Facilitating realistic self-appraisal by increasing the intensity of self-confrontation within growth groups

Material Information

Facilitating realistic self-appraisal by increasing the intensity of self-confrontation within growth groups
Mehlman, Anita Cooper, 1946-
Publication Date:
Physical Description:
x, 93 leaves. : ; 28 cm.


Subjects / Keywords:
Analysis of variance ( jstor )
Anxiety ( jstor )
Control groups ( jstor )
Magnetic storage ( jstor )
Observational research ( jstor )
Philosophical psychology ( jstor )
Psychotherapy ( jstor )
Self ( jstor )
Self concept ( jstor )
Self image ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Group relations training ( lcsh )
Psychology thesis Ph. D ( lcsh )
Self-perception ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )


Thesis -- University of Florida.
Bibliography: leaves 90-93.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
022666899 ( ALEPH )
13981462 ( OCLC )


This item has the following downloads:

Full Text


FACILITATING REALISTIC SELF -APPRAISAL BY INCREASING THE INTENSITY OF SELF -CONFRONTATION WITHIN GROWTH GROUPS By Anita Cooper Mehlman A Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Reouirements for the Degree of Doctor of Philosophy UNIVERSITY OF FLORIDA 1972


To Mother and Father with love, 'pride, and gratitude .


ACKNOWLEDGEMENTS The author expresses sincere appreciation for the many long hours of challenqina inquiry made by Dr. Hugh Davis as sunervisory chairman, nrofessor, and friend. Each of the supervisory-committee members stimulated this research with their own unique contributions. For expert assistance in the analysis of the data, special thanks are due to Dr. John Thornby, Dr. Ramon Littell , Dr. Carolyn Hursch, and Mr. Donald Cruse of the University of Florida Computing Center. Without the technical assistance and considerable filming done by Richard Ward, an undergraduate student, this project could not have been undertaken. To Dale Fowler and Ron Boddicker, group leaders, and Dr. Wiley Rasbury and Mary Ann Cruse, raters, a heartfelt thank you. Most importantly, for patience and prodding, each at the right time, for faith, understanding, and love, I am deeply grateful to my husband, Henry, without whom my life and life's work would be incomplete. iii


TABLE OF CONTENTS Page ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES vii ABSTRACT viii Chapter I Introduction 1 II Historical Perspective 10 III Development of the Problem 25 IV Method 31 V Results 39 Analysis of Variance 41 Analysis of Self-Concept Scales 41 Analysis of Anxiety Measures 45 Additional Comparisons 53 Discriminant Function Analysis 59 VI Discussion 60 Appendixes A. Instructions 72 B. Group Means of MSGO and STAI Scales 77 IV


TABLE OF CONTENTS — Continued Appendixes — (Cont.) Page C. Composite Profiles of the POI 79 P. Discriminant Function Analysis 84 E. Miskimins' Self -Goal-Others Discrepancy Scale 88 BIBLIOGRAPHY 90


LIST OF TABLES Table Pago 1. Analysis of Variance and Pairwise Comparison of Group Means for Self -Concept Subscale 42 2. Analysis of Variance and Pairwise Comparison of Group Means for Goal-Self-Concept Subscale 43 3. Analysis of Variance and Pairwise Comparison of Group Means for Perceived Rating of Others Subscale 4 5 4. Analysis of Variance and Pairwise Comparison of Group Means for State Anxiety 4 6 5. Analysis of Variance and Pairwise Comparison of Group Means for Trait Anxiety 47 VI


LIST OF FIGURES Figure Page 1. State Anxiety Profile for Group I by Subject 4 8 2. State Anxiety Profile for Grouo II by Subject 49 3. State Anxiety Profile for Grouo III by Subject 50 4. State Anxiety Profile for Group IV by Subject 51 5. Trait Anxiety Profile for Group I by Subject , 54 6. Trait Anxiety Profile for Group II by Subject 55 7. Trait Anxiety Profile for Group III by Subject 56 8. Trait Anxiety Profile for Group IV by Subject 57 VII


Abstract of Dissertation Presented to the Graduate Council o* the University of Florida in Partial Fulfillment of the TCe3auiroments ^or the Dearee of Doctor of Philosophy FACILITATING REALISTIC SELF -APPRAISAL BY INCREASING THE INTENSITY OF SELF -CONFRONTATION WITHIN GROWTH GROUPS By Anita Cooper Mehlman March, 1972 Chairman: Huqh C. Davis, Jr. Faior Department: Psvcho] or?y This study was desioned to exolore the effects of varied modes of self-confrontation. Four groups of normal subjects (Groups I-TII, five 5s each, Group IV, seven 5s) met for five sessions of a qrowth group interaction. Each group was selectively video-taped on each day by means of dual cameras, one fixed and one zoom. The instrumentation oermitted the superimposition of individual close-ur>s unon the fixed groun imaoe . The video-tapes were each 24 minutes in length. Each video-tape was viewed by two trained iuderes who rated each 5 on each day. Vlll


The rating instrument was the Actual Rating of Others (ARO) subscale of the Miskimins' Self-Gcal-Other-Discrepancy Scale (MSGO) . At a preexperimental meeting and again at sessions 2-5 all 5s completed the Self-Concept (SC) , Goal-SelfConcept (GSC) , and Perceived Rating of Others (PRO) subscales as well as the Spielberger State-Trait Anxiety Inventory (STAI) . The Personal Orientation Inventory (POI) was also completed at the preexperimental meeting and at the end of session five. Group I was designated the control group and did not receive structured feedback. Group II viewed the video-tape of the immediate past session at the beginning of each meetincr. Group III reviewed printed copies of the judges ' ratings made on the video-tape of the immediate past session. Group IV both watched the videotape and reviewed printed copies of the judges' ratings of the immediate past session. At the conclusion of the feedback period (controlled to last 24 minutes) , all 5s completed the above-named instruments. The 5s in Group I halted their interaction at the end of 24 minutes and then completed the test instruments. At the end of the test IX


period all groups interacted for one additional hour during which the filinina of the next feedback tape was made. Results indicate that video-taped feedback is less anxiety-producing and yields significantly greater positive changes in reported SC, GSC, and PRO than do judges' ratings used as feedback. The combined treatment of video-tape and judges' ratings falls between either modality used alone as measured by all instruments. It appeared that the 5s in both groups which received judges' ratings as feedback ignored and/or discredited the validity of these ratings. Further investigation of the effects of judges' ratings used as feedback is sugaested. Additionally, the use of logarithmic transformations of the data obtained from small-group interaction research was discussed. x


CHAPTER I INTRODUCTION "O wad some oower the gif tie gie us To see oursels as ithers see us" In its original context this passionate statement acquires a decidedly humorous note as the poet traces the journey of a "wee lousie" through the hair of an otherwise immaculately groomed woman, while the more intensive aspects of being able to objectively view ourselves are momentarily lost. The desire, even need, to know how we appear to others is evident throughout our history. Written records of the use of artificial mirrors date back to 1500 B.C. (Swallow, 1937), and surely, nature's own reflecting sources were relied upon long before that time. The question of how we appear to others goes beyond a mere evaluation of our physical being. The often fearfully asked "What did he tell you about me?" — "Did he like me?" — "Do you think he understands?" — "Did I make a good impression?" — all surpass a desire for an assessment of our physical bearing because that is one thing we can do ourselves, however 1 -


? subjectively, by looking into a mirror. The greater question involves coming to know tho irner us — our "soul," our "personality," our "realness" — and for this we do rely on external valuing sources. With the possible exception of a fully-functioning, self-actualized individual, if one could be found, each of us derives much of our awareness of our own being from observing reflections, that is, observing how we affect others and how we affect their behavior toward us. However distorted these reflections are, they are vitally important insofar as they are incorporated into our picture of ourselves, our self-image, and may lead to alterations in how we view ourselves. The extent to which "how others see me" affects "how I see me" undoubtedly differs from occasion to occasion and person to person, but, nevertheless, ^uch of our efforts to form our self-image are made through "the eyes of the beholder," or, more accurately, through our own understanding of how the other views us. This necessary reliance upon reflected identity forms the basis of at least two relatively distinct theories of personality development. Within the framework


3 of Sullivan's interpersonal theory of personality development (Sullivan, 1953), the very earliest selfevaluations are based on assessments of significant others' behavior toward us with an implied assumption that we are, in some vague way, causative agents of that behavior. Hence, the development of "good-me" and "bad -me." The "good-me" series of self-referents derives from situations in which our behavior has led to satisfaction, paired with tender and comforting behavior toward us by the mothering one. The "bad -me" constellation of self-referent responses are learned from situations in which specific behaviors brought about increasing levels of tension and anxiety. Because heightened tensions and anxiety are decidedly uncomfortable, we develop a series of security operations which enable us to avoid the occurrence of this anxiety within the scope of interpersonal relationships. There are occasions on which the provocation of anxiety can be mitigated by physically withdrawing from or avoiding threatening situations. When proximity cannot be avoided, anxiety can be tempered either by


