeled "successful" if her reduction index was greater
than 25.00 and "unsuccessful" if her reduction index was less than
25.00. Pretreatment and follow-up measures of these two groups are
summarized in Table 5.
Table 5
Means and Standard Deviations of "Successful" and
"Unsuccessful" Weight Reducers on Pretreatment
and Follow-up Measures
Successful Unsuccessful
(n
= 14)
(n =
14)
M
SD
M
SD
Pretreatment
Self-Concept
46.43
9.99
44.14
9.67
Complexity
47.47
15.54
37.43
12.41
Physical Self
37.00
12.41
33.71
6.62
Follow-up
Self-Concept
47.93
7.20
45.07
11.50
Complexity
46.93
16.93
35.64
14.33
Physical Self
40.07
8.00
36.57
10.30
Reduction Index
54.71
21.60
4.86
16.67
A two-group discriminant
analysis
of the "successful"
and "unsuc-
cessful" weight reducers revealed a canonical correlation
of 0.42, cor-
responding to a chi-
square of
4.80 (df
= 2, £ variables was analyzed controlling for
the effects of all
other


44
Table 2
Mean Reduction Indices for "Information" and "Role Playing"
Groups
Role
Playing
Information
M
SD
df t
M
SD
df
t
Posttreatment
20.00
24.91
11 -2.78**
14.06
17.51
15
-3.21**
Follow-up
18.58
32.34
11 -1.99*
38.19
29.37
15
-5.20***
* £ < .05
** £ < .01
*** £ < .001
Hypothesis 2: The "role playing" group will show a significantly
greater weight loss than the "information" group at the follow-up.
Hypothesis 3: The "role playing" group will have a higher self-concept
than the "information" group at the follow-up.
Hypothesis 4: The "role playing" group will have a higher complexity of
the self than the "information"group at the follow-up.
Hypothesis 5: The "role playing" group will have a higher physical self-
concept than the "information" group at the follow-up.
The following measures were used in order to test the above hypothe
ses: (1) the Total Positive subscale of the Tennessee Self-Concept Scale
(TSCS) was used to assess self-concept; (2) the Ziller Complexity of
Self Scale was used to assess complexity of the self; and (3) the Physi
cal Self subscale of the TSCS was used to assess physical self-concept.
Follow-up measurements for the two treatment groups are summarized in
Table 3.


28
Table 1
Pretreatment Measurements of "Role Playing" Group
and "Information" Group
Role Playing Information
(n = 13) (n = 16)
M
SD
M
SD
Initial Weight
176.23
44.49
166.94
25.20
Amount Overweight
45.62
37.41
36.19
23.20
Percent Overweight
34.15
26.04
27.75
18.07
Age
19.77
2.32
20.25
2.50
G.P.A.
2.74
0.42
2.77
0.50
Years Overweight
10.08
7.08
10.44
7.43
Class (l=freshman,
2=sophomore, 3=jr.,
4=sr.)
2.08
0.86
2.44
0.81
Ziller Complexity of
Self Scale
41.15
14.15
44.81
15.78
Tennessee Self-Concept
Scale (raw scores
converted to Std. T
scores: M = 50; SD
= 10)
Self-Criticism
48.23
10.17
52,69
12,48
Total Positive
43.46
11.46
46.69
7.74
Identity
45.85
13.09
50.25
6.93
Self-Satisfaction
44.69
8.53
46.25
7.71
Behavior
42.15
11.78
45.44
8.41
Physical Self
34.69
7.22
36.00
11.58
Moral-Ethical Self
45.31
12.05
48.69
6.73
Personal Self
44.77
11.74
47.81
8.73
Family Self
46.54
10.63
53.31
5.78
Social Self
51.31
14.03
50.50
10.15
Total Variability
54.08
7.65
55.25
9.23
Row Variability
53.92
8.64
54.69
8.22
Column Variability
52.62
6.70
53.81
10.69
Distribution
44.77
11.03
48.88
8.42


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EIQULZU25_MV917I INGEST_TIME 2014-12-05T23:58:38Z PACKAGE AA00026479_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


21
The research in weight reduction and concomitant changes in the
self-concept suggests that the self-concept does not change as quickly
as a person loses weight. Weight control programs which solely empha
size restricting your caloric intake seem to have little immediate ef
fects on the self-concept. These programs tend to ignore many of the
psychological or interpersonal problems associated with being overweight
Since they do not address themselves to all facets of the problem of
obesity they appear to be incomplete. While clients may lose weight,
they do not always change their attitudes toward themselves and will
often manifest a "phantom body size." According to Ziller (1973):
A change in the self-concept is the desired
outcome of personal change procedures but the
change processes must involve changes in
attitudes, values, behaviors, and roles.
... If the desired changes in the self-
other orientations have not been achieved, the
client is likely to revert to earlier attitudes,
behaviors, and roles, (p. 176)
Hence, the relative stability of the obese person's self-concept may
account for the poor long-term success of many weight reduction programs
The optimal strategy for personal change seems to be to change the
self-concept. Physical training programs and some experiences in group
therapy seemed to have some effect on the obese person's self-concept
(Collingwood & Willett, 1971; Rohrbacher, 1973; McCall, 1974), while
other group experiences yielded little or no self-concept change
(Suczek, 1955).
Role Playing and Self-Concept
Ziller's (1973) helical theory of personal change indicates that
changing a person's role may exert a strong influence on a person's


17
difference between their views of themselves as they impress others
through overt behavior and their conscious views of themselves (i.e., be
tween how they are and how they claim to be). In essence, they pre
sented themselves as conflict-free.
Subjects were assigned to groups of ten to fifteen individuals,
which met once a week for sixteen weeks. Half the groups were led by
psychiatric social workers whose aim was to promote change toward self
and others while the other groups were led by dieticians and nutritionists
whose aim was to guide spontaneous discussion and provide dietary infor
mation. Subjects in both types of groups were relatively successful in
losing weight. However, there was little or no change at posttreatment
on the IDPS. Thus, at the level of overt behavior and their conceptions
of themselves, the obese women resisted change through their experiences
in the group--their personalities did not change.
Suczek later compared the obese group with a sample of neurotics
undergoing group therapy and found that neurotics did undergo person
ality changes. Hence, changes in behavior and in self-conception were
achievable by means of group methods and measurable by the IDPS. However,
the obese women were not readily amenable to change and did not change
their attitudes about themselves to any appreciable degree in the weight
reduction groups. Unfortunately, no long term follow-up results were
reported to determine if weight losses were maintained.
Collingwood and Willett (1971) investigated the effects of physical
training on self-attitude changes of five male teenagers enrolled in a
special YMCA obese physical training program. The program consisted of
ten hours of exercise per week for three weeks, and a total of three
hours of group counseling-discussion. Two attitude scales were


3
to learn appropriate eating behaviors in order to lose weight. In addi
tion, they were trained in relaxation in order to counter tension in
situations which would ordinarily result in eating. The positive expecta-
tion--social pressure group utilized group pressure to help subjects lose
weight. The non-specific therapy group helped subjects understand their
underlying motives for being overweight. Wollersheim reported that all
three treatment groups lost significantly more weight than the control
group after a twelve week treatment program. Furthermore, after an
eight week follow-up subjects in the behavioral self-control group showed
a significantly greater weight loss (pretreatment to follow-up) than
either of the other two treatment groups.
Stuart (1971) treated six overweight women on an individual basis
in a fifteen week program which stressed environmental control of
overeating, nutritional planning, and regulated increase in energy
expenditure. Half the women started treatment immediately while the
other group started treatment fifteen weeks later. Approximately six
months following termination of treatment for the first group and three
months following termination of treatment for the second group, follow
up data were collected. Results indicated that subjects in the first
group lost an average of thirty-five pounds while those in the second
group lost an average of twenty-one pounds.
Research in the treatment of obesity has indicated that behavioral
approaches are generally effective in helping people lose weight.
However, outcomes vary across programs; that is, not all clients are
equally successful at losing weight during treatment or maintaining
weight losses over a follow-up period. This point is brought out in
the following studies.


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Seigfrd Fagerbergy
Associate Professor of
Health Education and Safety
This dissertation was submitted to the Graduate Faculty of the
Department of Psychology in the College of Liberal Arts and Sciences
and to the Graduate Council, and was accepted as partial fulfillment
of the requirements for the degree of Doctor of Philosophy.
August 1980
Dean, Graduate School


32
weight control manual, and were asked to take a baseline of their eating
and exercise behavior for the first week.
Week 2
Each subject weighed-in before the class period began. The first
forty-five minutes of this session was an interaction period in which
subjects discussed their reactions to taking a baseline (i.e., "In what
situations were you most likely to eat?"). The experimenter facilitated
discussion by reflecting the content and feelings being expressed.
During the next forty-five minutes of this session the experimenter
explained the food program to be used by subjects. An exchange system
was used where subjects selected from lists of six food groups (meat,
cereal, milk, fruits, vegetables, and miscellaneous). In addition, each
subject figured out her daily caloric intake limit in order to lose
one pound per week. Subjects were also asked to keep a graph of their
daily caloric intake, their daily exercise level, and their weekly
weight.
During the final part of the session subjects participated in an
exercise period. Exercise consisted of walking or running around the
Florida Field for six minutes. During the third week this time was
increased to nine minutes, and for weeks four through nine subjects
spent twelve minutes exercising.
Weeks 3-9
Standard Procedures
At the beginning of every class session each subject was privately


24
the greater is his potential for flexible adaptation. Ziller (1973)
supports this idea and states that
individuals with complex self-concepts may be
aware of or consider a greater number of
stimuli as being potentially associated with
the self. In terms of interpersonal perception,
the complex person has a higher probability
of matching some facet of the self with a
facet of the other person, since there are a
larger number of possible matches, (p. 79)
Hence, complex persons are less likely to be disturbed by new experi
ences or situations which appear to be incongruent with their self-
concepts.
The theories of Ziller (1973) and Horrocks and Jackson (1972),
taken together, suggest that role playing experiences facilitate role
taking behavior, and effect the complexity of the self-concept. Hence,
the individual perceives and incorporates new facets of the self; there
fore, he becomes more adaptive or less disturbed by change.
The value of role playing techniques in effecting role change has
been demonstrated by Lazarus (1966). He compared the effectiveness of
behavior rehearsal with two other techniques, direct advice and non
directive reflection-interpretation, in the management of specific
interpersonal problems. Seventy-five clients were randomly assigned to
one of the three procedures (twenty-five clients per subgroup) and a
maximum of four thirty-minute sessions was devoted to each treatment
condition. If there was no evidence of change or learning within one
month the treatment was regarded as having failed. The criterion of
change or learning was objective evidence that the client was behaving
adaptively in the area which had previously constituted a problem; e.g.
the socially awkward girl was going out on dates, the company executive


period. The scales used to measure these variables were the Tennessee
Self-Concept Scale and the Ziller Complexity of Self Scale.
Both groups had significant weight losses at posttreatment and
these losses were maintained after the four month follow-up period.
There were no significant differences in weight loss between the two
groups at posttreatment. However, at the follow-up, subjects in the
information" treatment group tended to maintain a greater weight loss
than subjects in the "role playing" treatment group.
At the follow-up, there were no significant differences in self-
concept, complexity, or physical self between the two groups. In
addition, there were no significant differences in self-concept,
complexity, or physical self between "successful" and "unsuccessful"
weight reducers at the follow-up. However, "successful" weight
reducers tended to have higher scores on the Ziller Complexity of Self
Scale at pretreatment than "unsuccessful" weight reducers. These
results are discussed with regard to ZiTier's Helical Theory of
Personal Change and future work in weight control.
v


53
Weight control programs can be greatly improved if they could
specify the type of client with whom they are more likely to be success
ful. In addition, treatment procedures may be enhanced if client
characteristics which hinder effective weight reduction efforts could be
delineated and then modified.
In the present study, pretreatment self-concept measurements
indicated that subjects presented a fairly positive view of themselves
(general self-concept); they maintained a poor opinion only of their
physical beings (physical self). Hence, they denied any dissatisfaction
with themselves except with their physical bodies.
Past research in obesity has revealed that obese individuals often
manifest other emotional and social problems (Held & Snow, 1972;
Querehi, 1972; Werkman & Greenberg, 1967; Wunderlich, Johnson, & Ball,
1973). This seemed to hold true of the subjects in the present study.
During the weekly meetings subjects disclosed conflicts regarding their
identities, social relationships, families, and other personal concerns.
However, research also indicates that obese persons may deny their
problems on a self-report questionnaire (Gottesfeld, 1962; Suczek, 1955;
Werkman & Greenbert, 1967). This may be the reason why subjects'
scores fell within the normal range on the total positive scale of the
Tennessee Self-Concept Scale. The only difficulty these subjects ad
mitted to was the obvious one; hence, their scores were below average on
the physical self subscale of the Tennessee Self-Concept Scale, but not
on the total positive scale. A projective instrument may have proven
to be a more valid indicator of subjects' general self-concepts.
At the follow-up no significant differences in self-concept were
found between the two treatment groups. Although total self-concept
scores may not have been accurate depictions of subjects' general level


5
behavior modification group. The five best reducers belonged to this
group as did the single least successful one, the only subject who
gained weight during treatment.
The Self-Concept
In order to explain the large variability of outcome in behavior
therapyInce (1972) suggests that the self-concept may be a moderator var
iable affecting outcome, but that it is largely ignored by behavioral
practitioners. The self-concept, according to Rogers (1951) consists of
an organized conceptual pattern of the "I" or "me" together with the
values attached to those concepts.
Ince (1972) cited two clients whom he treated using behavioral
techniques. In both cases, treatment of the overt symptoms by behavior
modification proved insufficient to effect change. Underlying each
client's difficulties was a poor self-concept which needed to be enhanced
in order for the treatment to be effective. Hence, the therapist began
verbally reinforcing each client for positive self-references with the
goal of increasing such verbalizations which would then become internal
ized and thus modify the self-concept. This approach met with success,
and suggested that the modification of an individual's self-concept can
be an important variable and one upon which success of therapy might
depend.
filler's (1973) helical theory of personal change supports
Ince's (1972) proposition that a change in the self-concept is a neces
sary and desired outcome of personal change procedures. This theory
suggests a hierarchy of potentially changeable characteristics which
include attitudes, values, behaviors, roles, and self-concepts. It


33
weighed on a balance scale. The first forty-five minutes of each class
session in both groups was an interaction period. Subjects spent the
time discussing difficulties and successes they had during the past
week while the experimenter facilitated the discussion by reflecting the
content and feelings being expressed by the subjects.
At the end of the interaction period each week the experimenter
explained a few of the behavioral techniques for subjects to practice.
Lesson plans followed the program developed by Moody and Schreiber
(1979a, 1979b). Techniques focussed on gradual habit improvement
through stimulus control, breaking or extending the chain of eating, and
positive reinforcement. Some of these techniques included:
1. Eat slower: Chew thoroughly and swallow before you pick up another
forkful. Take small bites and set a minimum time of twenty minutes for
which your meal must last. Enjoy your food; do not gobble it.
2. Use small dishes: Use smaller dishes and smaller food containers.
This will make small portions appear larger.
3. Use positive reinforcement: Silently praise yourself each time you
adhere to one of the guidelines. In addition, reward yourself points
each time you obey one of the principles, and when you accrue a certain
amount of points, administer a pre-selected reward.
4. Develop a routine schedule of eating: Eat three meals each day and
eat at the same time each day. Do not eat at other times! Do not eat
on impulse.
5. Never engage in any other activities while eating: Pay attention to
eating; do not watch T. V., read the newspaper, listen to the radio, or
engage in similar activities. Do not eat unless you are hungry.
6. Prepare only one portion of each course at a time: Take only one


APPENDICES


REFERENCES
Abramowitz, C. V., Abramowitz, S. I., Roback, H. B. & Jackson, C.
Differential effectiveness of directive and nondirective group
therapies as a function of client internal-external control.
Journal of Consulting and Clinical Psychology, 1974, 42, 849-853.
Abramson, E. E. A review of behavioral approaches to weight control.
Behavior Research and Therapy, 1973, Vl_, 547-556.
Buchanan, J. R. Five year psychoanalytic study of obesity. American
Journal of Psychoanalysis, 1973, ^33, 30-38.
Collingwood, T. R. & Willett, L. The effects of physical training upon
self-concept and body attitude. Journal of Clinical Psychology,
1971, 27, 411-412.
Fitts, W. H. Tennessee Self-Concept Scale Manual. Nashville, Tenn.:
Counselor Recordings and Tests, 1965.
Friedman, M. L. & Dies, R. R. Reactions of internal and external test-
anxious students to counseling and behavior therapies. Journal of
Consulting and Clinical Psychology, 1974, 42_, 921.
Glucksman, M. L. & Hirsch, J. The response of obese patients to weight
reduction: III. The perception of body size. Psychosomatic Medicine,
1969, 31_, 1-7.
Glucksman, M. L., Hirsch, J., McCully, R. S., Barron, B. A. & Knittle,
J. L. The response of obese patients to weight reduction. II. A
quantitative evaluation of behavior. Psychosomatic Medicine, 1968,
30, 359-373.
Gottesfeld, H. Body and self-cathexis of super-obese patients.
Journal of Psychosomatic Research, 1962, 6_, 177-183.
Harris, M. B. & Bruner, C. G. A comparison of a self-control and a
contract procedure for weight loss. Behavioral Research and Therapy,
1971,9,347-354.
Held, M. L. & Snow, D. L. MMPI, Internal-external control, and problem
check list scores of obese adolescent females. Journal of Clinical
Psychology, 1972, 28, 523-525.
69



PAGE 1

THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN OVERWEIGHT COLLEGE WOMEN By Fred M. Schreiber A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1980

PAGE 2

i ACKNOWLEDGEMENTS I would like to thank Dr. Harry Grater for the time he has contributed and the interest he has shown in seeing this study through to I its completion. His guidance and support throughout this entire study has been greatly appreciated. I would like to thank the following people for their interest in and contributions to this study: Dr. Afesa Bell-Nathaniel: for her suggestion and encouragement to study the self-concepts of overweight people. Dr. Sig Fagerberg: for his support in teaching weight control classes in the Department of Health Education and Safety. Dr. Linda Moody: for the opportunity to develop the weight control manuals used in this study. Dr. Paul Schauble: for his help in research methods and design, and his encouragement to examine various facets of the overweight person, Dr, Ted Landsman: for his personal support not only during this study, but throughout my entire graduate program. Special thanks to Dr. Al Kahn and Dr. Mike Omizo for contributions beyond the call of duty. They made the statistical portion of this study more manageable and understandable. Thanks to Dr. Bob Ziller for his stimulating ideas on the selfconcept and theory of personal change. n

