UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A task analysis of conflict resolution 1980

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1980_A8 T39.pdf [ 3.6MB ]
Metadata
JSON: 1.0054337.json
JSON-LD: 1.0054337+ld.json
RDF/XML (Pretty): 1.0054337.xml
RDF/JSON: 1.0054337+rdf.json
Turtle: 1.0054337+rdf-turtle.txt
N-Triples: 1.0054337+rdf-ntriples.txt
Citation
1.0054337.ris

Full Text

A TASK ANALYSIS OF CONFLICT RESOLUTION B.Ed., University of V ictor ia , 1974 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES* (Department of Counselling Psychology) We accept this thesis as conforming to the required standard UNIVERSITY OF BRITISH COLUMBIA October, 1980 (T) Lin Taylor, 1980 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Co lumb ia , I ag ree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i thout my w r i t t e n p e r m i s s i o n . Department o f C fiUlt* SELL (ft) fc IfvLOfr V The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date io/So (i) ABSTRACT Twenty four in therapy cl ient performances were analysed in order to describe patterns related to confl ict resolution. Twelve Gestalt events in which resolution was achieved were compared with twelve nonresolution events. Both groups, resolved and unresolved, were analyzed using three measures,-- 1) Experiencing scale; 2) Structural Analysis of Social Behavior, and 3) Vocal Quality Scale. Using these measures, the groups were divided into the three phases of confl ict resolution--opposition, merging and resolution. In Gestalt, therapy, the cl ient engages in a dialogue with himself to explore the two sides of the conf l ic t . He usualy moves from one chair to another, these chairs are known as the "experiencing" and "other" chair. In this study each chair was analyzed independently. The main hypothesis, that resolution events exhibit specif ic phase related behaviors that are not evident in non-resolution events, was tested using a number of comparisons. These comparisons between groups, between phases and between the two chairs indicate that the two groups are different and that there are three identif iable phases in a resolu- tion event. The pattern of resolution begins with the two chairs in the opposition phase rejecting each other. This is indicated by a difference in level of Experiencing and uncooperative interaction as measured by S.A.S.B. The merging phase is when the two chairs begin to engage in productive dialogue. The other chair "softens" toward the experiencing chair , as indicated by a change in the vocal quality at the merging point. Also during this phase, the level of experiencing of the other chair increases to the level of the experiencing chair. The resolution ph-ase is identif ied by the lack of difference between the two chairs; they apparently come together and function as one. ( i i ) TABLE OF CONTENTS List of Tabies Acknowledgement Chapter I INTRODUCTION Definition of the Problem Definition of Terms II LITERATURE REVIEW Task Analysis Research on the Two Chair Operation and Conflict Resolution Rationale for Hypothesis Hypotheses III METHODOLOGY Measuring Instruments 1 . The Experiencing Scale 2. Structural Analysis of Social Behavior S.A.S.B.. Client Vocal Quality Scale Target Complaints Discomfort Box Scale (Appendix D) Conflict Resolution Box Scale (Appendix Shift in Awareness Design Subject Selection Therapists ( i i i ) Table of Contents (cont'd.) Page Raters 36 Data Collection 37 Scoring Procedure 37 Procedure of Analysis 38 Tables 39 (a) to 39 (k) IV RESULTS Results of Between Group Comparisons 40 Results of Phase Hypotheses in the Resolution Group 42 V DISCUSSION, CONCLUSIONS, RECOMMENDATIONS 45 Discussion 45 Experiencing Scale Patterns 45 Vocal Quality Patterns 46 Structural Analysis of Social Behavior Patterns 47 Conclusions 48 Recommendations 50 Implications 51 Bibliography 52 Appendices 55 (iv) LIST OF TABLES Table .. Page I Phasing the non-resolution events 39 (a) II Example of Tabulation of Raw Data for Each Event 39 (b) III Mean Scores for each Chair in Each Phase of the Resolution Group on the Experiencing Scale 39 (c) IV Mean Scores for each Chair in Each Phase of the Non-Resolution Group on the Experiencing Scale 39 (d) V Proportion of Statements in each Quadrant of the Structural Analysis of Social Behavior Scale for the Resolution Group 39 (e) VI Proportion of Statements in each Quadrant of the Structural Analysis of Social Behavior Scale for the Non-Resolution Group 39 (f) VII Proportion of Statements in each Vocal Category for the Resolution Group 39 (g) VIII Proportion of Statements in each Vocal Category for the Non-Resolution Group 39 (h) IX Example of Data Tabulation for each Hypothesis 39 (i) X Frequency of A f f i l i a t i o n at the Merging Phase for the Two Groups 39 (j) XI Frequency of Focus Voice at the Merging for Resolution and Non-Resolution Events 39 (k) (v) ACKNOWLEDGMENTS Preparation and completion of this thesis would not have been possible without the support and guidance of Leslie Greenberg. I am especially grateful to him for his encourage- ment during the completion of this thesis. I would l ike to extend a special thanks to Steve Foster for his help in the research design and s tat is t ica l analysis of this project. I appreciate that Norm Amundsen was s t i l l available after two years to be on my committee. - 1 - CHAPTER I INTRODUCTION Resolution of a sp l i t by means of the two chair operation in Gestalt Therapy seems to be an important therapeutic process. This project is designed to study in detail the sequence of events that lead to reso- lut ion. In building on this information, an attempt wi l l be made to verify a proposed model of sp l i t resolution developed by Greenberg (1975) and Johnson (1979). If we as counsellors and therapists can learn the process of resolving a sp l i t or affective task, we wi l l be more effective in helping a cl ient resolve confl icts and experience the result ing, hopefully bene- f i c i a l , change. This understanding of human behavior wi l l assist us in helping people to experience l i f e more f u l l y . Thus, i t is important to examine intensively how spl i ts are resolved rather than be content with evaluation of the final result . The purpose of this project is to focus on the moment to moment process in the solving of s p l i t s . The goal of this study is to specify components of successfully resolved intrapsychic conf l ic ts , then, examine a number of successful performances for occurrence of these components. The f inal step is to compare the components in successfully resolved confl icts with those that are not resolved. It is hypothesized that in unsuccessful performances one or more of the identif ied components wi l l not be evident. Such a study is termed a task analysis and is defined by Schwartz and Gottman (1976) as follows, - 2 - "Such a study begins by specifying the l ike ly components of a competent response and then testing the extent to which performance on the components discriminates between competent and incompetent populations." (p. 18). Definition of the Problem The premise upon which this study is based is that thera- peutic change, spec i f i ca l l y , resolution of a s p l i t using the two chair technique, follows specific identif iable patterns. We can discover these patterns of behavior which occur across cl ients, by systematically analysing and specifying the components of successful performances in the process of psychotherapy. In fact , using this method, Greenberg (1975) and Johnson (1979) developed a theo- ret ical model identifying the necessary behaviors or conditions a cl ient wi l l experience as they approach resolution of a confl ict s p l i t . The purpose of this project is to verify aspects of the model by applying i t to actual therapeutic tasks. More spec i f ica l l y , success- ful or resolved tasks wi l l be compared with unresolved tasks to test the model. If the process of the resolved sp l i t follows the sequence of events identified by the model, this wi l l be evidence in support of the assumption of val id i ty of the model. With more precise information on the process of change in therapy, the therapist wi l l be better able to create optimal conditions for confl ict resolution. Exploring the problem from a task analytic approach w i l l result in a more concrete educational model for affective problem solving. This project is part of an ongoing research program applying a task analysis to the study of confl ict resolution in psychothera- peutic events. It focuses on steps six and seven in the following steps - 3 - of a task analysis as defined by Greenberg (1975). 1. The researcher has an intuit ive theory of how people function. Perls (1951) defines the Gestalt outlook as follows: "The average person, having been raised in an atmos- phere fu l l of s p l i t s , has lost his Wholeness, his Integrity. To come together again he has to heal the dualism of his person, of his thinking, and of his language. He is accustomed to thinking of contrasts— of infanti le and mature, of body and mind, organism and environment, self and rea l i t y , as i f they were opposing ent i t ies . The unitary outlook which can dissolve such a dual ist ic approach is buried but not destroyed and, can be regained with wholesome advantage." (p. 45). 2. Based on this theory, a task is selected and an obser- ver's description of the task is made. In this case resolution of the confl ict sp l i t is the task. Greenberg (1975) defines the sp l i t as " . . . the person is torn between alternatives. There is an experience of two parts, of the self sp l i t into partial selves in opposition, rather than the experience of single integrated self in process The sp l i t can be identif ied by i ts verbal markers... "I should do this but I "I want to stop but I...", etc. The technique used to reintegrate the two opposing parts is the two chair operation, defined by Greenberg as follows: " . . . the person plays the role of both sides of the conf l i c t , usually locating each side in a separate chair, and proceeds to have an encounter between them." (p. 8). 3. The existence of the event as potent and recurring is empirically ver i f ied. Greenberg (1975) found that use of the two chair technique lead to greater depth of experiencing and change in awareness when compared with the client-centered technique. - 4 - 4. Given the general mode and the task, a description of a subject's possible performances is made. - - A "thought experiment" or the idealized case. This step in the task analytic approach is the topic of a thesis by Johnson (1979) in which she "attempts to create and check models which i l l u s t r a t e , in specific de ta i l , what actually occurs during a psychotherapeutic event." More spec i f ica l l y , "resolution of a sp l i t by means of the two chair operation of Gestalt Therapy." 5. The subject's actual performance of the task is observed and described. Greenberg (in press) explored this phase by objective observation of events and analysis assisted by the clients using the two chair operation. He concluded, "Intensive research of this nature allowed a detailed elaboration of some of the subtleties of therapeutic process and by so doing has opened new avenues for research and for c l in i ca l practise." (p.2). Building on the previous five stages of the task analytic approach, the following two stages are the focus of this research. 6. A specific model is developed satisfying the general model and the task description. This is accomplished by comparing the idealized and actual performances of the task. Using the work of Johnson and Greenberg (1975) a refined, more specific model wi l l be presented. This is the postdictive step in this approach. 7. Based on this model, the f inal step of the task analysis is performed. The researcher hypothesizes the behavior of the subject in the task. This is the predictive phase in this approach. - 5 - Actual therapeutic tasks wi l l be analysed for elements speci- fied in the previous model. The predicted outcome being that the sub- ject responds as defined by the model. If the hypothesized and observed behaviors are equivalent in relevant ways to the predicted behaviors, c redib i l i ty is added to the specific model, the task description and to the general model. The overall prediction for this study is that cl ients progress through three sequential stages as they work toward resolution of a personal conf l ic t . These stages and related behaviors can be measured and identif ied with the use of scales designed to measure in-process data. The f i r s t stage is the opposition phase. One side of the confl ict is dominant and the dialogue indicates that this part t r ies to get i t s own way by aggressing and intimidating the submissive side. The second or merging phase is characterized by each side of the confl ict stating their position. And f ina l l y in the resolution phase there is mutual l is tening, understanding and acceptance of each other. Definition of Terms Gestalt Two Chair Operation Greenberg (1975) defines the two chair operation as follows: "(The) Operation is a series of suggestions and obser- vations made by the therapist or fac i l i ta to r to clearly separate two aspects or partial tendencies of the self process and to fac i l i ta te direct communication between these. The purpose of the experiment is to maintain a process of demarcation and contact between these parts. The following underlying principles are presented in an attempt to convey the structure of the operation—the nature of what can be done to achieve the process goal. These principles serve as guides to the therapist's behavior. The five principles are: - 6 - 1) Maintenance of a contact boundary: Maintaining clear separation and contact between the partial aspect of the se l f . 