4 selectively not attending to those situations which nurture anxious thoughts and feelings or by replacing the anxiety-producing responses with others that do not have the undesirable effect. The former is known as selective inattention and the latter as substitutive processes . The process of selective inattention permits us to deny to conscious awareness assessments of ourselves that are anxiety-provoking, whether the assessments are reflected from others (external evaluation) or a product of our own scrutiny (internal evaluation) . Thus, we need not accept all evaluations of our self-image, good or bad; however, even those qualities which we consciously reject or unconsciously fail to attend to become a part of our understanding of ourselves. In other words, our self-image includes the negation or absence of that unacceptable quality. Rogers (1959) makes a different assumption regarding the role and development of self-evaluative responses. He believes that all of us are innately evaluative and inherently capable of differentiating between desirable (effective) and undesirable (ineffective)


5 behaviors and, that if left alone to choose, we will quite naturally select those behaviors which comfort and satisfy and avoid those which yield discomfort, r&in, or dissatisfaction. This innate, evaluative process leading to the selection of satisfying behavior has been called the "self -actualizing tendency," that is, the tendency to maintain and enhance oneself. . . . there is no guarantee that the choice which is made will in fact prove to be self -actualizing . 3ut because whatever evidence exists is available to the individual, and because he is open to his experiencing, errors are correctable. If this chosen course of action is not self-enhancing this will be sensed and he can make an adjustment or revision. He thrives on a maximum feedback interchange and, thus, like the gyroscopic compass on a ship, can continually correct his course toward his true goal of self-fulfillment (Rogers, 1964; p. 164-165) . This discrimination may take place below the level of conscious awareness at which time it is labeled subception . While not identical to selective inattention, subception provides a similar means of avoiding incoming information that is not congruent with the existing self-conception. More specifically, Rogers appears to describe the self-conception as an organized totality of habitual patterns of thought about oneself into which new information is incorporated as long as it is not too


6 discrepant with the existing picture. The broader the initial self -conception, therefore, the more open to new experiencing will be the individual. A major difficulty in incorporating new aspects of the self -conception, especially those derived from the reflected self -conceptions others have of us, is in assessing their accuracy. The new information may be inaccurate due to distortions in the way it is transmitted to us from others or in the way we evaluate it upon reception. In fact, the new information may be objectively accurate but so discrepant with the existing elements of the self-conception that it is ignored, denied, or avoided. To maximize the incorporation of the message, the emotional components which trigger anxiety and thence avoidance behaviors (security operations or selective inattention and substitutive processes) must be circumvented whenever possible. Dissonance theory (Festinger, 1957) predicts that the arousal of anxiety will lead to efforts to reduce the discrepancy between the dissonant elements either by changing a behavioral cognitive element, changing an environmental cognitive element, or by the addition of


7 new cognitive elements. These active attempts at altering a dissonant situation are in addition to the attempts the individual may make to avoid situations and information that may tend to increase the state of dissonance. Security operations are such that when discrepant information is impinging on the existing self -conception, an assumption is made as to the accuracy of the established self-conception along the lines of "I'm right and they're wrong." This reliance on our own assessment of our selfconception in the face of discrepant information provides the final element necessary to enhance the probability that a "maximum feedback inter change" (Rogers, 1964) can be developed with the aim in mind of providing the individual with the opportunity to change his own selfconception within the most efficient framework possible. Every time the individual's self -conception is presented with new and discrepant information, a confrontation occurs. Thus, each of us experiences a multitude of confrontations both from external valuing sources and from within ourselves. The available security operations may moderate the intensity of the confrontation but, vith the possible exception of the most severe psychoses, do not totally insulate the self -conception .


8 The effectiveness of varying intensities of confrontation is likely measured in terms of a curvilinear relationship; that is, increased levels of confrontation above the threshold of awareness likely have a favorable impact on self-assessment until the degree of anxiety aroused acts to trigger the available security operations. The likelihood of distortion of the information message somewhere during the transaction remains fairly constant either as expressed initially by the individual, as interpreted by the recipient of the message, as reflected by the recipient, as received back by the individual, and then as reflected upon by the individual as he seeks to incorporate this new information into his existing self -conception . Undoubtedly, an aid to effecting a maximum feedback interchange would be to eliminate as many of the intervening steps as is possible without erasing the essential content and meaning of the message. Ideally, this would be managed by eliminating the "other person" and providing the "inquirer" with the necessary tools for objectivity as well as providing for him some "objectively realistic" framework in which to evaluate his own behavior and attitudes. Understandably,


_ 9 at some point this isolated self-exploration must be translated into an interpersonal behavioral setting so that the "inquirer" can assess his performance in social settings and reevaluate his skills in maintaining interpersonal relationships. Thus, maximum self-exploration and understanding are achieved within the natural framework of a confrontation model in which as many of the possibly distortive intervening steps are eliminated after which the newly incorporated information is utilized in an interpersonal setting where further feedback would be available. The utility of group interaction setting to effect this combination of individual self-exploration and then interpersonal experimentation and assessment should not be overlooked.


CHAPTER II HISTORICAL PERSPECTIVE The earliest mention of a deliberate attempt to confront a client directly with his own verbalized production is by Bierer and Strom-Olsen in 1948. The authors audio tape-recorded individual psychotherapy sessions which they later played back to the client. It was their conclusion that the playback of the therapy sessions helped to overcome resistance, "as useful in releasing associations and in shortening the length of treatment as well as aiding those clients who were prone to forgetting events that transpired in earlier sessions. Gill, Newman, and Redlich (1954) recognized in the audio-tape playback of therapy sessions a valuable instrument for confronting clients who utilize denial of past events and statements as a means of avoiding consequences. They also noted that while the clients readily accepted the tape recording and replay of their therapy sessions, the therapists reacted in a negative 10 -


11 manner, perhaps due to their perceived loss of direct control of the therapeutic session. Wolberg (1954), on the other hand, noted the benefits of audio-tape playback for both therapist and client. Specifically, the client could be presented with clear evidence of his progress and ". . .it (playback) tends to evoke further insight" (p. 686) . These early efforts at client confrontation included the rereading to clients of notes taken at earlier sessions (Pinney, 1955) as well as the transaction of entire therapy sessions by means of typewritten and longhand notes passed back and forth between the client and the therapist (Farber, 1965) . This latter technique was employed by Farber, a deaf psychiatrist, with individuals who had quite poor reality contact. He noted an increased self-awareness of confused thinking and a lessening of denial of the part of clients who occasionally disowned prior statements. Pinney (1955), who utilized "minutes" of previous sessions as stimuli, observed striking emotional responses in otherwise unresponsive schizophrenics. Golner, ~eddes, and Arsenian (1959) also confronted individuals within a group setting with


12 "minutes" of prior sessions of the therapy group which were read back to the group by the therapist. In general, the group members were described as becoming more "egoinvolved" subsequent to this confrontation. More specifically, of the eleven inpatients studied, ^ive were said to have been favorably affected, four not at all, and two unfavorably affected. Interestingly, it was again noted that the therapists were disaffected with the technique, likely in reaction to perceived loss of power and control of the sessions. In 1960, Cornelison and Arsenian inaugurated a photographic self-image confrontation regime within a nonrandom group of psychiatric inpatients. Their purpose was to investigate the responses of the patients, seven female and nine male, to their own photographs and motion pictures. In a four-step procedure the photograph was taken, ^hown to the S, discussed with him, and then observations were made of his response to the photograph. Long-term change, after repeated confrontation sessions, was measured in terms of Rorschach responses, sentence completion items, and drawings of people. It was apparent that the presence of inner turmoil enhanced


13 the S's responsiveness to his photograph. All patients recognized themselves; often expressions of shock and rejection were observed. The authors observed that, while this was not a controlled study, the information obtained indicated that photographic self-. Image confrontation might be a useful adjunct to more traditional modes of psychotherapy . Miller (1962) also obtained evidence that photographic self-image confrontation with chronic inpatients likely has a strong positive impact on them. With forty-two process psychotics, he observed that thirty-eight recognized themselves, most of them positively, An extension of the photographic self-image technique which relies most heavily on confrontation by means of still photographs is to utilize motion pictures. The greatest benefit of this confrontation method has been the immediacy with which a finished still photograph could be obtained (less than sixty seconds), but traditional motion pictures take days to process. Thus, the utility of video-tape recording has been explored, ind numerous studies investigating video_tape confrontation have been carried out in the past five years.