PAGE 3

TABLE OF CONTENTS ACKNOWLEDGEMENTS ABSTRACT INTRODUCTION Problem METHOD Measures Page i1 iv 1 1 Behavioral Approaches to Weight Control T The Self-Concept 5 27 Subjects 27 29 Procedures 3] RESULTS 43 Hypotheses 43 Pretreatment to Follow-up Change Scores 49 Correlations Between Variables 50 DISCUSSION 52 APPENDIX A INSTRUMENTS AND FORMS 62 APPENDIX B COURSE OUTLINE gg APPENDIX C LETTER SENT TO SUBJECTS AT THE FOLLOW-UP 68 REFERENCES gg BIOGRAPHICAL SKETCH 73 ni

PAGE 4

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN OVERWEIGHT COLLEGE WOMEN By Fred M. Schreiber August 1980 Chairman: Harry A. Grater Department: Psychology Research in psychology has indicated that the self-concept may be a moderator variable in the maintenance of behavioral changes. In addition, previous research has suggested that role playing techniques may facilitate a change in the self-concept. This study investigated the effectiveness of role playing techniques in a behavior modification weight reduction group in order to help clients enhance their self-concepts and to determine the effect this has on clients' abilities to lose weight and maintain their losses over a follow-up period. Two eight-week weight reduction groups were conducted. The goal of each group was gradual weight loss through improving eating habits. In addition, one group included an emphasis on using role playing techniques to facilitate clients' abilities to change their self-concepts. The other group was presented with theories and information on weight control. General self-concept, physical self-concept, and complexity of self were measured at pretreatment and after a four month follow-up TV

PAGE 5

period. The scales used to measure these variables were the Tennessee Self-Concept Scale and the Ziller Complexity of Self Scale. Both groups had significant weight losses at posttreatment and these losses were maintained after the four month follow-up period. There were no significant differences in weight loss between the two groups at posttreatment. However, at the follow-up, subjects in the i| "information" treatment group tended to maintain a greater weight loss than subjects in the "role playing" treatment group. At the follow-up, there were no significant differences in selfconcept, complexity, or physical self between the two groups. In addition, there were no significant differences in self-concept, complexity, or physical self between "successful" and "unsuccessful" weight reducers at the follow-up. However, "successful" weight reducers tended to have higher scores on the Ziller Complexity of Self i Scale at pretreatment than "unsuccessful" weight reducers. These results are discussed with regard to Ziller's Helical Theory of Personal Change and future work in weight control. / 7/ C'l^-A-/-'^-/ y ^/v^ov^ Chairman

PAGE 6

INTRODUCTION Problem The problem of obesity has reached such a high magnitude that the U.S. Public Health Service has classified it as "one of the most prevalent health problems in the United States today" (p. 547, Abramson, 1973) Stuart and Davis (1972) estimate that there are currently between 40 and 80 million obese individuals in this country alone. Traditional treatments for the problem have included medication, psychotherapy, and therapeutic starvation. However, these treatments have generally been unsuccessful, and Stunkard (1958) has concluded that Most obese persons will not remain in treatment. Of those that remain in treatment, most will not lose weight, and of those who do lose weight, most will regain it. (p. 79) Behavioral Approaches to Weight Control Recently, some successes have been reported using a behavioral approach to weight reduction. This approach views eating habits as learned behaviors and sees the overweight person as someone who has learned inappropriate patterns of eating. Behavioral approaches to weight reduction focus on helping the overweight person become aware of his eating patterns and helping him change inappropriate or problematic behaviors. It is presumed that by improving one's eating habits it will be easier to reduce one's caloric intake, and consequently, lose weight. In addition, by changing one's habits permanently, weight loss should also be permanent. 1

PAGE 7

Techniques used in behavioral approaches to weight reduction are derived from both operant and classical conditioning. Operant conditioning focusses on the antecedents and consequences which control a behavior. When applied to weight control, the overweight person is taught to bring his eating under appropriate situational cues and to reinforce improvements in his eating patterns. Classical conditioning involves the pairing of two stimuli so that eventually they bring about the same response in an organism. The overweight person has paired eating with a number of different external cues so that eventually the external cues alone elicit a response of "feeling hungry." The overweight person needs to disassociate eating from other activities in order to extinguish eating as a conditioned response to other activities. Stuart (1967) utilized operant and respondent conditioning techniques to help clients gain control over their eating behavior. These techniques included controlling the antecedent and consequent conditions of eating as well as record keeping and exercise. In addition, reinforcement was provided in the following three ways: (1) through the clients' experience of success in self-control; (2) through the reduction of the aversive consequences caused by a lack of self-control; and (3) through considerable reassurance by the therapist. Stuart treated eight women on an individual basis. Weight loss over a twelve month period for the eight women ranged from twenty-six to forty-seven pounds. Wollersheim (1970) compared the effectiveness of three group treatments (behavior self-control, positive expectation— social pressure, and non-specific therapy) with a no-treatment control group. The behavioral self-control group utilized operant conditioning techniques

PAGE 8

to learn appropriate eating behaviors in order to lose weight. In addition, they were trained in relaxation in order to counter tension in situations which would ordinarily result in eating. The positive expectation—social pressure group utilized group pressure to help subjects lose weight. The non-specific therapy group helped subjects understand their underlying motives for being overweight. Wollersheim reported that all three treatment groups lost significantly more weight than the control group after a twelve week treatment program. Furthermore, after an eight week follow-up subjects in the behavioral self-control group showed a significantly greater weight loss (pretreatment to follow-up) than either of the other two treatment groups. Stuart (1971) treated six overweight women on an individual basis in a fifteen week program which stressed environmental control of overeating, nutritional planning, and regulated increase in energy expenditure. Half the women started treatment immediately while the other group started treatment fifteen weeks later. Approximately six months following termination of treatment for the first group and three months following termination of treatment for the second group, followup data were collected. Results indicated that subjects in the first group lost an average of thirtyfive pounds while those in the second group lost an average of twenty-one pounds. Research in the treatment of obesity has indicated that behavioral approaches are generally effective in helping people lose weight. However, outcomes vary across programs; that is, not all clients are equally successful at losing weight during treatment or maintaining weight losses over a follow-up period. This point is brought out in the following studies.

PAGE 9

4 Jeffrey, Christensen, and Katz (1975) reported the results of four subjects whom they had treated for obesity using a behavioral approach. After six months of treatment weight losses ranged from twenty to thirtytwo pounds. However, after a six month follow-up period weight losses (pretreatment to follow-up) ranged from three to thirtytwo pounds; two subjects had maintained their weight loss while the other two subjects gained back part or most of their weight. Harris and Bruner (1971) reported two studies examining the effectiveness of a self-control behavioral program and a contract procedure for weight reduction. In the first study self-control procedures included positive reinforcement, stimulus control, and breaking or lengthening the chain of eating while the contract program consisted of subjects receiving a monetary reward for each pound lost. Both groups lost a significant portion of body weight after two months of treatment. Furthermore, the contract group had lost a larger proportion of initial weight than the self-control group. However, after a ten month follow-up period there were no significant weight losses for either group. In a second study, Harris and Bruner (1971) compared a self-control weight group with a no-treatment control group and found no significant weight losses for either group after sixteen weeks. Penick, Filion, Fox, and Stunkard (1971) compared a behavior modification weight control gorup with a control group which received supportive psychotherapy as well as instruction about dieting and nutrition. After three months of treatment the median weight loss for the behavior modification group was greater than that of the control group, but this difference did not reach significance. Penick et al attribute this result to the significantly larger variability of outcome in the

PAGE 10

behavior modification group. The five best reducers belonged to this group as did the single least successful one, the only subject who gained weight during treatment. The Self-Concept In order to explain the large variability of outcome in behavior therapy, Ince (1972) suggests that the self-concept may be a moderator variable affecting outcome, but that it is largely ignored by behavioral practitioners. The self-concept, according to Rogers (1951) consists of an organized conceptual pattern of the "I" or "me" together with the values attached to those concepts. Ince (1972) cited two clients whom he treated using behavioral techniques. In both cases, treatment of the overt symptoms by behavior modification proved insufficient to effect change. Underlying each client's difficulties was a poor self-concept which needed to be enhanced in order for the treatment to be effective. Hence, the therapist began verbally reinforcing each client for positive self-references with the goal of increasing such verbalizations which would then become internalized and thus modify the self-concept. This approach met with success, and suggested that the modification of an individual's self-concept can be an important variable and one upon which success of therapy might depend. filler's (1973) helical theory of personal change supports Ince's (1972) proposition that a change in the self-concept is a necessary and desired outcome of personal change procedures. This theory suggests a hierarchy of potentially changeable characteristics which include attitudes, values, behaviors, roles, and self-concepts. It

PAGE 11

is assumed that these characteristics are ordered according to their ease of change, with attitudes being the least difficult to change while self-concepts are the most resistant to change. Ziller's helical theory proposes that the self-concept is the anchoring characteristic of the system, or the ultimate level of the personal change hierarchy. Those who would effect change in attitudes, values, behavior, or roles are necessarily concerned with a change in the self-concept, for it is assumed here that unless a change in the self-concept is achieved which is congruent with changes at lower levels in the hierarchy, the lower level change is likely to be reversed and the organism returned to the initial state of equilibrium, (p. 174) Self-Concept and Body-Cathexis Secord and Jourard (1953) investigated the relationship between an individual's self-concept and body-cathexis. By body-cathexis is meant the degree of feeling or satisfaction with the various parts or processes of the body. In order to appraise body-cathexis subjects were presented with a list of forty-six items, each describing a different part or function of the body, and were asked to indicate on a five point scale their degree of satisfaction with each item. In order to measure selfcathexis, individuals were asked to rate in a similar fashion fifty-five items believed to represent a sampling of the various conceptual aspects of the self. Subjects were seventy college males and fifty-six college females. Intercorrelations between body-cathexis and self-cathexis scores were .58 for men and .66 for women. Hence, individuals had a moderate tendency to cathect their body to the same degree and in the same

PAGE 12

direction that they cathected their self. In addition, it was found that females cathected their bodies, irrespective of direction, more highly than did males, in that they did not assign as many threes (have no particular feeling one way or the other) to body items. Secord and Jourard suggested that women would be more likely than men to develop anxiety concerning their bodies because of the social importance of the female body. Rosen and Ross (1968) noted that in the correlations obtained by Secord and Jourard (1953) between body-cathexis and self-concept, they did not take into account that certain parts or processes may be more important to an individual than other parts or processes in evaluating his bodyand self-concepts. Hence, Rosen and Ross (1968) investigated the relationship between body image and self-concept, taking into account the relative subjective importance of the aspects being rated. Eighty-two undergraduates were presented with a list of twentyfour body parts and seventeen adjectives from the Adjective Check List. They were asked to indicate on a five-point scale for each body part or adjective how satisfied they were with that aspect and how important that aspect was to them. Correlations of subjects' mean satisfaction scores between body image and self-concept were; r = .52 for all items; r = .62 for all items above mean importance; and _r = .28 for items below mean importance. Hence, these findings support those of Secord and Jourard (1953), and in addition, they indicate that the relationship between body-cathexis and self-concept can be refined if the subjective importance of each component is considered. Lerner, Karabenick, and Stuart (1973) asked 118 male and 190 female college students to rate twenty-four body characteristics in terms of

PAGE 13

8 how satisfied they were with each characteristic of their own body and how important each part was in determining their own physical attractiveness. In addition, subjects responded to a short self-concept scale consisting of sixteen bipolar dimensions derived from the Adjective Check List. Results indicated that the degree of positive self-concept increased with the degree of positive attitude toward one's body characteristics (males: r = .33; females: r= .43). However, weighting satisfaction ratings by corresponding importance ratings did not significantly increase the satisfaction/self-concept relation (males: r = .33; females: r = .44). Hence, while this study supports the general relationship between self-concept and body-cathexis put forth by Secord and Jourard (1953), it failed to support the results of Rosen and Ross (1968) that importance ratings strengthened this relationship. Mahoney (1974) replicated the Lerner et al findings also using college students as subjects and self-report attitude measures. He found the correlation between mean unweighted (by importance) body-cathexis and self-esteem to be .45 for males and .37 for females. For weighted body-cathexis the respective correlations for males and females were .37 and .41. Further support for the relationship between body attitude and self-esteem has been provided by Weinberg (1960), Zion (1965), and Mahoney and Finch (1976). In each study, moderate correlations were found using college students and self-report instruments. The use of an identical scaling procedure to measure self-esteem and body attitude raises the question that the high correlation found between bodycathexis and self-esteem may represent method variance and not an intertrait relationship. The following study addresses this issue.

PAGE 14

Kurtz (1971) tested the hypothesis that body attitudes are related to feelings of self-esteem using two instruments which share very little method similarity. He used a self-report body attitude scale (methodologically similar to that used by Secord and Jourard, 1953) and the Ziller (1969) Self-Esteem Scale, which is a more indirect or projective technique for measuring a person's feelings toward himself. For the group of college students sampled, the hypothesis that a positive evaluative body attitude and a positive sense of self-esteem are related was confirmed. Of importance was the strength of this relationship considering that the scales used shared little methodological similarities. The relationship between body-cathexis and self-concept sheds some light on the self-concept of overweight individuals. It is assumed that people who volunteer to participate in weight reduction programs are dissatisfied with their bodies. Since people tend to cathect their bodies to the same degree that they cathect themselves, overweight individuals in weight control groups are probably somewhat dissatisfied with themselves. Hence, the preceding research on the relationship between body-cathexis and self-concept implies that obese individuals in weight reduction groups may be partly characterized by feelings of low self-esteem. Obese Self-Concept and Body-Cathexis Werkman and Greenberg (1967) compared eighty-eight obese adolescent girls at a medically oriented camp for overweight girls with forty-two normal -weight girls at an ordinary summer camp on a number of personality and interest measures (MMPI, Strong Vocational Interest Blank, Semantic Differential, and Sentence Completion Blank). One finding which cut

PAGE 15

10 across all the test results was the presence of a response set. The obese were defensive about revealing any psychological problems and, in short, presented themselves as "hypernormal ." In addition, there was a great difference between the obese and the control group on the social introversion scale of the MMPI, suggesting that the obese girls were more uncomfortable and anxious in social situations. Vocational interest patterns of the two groups also differed. While the control group identified with professions in which imagination, ambitiousness, creativity, and intellectual strivings were paramount, the obese girls' interests were more consistent with persons in nurturant professions. The obese showed a kind of "maturity"; that is, their interests were similar to those people whose occupations are stable and "realistic." These results suggest that the obese tend to restrict themselves both socially and vocationally. They appear to live within a conventional life pattern in which one does not attempt situations which might provoke anxiety. In addition, through their efforts to appear normal they may be sacrificing spontaneity and flexibility in many spheres of psychological functioning. This pattern may hamper them by preventing the full development of character through conflict and acceptance of challenges. Held and Snow (1972) studied twenty-three obese adolescent girls and twenty-three non-obese adolescent girls who were randomly selected from patients being seen for medical reasons at an out-patient clinic. Subjects were individually administered the MMPI, Mooney Problem Check List, and the Rotter Internal -External Locus of Control Scale. The obese group scored significantly higher on five of the ten clinical scales of

PAGE 16

n the MMPI than the non-obese group. These scales were D (depression), Pd (psychopathic deviate). Pa (paranoia), Pt (psychasthenia) and Sc (schizophrenia). The generally elevated profile of the obese group, particularly the Pd and Sc scales, suggested that many had serious psychological disturbances characterized by feelings of depression, alienation, and low self-worth. In addition, they tended to be non-conforming, had problems in impulse control, and were distrustful of others. Held and Snow failed to find the obese group defensive about revealing psychological problems in that they admitted to significantly more difficulties on the Mooney Problem Check List than the non-obese group. These findings suggested that obese adolescent girls may need help reconciling some of the feelings they have toward themselves and other people in addition to programs designed to help them lose weight. Wunderlich, Johnson, and Ball (1973) administered the Adjective Check List (ACL) and the Edwards Personal Preference Schedule (EPPS) to sixteen subjects ranging from 64.6% to 214.7% overweight. The obese group endorsed a significantly fewer number of favorable adjectives on the ACL than did the normative sample. In addition, the obese were characterized as having a low need for achievement, poor personal adjustment, and few preferences for close attachments to other people. They did not describe themselves as being dependable, cautious, stable, or inhibited. Instead, they described themselves as hasty, pleasure-seeking, headstrong, and rebellious. Order for them was low, as they described themselves as being careless, changeable, and disorderly, as well as having poor self-control. The obese scored low on endurance, suggesting that they were less willing to expend energy toward a goal or persist at a task. Hence, the

PAGE 17

12 obese tended to prefer sedentary activities or an inactive, passive lifestyle. The obese also scored low on dominance, indicating that they are seldom ambitious, determined, or assertive. They also characterized themselves as being manipulative, hostile, and aggressive. It seemed, therefore, that they would be likely to react to unexpected demands with sudden withdrawal and underlying hostility. Finally, the obese were very self-conscious about their size and expressed strong sexual conflicts and frustrations. This seemed to be produced by the desire to enjoy heterosexual contacts but the inability to engage in these behaviors due to the attitudes of themselves and others concerning their bodies. Quereshi (1972) studied 180 female members of TOPS (Take Off Pounds Sensibly) who had considerable difficulty with weight reduction, and on the average weighed over 200 pounds. He compared them with ninety-eight females who had been through the TOPS program and were successful over a six month period at staying within 5% of their ideal weight. These people were know as KOPS (Keep Off Pounds Sensibly). All subjects were administered the Michil Adjective Rating Scale (MARS) which consisted of forty-eight adjectives such as "nervous," "talkative," and "ambitious," and was accompanied by a five-point scale ranging from "very atypical" to "yery typical." This scale yielded four personality factors, which were labeled as unhappiness (Factor 1), extraversion (Factor 2), selfassertiveness (Factor 3), and productive-persistence (Factor 4). The mean for TOPS' selfratings on Factor 1, unhappiness, was significantly larger than that of KOPS, which indicated that TOPS perceived themselves as generally unhappy, nervous, tense, and dissatisfied. TOPS also had a significantly larger mean than KOPS on Factor 2, extra-

PAGE 18

13 version, which suggested that the obese considered themselves to be more outgoing and friendly than KOPS. However, Quereshi noted that obese persons' perceptions of themselves as extraverts did not necessarily mean that they were since previous research (Mayer & Thomas, 1967, cited in Quereshi, 1972) utilizing projective techniques with obese females evidenced traits such as passivity and withdrawal accompanied by feelings of rejection. A more realistic explanation may be gleaned from considering the high self-ratings on Factors 1 and 2 together. This suggests that obese females, despite their attempts to gain approval from others by means of friendliness and congeniality, perceive themselves as lonely and rejected. This explanation would also support the findings of Werkman and Greenberg (1967) and Held and Snow (1972). Gottesfeld (1962) compared self-drawings of thirty super-obese subjects with thirty neurotics" self-drawings in terms of each group's body-cathexis, or degree of satisfaction with the parts and processes of their bodies. The super-obese subjects showed a more negative bodycathexis than the neurotic group on the following three criteria; (1) the super-obese were judged as having more negative body-cathexis by a group of clinicians; (2) their drawings had more major parts of the body missing; and (3) their drawings were less differentiated. Gottesfeld (1962) also gave a list of twenty-eight personal traits to the same group of super-obese individuals and neurotics and asked them to rate themselves first as how they are and second as how they would like to be. The discrepancy between self and ideal self served as a measure of the degree of satisfaction with the self.