2) Responsibility: Directing the person to use their ab i l i t i es to respond in accordance with the true nature of their experience. 3) Attending: Directing the person's attention to particular aspects of his present functioning. 4) Heightening: Highlighting aspects of experience by increasing the level of arousal. 5) Expressing: Making actual and specific that which is intel lectual or abstract. Particularizing experience by moving from theory to praxis." (p. 10). Greenberg (1976) separates and identif ies the characteristic behaviors of the two chairs. The "Experiencing chair" is the experien- cing part of the person, and in-process dialogue moves from whining and excusing to inner exploration and deeper levels of experiencing. Technically, the cl ient in this chair dialogues at deeper levels of experiencing than in the "other" chair when measured on the Experiencing Scale. (Klein, et a l . 1969). It also uses more focused and expressive voice as measured by Rice's (1967) voice quality system. The "other chair" is f i l l e d with other parts of the personality, other people and things. Typically, the person in this chair engages in low levels of experiencing and uses a lecturing voice. Spl i t Greenberg (1979) discusses the sp l i t as follows: "Instead of a single clear preference ar is ing , the person is torn between alternatives. There is an experience of two parts, of the self sp l i t into partial selves in opposition, rather than the experience of a single integrated self in process. Clearly identifying this sp l i t and sensing the opposed forces within, becoming aware of the confl ict between the two parts, represent the fundamental task for the cl ient in the experiment." (p. 5). Although this project is concerned with only one type of sp l i t (Confl ict) , a brief definit ion of the other two types wi l l be given since - 7 - they sometimes are transformed into Conflict s p l i t s . 1) Confl ict : Two partial aspects of the self are in opposi- tion to each other. For example would be, "I want the security of marriage but I also want the freedom to do whatever I want." 2) Subject-Object: One partial aspect of the self (the subject, I) does something which the other aspect (the object, self) is the recipient or observer. Example: "I judge myself." 3) Attr ibut ion: A feeling that is actually a tendency or part of the self is attributed to an outside object or person. Example: "My father says I should f inish the four years, but I keep trying to te l l him I'm not learning anything." Conflict Resolution Perls (1970) defines resolution as: "the reconcil iation of opposites so that they no longer waste the energy in useless struggle with each other but can join in productive combination and interplay." (p. 6 7 ) . More spec i f ica l l y , Greenberg discovered that resolution is typified by " . . . a shift at some point in the dialogue in the "other chair" to higher levels of experience and more focused, expressive voice, much as though the person in the "other chair" becomes less c r i t i c a l , softer and more understanding or accepting of the self."(1979, p. 321). Resolution Phase In keeping with the preceding definition of confl ict resolution, for the purpose of this research, the resolution phase for any individual - 8 - is defined as that portion of the Gestalt event beginning with a level six score on the experiencing scale in either the other or the experien- cing chair. Merging Phase The merging phase occurs when the other chair begins to a f f i l i a t e with the experiencing chair. As defined by the S.A.S.B. scale scores, merging occurs when the other chair responds with more than two responses from quadrants one and/or four and continues with a higher pro- portion of quadrant one and four responses. The merging phase ends when the resolution phase begins. Opposition Phase The opposition phase is the portion of the Gestalt event which precedes the merging phase. Typically, i t has lower levels of experien- cing, more external voice and more responses from quadrants two and three of the S.A.S.B. scale in both the other and experiencing chairs. Experiencing Klein, et a l . (1969) refer to experiencing as the quality of a person's experiencing. It is the extent to which a person is aware of and can communicate about "their bodily fe l t flow of experiencing and the extent to which this is integrated with the person's action, and thought." Low level experiencing is typif ied by a lack of description of feelings and by impersonal, superficial dialogue. At an intermediate level of experiencing, the clients may describe and talk about their feelings. The greatest depth of experiencing is when the cl ients explore their immediate feelings in the here and now with the result of increased awareness and resolution of a problem situation. - 9 - Voice Quality The four categories for assessing voice are: A. Focused B. Externalized C. Limited D. Emotional These four categories include the following six features in varying quantities (Rice and Wagstaff 1975). 1) Energy 2) Primary stress 3) Regularity of stress 4) Pace 5) Timbre 6) Contours Greenberg and Rice (1979) describe voice quality as a measure of involvement and processing levels in the moment. They suggest that we can expect more focused voice in a good hour of therapy. (See Appendix A). Task Analysis Greenberg (1975) describes task analysis as " . . .an evolving technique for describing and analyzing human behavior in problem solving tasks. Task analysis is a method of analyzing a specific performance situat ion, in l ight of a general model of a psychological system, in order to construct a specific model which could generate the particular performance." (p. 12). A fu l ler explanation of task analysis and i ts relevance to research in the f ie ld of psychotherapy, follows in the l i terature review. - 10 - Structural Analysis of Social Behavior (S.A.S.B.) Benjamin (1977) has developed a model which appears to be effective in analysis and description of pathological and constructive social process. Using this system as a measure of social interaction, in this project, wi l l hopefully reveal that as a person works towards a resolution, the interaction between the two chairs changes. The S.A.S.B. model is an extension of Leary's c lass i f icat ion system for the study of interpersonal transactions in a c l in ica l setting. Benjamin (1977) describes the model as " . . . a mathematically defined, empirically substantiated extension of the work by Leary (1957), Schaeffer (1965) and many othersi1 1 (p. 3). This model is suff ic ient ly complex so as to encompass c l in ica l concepts in most therapies - psychoanalysis, family therapies, Gestalt, etc. It is not restricted to a particular theoretical approach and can be used in many contexts such as therapist -c l ient , parent-child or in t ra - psychically, experiencing self-other se l f . Benjamin describes the purpose of such a system, " . . . t o organize the c l in i ca l and folk wisdom in a way which wi l l make this knowledge more amenable to sc ient i f i c procedures." (p. 18). Benjamin (1977) demonstrated the use of the S.A.B.C. "in setting psycho-social treatment goals and in using before, during and after self ratings to establish efficacy of the therapy." (p. 22). The main question to be studied is whether the two groups can be discriminated according to measures of vocal quality, experiencing and structural analyses of social behavior. When this has been determined the resolution group wi l l be further investigated to see more speci f ical ly how - 11 - i t differs from the non resolution event. Considering this study focuses on behavioral events rather than individuals there is l i t t l e known about the effect of individual differences on confl ict resolution. Further investigation needs to be done to deter- mine i f individual differences are a factor. Although i t seems unlikely that an individual difference variable can explain the pattern given that the pattern was consistent across a l l the clients in a previous study. (Greenberg 1 975). - 12 - CHAPTER II LITERATURE REVIEW There are three sources of information that form the basis of this research project. The theory and basic premises come from the theory and practise of Gestalt therapy. More spec i f i ca l l y , for this project the writings of Perls (1951), Polster (1973), Zinker (1977), Grinder and Bandler (1975), Baumgardner (1975) and Greenberg (1977) on confl ict sp l i ts and resolution,,form the foundation. Concerning the new approach of applying a task analysis to research on psychotherapy, the l iterature is l imited. The recent works of Greenberg (1979), Rice, Pascual-Leone and Gottman (1975), but a few ar t i c les , are the total body of knowledge on the subject. Even more limited is the third area of interest to this research, that being studies on the Gestalt two chair technique in psycho- therapy. This information comes mainly from Bohart, Greenberg and U.B.C. students presently exploring and researching the process of psychotherapy. Part icular ly , the process of Gestalt Therapy using the two chair approach. Perls' apt description of the nature of modern man gives some of the background for Gestalt theory. "Modern man lives in a state of low-grade v i t a l i t y . Though generally he does not suffer deeply, he also knows l i t t l e of true creative l i v i n g . Instead of i t , he has become an anxious automaton He does not approach the adventure of l i f e with either excitement or zest. He seems to feel that the time for fun, for pleasure, for growing and learning, is childhood and youth, and he abdicates l i f e i t s e l f when he reaches maturity." He seems to have lost a l l spontaneity, - 13 - a l l capacity to feel and express directly and crea- t ively He spends endless time trying either to recapture the past or to mold the future. His present act iv i t ies are merely bothersome chores he has to get out of the way. At times he is not even aware of his actions at the moment. (Perls 1973,p. 27). Perls assumed that this blockage of energy is the result of being "fractionalized people". Every individual consists of innumerable unresolved conf l ic ts , spl i ts or polarit ies that when le f t unidentified and unfinished, leave the individual immobilized in a state of unpro- ductive confusion. "As long as the cl ient experiences only the extremes of any continuum, he has no centre, no experiencing of a self which gives and takes on i ts own terms with the world." (Baumgardner 1973, p..67). Perls (1951) refers to this impasse as the top dog and under dog being at war and reaching a stalemate. Only when the cl ient gives up the struggle for control of their parts and begins to l isten to both sides or accept their polarit ies are they free to act or grow. In any s p l i t , both roles usually emerge. The top dog is t yp i - cal ly control l ing, bullying, lecturing, threatening, omnipotent and moralist ic. On the contrary, the underdog controls passively. Underdog is helpless, passive, v indict ive, confused, apologetic, uncommited and procrastinating. Both roles attempt to manipulate to get what they want. Although the top dog appears to be more powerful, the underdog controls by being passive and avoiding, leaving the top dog frustrated. The result of this struggle is that nobody gets anything rea l . There is no integration or communication, only the struggle for control remains. The dialogue, using the two chair technique might go something 1 ike th i s : - 14 - Top dog: You should stand up for yourself. Underdog: I can't . I feel helpless. I can't change the way I deal with people. I don't know how to stand up for myself. I t 's just pointless. Top dog: What a cop'outl You make me sick. You complain about being pushed around but you don't do any- thing about i t . I think you l ike i t . There must be an integration of the two characteristics for the cl ient to be able to move. Polster emphasizes the need for integration rather than control of one of the sides by saying, "The effort devoted to keeping the squelched characteristic servile or si lent is a doomed e f f o r t — i t wi l l pop up in inconvenient ways to assert whatever val id i ty i t can muster, l ike a l l resistance forces which have been compelled to go underground'.'