14 Nielson (1964) tested the hypothesis that videotape playback would enhance recall of subjective experience by stimulating S to relive events. The evidence he obtained appears to support his hypothesis and he noted that Ss demonstrated a unique responsiveness in regard to their own self-image and a particular interest in understanding themselves as well as a willingness to associate to the material uncovered by the film. One of the earliest video-tape studies was in 1965 (Moore, Chernell, =ind West) and was largely based on Cornelison and Arsenian's 1960 work. In the present study all psychiatric inpatients were video-taped during a twelve-minute structured interview within twenty-four hours of admission. Half of the eighty patients saw a replay of the interview immediately after it occurred while the remaining patients comprised the control group and never saw their video-taped interviews. Additionally, the experimental group (E group) was involved in fiveminute follow-up sessions weekly which were also taped and each S reviewed all of his prior tapes. Upon discharge all patients were rated on the degree of improvement subjectively observed by a psychiatrist (one


15 of the authors), a resident, and the resident's supervisor Fifty percent of the E qrouo were rated as maximally improved while 50 percent of the C qroup were described as minimally improved. Althouqh this difference did yield a significant X 2 , several methodoloaical inadequacies limit the precision of this test. Firstly, the E qrouo members tended to remain hospitalized longer (24 versus 18 days) than did the C qrouo: next, it is quite likely that at least one of the raters (the psychiatrist-author) knew the qroup-membershio identity of the individuals he rated; and lastly, the video-tape confrontation sessions were just a supplemental procedure to the onqoincr therapy program and controls over each asoect of the proaram were not utilized. Geertsma and Reivich (1965) employed repetitive self-observations by means of video-tape playback with a single female client specifically selected because of her rioidity and extensive use of projection and rationalization. For six weeks the 5 completed a Cattell Checklist of Source Traits before and after viewing the video-tape of her last therapy session. Eiqht student


16 nurses also viewed all the tapes and rated S. Following treatment, 5*s self -ratings became more congruent with the nurses' ratings indicating a greater objectivity and more realistic self -appraisal. Boyd and Sisney (1967) designed a video-tape self-image confrontation study to explore the hypothesis that the dissonance created when discrepancies in the selfand ideal-self-concept become apparent would lead to a shift in the self-concept toward a more realistic appraisal. An alternative hypothesis, considered less likely by the authors, was that the dissonance would be resolved by perceptual distortions during the replay of the video-tape recordings. Two groups of seven each individually received a ten-minute structured interview which was video-taped. The E group then reviewed the tape while the C group watched a television comedy. Each 5 completed the Leary Interpersonal Checklist prior to the interview, after viewing the replay, and again two weeks later. The results support the hypothesis that after only a single session the self-concept scores of the E group shifted significantly away from pathology and this positive change was maintained during the two-week interval period.


17 Again, -nethodological deficiencies limit the usefulness of the information obtained. The authors recognized that their sample was small, their measurement of questionable validity, and that the identity of specific group membership was not a secret during the interview. Stoller (1967) publicly broadcast televized therapy sessions throughout a hospital and noted that the therapeutic effect on the group was high and long lasting. Most notable was the beneficial effect derived from the recognition of the group members by their own ward mates which continued during the intra-session periods. In 1968, video-tape playback beqan being used within a group therapy framework. Berger, Sherman, Spalding, and Westlake (1968) ran two groups of Ss under different record and playback schedules and determined that the playback of preceding sessions heightened feelings of group cohesiveness and intimacy. Additionally, it was observed that repeated confrontation led to an awareness of patterns of behavior. The first playback session was described as being very important primarily as a selfimage experience, but notice was also taken of its


18 threatening nature as a precipitate reducer of traditional barriers against intimacy. Hoqan and Alger (1969) were interested in studying the effects of video-tape playback on insight in group therapy. It was their contention that insight's critical function is to make freedom o* choice possible. Their procedure is to make it possible for any group member to stop the vidf?o-taoe playback at any point to make comments. They concluded that at least 90 percent of all office patients could benefit from this form of selfconfrontation and that client defensiveness is often decreased while motivation for therapy is increased. They have employed this technique with various groups, individuals, and families and credit it with increasing self -insight, shortening the lenqth of psychotherapy, and facilitatinq understanding of double-bind situations. Hum (1969) concluded from a study of high school counseling grouDS which utilized video-tape versus traditional verbal feedback as it arose within the group that the former may permit s to assume more of his own identity and less of the orouo's. As one of the few studies with adequate controls and sufficient procedural


19 integrity, these results offer some of the earliest reliable assessments of the benefits of video-tape playback. Danet (1969) , in a carefully controlled study of students diagnosed as neurotic or as having a character disorder, noted the marked emotionality and explosiveness in the E group both during the original session and the playback period. Also evident was a greater degree of extra-group contact among E group members and group cohesiveness. In an earlier study (1968) Danet observed, contrarily, that "a variety of data supported the contention that introduction of playback . . .had a disrupting influence on the group's processes" (p. 254). These contradictory conclusions drawn from similarly designed studies carried out by the same investigator point up the necessity of further investigation and refinement of the technique, especially insofar as some individuals may even by harmed by its use. Braucht (1970) , in a unique design which established positive standards of change for each S, determined that self -confrontation via video-tape playback did improve the accuracy of the Ss' self -concepts.


20 Seventy-three inpatients were divided into four-to-six-man groups and then half of the membership of each group observed the taped replay of their therapy sessions. All 5s rated themselves preand post-therapy and each was rated independently by several judges. While long-term patients appeared to benefit the most, generally all patients improved in terms of more accurate selfdescriptions relative to the individually determined positive change indicators. Miller (1970) investigated the relative effectiveness of audio-tape and video-tape playback both immediately and two days after each counseling session. Six groups of six students each entered into a six-v.'eek counseling program. Two groups each received audio-tape replay (immediate and delayed) and two video-tape replays (immediate and delayed) . There were also two control groups. The author concluded, based on Q-sort measures preand post-playback, that video-tape was superior to audio-tape for enhancing self-ideal-self -concept. Schloss (1970) attempted to assess the change in the level of self-knowledge among normal 5s as a function of both video-tape and audio -tape playback. Using the


21 Butler-Haigh Q-sort preand post-eight weeks of counseling, he found that only among those 5s whose experimental condition was traditional group therapy without any artificial playback was there a significant decrease in the discrepancy between real and ideal self. He concluded that video-tape and/or audio -tape may be a distraction or threaten a tightly held self -perception. Nonsignificant differences in reduction of the discrepancy between selfand ideal-self as a function of video-tape playback were also found by Fadale (1970) and Elbert (1970) . Elbert did find, however, within the group receiving video-tape playback that there was a significant difference in preand post-measures of selfconcept even though there was no significant difference between the experimental and control groups. A quite specialized form of self-confrontation has been developed by Kagan, Schauble, Resnikoff, Danish, and Krathwohl (1969) . In this individual therapy design, the client and the therapist are video-taped during the therapy hour after which the tape is replayed while the client is "interrogated" by a new therapist. The "interrogator's" concern is with teaching the client to


22 interrogate himself and how to gain insight from self-confrontation. They believe that positive client growth occurs when the client admits his discomfort and commits himself to change in addition to being able to differentiate stimuli and effectively behave differently. The original therapist does not interact with his client as recall interrogator because the same feelings which arose in the therapy session and were not dealt with may arise again, further blocking communication. A further extension of this technique is currently being investigated by Kagan and Schauble (1969) in which the client is confronted with his own video-taped reactions to a series of planned threatening behaviors by others presented to him on film. While several of the studies cited above were replete with methodological inadequacies and some contradictory evidence was assembled, the general body of the literature indicates that confronting clients, psychotic or neurotic, with objectively precise descriptions of their behavior has a positive effect on behavioral change. Defensiveness is minimized, denial and avoidance made very difficult to engage in, and nonverbal behavior


23 (in terms of video-tape replay) made available for study. Some clients are initially threatened by their videotaped image but probably not more so than the threat engendered by hearing oneself for the first time. In fact, r.hock and astonished queries of "Is that really me?" are far more common than fearful avoidance reactions, even among markedly psychotic patients. Sanford (1969), in reference to the usefulness of audiotape playback as a means of improving perception, offers as an hypothesis that "... any means of improving perception might result in accelerated awareness, mobilization of defenses, and reorganization of mental mechanisms with consequent change in behavior . . ." (p. 682). He found that clients were more willing to accept insightful messages after they were exposed to their own statements as opposed to interpretations offered by the therapist. Video-tape playback offers an additional dimension in the process of maximizing the feedback interchange by making readily available the nonverbal communications that occur, often subliminally . In a group situation, the roving camera eye can record the overt results of the immediate impact of new information and provide the


24 individual with an understanding of how what he says and expresses affects others and of how he, in turn, is affected, even by just a nonverbal reaction.