PAGE 19

14 Results indicated that the super-obese reported to be more satisfied with themselves than the neurotics. Later, however, while the superobese subjects were hospitalized for a two week evaluation, three independent judges observed their interactions with other patients and staff, and rated them on the same list of twenty-eight personal traits. The differences between self-ratings and observer (objective) ratings were significantly greater than the differences between self-ratings and ideal ratings. Gottesfeld concluded that the super-obese patients seemed to present a facade of satisfaction. They denied that they were dissatisfied with themselves on a self-report trait list, but they could not as easily guard against a negative self-picture in their drawings (projective test) or in observers' ratings. Stunkard and Mendel son (1967) have found very low body-concepts in some overweight people. These people felt that their bodies were grotesque and loathesome and that others viewed them with hostility and contempt. This feeling was associated with self-consciousness and impaired social functioning. In addition, the person took a very narrow view of himself, expecially during times of misfortune or unhappiness. All the unpleasant aspects of his life became focussed on his obesity; that is, his body became the explanation and the symbol of his unhappiness, Buchanan (1973) stated that the body was the receptacle for selfhate for obese individual; when they felt self-hate they complained they felt fat. Negative body-concepts have been found to be most prevalent among persons who became obese during childhood or adolescence (Stunkard and Mendelson, 1967; Stunkard and Burt, 1967). In addition, this condition seemed to persist despite weight reduction and prolonged maintenance of normal body weight (Stunkard and Burt, 1967).

PAGE 20

15 The research review thus far suggests that there are many psychological and social problems associated with obesity. The obese have problems with impulse control and tend to become easily angered, irritated, or resentful. They also describe themselves as disorderly, pleasure-seeking, and rebellious. In addition, they are seldom active or competitive and hardly expend a great deal of effort to accomplish a difficult task. They sometimes take a very unrealistic view of themselves and tend to blame many of their difficulties on their obesity. Obese adolescents and young adults seem to have a poor self-concept which is reflected in their dissatisfaction with their bodies as well as with themselves. They harbor deep feelings of insecurity and have exaggerated needs for attention and social approval. The world is seen as a threatening and rejecting place and their response is to withdraw as a defense against being hurt. As a result, they often feel lonely and alienated, distrust other people, and avoid close interpersonal relationships. These obese persons are self-conscious about their size and seem to feel uncomfortable and anxious in a variety of situations. Consequently, they tend to restrict their interests both socially and vocationally and they may be deficient in basic social skills. In addition, these individuals seem to have serious concerns about their masculinity or feminity, feeling afraid that they cannot perform adequately in sexual situations. Weight Reduction and Self-Concept Glucksman and Hirsch (1969) compared obese subjects' performance on a body size estimation task before, during, and after weight loss with that of a normal control group. Six obese subjects with a mean initial

PAGE 21

16 weight of 334 pounds were hospitalized for eight months and lost an average of 86.7 pounds. Using an adjustable distorting image apparatus, subjects were requested to make a distorted screen image of themselves correspond to their body size as they perceived it at that moment. Obese subjects increasingly overestimated their own body size from pretreatment, during fifteen weeks of weight loss, and after a six week maintenance period. In effect, despite their weight loss, they perceived themselves as if they had lost almost no weight. In addition, three of the obese subjects in this study were retested after an additional year of weight loss, and they continued to overestimate their actual body size. This "phantom body size" phenomenon was accompanied by supportive clinical and figure-drawing data. For example, these same subjects drew progressively larger figure-drawings during weight loss and at the end of the final weight maintenance period. In addition, they reported that they continued to feel obese despite weight loss (Glucksman, Hirsch, McCully, Barron, and Knittle, 1968). Suczek (1955) tried to delineate psychological aspects of weight reduction. He administered the Interpersonal Dimension of Personality System (IDPS) to 100 obese women before and after a sixteen week treatment program. This instrument analyzes behavior in terms of five discrete levels: I, the level of public communication; II, the level of conscious description; III, the level of private symbol izati on, IV, the level of unexpressed unconscious; and V, the level of ego ideal. In this study, only data from levels I and II (i.e., overt, facade behavior) were described. There was little variability or inconsistency between level I and level II data at pretreatment. In other words, obese women saw little

PAGE 22

17 difference between their views of themselves as they impress others through overt behavior and their conscious views of themselves (i.e., between how they are and how they claim to be). In essence, they presented themselves as conflict-free. Subjects were assigned to groups of ten to fifteen individuals, which met once a week for sixteen weeks. Half the groups were led by psychiatric social workers whose aim was to promote change toward self and others while the other groups were led by dieticians and nutritionists whose aim was to guide spontaneous discussion and provide dietary information. Subjects in both types of groups were relatively successful in losing weight. However, there was little or no change at posttreatment on the IDPS. Thus, at the level of overt behavior and their conceptions of themselves, the obese women resisted change through their experiences in the group--their personalities did not change. Suczek later compared the obese group with a sample of neurotics undergoing group therapy and found that neurotics did undergo personality changes. Hence, changes in behavior and in self-conception were achievable by means of group methods and measurable by the IDPS. However, the obese women were not readily amenable to change and did not change their attitudes about themselves to any appreciable degree in the weight reduction groups. Unfortunately, no long term follow-up results were reported to determine if weight losses were maintained. Collingwood and Willett (1971) investigated the effects of physical training on self-attitude changes of five male teenagers enrolled in a special YMCA obese physical training program. The program consisted of ten hours of exercise per week for three weeks, and a total of three hours of group counseling-discussion. Two attitude scales were

PAGE 23

18 administered at pretreatment and again at posttreatment. A Body Attitude Scale, containing fifteen body concepts, measured subjects' attitudes toward their bodies along the following three dimensions: (1) Evaluative dimension with good-bad, awkward-graceful, and beautiful-ugly bipolar adjectives; (2) Potency dimension with weak-strong, hard-soft, and thinthick bipolar adjectives; and (3) Activity dimension with activepassive, cold-hot, and fast-slow bipolar adjectives. In addition. Bills' Index of Adjustment and Values (lAV) containing twentyfour adjectives, measured subjects' attitudes toward themselves along the following three dimensions: (1) Self-Concept ("seldom like me" to "most of the time like me"); (2) Self-Acceptance ("I very much dislike being as I am in this respect" to "T verymuch like being as I am in this respect"); and (3) Ideal Self ("seldom would I like to be that way" to "most of the time I would like to be that way"). Results at posttreatment indicated a significant weight decrease and a significant increase on both the evaluative and potency dimensions of the Body Attitude Scale. In addition, on the lAV there were significant increases on the Self-Concept and Self-Acceptance dimensions, and a significant decrease on the discrepancy between Self-Concept and Ideal Self. These results indicated that physical training experiences can help teenagers lose weight and enhance their attitudes toward themselves. However, long-term follow-ups are needed to determine the stability or permanency of these changes. Rohrbacher (1973) studied 204 overweight boys between the ages of eight and eighteen in an eight week weight reduction camp program. Body image and self-concept assessments (Secord and Jourard, 1953, scales) were made before and after the camp program, and sixteen weeks

PAGE 24

19 after the camp program had ended and subjects had returned home. Significant weight losses were recorded at posttreatment and at the followup, although subjects did regain some weight during the follow-up period. In addition, body image showed a significant positive change as a result of the camp program, but self-concept remained unchanged. Rohrbacher suggested that changing body image may be an important factor related to the program's success, but that longer follow-ups are needed to accurately assess the program's effectiveness. Weller, Arad, and Levit (1977) compared the self-concepts of twenty-five women who had successfully reduced their weight with those of twenty-five women who were unsuccessful at weight reduction. On the Tennessee Self-Concept Scale, successful dieters revealed a better self-image (general self-concept score) and a better physical self-image (physical self score) than did the unsuccessful dieters. These results suggested that those individuals who were most successful at weight reduction had also modified their self-concepts in the direction of feeling better about themselves and their bodies. McCall (1973) studied MMPI profiles of two groups of women belonging to TOPS (Take Off Pounds Sensibly). One group had considerable difficulty losing weight and/or maintaining whatever weight loss they had achieved. Such refractorily or irremediably obese were designiated R-TOPS. The other group had successfully lost weight and had maintained their ideal body weight for at least six months. These women were referred to as KOPS (Keep Off Pounds Sensibly). On nine of the ten MMPI clinical scales, R-TOPS women were significantly more deviant than KOPS women. On six of the nine scales, the differences were significant at the .01 level; on three scales the level

PAGE 25

20 of significance was .05. From the six scales on which the relative differences were greatest R-TOPS women exhibited more body concern (hypochondriasis), psychic "hurting" (depression), somatization (hysteria), rebelliousness (psychopathic deviate), compulsive and ruminative tendencies (psychasthenia), and bizarre or confused thinking (schizophrenia). McCall (1974) studied nineteen women TOPS club members with a mean weight of 204 pounds. He broke them down into three experimental subgroups as follows: One group was chosen because their MMPI profiles closely resembled those of successful weight reducers previously studied (KOPS); another group consisted of women whose MMPI profiles closely resembled those of the resistively obese (R-TOPS); and a third group fell in-between these two extreme groups. The question asked was whether the MMPI profiles that distinguished between these three subgroups would have any bearing on success in group therapy and weight reduction. Subjects were randomly assigned to one of three groups which met for sixteen weeks and were oriented toward the development of self-control. Preand posttherapy data showed that only the R-TOPS-like subgroup showed significant changes after therapy. They changed on the following six clinical scales: Hs (hypochondriasis), D (depression), Hy (hysteria), Pd (psychopathic deviate), Pt (psychasthenia), and Sc (schizophrenia). These six scales had previously most differentiated the refractorily obese (R-TOPS) from the remediated obese (KOPS). In addition, only the R-TOPS-like subgroup showed a significant weight loss at posttreatment. Hence, obese women who had the "worst" MMPI profiles tended to benefit most from group therapy as indicated both by weight loss and profile improvement.

PAGE 26

21 The research in weight reduction and concomitant changes in the self-concept suggests that the self-concept does not change as quickly as a person loses weight. Weight control programs which solely emphasize restricting your caloric intake seem to have little immediate effects on the self-concept. These programs tend to ignore many of the psychological or interpersonal problems associated with being overweight. Since they do not address themselves to all facets of the problem of obesity they appear to be incomplete. While clients may lose weight, they do not always change their attitudes toward themselves and will often manifest a "phantom body size." According to Ziller (1973): A change in the self-concept is the desired outcome of personal change procedures but the change processes must involve changes in attitudes, values, behaviors, and roles. ... If the desired changes in the selfother orientations have not been achieved, the client is likely to revert to earlier attitudes, behaviors, and roles, (p. 176) Hence, the relative stability of the obese person's self-concept may account for the poor long-term success of many weight reduction programs. The optimal strategy for personal change seems to be to change the self-concept. Physical training programs and some experiences in group therapy seemed to have some effect on the obese person's self-concept (Collingwood & Willett, 1971; Rohrbacher, 1973; McCall, 1974), while other group experiences yielded little or no self-concept change (Suczek, 1955). Role Playing and Self-Concept Ziller's (1973) helical theory of personal change indicates that changing a person's role may exert a strong influence on a person's

PAGE 27

22 self-concept; therefore, focussing on taking on new roles may be a valuable dimension in the treatment of obesity. In review, Ziller's theory suggests a hierarchy of potentially changeable characteristics which include attitudes, values, behaviors, roles, and self -concepts. It is assumed that these characteristics are ordered according to their ease of change, with attitudes being the least difficult to change while selfconcepts are the most resistant to change. If one of the components in the system is changed a state of disequilibrium within the system ensues. It is assumed that in this imbalanced state there is a tendency for the components lower in the hierarchy to change in a way which will render the components congruent. In addition, a change in one component will exert some press toward change in components higher in the system. A change in behavior, for example, will exert some press toward changes in attitudes, values, roles, and self-concepts, but with diminishing force with regard to the higher level components. Behavioral approaches to weight control are concerned with changing behaviors, and this will cause the system to be in a state of disequilibrium. However, a change in behavior will not exert a very strong press on the individual's self-concept. Hence, in order for the system to regain equilibrium the person is likely to revert back to old ways of behaving (i.e., old eating habits) which are congruent with the self-concept, rather than change the self-concept. Consequently, weight maintenance is unlikely. A change in a person's role will exert a stronger press on a person's self-concept than a change in behavior since roles are closer than behaviors to self-concepts on the personal change hierarchy. This

PAGE 28

23 suggests that weight control clients should be more likely to change their self-concepts, and therefore, be more likely to maintain weight losses if treatment foccusses on changing their roles in addition to changing their behaviors. Role playing techniques or behavioral rehearsal may facilitate role changes. This, in turn, may have a strong effect on a person's self-concept. Horrocks and Jackson (1972) explain this process. First, they distinguish between role taking and role playing behavior. Role taking is a concrete manifestation or implementation of a hypothesized identity and presents an observable product of the self-process. Role playing, however, represents performance by an individual of prescribed, demand behavior determined by a situational context. The role player behaves according to situational expectations, but the behavior does not represent anything the person believes himself to be in that context. Role playing behavior, therefore, is the manifestation by an individual of anti-identities, or those roles not conceptualized by an individual of himself at that moment. When an individual role plays anti-identities, locus of control of his behavior is external to him, imposed upon him by situational or behavioral demands. When the person receives feedback on his performance, he assimilates into his cognitive structures the effectiveness and appropriateness of these actions. In this manner, cognitive dissonance may I alter previous cognitions and change a conception of the self to include aspects of the anti-identity. As a result, new meanings of the self may be conceptualized, and role taking behavior replaces role playing. Horrocks and Jackson suggest that the greater the array of identities an individual incorporates into his conceptualization of himself

PAGE 29

24 the greater is his potential for flexible adaptation. Ziller (1973) supports this idea and states that individuals with complex self-concepts may be aware of or consider a greater number of stimuli as being potentially associated with the self. In terms of interpersonal perception, the complex person has a higher probability of matching some facet of the self with a facet of the other person, since there are a larger number of possible matches, (p. 79) Hence, complex persons are less likely to be disturbed by new experiences or situations which appear to be incongruent with their selfconcepts. The theories of Ziller (1973) and Horrocks and Jackson (1972), taken together, suggest that role playing experiences facilitate role taking behavior, and effect the complexity of the self-concept. Hence, the individual perceives and incorporates new facets of the self; therefore, he becomes more adaptive or less disturbed by change. The value of role playing techniques in effecting role change has been demonstrated by Lazarus (1966). He compared the effectiveness of behavior rehearsal with two other techniques, direct advice and nondirective reflection-interpretation, in the management of specific interpersonal problems. Seventy-five clients were randomly assigned to one of the three procedures (twenty-five clients per subgroup) and a maximum of four thirty-minute sessions was devoted to each treatment condition. If there was no evidence of change or learning within one month the treatment was regarded as having failed. The criterion of change or learning was objective evidence that the client was behaving adaptively in the area which had previously constituted a problem; e.g., the socially awkward girl was going out on dates, the company executive

PAGE 30

25 was effecting a promising merger, or the considerate husband had persuaded his wife to move out of her parents' house into a home of their own. Results indicated evidence of learning in eight (32%) of the clients treated by reflection-interpretation, eleven (44%) of the clients by advice, and twenty-three (92%) of the clinets treated by behavior rehearsal. In addition, of the thirty-one clients who did not benefit from reflection-interpretation or from advice, twenty-seven were then treated by behavior rehearsal and there was evidence of learning in twenty-two (81%) of them. Thus, the overall effectiveness of behavior rehearsal in fiftytwo cases was 86.5%. Hence, behavior rehearsal appeared to be significantly more effective in changing behavior outside the therapy session than direct advice or non-directive therapy. The literature review revealed that the self-concept of obese adolescents and young adults tends to be characterized by feelings of low self-esteem, a restricted view of themselves focussed mainly on their bodies or their size, and social isolation or social anxiety. Moreover, a change in this self-concept may be a factor facilitating long-term maintenance of weight loss. It appears that adapting new roles allows a person to see himself in a variety of different ways, or expands a person's view of himself, and makes him more adaptive to change. In addition, role playing techniques may facilitate role taking behavior, and may exert a strong press on changing a person's self-concept. Based on these premises, the present study was devised to test the efficacy of role playing techniques in a behavior modification weight reduction group in order to help clients modify their self-concepts, and to determine the effects this

PAGE 31

26 on clients' abilities to lose weight and maintain their losses over a follow-up period. In order to carry out this study, two weight reduction groups were conducted. Both groups emphasized behavioral techniques in order to lose weight. In addition, one group included an emphasis on using role playing techniques to facilitate clients' abilities to take on new roles and change their self-concepts. The other group was presented with theories and information on weight control. The following served as the hypotheses: 1. Both groups will show significant reductions in weight at posttreatment. 2. The "role playing" treatment group will show a significantly greater weight loss than the "information" treatment group at the follow-up. 3. The "role playing" treatment group will have a higher self-concept than the "information" treatment group at the follow-up. 4. The "role playing" treatment group will have a higher complexity of the self than the "information" treatment group at the follow-up. 5. The "role playing" treatment group will have a higher physical selfconcept than the "information" treatment group at the follow-up. 6. At the follow-up, "successful" weight reducers will have a higher self-concept than "unsuccessful" weight reducers.