. ( P o l s t e r & P o l s t e r , 1973, p. 87). To be more speci f ic , Greenberg views identif ication of the s p l i t , sensing the opposed forces within and becoming aware of the two inner forces as the basic therapeutic task to be completed by the c l ient . (Greenberg, 1975). It is at this specific point in process, when the cl ient senses the sp l i t that the cl ient is most receptive to change. Recognition by the therapist of this particularly salient state may be the beginning of very productive work by the c l ient . A sp l i t can usually be identif ied by i ts verbal markers. Greenberg (1979) states, "The sp l i t is a verbal performance pattern mani- fested by one person (client) in interaction with another (.counsellor). The sp l i t is characterized by a division of the self process into two partial aspects of the se l f . These tendencies or partial aspects of the se l f , are related to each other in different ways and the d i f fe r - ent relationships between the tendencies define different types of s p l i t s " , (p. 317). A confl ict sp l i t typical ly is marked by the - 15 - statement, "I want to but I can't ." The person presents two obviously opposing sides of themselves that are in conf l i c t . It is at this point when the two sides make contact with each other, there is potential for resolution. The work of Bandler and Grinder (1976) is even more exten- sive on the topic of spl i ts or incongruences. More than a verbal performance pattern, the sp l i t is demonstrated in many "output channels" (body posture, movements, voice tempo, voice tonality and words). Bateson's comment..."the phenomenon familiar among humans where the friendliness of man's words may be contra- dicted by the.tension or aggressiveness of .his voice or posture. The man is engaging in a sort of dece i t . . . " (Bateson, 1976 - p. 136). implies that " . . . the nonverbal or analogical message is the one which fa i thfu l ly reflects the true nature of the person's feelings and intentions." On the contrary, Bandler and Grinder (1976) believe these numerous messages, termed paramessages, are a l l "true" and can generally be divided into two opposing or incongruent parts. At this point the therapist assists the cl ient in experiencing the polarity by encouraging the cl ient to be congruent in a l l their output channels on both sides of the s p l i t . This is accomplished by acute awareness on the part of the therapist, and his bringing any inconsistencies to the attention of the c l ient . The cl ient moves from one space or chair to another until the cl ient fu l ly experiences in a l l the output channels; feelings, gestures, voice quality, and postures of the two opposing parts. This method of treatment of a sp l i t is growing in popularity. It is a potent tool in helping the cl ient acknowledge both sides and - 16 - begin the process of integration. "By identifying the c l ient 's polar i - ties and then providing for the dialogue which can bring forth these two hostile roles, we create a place where the cl ient grows more wi l l ing to relinquish his struggle for control, at least for a moment, now and then, and to put some energy into l istening and hearing." (Baumgardner 1973 - p. 74). Although the very nature of the Gestalt approach defies definition and structure and depends on the creative intuit ion of the therapist, there are some basic principles used in the two chair tech- nique to assist the cl ient toward the process goal-resolution or inte- gration of two polar aspects of their characters. (Greenberg 1976). The therapist uses five basic techniques in Gestalt two chair work-- restoring contact, c l ient responsibi l i ty , attending, heightening, and expressing. The i n i t i a l and basic task is for the cl ient to restore contact between the opposing forces. The cl ient/in dialogue with himself as both sides of h i s . conf l ic t , begin to experience the d i f - ference and the val id i ty of each side. Polster (1973) notes, "Almost invariably, when contact is restored, the individual discovers that these disowned parts have many redeeming features and his l i f e expands when these are recovered." (p. 93). The second task is the cl ient taking responsibil ity for the conf l ic t . The therapist may intervene when the cl ient is not taking responsibil ity by avoiding, blocking awareness or ignoring feelings or experience. Speci f ica l ly , the cl ient is expected to "own" his experience by talking in the f i r s t person. He is encouraged to express honestly the true nature of both roles. He is asked also to identify - 17 - with a l l parts of his experience—the knot in his stomach, the tears or the high-pitched voice. Attending is another important principle used in the two chair approach. The therapist encourages increased awareness of a l l the c l ient 's experience. The therapist may ask the cl ient to stay with a particular feel ing, or draw his attention to some other behavior. The therapist might, for example, ask: "What are you doing with your foot?, or "Do you know what your voice is l ike?" Another principle used is called heightening. The therapist increases the impact of the experience by increasing arousal. This can be accomplished by encouraging the cl ient to exaggerate or repeat a statement. Or he might be asked to act out one aspect of his s p l i t . The therapist may also evoke a strong response by making expl ic i t some implicit message in the dialogue. Expressing is the technique of illuminating aspects of the experience by doing. The impact of an actual experience is often greater than the discussion about i t . This can be accomplished by having the cl ient expose the specific content of the inner dialogue. Have the cl ient express how he defeats himself in concrete terms. Greenberg (1977) found that when the previous "principles were applied to the two chair operation, there is an increase in scores on the Depth of Experiencing Scale; an index of productive psycho- therapy, and leads to resolutions with populations seeking counselling and with student volunteers." (Greenberg and Clarke, 1979). Gestalt therapists (Perls, Latner, Baumgartner, Polster, et al)) describe resolution, or a shift in awareness, or integration of the two warring parts in a kind of poetic way as i f i t were some mystical - 18 - happening. The polar aspects contact each other through dialogue and gradually integrate in a kind of synthesis, the result of the union being greater than the sum of the two parts. Perls (1970) described the process of resolution as "the reconcil iation of opposites so that they no longer waste energy in useless struggle with each other but can join in productive combination and interplay." (P- 7 2 ) The new trend is for Gestalt Therapy research "to move in the direction of finer discrimination of therapist interventions and more objective, illuminating measurement of c l ient process." C l a r k , 1977 (p. 8 ) . The result of this shi f t in focus is Greenberg's (1979) more technical description of resolution of a confl ict sp l i t in terms of scores on the Experiencing Scale (Klein, et a l . 1969) and on the Vocal Quality Scale (Rice and Wagstaff, 1967). "Resolution performances appear to be characterized by a shift at some point in the dialogue in the "other" chair to higher levels of experience and more focused- expressive voice, much as though the person in the "other" chair becomes less c r i t i c a l , softer and more understanding or accepting of the se l f . " (p. 13). Task Analysis Applying the task analytic approach to the study of process in psychotherapy is a sign of the move away from simple outcome studies in therapy. Rice, Greenberg and Pascual-Leone (1977) stress "the really interesting questions in psychotherapy research concern the step by step transactions between person and task situation. We need to sh i f t'our focus to study what takes place and how i t does so." What is task analysis? Gagne' (1974), the learning theorist , defined i t in terms of instruction, " . . . a procedure having the purpose - 19 - of identifying different kinds of performances which are outcomes of learning, in order to make possible the specification of optimal instruc- tional conditions for each kind of outcome a method of 'working back- wards' from intended learning outcome to the instructional s i tuat ion." (p. 8). When applied to Gestalt therapy, the outcome is confl ict reso- lution or a freeing up of the c l ient 's a b i l i t y to act or make a decision, and the instructional situation is the therapist intervention. As we can see with some refinements and modifications, task analysis can become an appropriate approach to the study of psychotherapy. Gottman and Swartz (1976) discuss the method of using task analysis in research as one in which one specifies the l i ke l y components of a successful response and then tests the extent to which performance on the components discriminates between successful and unsuccessful populations. Even more refined and specific to research in psychotherapy, Greenberg (1975) states the f i r s t step as breaking down the complex per- formance of a single therapy session into a series of events or tasks. For the purpose of this study, the confl ict sp l i t is the task under scrutiny. It is these performance events, not the individual cl ients involved, that are the focus. It is at this point that task analysis research differs from outcome research. (Rice and Greenberg, 1974). In this research, "the sp l i t " is a subtask in the Gestalt event to be studied. The sp l i t in this case is a kind of affective task, a problematic aspect of experience that cal ls for some kind of closure. This study examined twelve events that appear to have reached some kind of closure or resolution and compared them with twelve unresolved events. - 20 - Research on the Two Chair Operation and Conflict Resolution Rice stresses the importance of expanding the concept of the c l in ic ian -sc ient is t at a time when the trend is to research the effectiveness of various psychotherapeutic alternatives. They suggest the task analysis approach to research is appropriate to the study of psychotherapy because the cl ient in process "is actively working toward one or more goals". (Rice, et a l , 1978). Also, the steps of the task analytic approach can be adapted readily to the study of psychotherapy (as discussed previously in this paper). More spec i f ica l l y , Rice and Greenberg define the different strategy used in a task analytic approach to ensure that the behavior studied is homogenous. They claim that, "First we should ensure homo- geneity of behavior in the small groups selected for study and compari- son." "Secondly, and this is the crucial point, instead of selecting groups of c l ients , we should select for study homogenous groups of events in therapy. It is events not cl ients which we propose as the unit for study, events which can be recognized as having certain speci- f iable behavioral characteristics in common." (p. 2). This method is employed in this project. The studies in the area of Gestalt two chair work and reso- lution are l imited. Bohart (1976) found that in counselling analogue sessions, (established to study a particular effect using subjects) designed to resolve personal anger conf l i c ts , using a two chair role play was the only significant intervention. He states "role playing can be effective in modifying feelings, attitudes, and behaviors associated with interpersonal confl ict . . . the greater effectiveness of role play is in accord with the position that insight and emotion must - 21 - go hand in hand for change to occur." (p. 11). He also noted that role play seemed to be more effective because the cl ient began to become more accepting of the reasons for the provocateurs behavior when they were in role playing dialogue with themselves. In defense of the ana- logue procedure, Bohart argued that although the study was signif icantly different from actual counselling, this type of study-- in which pro- cedures are isolated—allows one to see the specific effects of that procedure. In another more convincing analogue study, Greenberg and Clarke (1979) found that "the 'two chair 1 operation is more effective than empathic reflection in deepening experiencing and bringing about changes in awareness when the cl ient is working on a s p l i t . " (p. 18). Depth of experiencing has been repeatedly shown to correlate with varied measures of successful outcome (Orlinsky and Howard, in press). Therefore, i f the therapist can fac i l i ta te high level experiencing, they are instigating change. In addition, Greenberg and Higgins (1980) found that the two chair operation when applied to a s p l i t , produced levels of experiencing signif icantly higher than the focusing inter- vention. in f u r t h e r s t u d i e s , Greenberg .