CHAPTER III DEVELOPMENT OF THE PROBLEM While the emphasis in the previous discussion has been on the utilization of confrontation techniques with nonnormal populations, its applicability to the normal population should not be overlooked. If we can add to the sensory and visceral experiencing which is characteristic of the whole animal kingdom, the gift of a free and undistorted awareness of which only the human animal seems fully capable, we have an organism x^hich is « beautifully and constructively realistic (Rogers, 1961; 105) . The effectiveness of confrontation techniques when used with neurotic or psychotic patients leads one to hypothesize on their utility vis-^-vis the normal population. Can these techniques, specifically videotaped feedback, either immediate or delayed, enable a relatively normally functioning individual to arrive at a more realistic, open, and aware perception of himself? The preceding review of the self-confrontation literature, with its emphasis on the utilization of a totally objective and unemotional "third party," the 25 -


26 video-tape recording, suggests that the necessary quality of awareness is best developed by presenting the individual with both a visual and auditory picture of himself as he appears to others. In this manner each individual can act as his own stimulus to the extent that the person, while viewing the tape, will be responding cognitively, emotionally, and even physiologically. In this fashion, the individual has the opportunity to respond to himself without, necessarily, having to interpret himself via others* behavior toward him. Reusch (1957) states that . . . the foremost criterion of successful * communication consists of the presence of feedback circuits which provide an opportunity to relay back to the original sender the effects that a statement has had upon other participants (p. 34) . While it is possible for the individual to be told how he has affected another and to, in fact, observe how he has affected another, it is a different problem to make him understand what it is about him that produces the observed reactions. In other words, discrepancy exists, to a variable degree, between what he meant to say, what he in fact said, and between how he believed his intention would be interpreted by another person, and how it was in fact interpreted. By enabling


27 the individual to react to himself as an outsider and by helping him to minimize perceptual distortions of this new information, openness to experiencing himself will be enhanced. In Rogerian theoretical interpretation, the more an individual is able to attend to, think about, and accept as a part of himself and his whole range of responses, the better adjusted he is likely to be. Intensive self-exploration could be expected to engender a considerable degree of anxiety as discrepancies between the self -conception, ideal-self -conception, and the results of realistic self -appraisal become evident. It is suggested that the dissonance aroused would be reduced by appropriate adjustments in the individual's perception of how others view him, or in terms of his viewing himself, in the direction of a more realistic self -appraisal. The recent attempts to evoke within individuals confrontations between their self-perceptions and objectively "real" perceptions about themselves has utilized, almost exclusively, the video-tape. Another form of objective feedback is to obtain the ratings of an individual made by a group of independent judges on


28 some easily understandable and interpretable instrument. By then making these ratings available to the individual, along with an explanation or interpretation of how they were derived, he would have more objective information about himself than he could obtain from simply interpreting his interactions with others. Compared with video-tape in which the individual "rates" himself, the obtained judges* ratings still involve the potential for a modicum of perceptual distortion in that the judges must still make assumptions and interpretations about the observed behavior of the individual. To the extent, however, that the judges are free from possible censure, disagreement, or emotional responsiveness by the individual, their ratings should be considerably more objective than those provided by the person with whom the individual is interacting. The entire field of research involving videotape as a mechanism of confrontation is fraught with methodological weaknesses and a generally sparse amount of statistically acceptable information. Also, the observed results may have been due to the novelty effect afforded by any kind of direct, concrete feedback to the


29 experimental subjects, especially within the therapy framework. Therefore, the present study will be directed toward a careful and systematic examination of the utility of video-taped feedback as a facilitator of more realistic self-appraisal within growth groups comprised of normal individuals. To afford some insight into the hypothesized effects of the novelty of direct feedback, ratings from objective iudges will be used alternatively in some treatment conditions. No definitive hypotheses regarding the effects of the objective judges' ratings will be made as this treatment mode is largely exploratory. The major hypothesis to be explored is that videotaped feedback, when added to a traditional feedback mode in which the group leader and other group members attempt to deliver objective messages of assessment to the individual, "ill induce greater change in the assessed self-concept than the traditional mode alone. Secondly, it is hypothesized that over the course of the interaction, greater congruence between selfand ideal-self -conceptions will develop. It is suggested that the video-taped feedback will again facilitate this outcome to the greatest extent. "


30 Thirdly, a more accurate self -appraisal, as measured in terms of the discrepancy between the judges* ratings and the self -ratings for each individual, should derive from the video-taped treatment. Lastly, anxiety, as a function of the confrontation experience, will initially increase and then level off as the individual habituates to the task of self-confrontation. As this study is to a considerable degree exploratory of relationships obtaining among the principle treatments, analysis on measures less directly concerned with the conditions of objective self-appraisal will be done. Hence, perceived ratings of others and trait anxiety will each be analyzed.


CHAPTER IV METHOD Subjects: Twenty-eight volunteers from the undergraduate population at the University of Florida were obtained by means of publicly posted signs telling of a five-session growth group experience being offered in conjunction with research for a doctoral dissertation. Those 5s (20) who signed up for the experiment as part of a course requirement received six experimental credit hours which counted as points earned toward the class grade. The remaining eight 5s were not compensated in any way. The average age of the 5s was 21.0 years with a range of 19.0 to 27.0 years. Leaders: Two graduate -level students in clinical psychology were hired to lead two groups each. They were each paid $25.00 for their involvement. The leaders were both male, above the third year in the program, and had had considerable prior experience in leading groups. Leader One interacted with Groups I and IV while Leader Two was involved with Groups II and III. Thus, each 31 -


32 leader experienced both a video-tape and judges' feedback condition. They were both instructed to lead their aroups in the way in which they felt most comfortable. The two constraints imposed by the experimental design were that there be no excessive movement about the room which mioht interfere with the filming and that the leader should not initiate specific feedback relative to the immediate past group session. Judges: Two staff members, one an assistant professor of psychology and the other a psychological assistant, acted as obiective iudges. Their task was to view each of the video-tapes and then rate each 5 on a 15-item bipolar scale. After trainina on the instrument, they reached a high level of concordance in their assessments as measured by the Kendall Coefficient of Concordance (Kendall, 1948) . The degree of concordance was assessed at each ratina and ranoed from .69 to .98 with the preponderance of scores falling above .85. Test Instruments: The Personal Orientation Inventory (POI) (Shostrom, 1962) , a 150-item questionnaire which yields a ten-scale profile of the individual, "as


33 completed at the pretest session and at the conclusion of the last experimental session. The Spielberoer State-Trait Anxiety Inventory (STAI) (Spielberger , Gorush, and Lushene, 1970), a twoscale forty-item instrument which yields a percentile score, was completed at the pretest session and at each experimental session. Scale Xj (State Anxiety) is endorsed with the S reporting his current, right-then feeling, "hich gives an indication of the anxiety enaendered in the experimental setting. Scale X 2 (Trait Anxiety) is endorsed from the qeneral or usually felt frame of reference and, thus, qives a measure of the overall level of anxiety. The Miskimins' Self-Goal-Other Discrepancy Scale (MSGO) (Miskimins, 196«) is composed of four subscales. The 5s completed three of the subscales at the pretest session and again at each experimental session. The judges completed the fourth scale for each S at the time of viewing each video-tape . The 5-endorsed subscales were the Self-Concept Scale (SC) , the Goal-Self-Concept Scale (GSC) , and the Perceived Ratinq of Others Scale (PRO) . The SC was to be completed in terms of where the S saw himself on the scale while the GSC measured where


34 he wanted to be on the scale. The PRO enabled S to record where he thought others would see him on the scale. The judges completed the Actual Rating of Others Scale (ARO) . For each subscale the items are identical. They are comprised of 15 bipolar pairs of statements with a nine-point range between the poles. The different scales are derived by instructing the endorser to respond with a different frame of reference each time he checks off the items . The MSGO is designed to arrive at a measure of the 5's realistic self -appraisal by testing the change over time in the observed discrepancy between the SC and ARO scales. This test requires that an assumption be made regarding the judges' objectivity, hence reality orientation. Then, a measurement of how close 5 comes to rating himself as the judges did will yield a score consistent with the 5's level of realistic self -appraisal. By deriving the change over time between the PRO-ARO scales, it can be assessed how accurately S understands how others see him. At the conclusion of the experiment each S was asked to rate his group on the dimension of group


35 cohesiveness which was defined as closeness, toqetherness, trust, or feeling of mutual understanding on a fivepoint scale. A score of 1 indicated a "total absence of cohesiveness" while a 5 indicated "very cohesive". Procedure: All 5s met one week prior to the first experimental session for the purpose of providing pretest data for all instruments. At this time, each 5 was briefed as to the nature of the experiment and given the opportunity to withdraw from the study without penalty (Appendix A). After completing the pretests, the Ss were randomly assicmed to one of four treatment groups (seven to a group) and these assignments, as well as the group meeting times, '-ere posted in the Psychology Building on the campus. Each aroup was to meet a total o^ five times, the first session lasting an hour, the next three, ninety minutes, and the final one, two hours. At the beginning of the first session, all 5s were informed of their feedback condition. Group I was desionated as the control group which interacted in a traditional growth-aroun framework. Group II experienced the traditional growth-group interaction with the addition of a twenty-four minute period of video-taped feedback intersected in each session 2-5.