PAGE 32

METHOD Subjects The subjects in this study were twenty-nine female college students enrolled in a weight reduction course taught through the Department of Health Education at the University of Florida. Two sections of the course were offered-one met Monday afternoon ("information" group) and one met Thursday afternoon ("role playing" group). Sixteen women enrolled in the Monday afternoon section ("information" group) and thirteen women enrolled in the Thursday afternoon section ("role playing" group). Subjects signed up for the section they preferred, mostly depending on their schedule of classes. Subjects were unaware of any difference between the two treatment groups at the time of registration. Pretreatment measurements for the two groups are summarized in Table 1. In order to determine if there were any differences between the two groups at pretreatment a two-group discriminant analysis was performed using all pretreatment variables. This analysis indicated that the "family self" subscale of the Tennessee Self-Concept Scale had the most discriminating power of all the variables entered (F_ = 4.79', df = 1,27, £ <.05). The analysis produced no additional significant discriminating variables. Since twenty-two variables were put into the analysis, it is likely that some differences may appear by chance and may not reflect actual differences between the two groups. In addition, "family self" appears 27

PAGE 33

28 Table 1 Pretreatment Measurements of "Role Playing" Group and "Information" Group Role Playing (n = 13) Information (n = 16) M SD M SD Initial Weight 176.23 44.49 Amount Overweight 45.62 37.41 Percent Overweight 34.15 26.04 Age 19.77 2.32 G.P.A. 2.74 0.42 Years Overweight 10.08 7.08 Class (l=freshman, 2=sophomore, 3= jr. 4=sr.) 2.08 0.86 Ziller Complexity of Self Scale 41.15 14.15 166.94 25.20 36.19 23.20 27.75 18.07 20,25 2.50 2.77 0.50 10.44 7.43 2.44 0.81 44.81 15.78 Tennessee Self-Concept Scale (raw scores converted to Std. T scores: M = 50; SD = 10) Self-Criticism 48.23 10.17 Total Positive 43.46 11.46 Identity 45.85 13.09 Self-Satisf action 44.69 8.53 Behavior 42.15 11.78 Physical Self 34.69 7.22 Moral -Ethical Self 45.31 12.05 Personal Self 44.77 11.74 Family Self 46.54 10.63 Social Self 51.31 14.03 Total Variability 54.08 7.65 Row Variability 53.92 8.64 Column Variability 52.62 6.70 Distribution 44.77 11.03 52,69 12.48 46.69 7.74 50.25 6.93 46.25 7,71 45.44 8.41 36.00 11.58 48.69 6.73 47.81 8.73 53.31 5.78 50.50 10.15 55.25 9.23 54.69 8.22 53.81 10.69 48.88 8.42

PAGE 34

29 to be unrelated to the focus of this study, and therefore, should have little effect on the outcome. Hence, it was assumed that there were no differences between the two groups at pretreatment. Students in this class were graded on a pass/fail basis depending solely on their record of attendance; those who attended eight out of nine sessions received a passing grade. Only female subjects were used because weight reduction programs attract mostly female clients. Subjects met a number of criteria for acceptance into the program: (1) a need to lose at least twenty pounds; (2) not currently enrolled in another weight reduction program; (3) not taking medication for weight; and (4) having no history of a metabloic or hormonal imbalance that could affect weight. Measures Tennessee Self-Concept Scale (TSCS)--Counselinq Form The TSCS (Fitts, 1965) was administered to all subjects at pretreatment and at the follow-up. The Scale is a self-administered paper and pencil test consisting of 100 self-descriptive statements which the subject uses to portray a picture of herself. For each item the respondent chooses one of five response options labeled from "completely false" to "completely true." Fourteen scores are derived from these items: self-criticism; nine self-esteem scores (identity, self-satisfaction, behavior, physical self, moral-ethical self, personal self, family self, social self, total); three variability of response scores [variation across the first three self-esteem scores, variation across the next five self-esteem scores, total variation); and a distribution score. Raw scores on the TSCS have been converted to Standard T-Scores; i.e., M = 50, SD = 10.

PAGE 35

30 The TSCS was normed on a sample of 626 persons of varying age, sex, race, socioeconomic status, intellectual level, and educational level. The standardization group was overrepresented in number of college students, white subjects, and persons in the twelve to thirty year age bracket. Test-retest reliability with sixty college students over a two week period ranged from .80 to .92 for nine self-esteem scales and ranged from .60 to .89 for the other five scales. Complexity of Self Scale (see Appendix A) The Ziller (1973) Complexity of the Self Scale was administered to all subjects at pretreatment and at the follow-up. This scale measures the degree of differentiation of the self-concept. A person with high complexity of the self-concept requires more words to describe herself. The Ziller Complexity of the Self Scale consists of a list of 109 adjectives and the task is to check the works that describe yourself. The score is the number of words checked. This scale has a range of zero to 109 and is scored in the direction of high complexity. In a study involving 100 randomly selected students from grades seven through twelve, the split-half reliability was .92. Test-retest reliability after one month for college sophomores was .72. Other Instruments A questionnaire asking for specific demographic information was given to all subjects at the initial meeting (see Appendix A). In addition, "role playing" subjects were asked to fill out a second questionnaire regarding how they thought weight reduction would affect

PAGE 36

31 their interpersonal relationships (see Appendix A). The information gathered from the second questionnaire was used to develop appropriate role training situations. Dependent Variable The dependent variable used was a standardized index called the weight reduction index (RI), which is equal to the percent of excess weight lost multiplied by the relative initial obesity. In addition to weight, this index takes into account height, amount overweight, and weight goal, and is expressed by the following formula: RI = (Wl/Ws) X (Wi/Wt) X 100 where Wl = weight lost, Ws = surplus weight, Wi = initial weight and Wt = target weight. Procedures The experimenter served as the moderator or therapist for each group. The emphasis in both groups was on gradual weight loss using self-control techniques designed to improve eating habits. Each group met for two hours, once a week for nine weeks. In addition, there was a four month follow-up. Week 1 At the initial meeting subjects were given a course outline (see Appendix B) and the experimenter explained the principles of the program. Subjects filled out the TSCS, Complexity of Self Scale, and Questionnaire #1. In addition, "role playing" subjects filled out Questionnaire #2. At the end of the initial meeting, subjects were weighed, were given a

PAGE 37

32 weight control manual, and were asked to take a baseline of their eating and exercise behavior for the first week. Week 2 Each subject weighed-in before the class period began. The first forty-five minutes of this session was an interaction period in which subjects discussed their reactions to taking a baseline (i.e., "In what situations were you most likely to eat?"). The experimenter facilitated discussion by reflecting the content and feelings being expressed. During the next forty-five minutes of this session the experimenter explained the food program to be used by subjects. An exchange system was used where subjects selected from lists of six food groups (meat, cereal, milk, fruits, vegetables, and miscellaneous). In addition, each subject figured out her daily caloric intake limit in order to lose one pound per week. Subjects were also asked to keep a graph of their daily caloric intake, their daily exercise level, and their weekly weight. During the final part of the session subjects participated in an exercise period. Exercise consisted of walking or running around the Florida Field for six minutes. During the third week this time was increased to nine minutes, and for weeks four through nine subjects spent twelve minutes exercising. Weeks 3 9 Standard Procedures At the beginning of eyery class session each subject was privately

PAGE 38

33 weighed on a balance scale. The first forty-five minutes of each class session in both groups was an interaction period. Subjects spent the time discussing difficulties and successes they had during the past week while the experimenter facilitated the discussion by reflecting the content and feelings being expressed by the subjects. At the end of the interaction period each week the experimenter explained a few of the behavioral techniques for subjects to practice. Lesson plans followed the program developed by Moody and Schreiber (1979a, 1979b). Techniques focussed on gradual habit improvement through stimulus control, breaking or extending the chain of eating, and positive reinforcement. Some of these techniques included: 1Eat slower : Chew thoroughly and swallow before you pick up another forkful. Take small bites and set a minimum time of twenty minutes for which your meal must last. Enjoy your food; do not gobble it. 2. Use small dishes : Use smaller dishes and smaller food containers. This will make small portions appear larger. 3. Use positive reinforcement : Silently praise yourself each time you adhere to one of the guidelines. In addition, reward yourself points each time you obey one of the principles, and when you accrue a certain amount of points, administer a pre-selected reward. 4. Develop a routine schedule of eating : Eat three meals each day and eat at the same time each day. Do not eat at other times I Do not eat on impulse. 5. Never engage in any other activities while eating : Pay attention to eating; do not watch T. V, read the newspaper, listen to the radio, or engage in similar activities. Do not eat unless you are hungry. 6. Prepare only one portion of each course at a time : Take only one

PAGE 39

34 helping, one course at a time. Take small portions. Do not feel that you must complete a meal, especially if you planned the meal when you were hungry. If you are not hungry, leave the remaining food on your plate. Be guided by your feelings of hunger, not by the amount of food on the table. 7. Do not buy prepared foods : Limit your diet to foods which must be prepared before eating. Eat low calorie foods or foods that are difficult to eat (e.g., eat five carrots or crackers rather than one piece of candy). Do not tempt yourself. 8. Eat only in one place in one room : Only eat while sitting at the table. Do not take snacks into other rooms; do not eat at your desk; don't eat standing up. Do not eat casually. 9. Eat before a party : If you plan to attend a social event at which there will be a great temptation to eat high calorie foods, eat a small, low calorie meal before you go. 10. Do not prepare food while you are hungry : If you tend to nibble while preparing meals, prepare the meal at a time when you are not hungry (e.g., prepare dinner soon after lunch). If you go grocery shopping, do that right after eating. 11. Develop a prepotent repertory : Think of times you have a hard time resisting food. Plan other activities you can be involved in during these times so that you won't be around food. ,The last part of each class session in both groups was spent exercising. Subjects were also encouraged to exercise during the week between classes and to note improvements in the number of laps they completed during the twelve minute walk/run period.

PAGE 40

35 Variable Procedures Role Playing Group During the second fortyfive minute period of the third week "role playing" subjects were introduced to role playing. The experimenter explained to the class that when people lose weight other people may begin reacting to them differently. Hence, as a thin person they may be faced with new situations which will require new roles. As a result, part of this class was used to help students practice these new roles. Week 3 Exercise 1 : Getting Acquainted Purpose a. To help group members get acquainted in a relatively non-threatening manner. b. To explore feelings generated by "becoming another person." c. To explore the dimensions of a brief encounter. d. To emphasize the need for careful, active listening as well as self-disclosure during conversation. Instructions 1. Group members are paired in dyads, and the facilitator instructs participants to "get to know your partner" for the next few minutes (5 to 10 minutes). Participants are instructed to listen to the "free information" or clues their partners give about themselves and to follow-up on this free information. In addition, participants are instructed to "self-disclose" or give information about themselves to their partners.

PAGE 41

36 2. After the interviewing phase, group members reassemble in the larger group. The facilitator indicates that they now have the responsibility of introducing their partner to the group. Each group member, in turn, is to introduce her partner by standing behind her and speaking in the first person, as if she were that partner. There should be no rechecking between partners during this phase. The individual who is being introduced should hold her comments for the discussion period. 3. After all the introductions have been made, the facilitator leads a discussion of the exercise, focussing on feelings generated and/or the issues inherent in the goals of the exercise. Exercise 2 : On Being Fat and Thin Purpose a. To examine stereotypical differences between fat and thin people. b. To get subjects thinking about their feelings concerning being fat and being thin. Instructions 1. Have the group generate a list of generalizations about fat and thtn people. Brainstorm in this manner for five to ten minutes, creating an arsenal of stereotypes. s. Discuss this exercise, focussing on subjects' feelings about the list. Homework: Think about new situations that you may encounter as a result of losing weight. What new roles will you become involved in as a thin person? Purpose a. To encourage students to expand their views of themselves, i.e..

PAGE 42

37 to increase their complexity of themselves. b. To have students think about situations they could role play. Week 4 Exercise 1 : Giving Feedback Purpose : To teach students how to give helpful, appropriate feedback. Instructions 1. The facilitator defines "feedback" to the group as: A way of helping another person to consider changing her behaviorIt is communication to a person (or group) which gives that person information about how she affects others. Feedback helps an individual keep her behavior on target and thus better achieve her goals. 2. Group members discuss various criteria for useful feedback. Exercise 2 : Role Playing Fat and Thin People Purpose a. To compare how it feels to be a fat person with how it feels to be a thin person. b. To rehearse how to act in different situations. c. To develop an awareness of oneself in relation to other people. d. To see oneself in a variety of different ways. e. To learn how to give appropriate, helpful feedback. Role Playing Instructions 1. Make up a situation that you would like to role play. Discuss the incident carefully with the group and how the different characters might feel in the situation. Discuss the main character's goal and the best way to achieve this goal considering the way in which other people are feeling. Next, have group members volunteer to act in the different

PAGE 43

38 roles. Now dramatize the situation for five to ten minutes, or until it comes to a natural conclusion. Do the same situation twice: first, as you would act as an overweight person, and second, as you would act as a person of normal weight. 2. After you have enacted the situation both ways (i.e., overweight and normal weight roles) discuss the situation and give each other feedback in the following order: (1) main character: hwo did you feel about acting overweight and about acting normal weight? (2) other person(s) in the exercise: how did you feel about yourself in relation to the main character? (3) non-participant observers: what did you observe was going on? During the feedback session other group members may role play alternate ways of responding to the situation. 3. Think of various ways of responding no matter how strange they seem. Contrast responses and see what "fits." 4. There are no right or wrong answers in this exercise, so don't be afraid of making mistakes. We are practicing and comparing ways of acting as overweight and normal weight people so feel free to try out any role you like. Weeks 5 to 9 During the second forty-five minute period of each week "role playing" subjects continued the role playing exercise. The following is a list, of various situations that were role played. 1. You see a friend whom you haven't seen in a long time. He remarks, "You look different. Have you been sick?" 2. You're home on vacation and your mother keeps "pushing" food on you. You'd love to taste some of the food, but you are already full.

PAGE 44

39 Your mother feels very hurt because you don't want to eat her food, 3. You're at a party with some of your girlfriends. A guy you've been attracted to for awhile notices you and strikes up a conversation with you. 4. You've lost some weight but haven't yet reached your goal. You are sitting on a bus next to a guy whom you've just met. After talking with him for awhile he remarks, "You have a beautuiful face! Do you realize that if you lost some weight you'd really be a knock-out:" 5. You're with a girlfriend on campus and meet a guy that she knows. He ignores you and speaks only to your girlfriend. Even after your girlfriend has introduced you and has made several references to you during the conversation he continues to address his attention solely to your girlfriend. 6. You are at a bar and strike up a conversation with a guy. He seems friendly but as time goes by he becomes overly aggressive. 7. A guy calls you up and wants to take you out to eat. You want to go out with him but have already eaten. 8. You are with a friend and you are both anxious about an exam. Your friend says, "Let's go get something to eat." 9. You've lost some weight and you're walking home across campus. You meet a guy that you know and he tells you now great you look and insists on taking you to the Arrendondo Room for lunch. 10. .You've lost weight and your roommate, who is a little overweight, begins making snide remarks about your body. You suspect she may be feeling a bit jealous of you.

PAGE 45

40 Information Group During the second forty-five minute period of each week the "information" treatment group was presented with various weight control theories. The following is an outline of the topics discussed each week. Week 3 Hypnotherapy: Dr. Sig Fagerberg A. Conscious vs. subconscious B. The power of suggestion to the subconscious C. Hypnotic state D. Suggestibility and self-hypnosis I 1. Daily use of suggestion 2. Simple, repetitive, positive imagery pertaining to the future E. Books 1. LeCron, Lesley. SelfHypnosis 2. Peale. Power of Positive Thinking Week 4 External Cue Sensitivity A. Perceived time B. Food cues C. Response cost D. Taste E. Emotional state F. Feedback and competing cues '^eek 5 Nutrition and Dieting: Stephanie Fredette, Graduate Student in Nutrition A. Basic four food groups \

PAGE 46

41 B. Definition of obesity C. Adipose cell theory D. Basal metabolic rate E. Additives and preservatives F. Pinch test Week 6 Surgical Methods of Weight Reduction: Dr. Joun Kuldeau, J. Hi 11 is Miller Health Center A. Interstinal bypass surgery I 1. Diagram and description of operation :| 2. Adverse physical side effects I 3. Benefits of operation B. Stapling the stomack shut i' 1. Diagram and description of operation I 2, Advantages of this procedure over intestinal bypass operation J C. Case reports of super-obese clients D. Mood changes during rapid weight reduction Week 7 Hereditary and environmental factors in obesity A. Family research B. Twin studies 1. Variability in weights in identical twins 2. Identical twins reared in same environment vs. identical twins reared in different environments 3. Identical twins vs. fraternal twins vs. non-twin siblings C. Adopted children research Week 8 TOPS (Take Off Pounds Sensibly): Marilyn Poss, Gainesville

PAGE 47

42 Chapter of TOPS A. History and background of TOPS B. TOPS basic program 1 No prescribed diet 2. Changing eating habits 3. Group therapy approach 4. Competition— honors and rewards C. KOPS (Keep Off Pounds Sensibly) Maintenance of weight loss D. Case reports of successful clients Week 9 Other Approaches to Weight Control A. Aversive conditioning B. Counting mouthfuls C. Fad diets and crash diets D. Scarsdale diet Follow-up During the week of July 9, 1979, all subjects were contacted by mail (see Appendix C). They were asked to report their current weight and to fill out the Tennessee Self-Concept Scale and the Ziller Complexity of Self Scale.