(1975 and .1980) treated the two chairs as independent systems and found a consistent pattern of experien- cing in nine events. In the preresolution phase, the 'experiencing chair 1 functions at consistently higher level of experiencing (level four or above) than the 'other chai r ' . Then at 'the merging1 point the 'other chair' increases to levels similar to the 'experiencing chai r ' . In the resolution phase both chairs reach levels higher than four. - 22 - Greenberg (1980) claims that "this attainment of the 'merging point' by the 'other chair 1 can therefore be regarded as a suff icient con- dition of a resolution and a signal of the resolution phase." These same data were analysed for voice quality and found that the 'other chair 1 uses more of a combination of an energetic, out directed voice (external: X) and an energyless, restricted voice ( l imited: L) , whereas the 'experiencing chair* uses more of a combination of a high energy emotionally expressive voice (emotional: E and focused: F). From the voice data, i t can be concluded that the 'other chair 1 is less involved and makes poor contact with i t s e l f and the 'experiencing cha i r ' , although the proportion of this voice in the 'other chair 1 decreases in the resolution phase, significant at the .05 leve l . This change appears to occur around the merging point, although there is sometimes focused voice before and, more often, after . Greenberg con- cludes: "Change to focused voice does, however, appear to be a neces- sary condition for resolution. This change of voice by the other chair i s , therefore, an important therapeutic cue. When this voice change in the 'other chair' is accompanied by an increase in experiencing to the level of the 'experiencing cha i r 1 , the demarcation event has entered the. resolution phase." (p. 14). The findings of this study seem to support the concept of "a reconciliation of two parts by integration. In the resolution phase, the 'other chair' appears to soften, i t becomes similar in style to the experiencing chair, is more involved and subjective and describes i t s own feelings more personally." (p. 27). The l e v e l s i n t h e d e p t h of e x p e r i e n c i n g i n the two..chairs b e g i n t o i n t e g r a t e . The v o i c e d a t a r e p r e s e n t - 23 - the change from a person lecturing at_ themselves to one who is in pro- ductive dialogue with themselves. Greenberg's (1975 and 1980) work is the beginning of the model building phase in a task analysis. Another study, presently in progress, breaks down the pro- cess of resolution to even more minute components. (Johnson, 1980). The model is developed in a concrete way such that each step along the way is ident i f ied, isolated, and l ike a task, needs to be completed before moving on to the next. There are thirteen steps in the model of con- f l i c t resolution developed by Johnson to be tested later by applying i t to a number of single cases. The following three stages have been abstracted from her more complex thirteen step process, including the major structures underlying resolution performance. The f i r s t stage is characterized by win-lose opposition. The top dog intimidates or threatens to get i t s way and the underdog passively resists with helplessness and avoidance. During the second or merging phase, both sides of the confl ict are stated with feeling and empathy for the other, and f ina l l y in the reso- lution phase there is mutual l i s tening, understanding and acceptance of each other. A kind of acknowledgement of the value of integrating both characterist ics, which were once in deadlock conf l i c t , in preparation for some concrete action or movement beyond the stalemate. Conclusively, considering the studies designed to employ the two chair technique, i t appears that i t may be a potent therapeutic tool in confl ict resolution. Rationale for Hypotheses Previous and ongoing research suggests that in successful - 24 - confl ict resolution, a cl ient moves through three identif iable phases. The premise of this study is that with better understanding of the pro- cess of therapy, the therapist w i l l be able to recreate the most optimal conditions for affective problem solving and the resulting change. The experiencing and voice quality scales used to extract the patterns and nuances of the process of therapy have been related to positive outcome in affective problem solving. (Klein, et a l . 1969) (Rice and Wagstaff, 1967). Also, the'very specific nature of the analy- sis of the S.A.S.B. and i ts wide application make i t a useful tool in understanding and identifying the therapeutic process. (Benjamin 1979). Findings for or against the hypotheses comparing the resolution and non-resolution groups, gives evidence concerning the val id i ty of the concept of the three phase process in confl ict resolution. Another aspect of the rationale is the present trend toward process and single case research rather than simple outcome studies. The adaptation and creation of analysis, which provide useful information for c l in ica l practice, are of growing importance for psychotherapy, research. (Bergin et a l . , 1978). Hypotheses Comparison Hypotheses The f i r s t set of hypotheses and questions compare the two groups, resolution and non-resolution. Hypotheses 1(a) The other chair of the resolution group wi l l achieve a f f i l i a t i o n signif icantly more often than wi l l the non-resolution group. - 25 - Hypothesis 1(b) The other chair of the resolution group wi l l have a higher proportion of quadrant one and four behaviors of the S.A.S.B. after the merging point (as defined by hypothesis l a ) . Assuming that there are two dist inct groups, as suggested by hypothesis 1(a) and 1(b), the following hypotheses are addressed. Hypothesis 2(a) In the resolution phase of the two groups, the resolution group wi l l have a signif icantly different proportion of F plus E voice in the other chair. Hypothesis 2(b) In the resolution phase of the two groups, the resolution group wi l l have a s ignif icant ly d i f - ferent proportion of F plus E voice in the experiencing chair. Hypothesis 3(a) In the merging phase of the two groups, the resolu- tion group wi l l have a s ignif icant ly different proportion of F plus E voice in the experiencing chair. Hypothesis 3(b) In the merging phase of the two groups, the resolu- tion group wi l l have a s ignif icant ly different proportion of F plus E voice in the other chair. Hypothesis 4 The other chair changes from external to focused at the beginning of the merging phase (three statements before and/or after) in the resolution group but not in the non-resolution group. As well as testing the previous hypotheses, the data were gathered and inspected in an effort to answer the following questions. - 26 - In further considering the relationship between the groups, i t is expected that 1. In the opposition phase of the two groups there wi l l be no significant difference in the proportion of F plus E voice in the other chair , and 2. In the opposition phase of the two groups there wi l l be no significant difference in the proportion of F plus E voice in the experiencing chair. Phase Hypotheses The second set of hypotheses and questions establish spec i f i - cal ly the difference in the phases in confl ict resolution performances. The hypotheses are stated in terms of expected scores, changes in scores and comparison of scores between phases using the three main measuring instruments .. . .experiencing scale, vocal quality scale, and structural analysis of social behavior, S.A.S.B. Hypotheses about Experiencing Hypothesis 5 In the opposition phase the experiencing chair scores at a signif icantly different level of experiencing than the other chair on the experiencing scale. Hypothesis 6 In the merging phase the other chair scores at a signif icantly different level of experiencing than in the opposition phase on the experiencing scale. As well as testing the previous hypotheses, the data were sc ru t i - nized to consider a further question. Considering the expected change - 27 - in level of experiencing in the other chair, i t is expected that 1. In the merging phase the level of experiencing of the other chair is not s ignif icantly different than that of the experiencing chair. Hypotheses about Vocal Quality Hypothesis 7 The experiencing chair uses a s ignif icant ly different proportion of F plus E voice than the other chair in the opposition phase. Hypothesis 8(a) The other chair uses a s ignif icantly different proportion of F plus E voice in the merging phase than in the opposition phase. Hypothesis 8(b) The other chair uses a s ignif icantly different proportion of F plus E voice in the resolution phase than in the opposition phase. The data were also scrutinized to consider the following question. 1. It is expected that in the non-resolution group there wi l l be no significant difference in the proportion of F plus E voice between the opposition and merging phases. Hypotheses about S.A.S.B. Hypothesis 9(a) The experiencing chair in the opposition phase wi l l have a s ignif icant ly different proportion of quadrant 2 and 3 behaviors than in the resolution phase. Hypothesis 9(b) The experiencing chair in the opposition phase wi l l have a signif icantly different proportion of quad- rant 2 and 3 behaviors than in the merging phase. - 28 - Hypothesis 10 The experiencing chair during the resolution phase wi l l have a signif icantly different proportion of quadrant 1 behaviors than from any other quadrant. - 29 - CHAPTER III METHODOLOGY Measuring Instruments 1. The Experiencing Scale The Experiencing Scale (Klein et a l . , 1969) (see Appendix A) wi l l be used to measure and compare the in-process level of experiencing in the two chairs. The scale was developed to evaluate the quality of patient involvement or "experiencing" in psychotherapy. "Experiencing refers to the quality of an individual's experiencing of himself, the extent to which his ongoing, bodily, fe l t flow of experien- cing is the basic datum of his awareness and com- munication about himself, and the extent to which this inner datum is integral to action and thought". (Klein et a l . , 1969, p.l) This seven point scale rating device is particularly useful in psychotherapy research because of i ts sensit iv i ty to changes in cl ient involvement within a single therapy hour. This is particularly useful for minute by minute process studies such as t h i s . The low levels on this scale are characterized by impersonal or superficial references to the se l f . "Moving up the scale, there is a progression from simple, limited or externalized self references to inwardly elaborated descriptions of feelings" (p. 1). At the highest levels of experiencing, exploration of feelings and new awareness, lead to prob- lem solving and greater self understanding. When the experiencing scale was applied to various settings, the overall trend was consistently related to therapeutic outcome, affirming the val id i ty of the scale and the concept of experiencing. On seven studies the rating r e l i a b i l i t i e s were s ignif icant , ranging - 30 - from rk .76 - .91 modes and .75 - .92 peaks using the Ebel Inter-class Rel iab i l i ty method which yields an estimate of the r e l i a b i l i t y of the average of the judges ratings. 2. Structural Analysis of Social Behavior S.A.S.B. The S.A.S.B. scale is used in this paper to measure the quality of the dialogue in the two chairs. It is expected that as the therapy hour progresses, the quality of interaction between the two chairs changes. The S.A.S.B. scale is particularly useful in this instance because of i ts spec i f ic i ty . The dialogue can be analyzed statement by statement and each identif ied as one of 36 character- i s t i cs on one of three grids, thus allowing the identif ication of pat- terns and subtle changes in interaction between the two chairs. The S.A.S.B. is an extension of the Leary c lass i f icat ion system. This model is divided into three two dimensional grids. The f i r s t grid measures behaviors focusing on others, i . e . , "118 encourage separate identity". For analyzing Gestalt, two chair process, the behaviors or dialogue of both chairs wi l l be rated on this f i r s t gr id , when the focus of the statement is on the other part in the dialogue. The second grid measures behaviors which focus on "se l f " , in the two chair dialogue the behaviors of the experiencing chair and the other chair wi l l be rated on this grid when the com- ments are about themselves. The corresponding behavior to 118 is 218 "own identity standards". The third set of behaviors is in t ra - psychic or introject of other to self - 318 "let nature unfold." This third grid w i l l not. be used for the purpose of this study. As is demonstrated, the three grids correspond and are interlated. The f i r s t number of the three digit behavior code refers to the gr id , 1 for - 31 - "focus on other" 2 for "focus on se l f " , 3 for "intrapsychic". The second number refers to the particular position on the gr id. The third number identif ies each specific behavior. (See Appendix B). Each of the quadrants is two dimensional. The horizontal axis runs from d isa f f i l i a t ion to a f f i l i a t i o n and the vertical axis from maximal dependence to maximal independence. "Each of the points within the quadrants is made up of mathematically defined proportions of the behaviors described by the axis. For example, chart point 118, encourage separate identity appears in the f i r s t (upper right hand) quadrant of the focus on other surface, i t consists of 1 unit of a f f i l i a t i o n and 8 units of endorsing of freedom (+1, +8) . . . "The 36 pairs of complements described by the f i r s t two surfaces allow def i - nition of complementarity in relationship and have clear and specific implications for the relevance of the patients' significant others to interpersonal diagnosis and treatment". (Benjamin, 1977, p. 7) . S t a t i s t i c a l l y , the S.A.S.B. has been found to be a sound measuring device. "Validity has been established by factor analysis, circumplex analysis, auto correlation techniques and dimensional ratings." (Benjamin, 1977, p. 5). The S.A.S.B. was used to set psycho- social treatment goals and in "before during and after" self ratings proved to be effective as a measure of the effectiveness of the therapy. When applied speci f ical ly to analysing Gestalt dialogue, r e l i a b i l i t y of interjudge agreement was tested using Cohens Kappa, and found to be .911. Benjamin concludes, "These high Kappas between inde- pendent judges establish that despite the complexity, the rules for applying the S.A.S.B. model to therapy transactions are communicable - 32 - and can yield consistent judgements among careful independent obser- vers." (p. 20). Client Vocal Quality Scale The C.V.Q. has four voice patterns - - focused, externalized, l imited, and emotional, each identif ied in terms of six features. 