36 Group III experienced the traditional growth-group interaction with the addition of a twenty-four minute period of review of the objective ratings made by the judges. This reviev; occurred in sessions 2-5. Group IV experienced the traditional growth-group interaction with the addition of a twenty-four minute period in which both video-tape and judges' ratings were reviewed in sessions 2-5. Description of the Group Sessions: Each group met for one hour at the first session. During that meeting and each successive one, the group was selectively videotaped in the following fashion. Dual cameras (one fixed on the entire group and one zoom free to move about) , situated behind a one-way mirror, were activated once each quarter-hour for six minutes. The middle six minutes of each quarter-hour were the target times. During each period of the filming the zoom camera was focused successively on each 5 for fifty seconds. Thus, each S appeared in a close-up for just over three minutes of each total film sample. The remaining time was used up in the repositioning and refocusing of the zoom lens for each 5. The order in which the close-ups were shot


37 was drawn from a random number table. The replay equipment v/as so designed that the entire group was seen on the monitor at all times and the particular close-up was superimposed in the lower left corner of the screen. The video-taped segments of each group session were viewed the following morning by the two judges who then completed the ARO scale of the MSGO for each subject. In sessions 2-5, for each of the three experimental groups (II, III, IV), the sequence of events was as follows. The initial twenty-four minutes of the session were concerned with reviewing the feedback from the immediate past session. That is, in Group II, the initial twenty-four minutes were spent watching the videotape from the immediate past session. Group III members received printed copies of the judges' ratings of them on the ARO scale of the MSGO and read and/or discussed these for twenty-four minutes. The 5s in Group IV both watched the video-tape and read/discussed the judges' ratings during the feedback time. The 5s in the control group began their traditional interaction immediately upon entering the group room. At the end of twenty-four minutes the leader halted the


38 interaction. At this point in time all 5s in all groups filled out the MSGO and the STAI . Thii; took approximately five minutes. Once the instruments were collected the groups interacted for one additional hour during which the filming of the next feedback was done according to the above-described procedure. Session 5 lasted an extra thirty minutes to provide for the completion of the POI which was done at the conclusion of the session.


CHAPTER V RESULTS Inspection of the raw data plots revealed considerable variability both within and across groups; therefore, the experimental data were analyzed by two different methods. Initially, all data were subjected to a stepwise discriminant analysis (Bimed Program, BMD 07M, 196R) . This analysis utilized the raw score differences between the pretest data and each succeeding day's data for each scale. As a parametric test, the discriminant analysis requires certain assumptions about the underlying distribution of scores from which the sample was drawn, i.e., a normal distribution about the mean and homogeneity of the variance. Questions about the first assumption must be raised in light of the smallgroup requirement of the experimental design. It also seemed likely that in this sample of 5s (a college population) the obtained scores might in fact be skewed in the more positive direction on the instruments. Additional doubt was raised about the nature of the 39 -


40 variance in terms of the effects of the random error component. Hctys (1963) suggests that a logarithmic transformation of the raw scores, which makes them conform to the linear model, be used when the effect of the random error is in doubt. Once the transformation is made, a standard analysis of variance test can be performed. While both analyses will be presented, all discussion will center around the analysis of variance tests on the transformed data as the E believes it is the more reliable test for the obtained data. Although 28 5s were assigned to the four groups, five failed to meet with their respective groups at the first session and the data from one subject had to be discarded when it was determined that he had not taken the pretests himself but had had a friend take them in his place. Thus, the group composition was five members each in the first three groups and seven members in the fourth group. Analysis of variance performed on the raw score pretest data, after the 5s had been randomly assigned to their respective groups, indicated no significant between-group differences. This analysis, coupled with


41 the random assignment of 5s to the treatment conditions, would appear to insure that there were no pre -treatment differences between crroups . Analysis of Variance Analysis of Self-Concept Scales The MSOO subscales (SC, GSC, PRO, and ARO) were analyzed individually and then in pairs which represented the discrepancy between the self -concept and the actual ratinas of others and the perceived ratinas of others and the actual ratings of others. Rimed Prooram BMD 01V (1968) was utilized in this stacre of the data analysis. On the self -concept dimension, across all days, there was a sioni^icant difference among all groups (R3 loe ~ 4.20, p<.01). Pairwise t-tests were then run to test for significant differences between the means of all groups taken two-at-a-time. The means for all subscales are presented in Aopendix B. CrouDS I, II, and IV all differed significantly from GrouD III as illustrated by Table 1. That is, on the self -concept dimension, the mean for Group III was significantly higher than that of


42 Table 1 Analysis of Variance and Pairwise Comparison of Group Means for Self-Concept Subscale Source Between Groups Within Groups Total Sum of Squares df ms 0.1148 0.9665 1.^813 3 106 109 0.0383 0.0091 4.20** Group I II hi II n.s. III 2.91** 4.34** IV n.s. n.s. 2.29* *p<.05 **p<.01 the other groups. The ranking of the groups from highest mean to lowest was Group III followed in order by IV, I, and II. Because of the scaling system used in the MSGO, a higher score or mean indicates a less favorable endorsement of the items. In reference to the SC scale, then, the members of Group III reported the lowest selfconcept scores while those in Group II reported the highest. Although neither video-tape condition differed significantly from the control group, the video-tape


43 conditions do differ siqnif icantly from the group which received only judges' ratings feedback. In addition, the rank-order of the groups is in line with the hypothesis that video-tape feedback would yield a more favorable self-concept. A similar pattern of significant differences emerged in the analysis of the GSC subscale (Table 2) . Table 2 Analysis of Variance and Pairwise Comparison of GrouD Means for Goal-Self -Concent Subscale Source Sum of Squares df ms Between Groups Within Groups Total 0.4008 3.2585 3.6593 3 106 109 0.1336 0.0307 4.35** Group I II Ill II n.s. III 2.95** 3.13** IV n.s. n.s. 2.29* *p<.05 **p<.01


44 Again, the observed mean vras sianif icantly higher for Croup III than the others when analysis was completed, followed in descending order by Croup IV, II, and I. While the control aroup fared the best on this subscale, nevertheless, the video-taped conditions did yield significantly more favorable mean scores than did the erroup which received the iudaes' ratinas alone. Analysis of the SCCSC discrepancy between groups did not yield sianif icance? thus, hypothesis two was not supported. Analysis of variance computations for the differences between SC-ARO and PRO-ARO across all days failed to yield significant differences between groups; thus, hypothesis three was not suooorted. Analysis of the PRO subscale revealed several additions to the established Dattern of sianificant differences (Table 3) . Inspection shows that the only pairwise comparisons not significant were between Croups I and II and between II and IV. Thus, the significant differences among means were obtained only when at least one group in the pair received obiective (-iudaes' ratings) feedback. Once again, the mean for Group III


45 Table 3 Analysis o^ Variance an^ Pairwise Comoarison of Grouo Means for Perceived Patina of Others Subscale Source Sum of Sauares df ms Between Grouns VTithin Grouos Total 0.1424 1.2540 1.3Q65 3 106 log 0.0475 O.^llR 4.01** Group I II Ill II n.s. III 2.80** 4.06** IV 2.09* n.s. 2.65' *p<.05 **p<.01 was higher than in each of the other groups, followed by I, IV, and II. Analysis of Anxiety Measures The Xj scale of the STAI (State Anxiety) yielded the following sianificant values (Table 4) . The group which received the video-taped feedback alone (II) did not


46 Table 4 Analysis of Variance and Pairwise Comnarison of Croun Means ^or State Anxiety Source Sum of Squares df ms P Between Grouos 0.°759 3 0.0920 7.68** Within Groups 1.26<>6 106 0.^120 Total 1.5455 109 Group I II Ill II 2.95** III 4.69** n .s. IV 2.34* n. s. 2.5' *p<.05 **p<.01 differ siqnif icantly from the other two experimental groups. Of interest is that the obtained means were hiohest in the control grouo followed by Group IV, II, and III. Once again, the higher mean indicates a more negative endorsement of the items or more experienced anxiety in the experimental setting. The X 2 scale of the STAI (Trait Anxiety) produced a similar array as shown in Table 5. As with the State


47 Table 5 Analysis of Variance and Pairwise Comparison of Croup Means *or Trait Anxiety Source Sura of Sauares df ins Between Crouns 3037 3 0. 1012 14 .58** Within Croups o .735* 106 0. "069 Total 1 0395 109 Croup I II Ill II 2.12* III 6.45** 4.20** IV 2.11* n.s. 4.72** *p<.05 **p<.01 Anxiety, the hiahest mean was in Croup I, strain followed by IV, II, and III. Inspection of the crroup profiles of changes on the Xi dimension across days (Fiaures 1-4) indicates that there were no consistent effects noted across all groups. Each group, rather, appears to respond to the experimental condition in a unigue fashion. For Ss in the control group (I) , the State Anxiety continued to increase through


C 4J C CD O u o 80 60-40' 20"/ T P Figure 1, 2 "T — 3 Day T 4 State Anxiety Profile for Group I by Subject. 48 -


80 60 -H •H 4-1 C C U u a 40 •• / \ j \ 20 -\ \ v.. T P Figure 2 2 3 Day 4 State Anxiety Profile for Group II by Subject. 49 -


80 60 o H •H P c o o 40 \ \ \ 20 -\ \ V \ T P Figure 3 . T 2 3 Day T 4 T 5 State Anxiety Profile for Group III by Subject. 50 -


80 60 rH c c; 0) a, 40 -20 Figure 4 . Day State Anxiety Profile for Group IV by Subject. 51 -