PAGE 48

RESULTS The sample at posttreatment consisted of 29 subjects; 13 were in the "role playing" treatment group and 16 were in the "information" treatment group. The attrition rate during the program was 0%. The follow-up sample consisted of 28 subjects; 12 were in the "role playing" group and 16 were in the "information" group. One subject from the "role playing" group had left the country and could not be contacted. Hypotheses Hypothesis 1 : Both groups will show significant reductions in weight at posttreatment. The mean reduction index at posttreatment was 20.00 for subjects in the "role playing" group and 14.06 for subjects in the "information" group. Additional analysis revealed that the mean reduction index at the follow-up was 18.58 for subjects in the "role playing" group and 38.19 for subjects in the "information" group. One-way t^ tests indicated that all four of these indices were significant at the .05 level or greater (Table 2). Hypothesis 1 was accepted. 43

PAGE 49

44 Table 2 Mean Reduction Indices for "Information" and "Role Playing" Groups Role Playing Information M SD df t M SD df t Posttreatment 20.00 24,91 11 -2.78** 14.06 17.51 15 -3.21** Follow-up 18.58 32.34 11 -1.99* 38.19 29.37 15 -5,20*** £ < ,05 ** £ < .01 *** £< .001 Hypothesis 2 : The "role playing" group will show a significantly greater weight loss than the "information" group at the follow-up. Hypothesis 3 : The "role playing" group will have a higher self-concept than the "information" group at the follow-up. Hypothesis 4 : The "role playing" group will have a higher complexity of the self than the "information" group at the follow-up. Hypothesis 5 : The "role playing" group will have a higher physical selfconcept than the "information" group at the follow-up. The following measures were used in order to test the above hypotheses: (1) the Total Positive subscale of the Tennessee Self-Concept Scale (TSCS) was used to assess self-concept; (2) the Ziller Complexity of Self Scale was used to assess complexity of the self; and (3) the Physical Self subscale of the TSCS was used to assess physical self-concept. Follow-up measurements for the two treatment groups are summarized in Table 3.

PAGE 50

45 Table 3 Means and Standard Deviations for "Role Playing" and "Information" Treatment Groups on all Measures at the Follow-up Role Playing Information (n = 12) (n = 16) M SD M SD Self-Concept 45.50 11.39 47.25 8.18 Complexity 39.42 15.40 42.69 17.53 Physical Self 37.67 8.60 38.81 9.92 Reduction Index 18.58 32.34 38.19 29.37 In order to test Hypotheses 2, 3, 4 and 5 Multivariate Analysis of Variance (MANOVA) and Discriminant Analysis techniques as discussed by Kerlinger and Pedhazur (1973) were used. MANOVA procedures to determine significant differences between the "role playing" and "information" groups on the four follow-up measures as a group (self-concept, complexity, physical self-concept, and reduction index) revealed an F ratio of .95, df = 4,24, ^ ^ -05. Univariate £ values, standardized discriminant coefficients (SDC), and £ values for SDC for the "role playing" and "information" treatment groups on all follow-up measures are presented in Table 4. Results from the discriminant analysis revealed that there were no significant differences between the treatment groups relative to self-concept, complexity, and physical self-concept as noted by the respective £ values of 1.23, 0.89, and 0.88. The reduction index approached

PAGE 51

46 significance in the opposite direction than predicted as indicated by an F value of 3.46 (^ < .10; df = 4,24). When each variable was analyzed controlling for the effects of all other variables, none of the follow-up measures proved to be significant discriminators. In other words, none of the follow-up measures were significantly different for the two groups (Table 4, £ values for SDC). Hypotheses 2, 3, 4, and 5 were rejected. Table 4 Univariate £ Values, Standardized Discriminant Coefficients (SDC), and F Values for SDC for the "Role Playing" and "Information" Treatment Groups on all Follow-up Measures Standardized Univariate Discriminant £ Values £ Values Coefficients (SDC) for SDC Self-Concept Complexity Physical Self Reduction Index 1.23 0.89 0.88 3.46* 0.29 0.19 0.07 0.80 0.05 0.05 0.01 0.30 *£ < .10 Hypothesis 6 : At the follow-up, those subjects who have maintained the greatest weight loss will have a higher self-concept than unsuccessful weight reducers. Subjects in both treatment groups were combined at the follow-up and divided into "successful" and "unsuccessful" weight reducers. A median split of the reduction indices was used to divide the two groups.

PAGE 52

47 A subject was labeled "successful" if her reduction index was greater than 25.00 and "unsuccessful" if her reduction index was less than 25.00. Pretreatment and follow-up measures of these two groups are summarized in Table 5. Table 5 Means and Standard Deviations of "Successful" and "Unsuccessful" Weight Reducers on Pretreatment and Follow-up Measures Successful Unsuccessful (n = 14) (n = 14) M SD M SD Pretreatment Self-Concept 46.43 9.99 44.14 9.67 Complexity 47.47 15.54 37.43 12.41 Physical Self 37.00 12.41 33.71 6.62 Follow-up Self-Concept 47.93 7.20 45.07 11.50 Complexity 46.93 16.93 35.64 14.33 Physical Self 40.07 8.00 36.57 10.30 Reduction Index 54.71 21.60 4.86 16.67 A two-group discriminant analysis of the "successful" and "unsuccessful" weight reducers revealed a canonical correlation of 0.42, corresponding to a chi-square of 4.80 (df = 2, £ <. .10). When each of the variables was analyzed controlling for the effects of all other

PAGE 53

48 variables pretreatment complexity and treatment group accounted for the most variance between the two groups, as indicated by the Standardized Discriminant Weights in Table 6. The variable self-concept failed to be a significant discriminator between the two groups (hypothesis 6 was rejected). When the type of treatment and each of the pretest and follow-up measures were analyzied independently pretreatment and posttreatment complexity scores approached a significant differency between the two groups (£ < .10; df = 1,26) as indicated by the univariate F values in Table 6. Table 6 Univariate F Values and Standardized Discriminant Weights for "Successful" and "Unsuccessful" Weight Reducers on Treatment Group and all Pretest and Follow-up Measures Standardized Discriminant Weights Univariate F Values Pretreatment Self-Concept 0.03 0.38 Physical Self 0.04 0.76 Complexity 0.74 3.64* Treatment Group 0.55 2.36 Follow-up Self-Concept 0.50 0.62 Physical Self -1.01 Complexity 0.32 3.62* Standardized Discriminant Weights greater than 0.50 are underlined £ < .10

PAGE 54

49 Pretreatment to FoT low-up Change Scores Paired t^ tests on mean pretreatment and follow-up measures for each group indicated no significant differences on any of the following variables: self-concept, complexity, and physical self (Table 7). Table 7 Comparison of Pretreatment and Follow-up Scores for "Role Playing" and "Information" Treatment Groups on all Measures Pretreatment Follow-up M SD M SD df "Role Playing" Group Self-Concept 43.46 11.46 45.50 11.39 11 -1.15 n.s, Complexity 41.15 14.15 39.42 15.40 11 0.58 n.s. Physical Self 34.69 7.22 37.67 8.60 11 -1.85 n.s, "Information Group" Self-Concept 46.69 7.74 47.25 8.18 15 -0.27 n.s. Complexity 44.81 15.78 42.69 17.53 15 1.00 n.s. Physical Self 36.00 11.58 38.81 9.92 15 -1.15 n.s.

PAGE 55

50 Correlations Between Variables Correlation coefficients were calculated between each of the following variables: initial weight, percent overweight, G.P.A., years overweight, pretreatment self-concept, pretreatment complexity, pretreatment physical self, treatment group, follow-up self-concept, follow-up complexity, follow-up physical self, and follow-up reduction index (Table 8).

PAGE 56

51 CO (0 in O) to i. (O O) O) 2 +J O) OQ CO o +J fO so C_3 o c 1— 1 Q. 3-TD 1 0) Ll! Qi 1 MI— O) CLOO 1 3 • 1/5 u. >> .c Q. a. a. 3 F • O Ll. o O. 30 1 • • bt/1 -l-J Q. S 3 +J O ss1— CD 1 4-!-> 1 — fO a % i--£: 1 a. Q. -p i+J Q. 0) ^ !^ O D. C_) 1 +-> C fO o CI) o i. 1 -(-> >41 0) I— S >o • D. e: +-> c -M 0) ^ o QO (O •t— (-> -(-> •r— 3 I — [ o I LO o 1 LO CM O O I CO o I o o I CM o o I o o o o I o o o CM CM o o I en un o CO CD O CO O ID O O O O O o CD OO p— CM o I o o CM r— CD en o o CO o n o o I o en en CO o I en o o I CO o CM CM o 1^ CM CO O I cn en I— o o .— o o O I— o o o I— o o o CM o o r— CD CO en cn en o o o o o o o 'CO o o CO o o o 00 CO o o CO CO o LO CO o o o o o o o O I— — o o o o •— o o LO o o O I— o o O I— o o > o CD 0) cu c +J (/I 2 O II i• Q. CO S~ CM cu CJ E >, II > o J= ^— r o Ul C-3 Q. — >> to CO OJ CD 0) Q-r— s~ ii. s_ i. Q. >Cu Q. Q. CD CyO o CJ Q. Z5 C/1 E I — r CO >> -o ci_ en CL CL 3 r! CO cu 4-> CO cu +-> cn QCL (O 11 (U > o <+o cu CO fO o cu J2 CJ) to O z. 3 Q. CjO

PAGE 57

DISCUSSION The findings of the present study indicate that a behavioral self-control approach to overeating produces significant weight losses during the treatment period. In addition, this study supports Jeffrey's (1974) findings that a self-control approach produces significant long-term weight losses. In the present study mean reduction indices for both treatment groups were significant after a four month follow-up period. At the follow-up, a comparison of the mean reduction indices between the two treatment groups approached significance in the opposite direction than predicted; that is, there was a trend toward greater weight loss for subjects in the "information" treatment group. A possible reason for this outcome may be related to the issue of client/ treatment compatibility. Several studies have indicated that outcome in psychotherapy is a function of the interaction between treatment parameters and client variables (Abramowitz et al., 1974; Friedman & Dies 1974; Kilman, Albert, & Sotile, 1973). In other words, clients differ in the type of treatment to which they best respond. Subjects in the "information" treatment group were exposed to a broader range of theories and treatment concerning obesity than subjects in the "role playing" treatment group. As a result, they may have had a greater opportunity to select an appropriate or compatible treatment approach. 52

PAGE 58

53 Weight control programs can be greatly improved if they could specify the type of client with whom they are more likely to be successful. In addition, treatment procedures may be enhanced if client characteristics which hinder effective weight reduction efforts could be delineated and then modified. In the present study, pretreatment self-concept measurements indicated that subjects presented a fairly positive view of themselves (general self-concept); they maintained a poor opinion only of their physical beings (physical self). Hence, they denied any dissatisfaction with themselves except with their physical bodies. Past research in obesity has revealed that obese individuals often manifest other emotional and social problems (Held & Snow, 1972; Querehi, 1972; Werkman & Greenberg, 1957; Wunderlich, Johnson, & Ball, 1973). This seemed to hold true of the subjects in the present study. During the weekly meetings subjects disclosed conflicts regarding their identities, social relationships, families, and other personal concerns. However, research also indicates that obese persons may deny their problems on a self-report questionnaire (Gottesfeld, 1962; Suczek, 1955; Werkman & Greenbert, 1967). This may be the reason why subjects' scores fell within the normal range on the total positive scale of the Tennessee Self-Concept Scale. The only difficulty these subjects admitted to was the obvious one; hence, their scores were below average on the physical self subscale of the Tennessee Self-Concept Scale, but not on the total positive scale. A projective instrument may have proven to be a more valid indicator of subjects' general self-concepts. At the follow-up no significant differences in self-concept were found between the two treatment groups. Although total self-concept scores may not have been accurate depictions of subjects' general level

PAGE 59

54 of self-esteem, the failure of role playing techniques to affect selfconcepts is probably a valid result. This is supported by other results which indicated that role playing techniques had no significant affect on those parts of the self-concept concerning subjects' complexity or physical selves. The physical self subscale of the Tennessee SelfConcept Scale dealt with concerns subjects were more open about and the Ziller Complexity of Self Scale is more indirect in measuring complexity. Hence, these scores are more likely to be valid, yet they too remained unaffected by role playing techniques, A possible explanation for the failure of role playing techniques to effectuate self-concept changes involves the relative commitment an individual has to each of the five components of Ziller's (1973) helical theory of personal change (attitudes, values, behaviors, roles, and self-concept). According to Roby (1960; cited in Ziller, 1973), "commitment refers to a reduction in further alternatives associated with a particular choice," (p. 151) The amount of commitment an individual has to each of the components of Ziller's system increases as one goes up the hierarchy. For example, having reported certain attitudes is less restricting than having accepted a certain role, such as group leader. Violation of the expectations associated with certain attitudes is more acceptable to the self and others than violation of role expectancies. The greatest commitment is associated with the self-concept; these constructs are associated with the stability, regularity, and consistency of personal behavior. Since attitudes require the least amount of commitment, they are subject to the most rapid change. A change in attitudes permits some experimentation within the personal system; little commitment is

PAGE 60

55 involved and the personal system may return to its previous state with relatively little difficulty. Increasing commitment is made as changes in the other components proceed. The hierarchy suggests an orderly progression up the scale. Attitude changes are tried first, then values, behaviors, roles, and finally the self-concept, in turn, under favorable circumstances (p. 154, Ziller, 1973) In the present study, the "role playing" treatment group was a social learning approach with an emphasis on practicing or rehearsing new social roles. However, subjects may not have been ready to commit themselves to role changes; more likely, they were at the stage of experimenting with new attitudes, values, and behaviors. There was evidence of this during the role playing sessions. During the discussions subjects often expressed their feelings on such issues as jealousy, intimacy, respect, and assertiveness. Hence, it seems that subjects were concerned with the lower components of Ziller's hierarchy. Learning new social roles may not become a need for clients until after they have changed their behavior. At this point they would have lost a significant amount of weight, would be confronted with new situations, and may feel a social role deficit. In the present study, the intervention of role playing may have been pre-mature and may have even served to arouse some anxiety about the prospect of losing weight. Future work in weight control might postpone the intervention of role playing until after clients are more committed to new attitudes, values, and behaviors. Then, it is more likely that role playing techniques would facilitate role changes and exert a strong press toward a change in the self-concept. A longer term treatment intervention would be required.

PAGE 61

56 Although role playing techniques did not have an immediate impact on subjects' reported self-concepts, these techniques may have facilitated subjects' own awareness of themselves. This additional selfawareness may eventually prove helpful in the process of losing weight. Previous research indicates that young overweight people are selfconscious about their size and seem to feel uncomfortable in interpersonal or social situations. Consequently, they may become socially withdrawn. Thus, while overweight people remain overweight because of poor eating habits, social problems may develop as well. However, these problems are often denied; hence, many overweight people may remain unaware of their feelings and needs. In the present study, role playing scenarios partly served to help subjects become aware of their feelings in interpersonal situations. In addition, the feedback sessions following role playing gave subjects an arena in which to share their feelings. Remaining overweight may become a defense against having to deal with personal concerns. Losing weight removes this defense. Some people may not want to deal with this issue and may resist losing weight. In this respect, losing weight is similar to any other type of personal change; i.e., it is often resisted. People may not be aware of their own defenses, and may not be aware of the consequences of losing them. In the present study, role playing techniques served to make subjects aware of personal issues. Role playing experiences brought out many anxiety provoking situations. As a result, these subjects became more aware of the challenges they might face while losing weight. These speculations would only hold true for overweight people with

PAGE 62

57 interpersonal anxiety. It is possible that only subjects who were not socially anxious were successful at losing weight. It was found that they tended to be more complex people. Since complexity is defined as the ability to match an aspect of oneself with an aspect of the environment, this may also imply more ease in social and interpersonal situations. It would be difficult to measure the relationship between social anxiety and success at weight reduction since social anxiety is difficult to measure; i.e., overweight people tend to deny personal problems. This line of reasoning implies that weight reduction groups may only work to the extent that feelings can be expressed and an awareness of personal problems can be developed. Groups that are restrictive in these areas will not help clients whose personal problems are entwined with their weight. If an awareness of personal concerns is fostered then success at weight control may depend upon the person's ability to deal with these concerns. The leader's responsibility would be to help a person mobilize or find appropriate support systems. An analysis of the self-concepts of "successful" and "unsuccessful" weight reducers revealed no significant differences between the two groups. Both "successful" and "unsuccessful" weight reducers presented a fairly positive total self-concept while their physical selfpicture remained poor. These results support previous research indicating that the physical self-concepts of obese persons do not change commensurate with weight loss (Glucksman & Hirsch, 1969; Glucksman et al 1968; Rohrbacher, 1973; Suczek, 1955). In addition, these results are compatible with Zi Tier's theory regarding change. According to this theory, people are more likely to change the self

PAGE 63

58 system starting with the lower components and gradually progressing to the higher components. The ultimate level of change is a change in the self-concept, and this needs to be achieved if lower level changes are to be maintained. Weight reduction reflects a change in behavior, the third component in Ziller's hierarchy. According to Ziller's theory, if weight loss (behavior change) is to be maintained it must be followed by changes in roles and self-concept. Since these additional changes take time, it follows that self-concept changes would lag behind weight loss. Hence, in the present study, follow-up measures were probably taken some time after "successful" subjects had changed their behavior, but before they changed their self-concepts. As a result, no selfconcept differences were found between "successful" and "unsuccessful" weight reducers at the time of the follow-up. A longer term follow-up might reveal one of the following results: (1) "Successful" weight reducers would eventually change their self-concepts; therefore, behavior changes would be retained and weight loss would be maintained, or (2) The self-concept of "successful" weight reducers would remain unchanged; therefore, behavior would return to its previous state and weight would be regained. A further analysis of "successful" and "unsuccessful" weight reducers revealed that those people who lost the most weight tended to be more complex people at pretreatment. This supports the theories of Horrocks and Jackson (1972) and Ziller (1973) who state that more complex people incorporate a greater array of facets or stimuli into their conceptualization of themself. Complex people are less likely to be disturbed by new experiences because there is a higher probability

PAGE 64

59 of matching some aspect of themself with an aspect of the situation. Hence, they have a greater potential for flexible adaptation. Ziller (1973) suggests that strategies of personal change directed toward the self-concept may be accomplished with greater facility if they begin with concern for complexity of the self-concept, in order to render the self system more adaptive, (pp. 155-156) Future work in weight control might investigate the usefulness of various techniques to help people expand their view of themselves and consequently have a greater potential for change. The present study suggests the need for longer term treatment of overweight clients. It is proposed that unless positive changes in the self-concept accompany changes in behavior following behavior modification procedures, the treatment should be continued until such changes are observed.

PAGE 65

APPENDICES

PAGE 66

APPENDIX A INSTRUMENTS AND FORMS

PAGE 67

Questionnaire #1 Name: Height: Address: Size of frame: small medium large Phone: (circle one) Where will you be during the week of July 9, 1979? Address: Phone: Are you taking medication for weight? Are you enrolled in another weight program now? Do you have a metabolic or hormonal imbalance which could affect your weight? What is your class? freshman, sophomore, junior, senior, grad. student (circle one) What is your age? What is your major area of study? What is your grade point average? For how may years have you been overweight? 62

PAGE 68

63 Questionnaire #2 1. When you lose weight, your physical appearance will change. Who do you think will notice? 2. How do you think people will react to you when you lose weight? 3. Do you think you will act differently as a thin person? If yes, please specify how you will act differentlv. 4. Do you expect to encounter new situations when you lose weight? If yes, please specify.