1) energy, 2) primary stresses, 3) regularity of stresses, 4) pace, 5) timbre and 6) contours. (See Appendix C). Greenberg (1979) describes vocal quality as a measure of involvement and processing level in the moment, and, "in a good hour of therapy we expect more focused voice." For the purpose of this study i t was hypothesized that the quality of voice is different in each of the three phases. Most important, more focused voice is expected at the merging point where the other chair softens and in the resolution phase when the two chairs usually begin to encounter each other in productive dialogue. Re l iab i l i t y for the C.V.Q. was tested in several ways. Rank order correlations between judges was found to be between .70 and .79 on the four categories. (Rice and Wagstaff, 1967). Percentage agreement was .70 and Cohens Kappa, a much more stringent measure, was .49 for the same study. Target Complaints Discomfort Box Scale (Appendix D) The T.C.D.B.S. is a thirteen point scale rating the amount of discomfort the cl ient is experiencing in relation to their present conf l ic t . For the purpose of this study, i t was administered before and after the sessions to identify any movement towards resolution. - 33 - One of the cr iter ion used to identify resolution events was at least a five point difference between the pre-test and post-test scores. In a previous study, Greenberg and Dompierre (1980) found that the T..C.D.B.S. was shown to discriminate between more and less effective psychotherapy sessions. Conflict Resolution Box Scale (Appendix E) The C.R.B.S. was created by Dompierre (1979) for use in her study comparing empathy versus the two chair operation at the s p l i t . The cl ient indicates the degree to which they feel resolved regarding the confl ict they have identif ied and explored in the session. This seven point scale ranges from "not at a l l resolved (in the f i r s t box, at the bottom) to "somewhat resolved" (fourth box), to tota l ly resolved in the seventh box. This scale was used as one of the cr i ter ion for closing resolution events in this study. The cl ient had to score at least 5 on the scale to be considered a resolution event. Shift in Awareness (Greenberg and Dompierre, 1980) The S.I .A. scale consists of two questions to be answered on a five point scale. The f i r s t question required the cl ient to identify i f any shift in awareness occurred, the second, whether the c l ient 's awareness of themselves increased as a result of the session. The S.I .A. was administered directly after the session and used in this study as one of the criterion in selecting the resolution events. Design Two groups of twelve Gestalt events were collected for com- parison. The Resolution Group, hypothesized to have a spec i f ic , - 34 - identif iable pattern of behavior was selected according to four c r i te r ia (discussed in Subject Selection). Any events which did not meet these c r i t e r i a - were considered to be non-resolution. These two groups of events - resolution and non-resolution, were then divided into three phases—opposition, merging and resolution and into two chair—"experiencing" and "other". Each of these phases and chairs were rated on the three process scales - - experiencing, S.A.S.B. and voice quality. The research design consisted of a set of comparisons between groups, between chairs and between phases using the scores of the three rating scales. I n i t i a l l y , the twelve resolution events were compared with the twelve non-resolution events to test i f in fact the two groups were different. Then, the other hypotheses were tested, using appropri- ate non parametric s t a t i s t i c s , to.test the pattern of confl ict resolu- tion in the resolution group. Subject Selection The Gestalt events collected for analysis were from actual sessions of clients involved in affective problem solving of real issues. Al l clients were basically well functioning people exploring basic problems in l i v ing . They were a sample of volunteers seeking therapy and pursuing their own growth. The cl ients were a l l in an ongoing relationship with the therapists, either in private practice or trainees, in a Gestalt awareness training group experience. The actual incidents chosen were selected on a specified c r i te r ia of resolution or non-resolution. To be considered a resolu- t ion , both therapist and cl ient had to say the cl ient had resolved - 35 - and report a significant shi f t in awareness leading to a desire to change some behavior. More spec i f ica l l y , the cl ient and therapist had to report after the session a score of 5 or more on a 7 point scale ranging from completely resolved to unresolved on the C.R.B.S. Scale. A second level cr i ter ion was that the level of experiencing during the event must reach level 6--"Feelings v iv idly expressed, integrate, conclusive or affirmative". The third cr iter ion was the target complaints box scale, measuring how bothered the cl ient was by the issue. There had to be at least a 5 point difference on the 13 point scale on a pre-test, a post-test session report. The fourth c r i - terion was that the therapist must report that, in his view, the cl ient resolved the specific confl ict that was worked on in the session by a process of integration. A non-resolution event was anyone that the cl ient and therapist said did not resolve and that did not reach level six on the experiencing scale. Our definition of resolution states that both cl ient and therapist must acknowledge that the cl ient has resolved, and an objective rater verif ies the resolution by rating i t at level six on the experiencing scale. Therapists Five therapists contributed events for this study. Two were professors in Counselling Psychology, one was a doctoral student and the other two were practising counsellors with masters degrees in counselling psychology. Al l the therapists have extensive training and practise in Gestalt therapy and the two chair technique. Each of the five therapists produced at least one resolution event for this : study. - 36 - Raters The six raters, two for each of the three measuring instru- < ments, were a l l students in counselling psychology at the masters leve l . The raters for the experiencing scale were trained for th i r t y - five hours according to "The Experiencing Scale, Training Manual, (Klein et a l . , 1969). They were then given specific instruction and rules governing the rating of two chair dialogue. After sixteen hours of practise, the raters were checked for inter-rater r e l i a b i l i t y . The Pearson Product Moment Correlation Coefficient produced an.: r of .89 on segments, which provided more information than the statement unit of measure. On the actual data, which was rated statement by statement, the raters .overlapped on one-third of the data and had a correlation of r = .77 on randomly selected scores. Training for the S.A.S.B. was done in accordance with the Coding Manual for Using SASB to Rate Typescripts, (Estfoff and Benjamin, 1979). The raters had twenty-five hours of instruction and fifteen hours of practise. The f inal sessions were group sessions of raters and trainers working together to create discussion helpful in under- standing the S.A.S.B. The raters were checked for r e l i a b i l i t y before being given the actual data and checked again on the data. The post- training correlation was r = .96, and on selected statements of one- third of the data, the correlation was r = .87. The voice quality raters.were .trained according to Manual For Client Vocal Quality, Volume II, Instructions for Raters (Rice et a l . , 1979). The training was about twenty-five hours of instruction followed by a r e l i a b i l i t y check. Voice quality is rated from both a tape recording - 37 - and a typescript. The percentage of agreement on randomly selected statements was 75. The Cohen's Kappa was .54. The actual data were rated so there was a one-third overlap which was rated by both raters. In the event that the two raters were discrepant on any segments, the trainers were called in to re-rate that portion of data for any of the three scales. Data Collection The Gestalt events were collected from the five therapists based on the criterion of resolution or non-resolution. These sessions were then transcribed from the point that the cl ient stated an affective problem or confl ict and proceeded to engage in the two chair dialogue. Then each cl ient statement was identif ied and numbered as a unit of measure. The events range from 42 to 132 segments. The average length of the resolution events was 96 statements and 81 statements for the non-resolution events. Scoring Procedure The twenty-four events collected, twelve resolution and twelve non-resolution, were identified by the therapists and researcher accord- ing to the c r i te r ia previously stated. The typescripts and tapes were then coded so raters could not identify the group to which they belonged. For Experiencing Scale rating the typescripts were broken into two-page segments. The raters randomly rated these segments, giving each cl ient statement a mode score. The Vocal Quality rating and the S.A.S.B. ratings were also done statement by statement, but each typescript was complete. Each of the six raters rated two-thirds of the data to provide a r e l i a b i l i t y - 38 - check. When the data were discrepant, the third rater's score was taken. The next step after the rating was complete was to identify the phases in the resolution group. According to the def in i t ion , merging begins with two or more quadrant 1 or 4 responses in the other chair on S .A .S .B . , and resolution begins with six on the experiencing scale in either chair. The same pattern did not occur in the non- resolution group, so in order to compare the two groups, this group was phased by taking the mean proportion of statements of the three phases in the resolution group and transposing i t on the non-resolution group. The resulting proportions were opposition phase .59, merging .31 and resolution .10. (See Table 1). At this point a l l the data were charted on graphs according to chair, phase, and rating on the three scales. Then the data were tabu- lated according to proportions. This compensated for the differ ing lengths of transcripts, phases and the varying number of scores for each chair or phase or scale. A chart was then made for each event containing the following information: 1. Mean experiencing score for both c h a i r s - experiencing and other, in each phase-opposition, merging and resolution. 2. The proportion of S.A.S.B. scores for each quadrant in each phase for both chairs; and 3. The proportion of responses in each category of the Voice Quality scale for each phase in both chairs. (See Table 2). Procedure of Analysis Due to the non-randomness, small sample size and the dependance of some of the data, the hypotheses were tested by appropriate non- - 39 - parametric tests . The hypothesis stating that the two groups were different, in that merging occured in the resolution group and not in the non-resolution group, was tested using Fishers Exact test (Hays, 1963) for independent random samples with a small n. The hypotheses comparing the two independent groups were tested using the Mann Whitney U test . This test was chosen because the samples were independent, had a small n and proportions are ordinal measurement. The Wilcoxin Rank Sign test was used to test the hypotheses about the relationships bet- ween phases and chairs within the groups for the S.A.S.B. and vocal quality comparisons. This test was used because of i ts application to small related samples. The matched pairs t ^ test (Marascuilo & McSweeney, 1977) was used to analyse the data of the experiencing scale. This test was applicable because of the interval scale of the experiencing rating. To simplify the analysis of the many hypotheses and questions, the data were tabulated onto graphs and charts for each event. (See Table 2 for example). The data were then compiled onto tables for each process scale (see Tables 3,4,5,6,7,8) and f i n a l l y , a table was made for each hypothesis.' (See Table 9 for example). The charts and data tables follow in this chapter, the results are presented in Chapter IV and a discussion in Chapter V. TABLE I PHASING THE NON-RESOLUTION GROUP Non-resolution events were divided into the three phases using the mean proportion of statements in each phase of the resolution group and applying i t to the non-resolution group PROPORTION OF SEGMENTS IN EACH PHASE NUMBER OF SEGMENTS IN EACH PHASE Resolution Group O p p o s i t i o n Merge Resolution Non-ResT Group T o t a l number O p p o s i - t i o n Merge R e s o l u t i o n 1 A .72 .15 .12 1 B 119 70 37 12 2 A .71 .20 .09 2 B 71 42 22 7 3 A .39 .57 .04 3 B 46 27 14 5 4 A .39 .33 .28 4 B 105 61 33 11 5 A .59 .32 .08 5 B 44 26 14 4 6 A .78 .18 .03 6 B 76 44 24 8 7 A .31 .66 .02 7 B 61 36 19 6 8 A .38 .48 .14 8 B 51 30 16 5 9 A .59 .25 .16 9 B 42 25 13 4 10A .71 .17 .12 10B 131 77 41 13 11A .70 .24 .06 11B 145 85 45 15 12A .81 .16 .02 12B 82 53 21 8 X .59 X .31 X .10 39 (b) TABLE II EXAMPLE OF TABULATION OF RAW DATA FOR EACH EVENT the two chairs are separated, the three phases are separated. graph of each chair according to level of experiencing score, £\/&)6T—| C AJo*> Rfc*oi_u.TnCr9 --4 TABLE III MEAN SCORES FOR EACH CHAIR IN EACH PHASE OF RESOLUTION GROUP ON THE EXPERIENCING SCALE INCLUDING NUMBER OF SEGMENTS IN EACH PHASE Number Number No. of . rofcal • . .in i n segs. C l i e n t -lumber-. phase OPPOSITION,, phase MERGE m phase. RESOLUTION Chair 2 Chair 1 Chair 2 Chair 1 Chair 2 Chair 1 '1 98 71 . 