52 the fourth experimental session and only two 5s recovered their pretest level of State Anxiety. Group II 5s were not as consistent in their pattern of responding to the treatment; however, the majority of the 5s reached their highest point of State Anxiety on the third experimental day. Once again, only two 5s regained their pretest level of anxiety. In Group III, two 5s were at their highest level of anxiety at the pretest and the other three 5s peaked on experimental day two. In this group, all 5s finished the experiment reporting less State Anxiety than they did at the pretest. For the 5s in Group IV, experimental day two proved to be the most anxiety-engendering, with all but one 5 reaching his highest level of anxiety on that day. The one 5 who did not follow the pattern steadily decreased in his reported anxiety throughout the experiment. Five of the 5s reported less State Anxiety at the conclusion of the experiment than they did at the pretest period. In terms of the fourth hypothesis, the results are supportive although no consistent pattern was observed across all groups. It appears that the judges* ratings conditions (Groups III and IV) produce the


53 earliest peaks in State Anxiety with apparently sharper decreases in the level o^ the anxiety over the remaining experimental days. The absence of direct feedback (Group I) produces a situation in which the level of State Anxiety continues to rise dropping sharoly in the final session. The video-taped feedback (Group II) yields a condition in which the State Anxiety appears to peak in the middle of the experiment and then decreases. The profile shapes for the four croups on the X2 scale do not yield as consistent a pattern (Fiaures 5-8) . Three 5s in Grouo I, three in Grouo II, all five in Grouo III, and -Four in Group IV decreased in reported Trait Anxiety over the course of the experiment . Thus, approximately two-thirds o* the experimental 5s decreased in Trait Anxiety while the remaining ones either did not chance or increased on this dimension. Additional Comparisons Both the pretest and posttest profiles for all 5s on the POI were within the normal range of standard scores. Composite croup profiles are presented in Appendix C and indicate that all croups appeared slightly


80 60 \ N 40 "20 N V X. / TV ' • • m i l. • • f^ T 2 i 3 Day \ \ Figure 5. Trait Anxiety Profile for Group I by Subject. \ T 5 54 -


80 60 QJ rH •H P C u

80 60 -

80 "60 .. o c C o p 40 20 1 1 1 "~l 1 p 2 3 Day 4 5 Figure 8. Trait Anxiety Profile for Group IV by Subject. 57 -


58 more elevated at the posttest than they did at the pretest. For comparison purposes, Shostrom's profile of normal college students is also presented in the Appendix C. It can be concluded that treatment did not affect the deviations of the 5s outside a normal range. The design does not permit a test for an interaction between the anxiety measures and the subscales of the MSGO; however, pome insight into their interaction can be obtained by inspection. The group with the lowest mean State Anxiety (III) demonstrated the highest (thus, more negative) mean SC score. Conversely, the group with the highest mean State Anxiety score (I) had the second lowest mean SC score (more favorable) . The two groups which had video-taped feedback (II and IV) were between the extremes on State Anxiety while II had the most favorable SC score and IV had the next-to-least favorable SC score. The responses to the cohesiveness question produced an interesting array. Group I members, when their scores were averaged, gave their group a rating of 3.6 indicating a fairly high level of cohesiveness. Group II gave itself a 4.0, somewhat more favorable than


59 did Group I. Group III, in marked contrast to all the others, ^ated itslef a 1.2 indicating that they felt there was no cohesiveness at all. Croup IV produced a 3.0 ratina, aqain a statement of a fair level of cohesiveness. The oreatest rancre obtained for any group was one (.1) . Discriminant Function Analysis Discriminant Function Analysis was completed on the profile shapes among the groups. The results of this analysis are given in Appendix D. As the data were judaed inapnrooriate for this type of analysis, the results, as expected, do not conform to the nrinciole findings given above .


CHAPTER VI DISCUSSION The usefulness of video-taped feedback to stimulate and bring about positive changes in reported self -concept was borne out by this study. Group III, the only experimental condition in which video-taped feedback was not employed, differed significantly from all other groups (including the control group) in reporting the most negative mean SC rating. The two video-taped feedback conditions and the control, '-'hile not differing significantly from one another, arrayed themselves with the combined treatment group (IV) just below Group III. That is, there may be an underlying trend which suggests that the addition of judges' ratings to a video-taped feedback condition mitigates the effectiveness of the video-tape itself. It is apparent, at least in terms of the SC dimension, that, when given a choice of responding to different stimuli, the 5s behaved as if only the video-tape were present. The array of means on the GSC scale parallel those of the SC scale with the interchanging of Groups 60 -


61 I and II. Of interest is the observation that while the groups differed significantly on both the SC and GSC dimensions when analyzed separately, there were no differences in the difference scores between SC and GSC across all groups. This suggests that the discrepancy between the two scales remains constant across individuals and groups so that if the SC scores go up (become more negative), so do the GSC scores. Perhaps the GSC acts as an anchor point for the individual which would allow only a certain degree of difference between the selfconcept as it is and the self-concept as desired. Too much discrepancy would likely prove exceedingly threatening by indicating to the individual that he was so far from his ideal-self as to make it an unattainable goal. It would be of interest to obtain the discrepancy scores from nonnormal populations to ascertain if this anchoring effect perhaps is missing from these groups. The failure of this experiment to produce changes in the discrepancy between SC and GSC may be due to the brevity of the project. Although the groups met five times, they did so over a period of 2% weeks. The long-established distance between the two scales would


62 likely take considerable pressure to yield. The resistance to reducing the disparity m?.y be a function of the self-exploration and multiple pressures toward change in the environment which are unique to this sample of 5s. Closure of the gap between SC and GSC implies the conclusion of a growth and exploration process. It suggests that the individual has settled upon a fairly definite understanding of who he is and, more importantly, who he can become. It is unlikely that this particular sample of 5s, or any other group of college undergraduates, would have attained this degree of closure. In fact, the author questions if complete congruence between SC and GSC would even be desirable at any age. Surely, growth must be a never-ending process because once it ceases a certain degree of torpor would set in, and, eventually, a deterioration of openness and spontaneity to new events would occur. In reference to the observed PRO scores, it is interesting to note that Group III, which recived only judges' ratings, reported the most negative PRO scores. Tangent ially, the author notes that, for the most part, the judges rated all 5s in all groups more negatively


63 than the 5s did themselves. Thus, the Group III 5s were receiving written information about how others viewed them and this information was, largely, of a more negative character than the self -ratings. When given the opportunity to rate how they believed others would see them, the Grouo III 5s directly reflected the judges' perception of them. Somewhat surprisingly, the 5s in Group IV did not produce negative PRO scores although they, too, received the judges' ratings. In fact, Groups II and IV did not differ from one another, indicating that when provided with both types of feedback, the judges' ratings were apparently disregarded. As with the results on the SC dimension, the Group IV 5s behaved as if only the videotape were present. It is evident that the message incorporated in the judges' ratings was too dissonant for the 5s so, vhen they could, they rejected it. The 5s in Group III could have ignored the judges' ratings but apparently found it less threatening than completing the PRO without any reference to the criterion at hand. There may also have been an unintended constraint in this technique whereby the 5s knew that the E would see both


64 the judges' ratings and the PRO ratings. Thus, the 5s may have felt they would appear naive or stubborn if they completely ignored the ARO feedback. The Group IV 5s, while rejectina the ARO, could at least claim to be responding to the video-taped feedback. The absence of significant differences across groups between SC-ARO and PRO-ARO suggests again that constant differences between the respective scales was maintained. Both the SC and PRO scales yielded significantly different scores between groups, xo it appears that they fluctuated parallelly to the ARO scale. More precisely, the 5-endorsed part of the specific comparisons between SC-ARO and PRO-ARO was obtained after the ratings of the judges were delivered to the 5s (or after they viewed the appropriate video-tape) . Thus, in Groups III and IV the 5s reported their SC and PRO after they had reviewed the judges 1 ratings. It would appear that the discrepancy between the scales was maintained as a result of information the 5s had about the judges' ratings. For the 5s in Groups I and II, who received no judges' ratings, the task of projecting how others might see oneself was decidedly more difficult,


65 •In the absence of any contradictory information, it would seem easiest to assume that "as I see myself, so do others see me." In other words, the 5s would be likely to report changes in the PRO that corresponded to their reported changes in the SC. If the 5 saw himself as a "better" person, he would likely attribute this perception to any other individual who might see him. The anxiety measures provide considerable contrast to the MSGO in terms of the rank-ordering of the groups. It was suggested that the level of anxiety would interact with changes in the self -concept in a curvilinear fashion. In other words, a minimum level of anxiety would be necessary to stimulate change, but excessive levels of anxiety would act to constrain the changes. Among Group III 5s, the reported level of State Anxiety was the lowest while the SC mean for that group was the most unfavorable. Thus, it appears that the group with the lowest anxiety level had a significantly more negative SC. As the groups did not differ from one another on any scale at the pretest, it appears that the change toward a more negative SC and the presence of a low State Anxiety occurred in tandem. Additionally, the


66 most negative PRO occurred in this group. The fact that the same 5s who reported that they believed others saw them negatively and who saw themselves more negatively than the 5s in any other group also reported the lowest State Anxiety is puzzling and difficult to understand. One possible explanation may lie in an "innoculation" effect. In Group III, the level of State Anxiety reached its peak on the second experimental day, representing the protocol that was filled out immediately after the first exposure to the judges 1 ratings. The effect of this unique and apparently quite threatening situation may have acted to numb the 5s to any further feedback from the judges for some period of time. A similar peak and then decline was evident in Group IV 5s who also received judges' ratings as feedback. The feeling among the members of Group III that their group lacked almost entirely any feeling of cohesiveness undoubtedly reflected (or influenced) the observed data. It may be that this group "gave up" in terms of participation in the experiment and parrotted the information qiven to them but did not attempt to incorporate it in any consistent fashion.