PAGE 69

64 Ziller Complexity of Self Scale INSTRUCTIONS: Here is a list of words. You are to read the words quickly and check each one that you think describes YOU. You may check as many or as few words as you like--but be HONEST. Don't check words that tell what kind of a person you should be Check words that tell what kind of a person you really are. l._ _able 29. delicate 57._ large 85. _serious 2._ _active 30.. delightful 58._ lazy 86. _sharp 3._ _afraid 31 ._ different 59._ little 87._ _silly 4._ _alone 32.. difficult 60._ lively 88._ _slow 5-_ _angry 33._ dirty 61 ._ lonely 89 ._ _smal 1 6-_ _anxious 34._ dull 62._ loud 90. smart 7._ _ashamed 35.. dumb 63._ lucky 91 •_ _soft 8._ _attractive 36.. eager 64._ mild 92._ _special 9-_ _bad 37.. fair 65._ miserable 93._ _strange 10._ _beautiful 38.. faithful 66 ._ modest 94._ _stupid l^-_ _big 39 ._ false 67._ neat 95. _strong 12._ _bitter 40 ._ fine 68._ old 96. sweet 13._ _bold 41 ._ fierce 69. patient 97. _terrible T4._ _brave 42. foolish 70. peaceful 98._ _ugly 15._ _b right 43.. friendly 71 •_ perfect 99. _unhappy 16._ _busy 44. funny 72 ._ ^pleasant 100._ unusual ^7._ _calm 45 ._ generous 73._ __polite 101 ._ useful 18._ _capable 46 ._ gentle 74._ _poor 102._ valuabl 19. _careful 47 ._ glad 75._ popular 103._ warm 20. _careless 48 ._ good 76._ ^proud 104._ weak 21 ._ _charming 49. great 77._ ^quiet 105. wild 22. _cheerful 50 ._ happy 78._ _quick 106._ wise 23. _clean 51 ._ humble 79 ._ _responsible 107._ wonderf 24. _clever 52._ idle 80. rough 108. wrong 25. _comfortable 53._ important 81 ._ _rude 109._ young 26. _content 54. independent 82._ sad 27. _cruel 55._ jealous 83._ _selfish 28._ _curious 56 ._ kind 84. _sensible

PAGE 70

APPENDIX B COURSE OUTLINE

PAGE 71

Course Outline Title: Hes 4905 Variable Topics in Health Education Weight Control Self-Managed Behavior Change Instructors: Fred Schreiber 377-7604 Sig Fagerberg Thursday Monday January 4 8 11 15 18 22 25 29 1 5 8 12 15 19 22 26 1 5 February March July 9, 1979: 4 month follow-up Required Reading: Toward Permanent Weight Balance: Student's Manual Course Description and Requirements: This is a nine week weight control program. You will be graded on an S/U basis. Grades will be given strictly on the basis of your attendance (not on how much weight you lose). In order to pass all you need to do is attend class. In cases of emergency, or other unforseen circumstances, you will be allowed to miss one class, but you must notify the instructor as to your reason for missing class. In addition, you are asked to participate in a 4 month follow-up. This will be held during the week of July 9, 1979, and will involve filT ing out a few questionnaires and weighing in. If you are going to be out of town at that time, please note where I may contact you. This weight control program is part of a research project designed to determine if certain techniques which are used to help individuals lose weight are more effective with one type of personality or another. Therefore, we are asking you to fill out some information forms, and your responses will be strictly confidential. Both groups will employ techniques that have been shown to be effective in helping oeonle lose weight. The groups will meet once a week for nine weeks with a trained leader. The techniques utilized in treatment will include various selfcontrol procedures which will be discussed during the sessions and practiced at home. No techniques which will cause pain or discomfort will be used during the program. 66

PAGE 72

APPENDIX C LETTER SENT TO SUBJECTS AT THE FOLLOW-UP

PAGE 73

1700 S.W. 16 Ct. #A-1 Gainesville, Fla, 32608 July 9, 1979 Dear It has been approximately four months since the end of our weight control class and now I am doing the follow-up. Please record your current weight at the bottom of this page, fill out both of the enclosed questionnaires, and return all the materials (this letter, 1 test booklet, 2 answer sheets) in the enclosed envelope. I realize that not all of you may have continued losing weight, and that some of you may have gained weight. However, please be honest in reporting your present weight and in answering the enclosed questionnaires so that my results will be valid. Please read all directions carefully. When you fill out the Tennessee Self-Concept Scale remember that the answer sheet is arranged so that you respond to e\/ery other item on it. Please do not omit any item. If you want to change an answer, do not erase it; rather, mark an X through the incorrect response and circle the response you want. Please send these materials back to me AS SOON AS POSSIBLE as I am very anxious to finish my paper before the end of the summer. If you are interested in discussing the results of my research with me please call me in mid-August. My phone number is 377-7604. Sincerely, Fred M. Schreiber 68

PAGE 74

REFERENCES Abramowitz, C. V., Abramowitz, S. I., Roback, H. B. & Jackson, C. Differential effectiveness of directive and nondi recti ve group therapies as a function of client internal-external control. Journal of Consulting and Clinical Psychology 1974, 42, 849-853. Abramson, E. E. A review of behavioral approaches to weight control. Behavior Research and Therapy 1973, JX, 547-556. Buchanan, J. R. Five year psychoanalytic study of obesity. American Journal of Psychoanalysis 1973, 33, 30-38. Collingwood, T. R. & Willett, L. The effects of physical training upon self-concept and body attitude. Journal of Clinical Psychology, 1971,27,411-412. Fitts, W. H. Tennessee Self-Concept Scale Manual Nashville, Tenn.: Counselor Recordings and Tests, 1965. Friedman, M. L. & Dies, R. R. Reactions of internal and external testanxious students to counseling and behavior therapies. Journal of Consulting and Clinical Psychology 1974, 42, 921. Glucksman, M. L. & Hirsch, J. The response of obese patients to weight reduction: III. The perception of body size. Psychosomatic Medicine, 1969, 31 1-7. Glucksman, M. L., Hirsch, J., McCully, R. S., Barron, B. A. & Knittle, J. L. The response of obese patients to weight reduction. II. A quantitative evaluation of behavior. Psyc hosomatic Medicine, 1968, 30, 359-373. Gottesfeld,..H. Body and self-cathexis of super-obese patients. Journal of Psychosomatic Research 1962, 6, 177-183. Harris, M. B. & Bruner, C. G. A comparison of a self-control and a contract procedure for weight loss. Behavioral Research and Therapy, 1971,9,347-354. Held, M. L. & Snow, D. L. MMPI Internal-external control, and problem check list scores of obese adolescent females. Journal of Clinical Psychology 1972, 28, 523-525. 69

PAGE 75

70 Horrocks, J. E. & Jackson, D. W. Self and Role: A Theory of SelfProcess and Role Behavior Boston: Houghton Mifflin Company, 1972. Ince, L. P. The self-concept variable in behavior therapy. Psychotherapy: Theory, Research, and Practice 1972, 9_, 223-225, Jeffrey, D. B. A comparison of the effects of external control and selfcontrol on the modification and maintenance of weight. Journal of Abnormal Psychology 1974, 83, 404-410. Jeffrey, D. B., Christensen, E. R. & Katz, R. C. Behavior therapy weight reduction programs: Some preliminary findings on the need for follow-ups. Psychotherapy: Theory, Research and Practice, 1975, 12, 311-3137^ Kerlinger, F. N. S Pedhazur, E. J. Multiple Regression in Behavior Research New York: Holt & Winston, Inc., 1973. Kilman, P. R. Albert, B. M. & Sotile, W. M. Relationship between locus of control, structure of therapy, and outcome. Journal of Consulting and Clinical Psychology 1975, 43, 588. Kurtz, R. M. Body attitude and self-esteem. Proceedings, 79th Annual Convention, APA 1971, 6^, 467-468. Lazarus, A. A. Behaviour rehearsal vs. non-directive therapy vs. advice in effecting behaviour change. Behaviou r Research and Therapy, 1966, 4, 209-212. Lerner, R. M. Karabenick, S. A. & Stuart, J. L. Relations among physical attractiveness, body attitudes, and self-concept in male and female college students. Journal of Psychology 1973, 85, 119-129, Mahoney, E. R. Body-cathexis and self-esteem: The importance of subjective importance. Journal of Psychology 1974, 88, 27-30. Mahoney, E. R. & Finch, M. D. Body-cathexis and self-esteem: A reanalysis of the differential contribution of specific body aspects. Journal of Social Psychology 1976, 99, 251-258. McCall, R. J. MMPI factors that differentiate remediably from irremediably obese women. Journal of C ommunity Psychology, 1973, 1, 34-36. ^^' McCall, R. J. Group therapy with obese women of varying MMPI profiles. Journal of Clinical psychology 1974, 30, 466-470. Moody, L. E. & Schreiber, FM. Toward Permanent Weight Control: Instructor's Manual Gainesville, Fla.: Florida Cooperative Extension Service, 1979fa). Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control : Student's Manual Gainesville, Fla.: Florida Cooperative Externsion Service, 1979(b).

PAGE 76

71 Penick, S. B., Filion, R. Fox, S. & Stunkard, A. Behavior modification in the treatment of obesity. Psychosomatic Medicine 1971, 33^, 49-55, Quereshi, M. Y. Some psychological factors that distinguish between the remediably and irremediably obese. Journal of Clinical Psychology 1972, 28, 17-22. Rogers, C. R. Client-Centered Therapy Boston: Houghton Mifflin Company, 1951. Rohrbacher, R. Influence of a special camp program for obese boys on weight loss, self-concept, and body image. Research Quarterly 1973, 44, 150-157. Rosen, G. M. & Ross, A. 0. Relationship of body image to self-concept. Journal of Consulting and Clinical Psychology 1968, 32, 100. Secord, P. F. & Jourard, S. M. The appraisal of body-cathexis: Bodycathexis and the self. Journal of Consulting Psychology, 1953, 17, 343-347. Stuart, R. B. Behavioral control of overeating. Behaviour Research and Therapy 1967, 5, 357-365. Stuart, R. B. A three-dimensional program for the treatment of obesity. Behaviour Research and Therapy 1971, 9^, 177-186. Stuart, R. B. & Davis, B. Slim Chance in a Fat World: Behavioral Control of Obesity Research Press: Champaign, 111., 1972. Stunkard, A. J. The management of obesity. New York State JournaV of Medicine 1958, 58, 79-87. Stunkard, A. & Burt, V. Obesity and body image; II. Age at onset of disturbances in the body image. American Journal of Psychiat ry, 1967, 223, 1443-1447. Stunkard, A. & Mendelson, M. Obesity and the body image: I. Characteristics of disturbance in the body image of some obese persons, American Journal of Psychiatry 1967, 123, 1296-1300. Suczek, R. F. Psychological aspects of weight reduction. In Eppright, E. S., Swanson, P. & Iverson, C. A. (Committee in charge). Weight Control: A collection of papers presented at the weight controf colloquium Ames, Iowa: Iowa State College Press, 1955. Weinberg, J. R. A further investigation of body-cathej 41-46.

PAGE 77

72 Werkman, S. L. & Greenberg, E. S. Personality and interest patterns in obese adolescent girls. Psychosomatic Medicine 1967, 29, 72-80. Wollersheim, J. P. Effectiveness of group therapy based upon learning principles in the treatment of overweight women. Journal of Abnormal Psychology 1970, 76, 462-474. Wunderlich, R. A., Johnson, W. G. & Ball, M. F. Some personality correlates of obese persons. Psychological Reports, 1973, 32, 1267-1277. — Ziller, R. C. The Social Self New York: Pergamon Press Inc., 1973. Zion, L. Body concept as it relates to self-concept. Research Quarterly, 1965, 36, 490-495. ^

PAGE 78

BIOGRAPHICAL SKETCH Fred M. Schreiber was born in New York City on May 2, 1953. He graduated from Baldwin Senior High School in Baldwin, New York, in June, 1971. He attended the University of Rochester in Rochester, New York, from which he received a B. A. degree in psychology in May, 1975. In September, 1975, he enrolled at the University of Florida for graduate study in psychology, and received an M. A. degree in psychology in June, 1977. Presently he is a Counseling Intern at the Counseling and Testing Service at the University of Houston, and expects to receive a Ph.D. in psychology from the University of Florida upon completion of his internship in August, 1980. 73

PAGE 79

I certify that I have read this study and that tn my opinion tt conforms to acceptable standards of scholarly presentation and ts fully adequate, in scope and quality, as a dts~sertation for the degree of Doctor of Philosophy. ^?rarfy A. ''Professor ol :er. Chairman Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. / / / :.4..^ H Ted Landsman Professor of Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Paul G. Schauble Professor of Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. 1 Afesa Bell-Nathaniel Assistant Professor of Psychology

PAGE 80

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. SeigiWd Fagerberg y Associate Professor of Health Education and Safety This dissertation was submitted to the Graduate Faculty of the Department of Psychology in the College of Liberal Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1980 Dean, Graduate School


INTRODUCTION
Problem
The problem of obesity has reached such a high magnitude that the
U.S. Public Health Service has classified it as "one of the most preva
lent health problems in the United States today" (p. 547, Abramson, 1973).
Stuart and Davis (1972) estimate that there are currently between 40
and 80 million obese individuals in this country alone. Traditional
treatments for the problem have included medication, psychotherapy, and
therapeutic starvation. However, these treatments have generally been
unsuccessful, and Stunkard (1958) has concluded that
Most obese persons will not remain in treatment.
Of those that remain in treatment, most will not
lose weight, and of those who do lose weight,
most will regain it. (p. 79)
Behavioral Approaches to Weight Control
Recently, some successes have been reported using a behavioral ap
proach to weight reduction. This approach views eating habits as
learned behaviors and sees the overweight person as someone who has
learned inappropriate patterns of eating. Behavioral approaches to
weight reduction focus on helping the overweight person become aware
of his eating patterns and helping him change inappropriate or problem
atic behaviors. It is presumed that by improving one's eating habits it
will be easier to reduce one's caloric intake, and consequently, lose
weight. In addition, by changing one's habits permanently, weight loss
should also be permanent.
1


70
Horrocks, J. E. & Jackson, D. W. Self and Role: A Theory of Self-
Process and Role Behavior. Boston: Houghton Mifflin Company, 1972.
Ince, L. P. The self-concept variable in behavior therapy. Psycho
therapy: Theory, Research, and Practice, 1972, _9, 223-225,
Jeffrey, D. B. A comparison of the effects of external control and self-
control on the modification and maintenance of weight. Journal of
Abnormal Psychology, 1974, 83, 404-410.
Jeffrey, D. B., Christensen, E. R. & Katz, R. C. Behavior therapy
weight reduction programs: Some preliminary findings on the need
for follow-ups. Psychotherapy: Theory, Research, and Practice,
1975, 12, 311-313.
Kerlinger, F. N. & Pedhazur, E. J. Multiple Regression in Behavior
Research. New York: Holt & Winston, Inc., 1973.
Kilman, P. R., Albert, B. M. & Sotile, W. M. Relationship between locus
of control, structure of therapy, and outcome. Journal of Consulting
and Clinical Psychology, 1975, 43, 588.
Kurtz, R. M. Body attitude and self-esteem. Proceedings, 79th Annual
Convention, APA, 1971, 6_, 467-468.
Lazarus, A. A. Behaviour rehearsal vs. non-directive therapy vs. advice
in effecting behaviour change. Behaviour Research and Therapy, 1966,
4, 209-212.
Lerner, R. M., Karabenick, S. A. & Stuart, J. L. Relations among
physical attractiveness, body attitudes, and self-concept in male
and female college students. Journal of Psychology, 1973, 85, 119-129.
Mahoney, E. R. Body-cathexis and self-esteem: The importance of sub
jective importance. Journal of Psychology, 1974, 88, 27-30.
Mahoney, E. R. & Finch, M. D. Body-cathexis and self-esteem: A
reanalysis of the differential contribution of specific body aspects.
Journal of Social Psychology, 1976, 99, 251-258.
McCall, R. J. MMPI factors that differentiate remediably from
irremediably obese women. Journal of Community Psychology, 1973, 1,
34-36.
McCall, R. J. Group therapy with obese women of varying MMPI profiles.
Journal of Clinical Psychology, 1974, 30, 466-470.
Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Instructor's Manual. Gainesville, Fla.: Florida Cooperative
Extension Service, 1979(a).
Moody, L. E. & Schreiber, F. M. Toward Permanent Weight Control:
Student1s Manual. Gainesville, Fla.: Florida Cooperative
Externsion Service, 1979(b).