3.25 - 2.36 15 . 3.28 3.67 12 4.83 2 132 94 3.58 2.31 26 4.52 12 5.71 6.00 3 109 43 2.89 2.24 62 4.00 3.86 4 6.00 :4 46 18 2.92 2.00 15 0 4.00 4.07 13 5.53 _ 5 118 70 3.35 38 4.13 10 5.67 2.86 3.77 6.00 125 98 3.23 23 3.75 4 5.60 2.86 • 4.29 7 83 . 26 3.00 2.11 55 2.83 3.54 2 6.00 8 69 26 3.20 • 2.56 33 3.56 4.79 10 5.50 5.00 9 95 56 3.32 2.70 24 4.00 4.53 15 5.78 4.50 10 09 77 3.58 2.68 19 4.86 3.60 13 5.46 11 71 50 2.92 2.00 17 4.25 4.00 4 6.00 6.00 12 97 79 3.66 3.00 16 4.10 4.67 2 6.00 TABLE IV MEAN SCORES FOR EACH CHAIR IN EACH PHASE OF NON-RESOLUTION GROUP ON THE EXPERIENCING SCALE INCLUDING NUMBER OF SEGMENTS IN EACH PHASE PHASE OPPOSITION MERGING RESOLUTION C l i e n t Segments Chair 2 Chair 1 Segments Chair 2 Chair 1 Segments Chair 2 Chair 1 T n r a l Tn pha<= In phase In phase 1 119 70 3.24 37 2.73 12 2.44 2.80 2.40 2.00 2 71 42 2.71 22 3.10 7 2.50 2.36 2.67 2.60 3 46 27 3.00 14 3.00 5 2.75 2.56 2.70 3.00 4 105 61 2.57 33 2.36 11 3.00 2.14 2.20 5 44 26 2.95 14 3.50 4 3.75 2.57 2.40 6 76 44 2.38 24 2.53 8 2.50 2.30 2.20 2.00 7 61 36 2.78 19 2.41 6 2.75 2.08 3.00 2.00 8 51 30 2.84 16 2.62 5 2.67 2.45 2.67 2.00 9 42 25 2.50 13 2.83 4 2.25 2.44 2,57 10 131 77 2.71 41 3.77 13 2.40 2.21 2.47 2.13 11 145 85 3.54 45 3.62 15 3.75 12 2.95 3.10 3.00 82 46 2.84 25 2.95 11 2.80 2.44 2.58 2.34 TABLE V PROPORION OF STATEMENTS IN EACH QUADRANT OF THE STRUCTURAL ANALYSIS OF SOCIAL BEHAVIOR SCALE FOR RESOLUTION GROUP OPPOSITION PHASE MERGING RESOLUTION Quadrant 1 2 3 4 1 2 3 4 1 2 3 4 A 1 Ch.2 .56 .09 .20 .15 .29 .71 1 .00 Ch. 1 .15 .08 .73 .04 .38 .63 2 .63 .17 .06 .23 .83 .08 .96 .04 .86 1 .00 .14 3 .46 .04 .06 .27 .94 .23 .76 .50 .03 .32 .21 .18 1 .00 4 .67 .17 .08 .25 .67 .17 1 .00 .67 .13 .20 .69 .31 5 .58 .08 .08 .27 .81 .06 .11 .63 .64 .04 .04 .37 .28 .87 .75 .13 .13 .13 6 .65 .06 .05 .44 .31 .50 .47 1 .00 .47 .05 1 .00 7 .35 .10 .24 .35 .80 .06 .10 .40 .62 .03 .08 .37 .08 .20 .23 1 .00 8 .50 .08 .85 .50 .08 .47 .60 .16 .13 .37 .27 .44 .50 .11 .44 .50 9 .86 .27 .07 .14 .63 .03 .83 .45 .17 .55 .92 .40 .08 .60 10 .79 .09 .05 .15 .10 .76 .07 .93 .80 .07 .20 1 .00 11 .57 .03 .30 1 .00 .19 .50 .90 .50 .10 1 .00 1 .00 12 .15 .05 .19 .26 .53 .63 .14 .05 .83 .83 .08 .08 .17 1 .00 TABLE VI PROPORTION OF STATEMENTS IN EACH QUADRANT OF THE STRUCTURAL ANALYSIS 'OF SOCIAL BEHAVIOR SCALE IN THE NON-RESOLUTION EVENTS PHASE • OPPOSITION MERGING RESOLUTION Event Quadran t 1 2 3 4 1 2 3 4 1 2 3 4 B 1 Chair 2 " 1 .45 .27 .07 .45 .67 .09 .45 .13 .07 .55 .80 - • 33' .33 1 .00 ;33 2 i. 2 1 .67 .15 .50 .18 .50 .90 .09 .10 .91 1 .40 .00 .60 3 2 .67 .22 .11 .67 .17 .17 .60 . 4 0 1 .25 .13 .63 .30, .50 .20 1 .00 4 2 1 .44 .08 .02 .Oa .31 .85 .23 .75 .04 .18 1 .00 .04 1 .00 5 " 2 1 ; .76 .20 .10 .05 .80 .10 .75 .25 .80 .20 .75 .25 6 2 1 .65 .20 .29 .80 .06 .55 .40 1 .00 .05 1 .00 1 .00 7 8 2 " 1 2 1 .57 1.00 .18 .30 .09 .92' .82 .04 .08 .50 ••.77 .33 .11 .33 .39 : 1 .15 .00 .33 .08 1 .00 .25 .50 1 1 .00 .00 .25 9 2 1 .50 .31 .13 .06 .75 .13 .13 .50 .29 .33 .71 .17 .50 .50 10 2 1 .37 .10 .13 .39 .88 .14 .59 .11 .41 .89 .80 1 .00 .20 11 2 1 .33 .13 .06 .33 .54 .50 .07 .04 .48 .23 .15 .08 .27 .46 .09 .23 .30 , .10 .60 1 .00 12 "• 2 " 1 .58 .19 .15 .20 .22 .50 .07 .11 .63 .12 .12 .13 .25 .50 .25 .75 .25 1 .00 TABLE VII PROPORTION OF STATEMENTS IN EACH VOCAL CATEGORY FOR RESOLUTION GROUP PHASE Cateqory F OPPOSITION E X L F MERGING E X L F RESOLUTION E X L A 1 .63 .45 .37 .55 .71 .7b .29 .2b 1 .OU 2 .59 .06 .33 .90 .07 .03 .63 .37 1 .00 .40 .60 3 .38 .35 .27 .27 .65 .08 .74 .57 .21 .21 .21 .06 .75 .25 4 .25 .17 .17 .08 .50 .50 .33 1.00 .36 .14 .36 .14 .85 .15 5 .21 .791 .00 .53 .32 .47 .68 .67 .80 .33 .20 6 .49 .44 .51 .56 1.00 1 .00 1 .00 7 .41 .711 .00 .38 .30 .59 .69 .03 1 .00 8 .60 .31 .30 .62 .10 .06 .78 .67 .22 .33 .88 .50 .13 .50 9 .26 .15 .74 .85 .20 .47 .20 .60 .53 .90 .50 .10 .50 10 .51 .32 .47 .56 .02 .12 .64 .80 .36 .20 .15 .85 11 .50 .07 .06 .42 .93 .03 .63 .78 .38 .22 1 .00 1 .00 12 .28 .14 .18 .14 .54 .71 .40 .50 .30 .33 . JO .17 Categories F - Focused E - Emotional X - External L - Limited TABLE VIII PROPORTION OF STATEMENTS IN EACH VOCAL CATEGORY FOR NON-RESOLUTION GROUP OPPOSITION PHASE MERGING RESOLUTION Category F E X L F E X L F E X L Chair 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 B 1 .38 .61 .62 .33 06 .46 .5 .54 .5 .4 .67 .6 .33 2 .14 .86 1.0 .3 .7 1.0 .5 .5 1.0 3 .5 .2 .5 .8 1.0 1 .0 .25 .75 1 .0 4 .23 .07 .74 .93 .02 .04 .20 .96 .80 .09 .91 5 .26 .74 1.00 .50 .50 1.00 .25 .25 .50 6 .65 .20 .26 .80 .09 .84 .80 .16 .20 .17 1.00 .83 7 .78 .85 .17 .08 .04 08 .77 .08 .15 1 .00 1 .00 .50 .50 8 .1 .9 1 .0 .39 .61 1.0 .33 .67 1.0 9 .13 .87 1.00 .43 .57 1 .0 .25 .75 10 .35 .24 .65 .76 .33 .14 .67 .8 .06 .3 .24 .7 .86 11 .30 .125 .04 .66 .875 .28 .09 .69 .91 .03 .37 .63 1.0 12 .64 .4 .36 .6 .48 .29 .52 .71 .6 .2 .4 .8 39 (1) TABLE IX EXAMPLE OF DATA TABULATION FOR EACH HYPOTHESES HYPOTHESIS 1 b OTHER CHAIR MERGING PHASE Resolution Group Non-Resolution Group Event p 1 1 .00 .13 R 0 2 1 .00 .09 P 0 3 .68 .30 R T 4 .87 .00 I 0 5 .92 .20 N 6 1 .00 .00 0 F 7 .85 .00 A 8 .87 .33 F F 9 1 .00 .29 I L 10 1 .00 .11 I A 11 1 .00 .46 T I 12 1 .00 .37 0 N 39 ( j ) TABLE X FREQUENCY OF AFFILIATION AT THE MERGING PHASE FOR THE TWO GROUPS HYPOTHESIS 1 a OTHER CHAIR Resolution Group Non-Resolution Group A 1 Yes No F 2 Yes No F 3 Yes No I 4 Yes No L 5 Yes No I 6 Yes No A 7 Yes No T 8 Yes No I 9 Yes No 0 10 Yes No N 11 Yes No 12 Yes No 39 (k) TABLE XI FREQUENCY OF FOCUS VOICE AT MERGING FOR RESOLUTION AND NON-RESOLUTION EVENTS HYPOTHESIS 4 OTHER CHAIR MERGE PHASE Resolution Non-Resol F 1 Yes No 0 2 Yes No C 3 Yes No U 4 Yes No S 5 Yes No 6 Yes Yes V 7 No. No 0 8 'i Yes No I 9 Yes No c 10 Yes No E 11 Yes No 12 Yes No - 40 - CHAPTER IV RESULTS This chapter presents a l l the s tat is t ica l analyses performed on the data. The tests for each of the hypotheses and questions are discussed individually in this chapter and summarized in Chapter V, to show the pattern of resolution events. The main research question is whether events identif ied as resolved are associated with different phase related behaviors, and whether the non-resolution events do not show this pattern. Results of Between Group Comparisons Hypothesis 1(a) The other chair of the resolution group wi l l achieve a f f i l i a t i o n signif icantly more often than wi l l the non-resolution group. It was found using Fishers Exact Test (Hays 1963) (Table 10) p <.001 with d.f. =11. It appears that the resolution group in this sample a f f i l ia ted signif icantly more often than the non-resolution group. Hypotheses 1(b) The other chair of the resolution group wi l l have a higher proportion of quadrant one and four behaviors of the S.A.S.B. after the merging point (as defined by hypothesis 1(a)). Using the Mann Whitney U Test (Marascuilo and McSweeney, 1977), i t was found U = 0, p < .01 and d.f. = 11. (Table 5 and 6). This s ignif icantly greater proportion of a f f i l i a t i v e behavior in the resolution group supports the f i r s t hypothesis that in fact the two groups were different. Since the two groups were different, i t was appropriate to consider the hypotheses based on that assumption. - 41 - Hypothesis 2(a) In the resolution phase of the two groups the resolution group wi l l have a s ign i - f icantly different proportion of F plus E voice in the other chair. The Mann Whitney U Test produced U = 21, p <.01, d.f . = 10 (Table 7 and 8), indicating that there was more focused voice in the resolution events than the non-resolution events in the other chair. Hypothesis 2(b) In the resolution phase of the two groups the resolution group wi l l have a s ign i - f icantly different proportion of F plus E voice in the experiencing chair. The Mann Whitney U Test produced U = 5, p<".01 with d.f . = 5, indicating that the experiencing chair in the resolution phase uses more F plus E voice in the resolution group than in the non-resolution group. Hypothesis 3(a) In the merging phase of the two groups, the resolution group wi l l have a s ignif icant ly different proportion of F plus voice in the experiencing chair. It was found U = 34, p <.05, and d.f . = 10. (Table 7 and 8). This indicates that in the resolution group the experiencing chair uses more F plus E voice in the merging phase. Hypothesis 3(b) In the merging phase of the two groups, the resolution group w i l l have a s ignif icant ly different proportion of F plus E voice in the other chair. The test produced U = 15, p<\01, with d.f . = 10, (Table 7 & indicating that the other chair in the resolution group also uses more F plus E voice. Hypothesis 4 The other chair changes from external to focused at the beginning of the merging phase (three statements before and/or after) in the resolution group but not in the non- resolution group. - 42 - Fishers Exact Test was used and produced p <.001 with d.f . = 11 (Table 11), indicating that the resolution group became focused at the merging point s ignif icantly more often than did the non-resolution group. Question 1 In the opposition phase of the two groups there wi l l be no significant difference in the propor- tion of F plus E voice in the other chair , or 2 - in the experiencing chair. The Mann Whitney U test produced U = 56.5, p ^ . 0 5 , with d.f . = 11. (Table 7 & 8). There was no significant difference in the proportion of F plus E voice in the other chair. The same test for question 2 was found to give U = 62, p N .05, with d.f . = 11, also, not s ta t i s t i ca l l y significant difference between the two groups in the amount of F plus E voice in the experiencing chair. Results of Phase Hypotheses in the Resolution Group Hypothesis 5 In the opposition phase the experiencing chair scores at a s ignif icantly different level of experiencing than the other chair on the experiencing scale. The Matched Pairs t Test produced t = 10.98, p^ .001, with d.f. = 11. (Table 3 and 4). This result suggests hypothesis 5, the experiencing chair scores higher on the experiencing scale than the other chair. Hypothesis 6 In the merging phase the other chair scores at a s ignif icant ly different level of experiencing than in the opposition phase, on the experiencing scale. With t = 8.32, p^ .001, and d.f . = 11 (Table 3 and 4) , i t appears that the level of experiencing in the other.chair during merging is higher than in the opposition phase. - 43 - Question 1 In the merging phase the level of experien- cing of the other chair is not s ignif icantly different than that of the experiencing chair. The t test produced t = - . 3 6 , p ^ . 0 5 , with d.f . = 11. (Table 3 and 4). There was no s ta t i s t i ca l l y significant difference between the two chairs in the merging phase. Hypothesis 7 The experiencing chair uses a s ignif icantly different proportion of F plus E voice than the other chair in the opposition phase. The Wilcoxam Matched Pairs Signed Ranks Test was applied to the data. It was found that T = 0, p<.01, with d.f . = 11. (Table 7 and 8). This score indicates that the experiencing chair uses a s ignif icantly higher proportion of F plus E voice than does the other chair. Hypothesis 8(a) The other chair uses a s ignif icantly different proportion of F plus E voice in the merging phase than in the opposition phase. With T = 0, p<.01, with d.f . = 11 (Table 7 and 8) , i t appears that the other chair uses higher proportion of F plus E voice in the merging phase. Hypothesis 8(b) The other chair uses a s ignif icantly d i f fe r - ent proportion of F plus E voice in the resolution phase than in the opposition phase. The test produced T = 0, p= .05, with d.f . = 5. (Table 7 and 8). This test was not signif icant at the .05 level because i t was performed with n = 6, the result of fewer instances of statements in the resolution phase for the other chair. Often during this phase one of the two chairs were not present, thus creating empty ce l ls in the analysis. - 44 - Question 1 It is expected that in the non-resolution group there wi l l be no significant d i f fe r - ence in the proportion of F plus E voice be- ctween the opposition and merging phases. It was found that T = 25, p^.0'5 and d.f. = 11. (Table 7 and 8). There appears to be no significant difference in the pro- portion of F plus E voice between the merging and opposition phase of the non-resolution group. Hypothesis 9(a) The experiencing chair in the opposi- tion phase wi l l have a s ignif icant ly different proportion of quadrant two and three behaviors than in the resolution phase. It was found when the Wilcoxin test was applied that T = 2, p ^ . 