67 The variability noted in the effect of the experimental situation across days from one group to another is of interest. It appears that the greatest impact, emotionally, is obtained by providing the Ss with concrete, readable feedback on where they appear to others, The State Anxiety rose earliest in Groups III and IV and appeared to decline most sharply. The video-tape alone appeared to induce a situation of gradually increasing anxiety through the middle of the experiment and then a somewhat more steep decline. The nofeedback control, however, experienced its high level of State Anxiety at the latest Doint in time and the recovery to pretest levels may only have been due to the perceived termination of the experiment. Thus, it appears that the 5s anxiety habituated to the judges' ratings faster than to the video-tape and that the absence of concrete feedback left the control individuals "up in the air" longer. The overall impression one gets from the data is that the video-tape is a useful means of providing feedback to individuals. When paired with judges' ratings, however, the latter is ignored. This may reflect the true interpretation that 5s give to the


68 written feedback. It may be too discordant with already-held perceptions or it may fall prey to a thought similar to "I can see myself right there on the screen and I don't look like that," where "that" refers to the way in which the judges reported their perceptions. The uniqueness of a direct confrontation between ideas held about the self and ideas others hold about oneself without the opportunity to question the rater, argue with him, or even understand much about him undoubtedly contributes to the apparent rejection of the feedback. At the end of the experiment a number of 5s questioned the E quite vigorously about the qualifications of the raters. This appeared to be an attempt to resolve some of the dissonance aroused by discrediting the judges. The extreme position of Group III on all dimensions poses a question regarding the real effects of the experimental treatment. If the treatment was as anxietyprovoking as it appeared and if the 5s cognitively rejected participation in the study, then the true nature of the effects of judges* ratings is still unknown. The results from the combined video-tape and judges' ratings group (IV) also seem to indicate a cognitive rejection of the latter form of feedback.


69 The reaction of the experimental Ss to a one-way authoritative communication about themselves, both as measured by the test instruments and as it developed in the attempts to understand and discredit the judges, gives some indication of how great an impact this mode of feedback can have. Further exploration of this feedback modality should undoubtedly be done, especially in view of the observed outcome of this study. As an aside to experimenters who may enter this area of research in the future, the E would like to establish the fact that the use of video-tape equipment is not as complex as it would appear. In fact with a brief period of training on the mechanism for superimposing one picture on another, the entire task becomes as simple as focusing a lens and turning on a tape recorder. The E highly recommends this new and highly versatile means of exploring human behavior. Additionally, an unexpected outcome of this study was the necessity of entering into a new method of data analysis. While the use of logarithmic transformations in statistics is not new, the author did not encounter a single reported instance of the use of this


-70 procedure in any of the prior research in the area. This, in spite of the fact that the transformation procedure seems ideally suited, in fact is essential to the type of data derived from small -group interaction research. Certainly any future small-group researcher should investigate the use of logarithmic transformations of his observed data.






INSTRUCTIONS Good evenincr. I'm Anita Mehlman, a doctoral candidate in clinical psychology. You have all volunteered to participate in a six-session experiment in group interaction, although some of you v;ill be receiving credit for your participation as required by your respective psychology classes. Let's talk about that experimental credit for a minute. As you know, the experiment is for six credit hours contingent upon your completing the experiment. You also know that you have the option of terminating the experiment at any time without penalty. I am going to make a special request in line with that option however. Because this experiment is concerned with group interaction and communication, it is essential that the composition of the crroups remain stable. I will be explaining in some detail the nature of the experiment so please listen with the idea in mind that if you don't like what you hear you are free to leave, no questions asked. If, however, you remain this evening and complete the pretests, I will expect you to honor your commitment to continue throughout the entire experiment. 73 -


74 One other bit of information is necessary at this time. The experimental design requires that you be randomly assigned to one of four groups. In other words you do not, unfortunately, have a choice about what group you'll be in. The groups will be meeting on one of two schedules, either one beginning next week. Two groups will be meeting on Monday, Wednesday, and Friday of next week and Monday and Wednesday of the following week. The remaining groups will meet on Tuesday, Thursday, and Saturday of next week and Tuesday and Thursday of the following week. All meetings, with the exception of the Saturday one, will be in the evening somewhere between 6:30 and 9:30. The Saturday meeting time will be decided by each group at the first session. If you will not be free for all of these times so that I can assign you randomly, then I will not be able to use you as a subject. Are there any questions so far? (There were none.) Okay. This experiment is comprised of four growth groups (some people call them encounter groups) . They will be led by graduate students in clinical psychology who have had considerable past experience in leading groups.


75 All of the grouns will be video-taped through a one-way mirror. The camera will not be running all the time but will be turned on occasionally durinq each group meeting. Each of these video-tapes will be viewed by two raters, staff members in the Department of Clinical Psychology. This experiment is designed to explore some aspects of group interaction and communication. No tricks will be played on you. You will not be required to reveal private thoughts and feelings. The only expectation is that each of you will enter into this experiment openly and with a genuine interest in the group's interaction and communication process. Are there any questions at this time? (There were none.) If any of you feel that you would not be comfortable participating in this study, please feel free to leave. (No one did.) All right. By your presence here now I assume you all intend to pursue the experiment to its conclusion. If at any time during the experiment you feel that you cannot continue, discuss it with either your group leader or me before dropping out. If you just stop


76 coming to the meetings, it will make it impossible to assess the outcome of the experiment. Please fill these rating instruments out in the order in which they are passed out. Your responses will be held in confidence and I will be the only one to see them. If you have any questions about the directions, please ask them now. You may leave as soon as you complete the forms. The group assignments will be posted in the Psychology Building this Friday. Thank you for your interest and participation.




GROUP MEANS OF MSGO AND STAI SCALES KSGO-SC MSGO-GSC MSGO-PRO I 1.6811 1.4115 1.6794 II 1.6644 1.4154 1.6681 III 1.7527 1.5684 1.7584 IV 1.6854 1.4615 1.6720 STAI-X, STAI-X 2 I 1.6519 1.6262 II 1.5605 1.5740 III 1.5052 1.4752 IV 1.5803 1.5783 78 -




70 60 -to o u o o w 50 -o u c to w 40 30 , . . Normal Population Pretest Post test i SAV — r Ex i Fr T S -i r Sr Sa "T" Nc 1— Sy T A -1 C Scales POI Composite for Group I 80 -


70 60 -" n o u o o w 'C u •0 c « p (A 50 -40 -30 -~ Normal Population Pretest Posttest 1™ Ex SAV I Fr i s "T" Sr Sa Nc Sy i A i C Scales POI Composite for Group II 91 -


70 60 10 O u o o w •a 50 u a >v a m -p w 40 -30 -Normal Population Pretest Posttest "— T i 1 SAV Ex Pr S Sr Sa Scales Nc Sy A C POI Composite for GrouD III 82 -


70' 60' w Q) U o w



DISCRIMINANT FUNCTION ANALYSIS A stepwise discriminant function analysis (Bimed Proaram BMD 07M, 196R) made comparisons amono the crroup profile shapes possible. Once again, the MSGO subscales were analyzed individually at first and then in terms of SC-ARO and PRO-ARO discrepancies. The orofile analysis provides insiaht into day-by-day differences among crroups as well as overall grouo differences. For the SC suhscale, the profile analysis indicated that both crroups II and III differed significantly from Group IV (F„ 15 = 3.52, p<.05; F H 15 = 3.67, d<. a 5) across all days, while on day 3, all groups differed sicrnif icantly from Croup IV. The significance values for the comparisons were: I-IV (F 3 is = 5.07, p<.05), II-IV (F s j 6 4.47, p<.05), and III-IV (F 3 i6 = 7.88, p<.01). The observed mean for Group IV was, in this case, siqnif icantly lower than for each of the other orouns, indicatina a more favorable self -concept . 85 -


Sfi No sicmi^icant profile shape differences were obtained for the GSC subscale. Profile analysis of the PRO subscale yielded the same array of sicmificant values across all days (I-IV: Fu is 4.10, p<.05; II-IV: F* is = 4 .38, p<.05; III-IV: Fi, 15 = 5. PR, p<. n D but no significant day-byday differences amoncr arouns . The Ss in Grouo IV reDorted the hiahest means, indicatina a less favorable endorsement of the PRO items. Testing the discrenancy between SC-ARO and PRO-ARO in reaard to profile shape differences yielded the following between-grouo differences. For SC-ARO, Groups III and IV were significantly different in shape (t?,, l5 = 4. "3, p<. n 5) where Group III demonstrated the greater reduction in discrenancy over time. On the PRO-ARO dimension, Croun II had a significantly Greater reduction in discrepancy than Group IV {F u j 5 = 3.07, p<. 05). The discriminant analysis orogram also yields a measure of within qrouo variance across all days. On the SC scale, only Group IV was siqnificant (Fi» is = 7.57, p<.01). There were no significant within grouo variances on the GSC scale. On the PRO scale, Group IV was


87 significant (F k is = 12.30, p<.01). For the discrepancy between SC-ARO, both HrouD II and IV were significant, the former at the .05 level (Fi^is 3.62) and the latter at the .01 level (P., is = 7.26). rroup IV was the only r one with significant within aroup variance on the PRO-ARO dimension (s^ is = 5.42, tx.Ol). As is readily apparent r from these fiaures, the greatest variability occurred in GrouD IV. Profile analysis of the state anxiety dimension (Xi ) did not yield significant F-values for differences across days, nor were there any significant within group variances. On the X2 scale (trait anxiety) , profile analysis indicated that Grouns I and II (Fi, is =3.44, n<.05), r I and IV (Fi, is = 6.91, p<.01), nnd II and IV (Pi, is = 3.88, p<.05) differed sianificantly from one another. It was also observed that significant within aroup variances occurred across all days in Group I (Ft, is 3.60, p<.05) and Group IV (F% is = 5.26, p<.01).