THE CONTRIBUTION OF ROLE PLAYING TECHNIQUES TO
SELF-CONCEPT ENHANCEMENT AND WEIGHT LOSS IN
OVERWEIGHT COLLEGE WOMEN
By
Fred M. Schreiber
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1930


6
is assumed that these characteristics are ordered according to their
ease of change, with attitudes being the least difficult to change while
self-concepts are the most resistant to change.
Ziller's helical theory proposes that the self-concept is the
anchoring characteristic of the system, or the ultimate level of the
personal change hierarchy.
Those who would effect change in attitudes,
values, behavior, or roles are necessarily
concerned with a change in the self-concept,
for it is assumed here that unless a change
in the self-concept is achieved which is con
gruent with changes at lower levels in the
hierarchy, the lower level change is likely
to be reversed and the organism returned to
the initial state of equilibrium, (p. 174)
Self-Concept and Body-Cathexis
Secord and Jourard (1953) investigated the relationship between an
individual's self-concept and body-cathexis. By body-cathexis is meant
the degree of feeling or satisfaction with the various parts or processes
of the body. In order to appraise body-cathexis subjects were presented
with a list of forty-six items, each describing a different part or
function of the body, and were asked to indicate on a five point scale
their degree of satisfaction with each item. In order to measure self-
cathexis, individuals were asked to rate in a similar fashion fifty-five
items believed to represent a sampling of the various conceptual aspects
of the self. Subjects were seventy college males and fifty-six college
females.
Intercorrelations between body-cathexis and self-cathexis scores
were .58 for men and .66 for women. Hence, individuals had a moderate
tendency to cathect their body to the same degree and in the same


64
Ziller Complexity of Self Scale
INSTRUCTIONS: Here is a list of words. You are to read the words quick
ly and check each one that you think describes YOU. You may check as
many or as few words as you like--but be HONEST. Don't check words that
tell what kind of a person you should be. Check words that tell what
kind of a person you really are.
1.
able
29.
delicate 57.
large
85.
serious
2.
active
30.
delightful 58.
lazy
86.
sharp
3.
afraid
31.
different 59.
1ittle
87.
silly
4.
alone
32.
difficult 60.
lively
88.
slow
5.
angry
33.
di rty 61.
lonely
89.
smal 1
6.
anxious
34.
dull 62.
loud
90.
smart
7.
ashamed
35.
dumb 63.
lucky
91.
soft
8.
attractive
36.
eager 64.
mild
92.
special
9.
bad
37.
fair 65.
miserable
93.
strange
10.
beautiful
38.
faithful 66.
modest
94.
stupid
11.
big
39.
false 67.
neat
95.
strong
12.
bitter
40.
fine 68.
old
96.
sweet
13.
bold
41.
fierce 69.
patient
97.
terrible
14.
brave
42.
foolish 70.
peaceful
98.
ugly
15.
bright
43.
friendly 71.
perfect
99.
unhappy
16.
busy
44.
funny 72.
pleasant
100.
unusual
17.
calm
45.
generous 73.
_pol ite
101.
useful
18.
capable
46.
gentle 74.
poor
102.
valuable
19.
careful
47.
glad 75.
popular
103.
warm
20.
careless
48.
good 76.
proud
104.
weak
21.
charming
49.
great 77.
quiet
105.
wild
22.
cheerful
50.
happy 78.
quick
106.
wise
23.
clean
51.
humble 79.
responsible
107.
wonderful
24.
clever
52.
idle 80.
rough
108.
wrong
25.
comfortable
53.
important 81.
rude
109.
young
26.
content
54.
independent 82.
sad
27.
cruel
55.
jealous 83.
sel fish
28.
curious
56.
kind 84.
sensible


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT iv
INTRODUCTION I
Problem I
Behavioral Approaches to Weight Control 1
The Self-Concept 5
METHOD 27
Subjects 27
Measures 29
Procedures 31
RESULTS 43
Hypotheses 43
Pretreatment to Follow-up Change Scores 49
Correlations Between Variables 50
DISCUSSION 52
APPENDIX A INSTRUMENTS AND FORMS 62
APPENDIX B COURSE OUTLINE 66
APPENDIX C LETTER SENT TO SUBJECTS AT THE FOLLOW-UP 68
REFERENCES 69
BIOGRAPHICAL SKETCH 73
i i i


42
Chapter of TOPS
A. History and background of TOPS
B. TOPS basic program
1. No prescribed diet
2. Changing eating habits
3. Group therapy approach
4. Competitionhonors and rewards
C. KOPS (Keep Off Pounds Sensibly) Maintenance of weight loss
D. Case reports of successful clients
Week 9 Other Approaches to Weight Control
A. Aversive conditioning
B. Counting mouthfuls
C. Fad diets and crash diets
D. Scarsdale diet
Follow-up
During the week of July 9, 1979, all subjects were contacted by
mail (see Appendix C). They were asked to report their current weight
and to fill out the Tennessee Self-Concept Scale and the Ziller
Complexity of Self Scale.


63
Questionnaire #2
1. When you lose weight, your physical appearance will change. Who do
you think will notice?
2. How do you think people will react to you when you lose weight?
3. Do you think you will act differently as a thin person?
If yes, please specify how you will act differently.
4.Do you expect to encounter new situations when you lose weight?
If yes, please specify.


12
obese tended to prefer sedentary activities or an inactive, passive
lifestyle. The obese also scored low on dominance, indicating that
they are seldom ambitious, determined, or assertive. They also character
ized themselves as being manipulative, hostile, and aggressive. It
seemed, therefore, that they would be likely to react to unexpected
demands with sudden withdrawal and underlying hostility. Finally, the
obese were very self-conscious about their size and expressed strong
sexual conflicts and frustrations. This seemed to be produced by the
desire to enjoy heterosexual contacts but the inability to engage in
these behaviors due to the attitudes of themselves and others concerning
their bodies.
Quereshi (1972) studied 180 female members of TOPS (Take Off Pounds
Sensibly) who had considerable difficulty with weight reduction, and on
the average weighed over 200 pounds. He compared them with ninety-eight
females who had been through the TOPS program and were successful over a
six month period at staying within 5% of their ideal weight. These peo
ple were know as KOPS (Keep Off Pounds Sensibly). All subjects were ad
ministered the Michil Adjective Rating Scale (MARS) which consisted of
forty-eight adjectives such as "nervous," "talkative," and "ambitious,"
and was accompanied by a five-point scale ranging from "very atypical"
to "very typical." This scale yielded four personality factors, which
were labeled as unhappiness (Factor 1), extraversin (Factor 2), self-
assertiveness (Factor 3), and productive-persistence (Factor 4).
The mean for TOPS' self-ratings on Factor 1, unhappiness, was sig
nificantly larger than that of KOPS, which indicated that TOPS per
ceived themselves as generally unhappy, nervous, tense, and dissatisfied.
TOPS also had a significantly larger mean than KOPS on Factor 2, extra-


72
Werkman, S. L. & Greenberg, E. S. Personality and interest patterns in
obese adolescent girls. Psychosomatic Medicine, 1967, 29_, 72-80.
Wollersheim, J. P. Effectiveness of group therapy based upon learning
principles in the treatment of overweight women. Journal of
Abnormal Psychology, 1970, 715, 462-474.
Wunderlich, R. A., Johnson, W. G. & Ball, M. F. Some personality
correlates of obese persons. Psychological Reports, 1973, 32,
1267-1277.
Ziller, R. C. The Social Self. New York: Pergamon Press Inc., 1973.
Zion, L. Body concept as it relates to self-concept. Research Quarterly,
1965, 36, 490-495.


10
across all the test results was the presence of a response set. The
obese were defensive about revealing any psychological problems and,
in short, presented themselves as "hypernormal." In addition, there
was a great difference between the obese and the control group on the
social introversion scale of the MMPI, suggesting that the obese girls
were more uncomfortable and anxious in social situations.
Vocational interest patterns of the two groups also differed.
While the control group identified with professions in which imagination,
ambitiousness, creativity, and intellectual strivings were paramount, the
obese girls' interests were more consistent with persons in nurturant
professions. The obese showed a kind of "maturity"; that is, their
interests were similar to those people whose occupations are stable and
"real is tic."
These results suggest that the obese tend to restrict themselves
both socially and vocationally. They appear to live within a con
ventional life pattern in which one does not attempt situations which
might provoke anxiety. In addition, through their efforts to appear nor
mal they may be sacrificing spontaneity and flexibility in many spheres
of psychological functioning. This pattern may hamper them by preventing
the full development of character through conflict and acceptance of
challenges.
Held and Snow (1972) studied twenty-three obese adolescent girls
and twenty-three non-obese adolescent girls who were randomly selected
from patients being seen for medical reasons at an out-patient clinic.
Subjects were individually administered the MMPI, Mooney Problem Check
List, and the Rotter Internal-External Locus of Control Scale. The obese
group scored significantly higher on five of the ten clinical scales of


20
of significance was .05. From the six scales on which the relative dif
ferences were greatest R-TOPS women exhibited more body concern (hypo
chondriasis), psychic "hurting" (depression), somatization (hysteria),
rebelliousness (psychopathic deviate), compulsive and ruminative tenden
cies (psychasthenia), and bizarre or confused thinking (schizophrenia).
McCall (1974) studied nineteen women TOPS club members with a mean
weight of 204 pounds. He broke them down into three experimental sub
groups as follows: One group was chosen because their MMPI profiles
closely resembled those of successful weight reducers previously studied
(KOPS); another group consisted of women whose MMPI profiles closely re
sembled those of the resistively obese (R-TOPS); and a third group fell
in-between these two extreme groups. The question asked was whether the
MMPI profiles that distinguished between these three subgroups would
have any bearing on success in group therapy and weight reduction.
Subjects were randomly assigned to one of three groups which met
for sixteen weeks and were oriented toward the development of self-con
trol. Pre- and posttherapy data showed that only the R-TOPS-like sub
group showed significant changes after therapy. They changed on the
following six clinical scales: Hs (hypochondriasis), D (depression), Hy
(hysteria), Pd (psychopathic deviate), Pt (psychasthenia), and Sc
(schizophrenia). These six scales had previously most differentiated
the refractorily obese (R-TOPS) from the remediated obese (KOPS). In
addition, only the R-TOPS-like subgroup showed a significant weight loss
at posttreatment. Hence, obese women who had the "worst" MMPI profiles
tended to benefit most from group therapy as indicated both by weight
loss and profile improvement.


BIOGRAPHICAL SKETCH
Fred M. Schreiber was born in New York City on May 2, 1953.
He graduated from Baldwin Senior High School in Baldwin, New York,
in June, 1971. He attended the University of Rochester in Rochester,
New York, from which he received a B, A. degree in psychology in
May, 1975. In September, 1975, he enrolled at the University of
Florida for graduate study in psychology, and received an M. A. degree
in psychology in June, 1977. Presently he is a Counseling Intern
at the Counseling and Testing Service at the University of Houston,
and expects to receive a Ph.D. in psychology from the University of
Florida upon completion of his internship in August, 1980.
73


35
Variable Procedures
Role Playing Group
During the second forty-five minute period of the third week "role
playing" subjects were introduced to role playing. The experimenter
explained to the class that when people lose weight other people may
begin reacting to them differently. Hence, as a thin person they may be
faced with new situations which will require new roles. As a result,
part of this class was used to help students practice these new roles.
Week 3
Exercise 1: Getting Acquainted
Purpose
a. To help group members get acquainted in a relatively non-threatening
manner.
b. To explore feelings generated by "becoming another person."
c. To explore the dimensions of a brief encounter.
d. To emphasize the need for careful, active listening as well as
self-disclosure during conversation.
Instructions
1. Group members are paired in dyads, and the facilitator instructs
participants to "get to know your partner" for the next few minutes
(5 to 10 minutes). Participants are instructed to listen to the "free
information" or clues their partners give about themselves and to
follow-up on this free information. In addition, participants are in
structed to "self-disel ose" or give information about themselves to
their partners.


38
roles. Now dramatize the situation for five to ten minutes, or until it
comes to a natural conclusion. Do the same situation twice: first, as
you would act as an overweight person, and second, as you would act as
a person of normal weight.
2. After you have enacted the situation both ways (i.e., overweight and
normal weight roles) discuss the situation and give each other feedback
in the following order: (1) main character: hwo did you feel about
acting overweight and about acting normal weight? (2) other person(s)
in the exercise: how did you feel about yourself in relation to the
main character? (3) non-participant observers: what did you observe
was going on? During the feedback session other group members may
role play alternate ways of responding to the situation.
3. Think of various ways of responding no matter how strange they seem.
Contrast responses and see what "fits."
4. There are no right or wrong answers in this exercise, so don't be
afraid of making mistakes. We are practicing and comparing ways of act
ing as overweight and normal weight people so feel free to try out any
role you like.
Weeks 5 to 9
During the second forty-five minute period of each week "role play
ing" subjects continued the role playing exercise. The following is a
list,of various situations that were role played.
1. You see a friend whom you haven't seen in a long time. He remarks,
"You look different. Have you been sick?"
2. You're home on vacation and your mother keeps "pushing" food on
you. You'd love to taste some of the food, but you are already full.


36
2. After the interviewing phase, group members reassemble in the
larger group. The facilitator indicates that they now have the responsi
bility of introducing their partner to the group. Each group member,
in turn, is to introduce her partner by standing behind her and speaking
in the first person, as if she were that partner. There should be no
rechecking between partners during this phase. The individual who is
being introduced should hold her comments for the discussion period.
3. After all the introductions have been made, the facilitator leads a
discussion of the exercise, focussing on feelings generated and/or the
issues inherent in the goals of the exercise.
Exercise 2: On Being Fat and Thin
Purpose
a. To examine stereotypical differences between fat and thin people.
b. To get subjects thinking about their feelings concerning being fat
and being thin.
Instructions
1. Have the group generate a list of generalizations about fat and thin
people. Brainstorm in this manner for five to ten minutes, creating
an arsenal of stereotypes.
s. Discuss this exercise, focussing on subjects' feelings about the
1 ist.
Homework: Think about new situations that you may encounter as a result
of losing weight. What new roles will you become involved in as a thin
person?
Purpose
a. To encourage students to expand their views of themselves, i.e.,


METHOD
Subjects
The subjects in this study were twenty-nine female college students
enrolled in a weight reduction course taught through the Department of
Health Education at the University of Florida. Two sections of the course
were offered--one met Monday afternoon ("information" group) and one met
Thursday afternoon ("role playing" group). Sixteen women enrolled in
the Monday afternoon section ("information" group) and thirteen women
enrolled in the Thursday afternoon section ("role playing" group).
Subjects siqned up for the section they preferred, mostly depending on
their schedule of classes. Subjects were unaware of any difference
between the two treatment groups at the time of registration.
Pretreatment measurements for the two groups are summarized in
Table 1. In order to determine if there were any differences between
the two groups at pretreatment a two-group discriminant analysis was
performed using all pretreatment variables. This analysis indicated
that the "family self" subscale of the Tennessee Self-Concept Scale
had the most discriminating power of all the variables entered (F_ =
4.79', df = 1,27, £ <.05). The analysis produced no additional sig
nificant discriminating variables.
Since twenty-two variables were put into the analysis, it is likely
that some differences may appear by chance and may not reflect actual
differences between the two groups. In addition, "family self" appears
27


15
The research review thus far suggests that there are many psycho
logical and social problems associated with obesity. The obese have
problems with impulse control and tend to become easily angered, irri
tated, or resentful. They also describe themselves as disorderly,
pleasure-seeking, and rebellious. In addition, they are seldom active
or competitive and hardly expend a great deal of effort to accomplish a
difficult task. They sometimes take a very unrealistic view of them
selves and tend to blame many of their difficulties on their obesity.
Obese adolescents and young adults seem to have a poor self-concept
which is reflected in their dissatisfaction with their bodies as well as
with themselves. They harbor deep feelings of insecurity and have
exaggerated needs for attention and social approval. The world is seen
as a threatening and rejecting place and their response is to withdraw
as a defense against being hurt. As a result, they often feel lonely and
alienated, distrust other people, and avoid close interpersonal relation
ships .
These obese persons are self-conscious about their size and seem to
feel uncomfortable and anxious in a variety of situations. Consequently,
they tend to restrict their interests both socially and vocationally and
they may be deficient in basic social skills. In addition, these indi
viduals seem to have serious concerns about their masculinity or feminity,
feeling afraid that they cannot perform adequately in sexual situations.
Weight Reduction and Self-Concept
Glucksman and Hirsch (1969) compared obese subjects' performance on
a body size estimation task before, during, and after weight loss with
that of a normal control group. Six obese subjects with a mean initial


APPENDIX C
LETTER SENT TO SUBJECTS AT THE FOLLOW-UP


54
of self-esteem, the failure of role playing techniques to affect self-
concepts is probably a valid result. This is supported by other results
which indicated that role playing techniques had no significant affect
on those parts of the self-concept concerning subjects' complexity or
physical selves. The physical self subscale of the Tennessee Self-
Concept Scale dealt with concerns subjects were more open about and the
Ziller Complexity of Self Scale is more indirect in measuring complexity.
Hence, these scores are more likely to be valid, yet they too remained
unaffected by role playing techniques.
A possible explanation for the failure of role playing techniques
to effectuate self-concept changes involves the relative commitment an
individual has to each of the five components of ZiTier's (1973) helical
theory of personal change (attitudes, values, behaviors, roles, and
self-concept). According to Roby (1960; cited in Ziller, 1973), "com
mitment refers to a reduction in further alternatives associated with a
particular choice." (p. 151)
The amount of commitment an individual has to each of the components
of Ziller's system increases as one goes up the hierarchy. For example,
having reported certain attitudes is less restricting than having accepted
a certain role, such as group leader. Violation of the expectations
associated with certain attitudes is more acceptable to the self and
others than violation of role expectancies. The greatest commitment
is associated with the self-concept; these constructs are associated
with the stability, regularity, and consistency of personal behavior.
Since attitudes require the least amount of commitment, they are
subject to the most rapid change. A change in attitudes permits some
experimentation within the personal system; little commitment is


40
Information Group
During the second forty-five minute period of each week the "infor
mation" treatment group was presented with various weight control
theories. The following is an outline of the topics discussed each
week.
Week 3 Hypnotherapy: Dr. Sig Fagerberg
A. Conscious vs. subconscious
B. The power of suggestion to the subconscious
C. Hypnotic state
D. Suggestibility and self-hypnosis
1. Daily use of suggestion
2. Simple, repetitive, positive imagery pertaining to the future
E. Books
1. LeCron, Lesley. Self-Hypnosis
2. Peale. Power of Positive Thinking
Week 4 External Cue Sensitivity
A. Perceived time
B. Food cues
C. Response cost
D. Taste
E. Emotional state
F. Feedback and competing cues
Week 5 Nutrition and Dieting: Stephanie Fredette, Graduate Student
in Nutrition
A. Basic four food groups


37
to increase their complexity of themselves.
b.To have students think about situations they could role play.
Week 4
Exercise 1: Giving Feedback
Purpose: To teach students how to give helpful, appropriate feedback.
Instructions
1. The facilitator defines "feedback" to the group as:
A way of helping another person to consider
changing her behavior. It is communication to
a person (or group) which gives that person
information about how she affects others.
Feedback helps an individual keep her behavior
on target and thus better achieve her goals.
2. Group members discuss various criteria for useful feedback.
Exercise 2: Role Playing Fat and Thin People
Purpose
a. To compare how it feels to be a fat person with how it feels to be
a thin person.
b. To rehearse how to act in different situations.
c. To develop an awareness of oneself in relation to other people,
d. To see oneself in a variety of different ways.
e. To learn how to give appropriate, helpful feedback.
Role Playing Instructions
1. Make up a situation that you would like to role play. Discuss the
incident carefully with the group and how the different characters
might feel in the situation. Discuss the main character's goal and the
best way to achieve this goal considering the way in which other people
are feeling. Next, have group members volunteer to act in the different