0 1 , with d.f. = 8. (Table 5 and 6). This indicates that the experiencing chair uses more F plus E voice in the resolution phase than in the opposition phase. Hypothesis 9(b) The experiencing chair in the opposition phase wi l l have signif icantly different proportion of quadrant 2 and 3 behaviors than in the merging phase. With T = 1, p<.01, with d.f . = 11, (Table 5 and 6) i t is apparent that the experiencing chair changes from quadrant 2 and 3 behaviors (unfriendly and controlling) in the merging phase. Hypothesis 10 The experiencing chair during the resolu- tion phase wi l l have a signif icantly d i f - ferent proportion of quadrant behaviors than any other quadrant. The Wilcoxin test produced T = 1, p<.01, with d.f . = 10. (Table 5 and 6). This indicates a s ignif icantly higher proportion of quadrant 1 responses (a f f i l i a t i ve and independent) in the resolution phase in the experiencing chair. CHAPTER V DISCUSSION, CONCLUSIONS, RECOMMENDATIONS Discussion This chapter describes the data in terms of the expected pat- terns of behavior in successful resolution events as compared to non- resolution events. The cl ient behavior patterns, as measured by each of the process measures, are discussed for the resolution events. The underlying premise of this study is that increased awareness of the process of resolution wi l l help therapists fac i l i ta te their cl ients in solving affective problems. The patterns of behavior wi l l be described in terms of each process measure. Experiencing Scale Patterns The resolution events were chosen because they were rated on this scale at level 6, (synthesis of readily accessible feelings and experiences to resolve personally significant issues). Then i t was found, as expected, that the beginning of a therapy session is charac- terized by confl ict between two opposing parts of the person. During this phase, the level of experiencing of the other chair is about level 2 (behavioral or intel lectual self -descript ion) , while the experiencing chair is at about level 4 (description of feelings and personal experi- ences). The "softening" or beginning of negotiation between the two chairs, apparently necessary for resolution to occur, is indicated by • the change in the level of experiencing of the other chair. During the merging phase, there is no s t a t i s t i c a l l y significant difference between - 46 - the two chairs; in level of experiencing measured by the Experiencing Scale, they both function at level 4 and 5 (problems or propositions about feelings and personal experiences). It is interesting that in the resolution phase the two chairs are often d i f f i c u l t to discriminate. In fact , in half of the resolution events, the final phase is dominated by responses from the other chair but are not s ignif icantly different in level of experiencing than those containing only responses from the experiencing chair. This leads support to the concept of resolution being the result of integration of the two parts in a conf l ic t . Although there was no apparent pattern in the non-resolution events, i t was interesting to note that as well as l i t t l e difference between the level of experiencing of the two chairs, there were fewer responses from the other chair. Perhaps the reluctance of the cl ient to acknowledge and articulate their disowned feelings and thoughts, in the other chair, or to engage in the two chair dialogue also hinders them from seeing and experiencing their strengths and thus, integration does not occur. Another observation about the experiencing data was that in the resolution group, even in the opposition phase, there seemed to be a higher level of experiencing in the experiencing chair. Perhaps clients that resolve confl icts are more bothered by their confl ict and/or more ready to be open with their feel ings, which seems to aid resolution, Vocal Quality Patterns This scale seemed to be the most sensitive in picking up the "softening" necessary for productive dialogue at the merging. The raters consistently agreed on statements rated focused at the beginning of - 47 _ the merging phase, and the data indicated that focused voice is an essential part of a true merging. Eleven of the twelve resolution events had focused voice at merging. As well as focused voice appearing at c r i t i - cal points in the resolution events, the pattern seems to,show an increase in the proportion of focused plus emotional voice quality. The propor- tion of F plus E increases from the opposition to merging to resolution phase for both chairs. Similar to the experiencing pattern, the soften- ing of the other chair is indicated by the frequency of F plus E voice at merging occurring signif icantly more often in resolution events. There was also a significant increase in the proportion of F plus E voice in the merging phase as compared to the opposition phase in the resolution group. As expected, the non-resolution group does not show signif icant difference in F plus E voice between the three phases. Contrary to the pattern of experiencing, both groups appear to start the sessions at similar levels in voice quality. Structural Analysis of Social Behavior Patterns The S.A.S.B. scale was found to be a most useful instrument in process analysis. Although not previously used to analyze Gestalt two chair dialogue, i t was found to help in defining the merging point and identifying a measurable pattern of c l ient responses. For the purpose of this study, only two categories of the system were used. Quadrant 2 and 3 responses (unfriendly and controlling) were combined, as were quadrant 1 and 4 responses (a f f i l i a t i ve and independent). As expected, the conf l ict between the two chairs in opposition was demonstrated by both chairs responding with more control l ing, unfriendly statements. Further scrutiny suggests that the other chair was more controll ing - 48, . (quadrant 3), while the experiencing chair was more whining or excusing (quadrant 2). Although the experiencing chair had a higher proportion of 1 and 4 behaviors in opposition i t was similar to the other chair in merging, suggesting more productive dialogue is necessary for resolution to occur. Then in the resolution phase, almost a l l the responses were from quadrant 1 (most a f f i l i a t i v e and most independent). It appears as though both chairs become one and do not respond with cr i t ic ism or mani- pulation. In the non-resolution group, there was no significant d i f - ference in the proportion of a f f i l i a t i v e behavior in either chair in any of the phases. Compared to the resolution group, both chairs had s i g n i f i - cantly more quadrant 2 and 3 behaviors in a l l the phases. In summary, the proposed model of conf l ict resolution using the Gestalt two chair approach was supported by results of this study. Low level experiencing, host i le , controlling response and external voice in the other chair in the opposition phase changes to greater openness, more focused voice and greater a f f i l i a t i o n in the merging phase. This trend culminates in even higher levels of these characteristics. The idea of integration of the two opposing sides is supported by the apparent cooperation in the merging phase and that difference in the two chairs in the resolution phase is often not discernible. This dist inct ident i - f iable pattern also adds support to the fact that there are signif icant differences between resolved and unresolved events. Conclusions Inspection of the raw data and the results of the s tat is t ica l analysis suggest that identif ication of spl i ts or affective problems and the subsequent application of the Gestalt two chair technique is a powerful tool in psychotherapy. There actually appears to be two separate systems working in opposition that when integrated, lead to better func- tioning. The task analysis approach applied to the study of psycho- therapy, appears to be successful. The idealized model of confl ict reso- lution was presented and then tested. Developing a theoretical system and then applying i t to actual events deepens the exploration and know- ledge of the numerous underlying influences in a therapy session. In this study, the data appeared to confirm the idealized model of confl ict resolution. When there is a clearer understanding of the characteristics of successful resolution events, the therapist can work to create optimum therapeutic interventions. The data of this study also supports the l i terature which suggests integration as a mechanism of confl ict resolution (Perls et a l . , 1951) and confirms the model i n i t i a l l y proposed by Greenberg (1980). The concept of a three-phase process in resolving a sp l i t has added support. It seems necessary that the person in confl ict experience and articulate both sides of the conf l ic t . Then the two parts need to l is ten to and acknowledge each other. At this point, the side that is usually control l ing, threatening and demanding must soften and allow the apparently more passive side to be heard. When the passive part asserts i t s e l f , and the aggressive side softens, productive dialogue or negotia- tion begins. This is characterized by both sides stating their feelings and needs with mutual respect, and often results in the cl ient apprecia- ting a l l parts of himself. This new awareness helps the cl ient resolve the present confl ict and hopefully, w i l l be followed by positive behav- ioral change. - 5G> - Considering the relative infancy of in-process research, i t is important to note that attempting to explore and expose the under- lying performance patterns from vast amounts of data appears to be successful. There seems to be potential for this type of research to develop to the point of practical application so i t can enhance c l in ica l practise in the area of therapeutic intervention and change, and i l l u - minate phenomenon previously thought to be intangible and too complex to understand. Recommendations A logical follow-up to the findings of this study would be to follow c l ients , who, according to the three process measures, have resolved real l i f e confl icts in therapy, and determine the actual effect of this resolution on their behavior. Another interesting and practical study would be to ask clients to identify the most potent therapist interventions in a therapy session. Now that some of the components of successful confl ict resolu- tion have been ident i f ied, i t would be helpful to develop appropriate therapist interventions that could expedite the resolution process. It would also be interesting to compare therapist behavior between resolved and unresolved events. Considering the wide application of the S.A.S.B. scale, i t could probably be used to i ts fu l l capacity, using a l l three surfaces and nine tracks, to give an even more in-depth understanding of in-process behavior. As well as analysing resolution events, i t would be helpful to destructure non-resolution events for characteristics which seem to inhibit resolution or components associated with productive process that are missing. Implications The results of this study have implications for c l i n i ca l prac- t i s e , therapist training and in-process research. For the c l i n i c i a n , knowledge of the essential components in successful resolution events would help them understand the process, and work toward appropriate interventions to promote the optimum problem solving climate. With the use of these process measures, the c l in ic ian wi l l be able to measure progress and change in c l ients . Perhaps these measures could be the basis of a therapist evaluation system. The implications for therapist training are that the ident i - f ication of a s p l i t and the two chair intervention have been found again to be a potent therapeutic too l . Training programs need to teach students measures such as the Experiencing Scale, S.A.S.B. and Vocal Qual i ty , - - associated with productive therapy. The model of conf l ict resolution defined by specific measures, i s a framework within which the student can experiment and learn. This study lends support to the present trend to intensive in-process research. The outcome of examining in detail large amounts of data for a single therapeutic event seems to provide an insight into the l i t t l e known realm of human behavior. It appears as though the development of a theoretical framework, in this case of confl ict resolu- t ion , and the subsequent testing of this theory,is a viable approach to exploring these new frontiers. The final implication being, that i n - process research must continue to be refined to meet the needs of the researcher and the c l in ic ian str iving for more effective therapeutic techniques. - 52 - BIBLIOGRAPHY Baker, E. The differential effects of two psychotherapeutic approaches on c l ient perception. Journal of Counselling Psychology, 1960, ]_, 46-50. Bandler, R. & Grinder, J . The Structure of Magic. Vol. 1 1 . , Palo Alto Science and Behavior Books, 1975. Batt le, C.C., Imber, S .D. , Hoehn-Saric, R., Stone, A.R., Nash, E.H. & Frank, J.D. Target Complaints as c r i te r ia of improvement. American Journal of Psychotherapy, 1966, 2fJ: 184-192. Baumgardner, P. Legacy from Fr i tz : Gifts from Lake Cowichan. Palo Alto Ca l i f : Science and Behavior Books, 1975. Benjamin, L.S. A biological model for understanding the behavior of individuals. Tn~: J . Westman (Ed.). Individual ~ differences in children. New York: Wiley, 1973. Benjamin, L.S. Structural analysis of social behavior. Psycho- logical Review 1974, 81, 392-425. Benjamin, L.S. Structural analysis of a family in therapy. Journal of Consulting and Cl inical Psychology. 