MISKTMINS 1 SELF-GOAL-OTHERS DISCREPANCY SCALE Intelligent Creative and Oriqinal Physically Attractive Successful in Life Competent for Many Jobs Friendly and Warm 123456789 123456789 Ignorant Not Creative and Oricrinal Physically 123456789 Unattractive Unsuccessful 123456789 in Life Not Fit for 123456789 Any Job Unfriendly 1234 5 6789 and Cold Prefer Being With People Good Relations With the Opposite Sex Socially Skillful Concerned for Others Prefer Beinq 1234 5 6789 Alone 123456789 1234 5 6789 1234 5 6789 Poor Relations With the Opposite Sex Awkward Socially Not Concerned for Others Happy Relaxed 123456789 Sad 123456789 Tense Hioh SelfConfidence 123456789 Lack SelfConfidence Handle Personal Problems 123456789 Can't Handle Personal Problems Alert and Active 1 2 6 7 8 9 Dull and Lifeless 89 -

PAGE 100

BIBLIOGRAPHY Berger, M. , Sherman, B. f Spalding, Janet, and Westlako, R. The use of videotape with psychotherapy groups in a community mental health service program. Int. J. Group Psychother. , 1968, 18, 504-515. Bierer, J. and Strom-Olsen, R. The recording of psychotherapeutic sessions, its value in teaching, research, md treatment. The Lancet, 1948, 254, 957-958. Biomedical Computer Programs (Dixon, W. J., Ed.), Berkeley, Univ. Calif. Press, 1968. Boyd, H. and Sisney, V. Immediate self-imaqe confrontation and changes in self-concept. J. Consult. Psychol. , 1967, 31, 291-296. Braucht, G. Immediate effects of self -confrontation on the self -concent . J. Consult. Clin. Psychol. , 1970, 35, 95-101. Cornelison, P. and Arsenian, Jean. A study of the response of psychotic patients to photographic self-imaae experience. Psychiat. Quart., 1960, 34, 1-8. Danet, B. Videotape playback as a therapeutic device in qrouo psychotherapy. Int. J. Croup Psuchother. , 1969, 19, 433-440. Elbert, W. Chancres in self -concept, self -actualization, and interpersonal relations as a result of video feedback in sensitivity traininq. dissert. Ahstr., 1970, 30, 12A, 5233. 90 -

PAGE 101

91 Fadale, V. An experimental study of the effects of videotape feedback in a basic encounter group. Dissert. Abstr., 1970, 30, 12A, 5234. Farber, D. Written communication in psychotherapy . Psychiatry, 1953, 16, 365-374. Festinger, L. A Theory of Cognitive dissonance , New York, Harper and Row, 1957. Geertsma, R. and Reivich, R. Repetitive self -observation by videotape playback. J. Nerv. Vent. Dis., 1965, 141, 29-41. Gill, M. , Newman, R. and Redlich, P. The Initial Interview in Psychiatric Practice , New York, International Universities Press, 1954. Golner, J. H., Geddes, H. H., and Arsenian, J. Notes on the use of recorded minutes in arouo therapy with chronic psychotic patients. Psychiat. Quart. , 1959, 33, 312-325. Hays, W. Statistics for Psychologists , New York, Holt, Rinehart, and Winston, 1963. Hoqan , P. and Alcrer , I. The impact of videotape recordina on insiaht in qroup psychotherapy . Int. J. Group Psychother . , 1969, 19, 158-164. Hum, S. An investiaation of the use of focused videotape feedback in hiqh school oroup counselina. Dissert. Abstr., 1969, 29, 12A, 4284. Kagan, N. and Schauble, P. Affect simulation in interpersonal process recall. J. Counsel. Psychol. , 1969, 16, 309-313. Kagan, N., Schauble, P., Resnikoff, A., Danish, S., ind Krathwohl, D. Interpersonal process recall. J. Nerv. Ment. Dis., 1969, 148, 365-374.

PAGE 102

92 Kendall, M. G. ffa?ifc Correlation Methods, London, Griffin Press, 1948. filler, D. The effects of immediate and delayed audiotape and videotape playback of qrouo counseling. Dissert. Abstr. , 1970, 30, 8R, °.R72. Miller, M. Responses of psychiatric patients to their photoqraohed se If -images . Diseases Nerv. Pyst. , 1962, 23, 296-298. Miskimins, R. Manual: Self -Goal-Others Discrepancy Scale, Ft. Collins, Rocky Mountain Behavioral Sciences Institute, 1968. Moore, P., Chernell, E., and West, M. Television as a therapeutic tool. Arch, Cener. Psychiat. , 1965, 12, 117-120. Nielson, G. Studies in Self -Confrontation , Copenhagan, Munkogaard, 1964. Pinney, E. L. The use of recorded minutes of group meetincrs in group psychotherapy: A preliminary report on a nev; technique. Psuchiat. Quart. Supp. , 195 r >, 29, 1-8. Reusch, J. Disturbed Communication, New York, W. W. Norton, 1957. Rogers, C. R. A theory of therapy, personality, and interpersonal relationships as developed in the client-centered framework. In S. Koch (Ed.) Psychology: A study of a Science, Vol. Ill, New York, McGraw-Hill, 1959, 184-256. Rogers, C. R. On Becoming a Person, Boston, HoughtonMifflin, 1961. Roaers, C. R. Toward a modern aporoach to values: The valuinq process in the mature person. J. Abnorm. Soc. Psychol., 1964, 68, 160-167.

PAGE 103

93 Sanford, E. An acoustic mirror in psychotherapy. Amer. J. Psychother. , 1969, 23, 681-695. Schloss, J. The effect of video and audio playback in aroup counseling on personality change. Dissert. Abstr., 1970, 30, RA, 32R5. Shostrom, E. L. Personal Orientation Inventory , San Dingo, Educational and Industrial Testing Service, 1962. Spielberger, C . , Gortisch, R., ind Lushene , R. State-Trait Anxiety Inventory , Palo Alto, Consulting Psycholoaists Press, 1970. Stoller, P. Groun psychotherapy on TV: An innovation with hospitalized patients. Amer. Psychol . , 1967, 22, 15R-162. Sullivan, H. Conceptions of Modern Psychiatry , Washinaton, D.C., Win. A. White Psychiatric Foundation, 1947. Sullivan, H. The Interpersonal Theory of Psychiatry , New York, Norton, 1953. Swallow, R. Ancient Chinese Bronze Mirrors, Peining, Henri Vitch, 1937. Wolberg, L. The Technique of Psychotherapy , New York, Grune and Stratton, 1954.

PAGE 104

BIOGRAPHICAL SKETCH Anita Cooper Mehlman, n£e Anita Helene Cooper, was born on August 6, 1946, in Brooklyn, New York. The daughter of a career Air Force officer, she attended public schools in California, Ohio, New Jersey, Utah, and Colorado. She was graduated from Air Academy High School at the United States Air Force Academy, Colorado, in 1963. In September 1963 she enrolled at the University of Florida from which she received a Bachelor of Arts degree with honors in 1967 and a Master of Arts degree, "ith a major in psychology, in 1969. Mrs. Mehlman is a member of Phi Beta Kappa and received both a three-vear National Defense Education Act Fellowship and a Graduate School Fellowship for her graduate studies. She is the wife of Henry Edwin Mehlman, an attorney, currently working in the Florida State Legislature.

PAGE 105

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation end is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. C?<
PAGE 106

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 deqree of Doctor of Philosophy. Richard McCee ku^g^. Associate 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. [JZM/W ^7-t*^v^ ^L David Lane Professor of Education This dissertation was submitted, to the Dean of the College of Arts and Sciences and to the Graduate Council, and was accented as partial -fulfillment of the reouirements for the degree of Doctor of Philosophy. March, 1972 Dean, College of Arts and Sciences Dean, Graduate School

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 E402FYK0S_JO3KJD INGEST_TIME 2014-06-17T18:22:50Z PACKAGE AA00022245_00001