7
direction that they cathected their self. In addition, it was found that
females cathected their bodies, irrespective of direction, more highly
than did males, in that they did not assign as many threes (have no
particular feeling one way or the other) to body items. Secord and
Jourard suggested that women would be more likely than men to develop
anxiety concerning their bodies because of the social importance of the
female body.
Rosen and Ross (1968) noted that in the correlations obtained by
Secord and Jourard (1953) between body-cathexis and self-concept, they
did not take into account that certain parts or processes may be more
important to an individual than other parts or processes in evaluating
his body- and self-concepts. Hence, Rosen and Ross (1968) investigated
the relationship between body image and self-concept, taking into account
the relative subjective importance of the aspects being rated.
Eighty-two undergraduates were presented with a list of twenty-
four body parts and seventeen adjectives from the Adjective Check List.
They were asked to indicate on a five-point scale for each body part or
adjective how satisfied they were with that aspect and how important that
aspect was to them. Correlations of subjects' mean satisfaction scores
between body image and self-concept were; r = .52 for all items; _r = .62
for all items above mean importance; and = .28 for items below mean
importance. Hence, these findings support those of Secord and Jourard
(1953), and in addition, they indicate that the relationship between
body-cathexis and self-concept can be refined if the subjective impor
tance of each component is considered.
Lerner, Karabenick, and Stuart (1973) asked 118 male and 190 female
college students to rate twenty-four body characteristics in terms of


18
administered at pretreatment and again at posttreatment. A Body Attitude
Scale, containing fifteen body concepts, measured subjects' attitudes
toward their bodies along the following three dimensions: (1) Evalua
tive dimension with good-bad, awkward-graceful, and beautiful-ugly
bipolar adjectives; (2) Potency dimension with weak-strong, hard-soft,
and thin-thick bipolar adjectives; and (3) Activity dimension with active-
passive, cold-hot, and fast-slow bipolar adjectives. In addition,
Bills' Index of Adjustment and Values (IAV) containing twenty-four
adjectives, measured subjects' attitudes toward themselves along the
following three dimensions: (1) Self-Concept ("seldom like me" to "most
of the time like me"); (2) Self-Acceptance ("I very much dislike being
as I am in this respect" to "I"very-much like being as I am in this
respect"); and (3) Ideal Self ("seldom would I like to be that way" to
"most of the time I would like to be that way").
Results at posttreatment indicated a significant weight decrease
and a significant increase on both the evaluative and potency dimensions
of the Body Attitude Scale. In addition, on the IAV there were sig
nificant increases on the Self-Concept and Self-Acceptance dimensions,
and a significant decrease on the discrepancy between Self-Concept and
Ideal Self. These results indicated that physical training experiences
can help teenagers lose weight and enhance their attitudes toward them
selves. However, long-term follow-ups are needed to determine the
stability or permanency of these changes.
Rohrbacher (1973) studied 204 overweight boys between the ages
of eight and eighteen in an eight week weight reduction camp program.
Body image and self-concept assessments (Secord and Jourard, 1953,
scales) were made before and after the camp program, and sixteen weeks


RESULTS
The sample at posttreatment consisted of 29 subjects; 13 were in
the "role playing" treatment group and 16 were in the "information"
treatment group. The attrition rate during the program was 0%. The
follow-up sample consisted of 28 subjects; 12 were in the "role play
ing" group and 16 were in the "information" group. One subject from
the "role playing" group had left the country and could not be contacted.
Hypotheses
Hypothesis 1: Both groups will show significant reductions in weight at
posttreatment.
The mean reduction index at posttreatment was 20.00 for subjects in
the "role playing" group and 14.06 for subjects in the "information"
group. Additional analysis revealed that the mean reduction index at
the follow-up was 18.58 for subjects in the "role playing" group and
38.19 for subjects in the "information" group. One-way t tests indica
ted that all four of these indices were significant at the .05 level
or greater {Table 2). Hypothesis 1 was accepted.
43


13
version, which suggested that the obese considered themselves to be more
outgoing and friendly than KOPS. However, Quereshi noted that obese
persons' perceptions of themselves as extraverts did not necessarily mean
that they were since previous research (Mayer & Thomas, 1967, cited in
Quereshi, 1972) utilizing projective techniques with obese females evi
denced traits such as passivity and withdrawal accompanied by feelings
of rejection.
A more realistic explanation may be gleaned from considering the
high self-ratings on Factors 1 and 2 together. This suggests that obese
females, despite their attempts to gain approval from others by means of
friendliness and congeniality, perceive themselves as lonely and rejected.
This explanation would also support the findings of Werkman and Greenberg
(1967) and Held and Snow (1972).
Gottesfeld (1962) compared self-drawings of thirty super-obese
subjects with thirty neurotics' self-drawings in terms of each group's
body-cathexis, or degree of satisfaction with the parts and processes of
their bodies. The super-obese subjects showed a more negative body-
cathexis than the neurotic group on the following three criteria; (1)
the super-obese were judged as having more negative body-cathexis by a
group of clinicians; (2) their drawings had more major parts of the body
missing; and (3) their drawings were less differentiated.
Gottesfeld (1962) also gave a list of twenty-eight personal traits
to the same group of super-obese individuals and neurotics and asked
them to rate themselves first as how they are and second as how they
would like to be. The discrepancy between self and ideal self served as
a measure of the degree of satisfaction with the self.


APPENDIX A
INSTRUMENTS AND FORMS


2
Techniques used in behavioral approaches to weight reduction are
derived from both operant and classical conditioning. Operant condi
tioning focusses on the antecedents and consequences which control a
behavior. When applied to weight control, the overweight person is
taught to bring his eating under appropriate situational cues and to
reinforce improvements in his eating patterns. Classical conditioning
involves the pairing of two stimuli so that eventually they bring about
the same response in an organism. The overweight person has paired eat
ing with a number of different external cues so that eventually the
external cues alone elicit a response of feeling hungry." The over
weight person needs to disassociate eating from other activities in
order to extinguish eating as a conditioned response to other activities.
Stuart (1967) utilized operant and respondent conditioning tech
niques to help clients gain control over their eating behavior. These
techniques included controlling the antecedent and consequent conditions
of eating as well as record keeping and exercise. In addition, rein
forcement was provided in the following three ways: (1) through the
clients' experience of success in self-control; (2) through the reduction
of the aversive consequences caused by a lack of self-control; and (3)
through considerable reassurance by the therapist. Stuart treated
eight women on an individual basis. Weight loss over a twelve month
period for the eight women ranged from twenty-six to forty-seven pounds.
Wollersheim (1970) compared the effectiveness of three group
treatments (behavior self-control, positive expectation--social pressure,
and non-specific therapy) with a no-treatment control group. The
behavioral self-control group utilized operant conditioning techniques


APPENDIX B
COURSE OUTLINE


71
Penick, S. B., Filion, R., Fox, S. & Stunkard, A. Behavior modification
in the treatment of obesity. Psychosomatic Medicine, 1971 213, 49-55,
Quereshi, M. Y. Some psychological factors that distinguish between
the remediably and irremediably obese. Journal of Clinical Psychology,
1972, 28, 17-22.
Rogers, C. R. Client-Centered Therapy. Boston: Houghton Mifflin
Company, 1951.
Rohrbacher, R. Influence of a special camp program for obese boys on
weight loss, self-concept, and body image. Research Quarterly,
1973, 44, 150-157.
Rosen, G. M. & Ross, A. 0. Relationship of body image to self-concept.
Journal of Consulting and Clinical Psychology, 1968, 32, 100.
Secord, P. F. & Jourard, S. M. The appraisal of body-cathexts: Body-
cathexis and the self. Journal of Consulting Psychology, 1953,
17, 343-347.
Stuart, R. B. Behavioral control of overeating. Behaviour Research and
Therapy, 1967, 1, 357-365.
Stuart, R. B. A three-dimensional program for the treatment of obesity.
Behaviour Research and Therapy, 1971, 1, 177-186.
Stuart, R. B. & Davis, B. Slim Chance in a Fat World: Behavioral Con
trol of Obesity. Research Press: Champaign, Ill., 1972.
Stunkard, A. J. The management of obesity. New York State Journal of
Medicine, 1958, 58, 79-87.
Stunkard, A. & Burt, V. Obesity and body image: II. Age at onset of
disturbances in the body image. American Journal of Psychiatry,
1967, 123, 1443-1447.
Stunkard, A. & Mendelson, M. Obesity and the body image: I. Char
acteristics of disturbance in the body image of some obese persons.
American Journal of Psychiatry, 1967, J_23, 1296-1300.
Suczek, R. F. Psychological aspects of weight reduction. In Eppright,
E. S., Swanson, P. & Iverson, C. A. (Committee in charge). Weight
Control: A collection of papers presented at the weight control
colloquiurn. Ames, Iowa: Iowa State College Press, 1955.
Weinberg, J. R. A further investigation of body-cathexts and the self.
Journal of Consulting Psychology, 1960, 24, 277.
Weller, L., Arad, T. & Levit, R. Self-concept, delayed gratification,
and field dependence of successful and unsuccessful dieters. Israel
Annals of Psychiatry and Related Disciplines, 1977, J_5, 41-46.


56
Although role playing techniques did not have an immediate impact
on subjects' reported self-concepts, these techniques may have facili
tated subjects' own awareness of themselves. This additional self-
awareness may eventually prove helpful in the process of losing weight.
Previous research indicates that young overweight people are self-
conscious about their size and seem to feel uncomfortable in interper
sonal or social situations. Consequently, they may become socially with
drawn. Thus, while overweight people remain overweight because of
poor eating habits, social problems may develop as well. However,
these problems are often denied; hence, many overweight people may
remain unaware of their feelings and needs. In the present study, role
playing scenarios partly served to help subjects become aware of their
feelings in interpersonal situations. In addition, the feedback ses
sions following role playing gave subjects an arena in which to share
their feelings.
Remaining overweight may become a defense against having to deal
with personal concerns. Losing weight removes this defense. Some
people may not want to deal with this issue and may resist losing weight.
In this respect, losing weight is similar to any other type of personal
change; i.e., it is often resisted.
People may not be aware of their own defenses, and may not be aware
of the consequences of losing them. In the present study, role playing
techniques served to make subjects aware of personal issues. Role
playing experiences brought out many anxiety provoking situations. As.
a result, these subjects became more aware of the challenges they might
face while losing weight.
These speculations would only hold true for overweight people with


Questionnaire #1
Name: Height:
Address: Size of frame:
small medium large
Phone: (circle one)
Where will you be during the week of July 9, 1979?
Address:
Phone:
Are you taking medication for weight?
Are you enrolled in another weight program now?
Do you have a metabolic or hormonal imbalance which could affect
your weight?
What is your class? freshman, sophomore, junior, senior, grad, student
(circle one)
What is your age?
What is your major area of study?
What is your grade point average?
For how may years have you been overweight?
62


DISCUSSION
The findings of the present study indicate that a behavioral
self-control approach to overeating produces significant weight losses
during the treatment period. In addition, this study supports Jef
frey's (1974) findings that a self-control approach produces significant
long-term weight losses. In the present study mean reduction indices
for both treatment groups were significant after a four month follow-up
period.
At the follow-up, a comparison of the mean reduction indices be
tween the two treatment groups approached significance in the opposite
direction than predicted; that is, there was a trend toward greater
weight loss for subjects in the "information" treatment group. A pos
sible reason for this outcome may be related to the issue of client/
treatment compatibility. Several studies have indicated that outcome
in psychotherapy is a function of the interaction between treatment
parameters and client variables (Abramowitz et al., 1974; Friedman & Dies,
1974; Kilman, Albert, & Sotile, 1973). In other words, clients differ
in the type of treatment to which they best respond.
Subjects in the "information" treatment group were exposed to a
broader range of theories and treatment concerning obesity than sub
jects in the "role playing" treatment group. As a result, they may
have had a greater opportunity to select an appropriate or compatible
treatment approach.
52


16
weight of 334 pounds were hospitalized for eight months and lost an
average of 86.7 pounds. Using an adjustable distorting image apparatus,
subjects were requested to make a distorted screen image of themselves
correspond to their body size as they perceived it at that moment. Obese
subjects increasingly overestimated their own body size from pretreatment,
during fifteen weeks of weight loss, and after a six week maintenance
period. In effect, despite their weight loss, they perceived themselves
as if they had lost almost no weight. In addition, three of the obese
subjects in this study were retested after an additional year of weight
loss, and they continued to overestimate their actual body size.
This "phantom body size" phenomenon was accompanied by supportive
clinical and figure-drawing data. For example, these same subjects
drew progressively larger figure-drawings during weight loss and at the
end of the final weight maintenance period. In addition, they reported
that they continued to feel obese despite weight loss (Glucksman, Hirsch,
McCully, Barron, and Knittle, 1968).
Suczek (1955) tried to delineate psychological aspects of weight
reduction. He administered the Interpersonal Dimension of Personality
System (IDPS) to 100 obese women before and after a sixteen week treat
ment program. This instrument analyzes behavior in terms of five
discrete levels: I, the level of public communication; II, the level of
conscious description; III, the level of private symbolization, IV, the
level of unexpressed unconscious; and V, the level of ego ideal. In this
study, only data from levels'I and II (i.e., overt, facade behavior) were
described.
There was little variability or inconsistency between level I and
level II data at pretreatment. In other words, obese women saw little


I certify that I have read this study and that in my opinion it
conforms, to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
farry A. Grader, Chairman
'Professor or Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Ted Landsman
Professor of Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
/.eld
Paul G. Schauble
Professor of Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Afesa Bel 1-Nathaniel
Assistant Professor of
Psychology


14
Results indicated that the super-obese reported to be more satisfied
with themselves than the neurotics. Later, however, while the super-
obese subjects were hospitalized for a two week evaluation, three inde
pendent judges observed their interactions with other patients and staff,
and rated them on the same list of twenty-eight personal traits. The
differences between self-ratings and observer (objective) ratings were
significantly greater than the differences between self-ratings and ideal
ratings. Gottesfeld concluded that the super-obese patients seemed to
present a facade of satisfaction. They denied that they were dissatis
fied with themselves on a self-report trait list, but they could not
as easily guard against a negative self-picture in their drawings (pro
jective test) or in observers' ratings.
Stunkard and Mendelson (1967) have found very low body-concepts in
some overweight people. These people felt that their bodies were gro
tesque and loathesome and that others viewed them with hostility and
contempt. This feeling was associated with self-consciousness and im
paired social functioning. In addition, the person took a very narrow
view of himself, expecially during times of misfortune or unhappiness.
All the unpleasant aspects of his life became focussed on his obesity;
that is, his body became the explanation and the symbol of his unhappi
ness. Buchanan (1973) stated that the body was the receptacle for self-
hate for obese individual; when they felt self-hate they complained
they felt fat.
Negative body-concepts have been found to be most prevalent among
persons who became obese during childhood or adolescence (Stunkard and
Mendelson, 1967; Stunkard and Burt, 1967). In addition, this condition
seemed to persist despite weight reduction and prolonged maintenance of
normal body weight (Stunkard and Burt, 1967).


48
variables pretreatment complexity and treatment group accounted for the
most variance between the two groups, as indicated by the Standardized
Discriminant Weights in Table 6. The variable self-concept failed to
be a significant discriminator between the two groups (hypothesis 6
was rejected). When the type of treatment and each of the pretest
and follow-up measures were analyzied independently pretreatment and
posttreatment complexity scores approached a significant differency
between the two groups (]3 variate F values in Table 6.
Table 6
Univariate £ Values and Standardized Discriminant
Weights for "Successful" and "Unsuccessful"
Weight Reducers on Treatment Group and
all Pretest and Follow-up Measures
Standardized
Discriminant
Weights
Univariate
£ Values
Pretreatment
Self-Concept
0.03
0.38
Physical Self
0.04
0.76
Complexity
0.74
3.64*
Treatment Group
0.55
2.36
Follow-up
Self-Concept
0.50
0.62
Physical Self
--
1.01
Complexity
0.32
3.62*
Standardized Discriminant Weights greater than 0.50 are underlined
* £ < .10


25
was effecting a promising merger, or the considerate husband had per
suaded his wife to move out of her parents' house into a home of their
own.
Results indicated evidence of learning in eight (32%) of the clients
treated by reflection-interpretation, eleven (44%) of the clients by ad
vice, and twenty-three (92%) of the clinets treated by behavior rehearsal.
In addition, of the thirty-one clients who did not benefit from reflec
tion-interpretation or from advice, twenty-seven were then treated by
behavior rehearsal and there was evidence of learning in twenty-two (81%)
of them. Thus, the overall effectiveness of behavior rehearsal in fifty-
two cases was 86.5%. Hence, behavior rehearsal appeared to be signifi
cantly more effective in changing behavior outside the therapy session
than direct advice or non-directive therapy.
The literature review revealed that the self-concept of obese
adolescents and young adults tends to be characterized by feelings of low
self-esteem, a restricted view of themselves focussed mainly on their
bodies or their size, and social isolation or social anxiety. Moreover,
a change in this self-concept may be a factor facilitating long-term
maintenance of weight loss.
It appears that adapting new roles allows a person to see himself
in a variety of different ways, or expands a person's view of himself,
and makes him more adaptive to change. In addition, role playing tech
niques may facilitate role taking behavior, and may exert a strong press
on changing a person's self-concept. Based on these premises, the
present study was devised to test the efficacy of role playing techniques
in a behavior modification weight reduction group in order to help
clients modify their self-concepts, and to determine the effects this


41
B. Definition of obesity
C. Adipose cell theory
D. Basal metabolic rate
E. Additives and preservatives
F. Pinch test
Week 6 Surgical Methods of Weight Reduction: Dr. Joun Kuldeau,
J. Hillis Miller Health Center
A. Interstinal bypass surgery
1. Diagram and description of operation
2. Adverse physical side effects
3. Benefits of operation
B. Stapling the stomack shut
1. Diagram and description of operation
2. Advantages of this procedure over intestinal bypass operation
C. Case reports of super-obese clients
D. Mood changes during rapid weight reduction
Week 7 Hereditary and environmental factors in obesity
A. Family research
B. Twin studies
1. Variability in weights in identical twins
2. Identical twins reared in same environment vs. identical twins
reared in different environments
3. Identical twins vs. fraternal twins vs. non-twin siblings
C. Adopted children research
Week 8 TOPS (Take Off Pounds Sensibly): Marilyn poss, Gainesville