1 977, 45, 391 -406. Benjamin, L.S. Structural analysis of differentiation fa i lure . Psychiatry, 1 979, 42_, 1-23. Bergin, A. & Lambert, M. Evaluation of Therapeutic Outcomes, Handbook of Psychotherapy and Behavior Change, Garf ield, F., Bergin, A. New York: Wiley, 1978. Bohart, A.C. Role-playing and interpersonal conf l ict reduction. Journal of Counselling Psychology, 1977, 30, 311-318. Fagan, J . & Shepherd, I. (Eds.). Gestalt Therapy Now. Palo Alto : Science and Behavior, 1970. Gottman, J . & Swartz Toward a Task Analysis of Assertive Behavior, Journal of Consulting and Clinical Psychology, 1976, 89-117. Greenberg, L.S. A task therapeutic approach to the study of psycho- therapeutic events. Unpublished doctoral dissertation: York University, 1975. - 53 - Greenberg, L.S. , & Clarke, K. The differential effects of the two-chair experiment and empathic reflections at a confl ict marker. Journal of Counselling Psychology, . 1 979. (a), 26s 1-8. Greenberg, L.S. , & Dompierre, L. Differential effects of Gestalt two-chair dialogue and empathic reflection at a s p l i t in therapy. (Unpublished paper 1979). Greenberg, L.S. Resolving s p l i t s : The two-chair technique. Psychotherapy, Theory, Research and Practice, 1979, 1_6, 310-318. Greenberg, L.S. An intensive analysis of recurring events from the practice of Gestalt therapy. Psychotherapy, Theory, Research and Practice, in press, 1980. Greenberg, L.S. & Rice, L. The specific effects of Gestalt intervention. Psychotherapy, Theory, Research and Practice. In press. Greenberg, L.S. Advances in c l in ica l intervention research: A decade review. Submitted to the Canadian Psychologist, 1980. Hatcher, C. & Himelstein, P. (Eds.). The Handbook of Gestalt Therapy New York: Jason Saranson Inc. 1976. Hays, W.L. Stat ist ics For Psychologists. New York: Holt, Rinehart & , Winston Inc. 1963. Hersen, M. & Barlow, D. Single case experimental designs: Strategies for studying behavior change. New York: Pergamon Press, 1976. Johnson, N. Model building and intensive analysis of Gestalt events. Unpublished Masters thesis 1980, U.B.C. Klein, M., Mathieu, P., Keisler, D. & Gendlin, E. The experiencing scale. Wiseman Psychiatric Institute: Madison Wise. 1969. Latner, J . The Gestalt therapy book. New York: The Julian Press, 1 973. Marascuilo, L. & McSweeney, M. Non Parametric Stat ist ics and Distribution Free Methods for the Social Sciences. Cal i fornia: Brooks/Cole Publishing Co., 1 977. Orlinsky, D. & Howard, K. The relation of process to outcome in psychotherapy. In S.L. Garfield & A.E. Bergin (Eds.) Handbook of psychotherapy and behavior change (2nd Ed.). New York: Wiley, 1978. - 54 - Perls, F.. Hefferline, R., and Goodman, P., Gestalt Therapy. New York: Delta, 1951. Perls, F., Four lectures in J . Fagan and I. Shepherd (Eds.) Gestalt therapy now. Palo Alto^Science & Behavior, 1970. Polster, E., & Polster, M. Gestalt therapy integrated. New York: Brunner/Mazel, 1973. Rice, L., Koke, C , Greenberg, L., & Wagstaff, A.. Manual for cl ient voice qua!ity. Toronto: York University Counselling and Development Centre, 1979. Rice, L., & Wagstaff, A. Client voice quality and expressive style as indexes of productive psychotherapy. Journal of Consulting Psychology, 1 967, 31_, 557-563. Rice, L., & Greenberg, L., & Pascual-Leone, J . , The c l i n i c i a n - sc ient is t : Redirecting a needed resource (submitted for publication, 1978). Wexler, D.A. Self Actualization and Cognitive Processes Journal of Consulting and CIinical.Psychology. 1974, 42, 47-53. Zinker, J . Creative process in Gestalt therapy. New York: Brunner/Mazel, 1977. - 55 - APPENDIX A Short Form of EXP Scale Stage 1 4 5 Content External events; refusal to participate. External events; behavioral or intel lectual self -descript ion. Personal reactions to external events; limited self -descriptions; behavioral descriptions of feelings. Descriptions of feelings and personal experiences. Problems or propositions about feelings and personal experiences. Synthesis of readily accessible feelings and experiences to re- solve personally significant issues. Treatment Impersonal, detached. Interested, personal se l f -part ic ipat ion. Reactive, emotion- a l l y involved. Self -descr ipt ive; associative. Exploratory, elabo- rat ive, hypothetical. Feelings v iv id ly expressed, integrative, conclusive or a f f i r - mative . 7 F u l l , easy presentation of ex- periencing; a l l elements con- fidently integrated. Expansive, i l l u m i - nating, confident, buoyant. - 56 - APPENDIX B Structural Analysis of Social Behavior (SASB) INTERPERSONAL 120 Endorse freedom OTHER Uncaringly let go 128 Forget 127 Ignore, pretend not there 126 Neglect interests, needs 125 Illogical Initiation 124 Abandon, leave in lurch 123 Starve, cut out 122 Angry dismiss, reiect 121 j- Annihilating attack 130 Approach menacingiy 131 Rip off, orain 132 Punish, take revenge 133 Delude, divert, mislead 134 Accuse, blame 135 Put down, act superior 136 Intrude, block, restrict 137 Enforce conformity 138 118 Encourage separate identiry 117 You can do it fine 116 Carefully, fairly consider 115 Friendly listen 114 Show empathic understanding 113 Confirm as OK as is 112 Stroke, soothe, C2lm 111 Warmly welcome 110 Tender sexuality 141 Friendly invite 142 Provide for. nurture 143 Protect, back up -I 144 Sensible analysis 145 Constructive stimulate 146 Pamper, overindulge 147 Benevolent monitor, remind 148 Specify what's best Manage, control 140 220 Freely come and go SELF Go own separate way 228 Defy, do opposite 227 Wall-off, nondisclose 226 Busy with own thing 225 Noncontingent reaction 224 Detach, weep alone 223 Refuse assistance, care 222 Flee, escape, withdraw 221 r— Desperate protest 230 —j— Wary, fearful 231 Sacrifice greatly 232 I Appease, scurry 233 Uncomprehending agree 234 Whine, defend, justify 235 Sulk, act put upon 236 Apathetic compliance 237 218 Own identity, standards 217 Assert on own 216 "Put cards on the table" 215 Openly disclose, reveal 214 Clearly express —1 213 Enthusiastic showing -*—< 212 Relax, flow, enjoy 211 Joyful approach — 210 Ecstatic response 241 Follow, maintain contact I I I 242 Accept caretaking 243 Ask, trust, count on 244 Accept reason 245 Take in, learn from 246 Cling, depend Follow rules, proper 238 Yield, submit, give in 240 247 Defer, overconform 248 Submerge into role INTRAPSYCHIC Introject of OTHER TO SELF 320 Happy-go-lucky Drift with the moment 328 Neglect options 327 Fantasy, dream 325 Neglect own potential 325 Undefined, unknown self 324 Reckless 323 Ignore own basic needs 322 Rejec:. dismiss self 321 Torture, annihilate self 330 " r Menace to self 331 " Drain, overburden self 332 Venqeful self punish 333 Deceive, divert self 334 Guilt, blame, bad self 335 Doubt, put sell down 336 Restrain, hold back self 3^7 Force propriety 338 318 Let nature unfold 317 Let self do it. confident 316 Balanced self acceptance •t 315 Explore, listen to inner self J — » 314 Integrated, solid core 313 Pleased with self 312 Stroke, soothe self 311 Entertain, enjoy self • 310 Love, cherish self 341 Seek best for self 2 Nurture, restore self 343 Protect self 344 Examine, analyze self 345 Practice, oecome accomplished 346 Self pamper, indulge 347 Benevolent eye on self 34S Force ideal identiry Control, manage self 340 - 56a- APPENDI.X C VOCAL QUALITY RATING SCALE The characteristics of the four different patterns are as follows: A. Focused 1. Energy The energy is fa i r l y high. Pitch is moderate to low, with appropriate loudness. 2. Primary Stresses Primary stresses are achieved more by an increase in loudness than by a r ise in pitch. Loudness/pitch is greater than 1. The stress may also be achieved by lengthening the stressed syllable (drawl). 3. Regularity of Stresses The stress pattern is irregular for English, and stresses sometimes occur in unexpected places. For instance, adjoining syllables sometimes receive almost equal stress. 4. Pace The pace is irregular. It is usually slowed, but there may be patches that are speeded up. 5. Timbre The voice is f u l l , and resting firmly on the platform. 6. Contours These may be unexpected in direct ion, but the effect is ragged rather than mellif luous. - 57 - B. Externalizing 1. Energy The energy is fa i r l y high. The pitch is moderate to high, but the volume is adequate. 2. Primary stresses These are achieved with pitch r ise as well as some increase in loudness. Loudness/pitch is equal to or less than 1. 3. Regularity of stresses The stress pattern is markedly regular for English. The melodic l ine may sound sing-song at lower energy levels and resounding at higher levels. 4. Pace The pace is fa i r l y even, though i t may be s l ight ly speeded as i t approaches a stress point, 5. Timbre The voice is fa i r l y f u l l , and resting on the platform. 6. Contours These may go up, down, or remain level at times when this would not be quite the expected pattern, although meaning is not usually distorted. The effect is oratorical rather than ragged. C. Limited 1 . Energy The energy is low. The volume is not adequate for the pitch. 2. Primary stresses The primary stresses are not very strong, and are achieved by - 58 - normal balance of pitch to loudness. 3. Regularity of stresses The stress pattern has about the normal irregularity of English. 4. Pace The pace is somewhat slowed, but tends to be quite regular. 5. Timbre This is one of the clearest distinguishing characterist ics. The voice is thinned from below, and the effect is that of a voice that is "not resting on i ts platform." 6. Contours Nothing notable here. D. Emotional Overflow Eo This subcategory is d i f f i c u l t to describe using the six features, because a variety of different emotions are put in the same class. The primary characteristic is a disruption of ordinary speech patterns. The voice may break, tremble, r ise to a shriek, etc. However, the mere presence of emotion does not put i t in this c lass, without disruption of speech patterns. For instance, laughter is often found in conjunc- tion with Externalizing, and would not push the response into Emotional unless i t real ly disrupts speech. This is not a very satisfactory class as i t now stands, but is not too d i f f i c u l t to recognize. Expressive Ee 1 . Energy - 59 - Very high. Pitch is generally higher and loudness greater than any of the other categories. Primary stresses These are generally achieved by substantial increases in both pitch and loudness--although one may have a larger relative increase than the other. Also, there is often a clipped sense to stressed syl lables, and a sl ight pause after each one. Expressive vs. external--aside from regu- la r i t y of stresses distinguishing expressive from external (see below), there is greater pitch and loudness r ise with expressive voice than with external. If X is generally at modal pitch and one step above, E varies between modal and two or three steps above, (or even higher). Expressive vs. Focused—similarly, focused generally stays on_ modal pitch and occasionally goes down, or there may be a pitch r ise without loudness increasing to any marked degree. Regularity of stresses The most distinguishing feature of this category is stressed, adjoining sy l lables, with higher pitch and greater loudness than found in focused, e .g . , the stressed adjoining syllables in the sentence below are 'I hate 1 . I hate you There may be a pitch rise on the second of the stressed syl lables, but there is a clear sense of adjoining stressed syllables as shown in the sentence below. I don't care about you. - 60 - Pace Regular over stressed syl lables, but not regular in general. Often a stacatto quality to stressed syllables (relates to the sl ight pauses after stressed syl lables) . Timbre Generally a very fu l l voice. - 61 - APPENDIX D TARGET COMPLAINTS DISCOMFORT BOX SCALE We are interested in how much discomfort your decisional confl ict is causing you right now. Please indicate with an (X) your present position. Couldn't be worse Very much Pretty much A l i t t l e None at a l l - 62 - APPENDIX E CONFLICT RESOLUTION SCALE We are interested in how resolved you feel right now about your decisional conf l ic t . Please indicate with an (X) your present position. Totally resolved Somewhat resolved Not at a l l resolved

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Japan 8 0
United States 7 1
China 2 9
Canada 1 0
France 1 0
Russia 1 0
City Views Downloads
Tokyo 7 0
Unknown 5 0
Beijing 2 0
Surrey 1 0
Milwaukee 1 0
Houston 1 0
Sacramento 1 0
Redmond 1 1
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items