STRATEGIES 0F MANEUVERS DURING THE ACQUAINTANCE PROCESS Thesis for the Degree of ‘Ph. D. MICHIGAN STATE UNIVERSITY i SUSANJANE CONLEY ‘ _ 1968 ' u~ I a» a #6.“ a» .‘ ‘1; L [B R A R Y ”a Michigan State University - I 7H E815 This is to certify that the ‘ thesis entitled STRATEGIES 0F MANEUVERS DURING THE ACQUAINTANCE PROCESS I I presented by Susan Jane Conley has been accepted towards fulfillment of the requirements for Ph.D. degree in Psychology [g / ) / Majo{ professor Date //%V//Q’ 0-169 Whine-a! 37%” “S 2797 ABSTRACT STRATEGIES OF MANEUVERS DURING THE ACQUAINTANCE PROCESS by Susan J. Conley This study served a three-fold purpose: (l) to investigate the nature of the acquaintance process; (2) to characterize different types of dyadic relationships in terms of the strategies or maneuvers employed by the members as well as selected personality characteristics believed to be important determinants of dyadic relationships; and (3) to generate research-based hypotheses for later studies on psychotherapeutic dyads. The basic conditions of the study involved selecting subjects from college students who were not previously acquainted, pairing males with females, and then randomly assigning the dyads to one of three types of relationships (to get acquainted, to be friends, or to really know the other). Each dyad met for nine half hour sessions which were observed and tape recorded. Each dyad member was observed and rated on the over-all levels of Genuineness, Warmth, and Empathy displayed in a session as well as the frequency of Moving Toward, Moving Away and Moving Against maneuvers. Following the ninth session each Susan J. Conley subject completed the Interpersonal Checklist as well as a rating sheet on his impressions of the experiment and a demographic questionnaire. The results indicate, as hypothesized, that the type of relationship defined did result in different fre- quencies of maneuvers designed to move toward, move away, or move against the other person in the dyad. ‘While all.dyads engaged predominately in Moving Toward maneuvers, dyads instructed to become acquaintances tended to display more Moving Away and Against tactics than dyads instructed to become friends or to know the other person. As hypothesized, females displayed higher levels of warmth and empathy but contrary to expectations, this sex difference was maintained over the nine sessions. At variance with the hypotheses were the findings that the level of Genuineness, Warmth, or Empathy that a subject displayedrtuinot vary with the type of relationship. While the results met the hypothesis that Genuine- ness, Warmth and Empathy bear a positive relationship to each other, the results were contrary to the hypotheses regarding a relationship between the Movement Scale and the Genuineness--Warmth--Empathy Scales. The subscales of the two main measures bear essentially no relationship to each other. Further, the Interpersonal Checklist data bore no relationship to the other two measures, contrary to expectations. Susan J. Conley The results were explained by considering the testing of outcomes and rewards that subjects seemed to perform. The type of relationship, e.g., friends or acquaintances, would suggest certain types of rewards or outcomes and pre- clude others. The findings of the study are consistent with the notion that in the early stages of a relationship the costs are reduced if one participates in socially pre- scribed roles consistent with the definition of the rela- tionship. The results further suggest that Genuineness, Warmth and Empathy may play more or less important roles at varying stages in the relationship. In relationship definitions such as those used in the study, Empathy and Warmth may be maneuvers important to the develoPment of trust, leading to increased genuineness. The unsupported hypoth- eses suggested that the dyads had not progressed to the development of such trust. Further speculations were generated from the data concerning the relationships of genuineness and openness. It was suggested that relationships may be on a continuum of these two factors and that the relationships in the study were at a superficial point on this continuum, indi- cating that the subjects were engaging in maneuvers heavily determined by social roles. It was further suggested that future research be directed toward exploration of the role of Openness. STRATEGIES OF MANEUVERS DURING THE ACQUAINTANCE PROCESS BY Susan Jane Conley A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1968 TO ACKNOWLEDGMENTS I give many thanks to my chairman, Gary Stollak for his unceasing interest, enthusiasm, ideas and confidence. He provided a light during many a dark moment of despair. I also offer my thanks to Lawrence Messé who clar— ified what would otherwise have been a maze of statistical procedures. For their numerous helpful comments regarding the design and interpretations, I thank Mary Leichty and Dozier Thornton. They often helped me see the trees in the forest. Many others provided suggestions, a willing ear and moral support along the way, especially Liz Force. Finally, I thank the coders and the subjects all of whom served above and beyond the call of duty. TABLE OF CONTENTS ACKNOWLEDGMENT S O O 0 O O O O O O O O O O O O O O O 0 LIST OF TABLES O O O O O O O O O O O O O O I O O O 0 LIST OF APPENDICES O O O O O O O O O O O O O O O O 6 CHAPTER I. II. III. INTRODUCTION . . . . . . . . . . . . . . . . METHOD . . . . . . . . . . . . . . . . . . . Overview . . . . . . . . . . . . . . . . . Subjects . . . . . . . . . . . . . . . . . Materials. . . . . . . . . . . . . . . . . The Movement Scale . . . . . . . . . . . G-E-W Scales . . . . . . . . . . . . . . ICL. . . . . . . . . . . . . . . . . . . Demographic Questionnaire. . . . . . . . Rating Scale . . . . . . . . . . . . . . Procedure. . . . . . . . . . . . . . . . . Observers. . . . . . . . . . . . . . . . Observational Measures . . . . . . . . . Subject Scales . . . . . . . . . . . . . Methods of Analysis. . . . . . . . . . . . RESULTS. . . . . . . . . . . . . . . . . . . Inter- Rater Reliability for Movement Scale and G— E- W Scales . . . . . . . . . . . The Definition of the Relationship and the Strategies of Moving Toward, Moving Away, and Moving Against . . . . . . . The Type of Relationship Defined and the Levels of Genuineness, and Empathy . . . Relationship Between Genuineness, Warmth, and Empathy and the Strategies of Moving Toward, Away, and Against. . . . . . . . iv Page iii vi vii 16 16 l6 17 17 18 18 l9 19 20 20 21 23 23 25 25 25 34 40 CHAPTER Page Post Hoc Analysis. . . . . . . . . . . . 43 Relationship of Lov and Dom to Genuine— ness, Warmth, and Empathy Levels and to Moving Toward, Away and Against . . . 43 The Rating Scale and the Demographic Scale. . . . . . . . . . . . . . . . . 44 Summary of Results . . . . . . . . . . . . . 44 IV. DISCUSSION . . . . . . . . . . . . . . . . . . 46 V. SUMMARY. . . . . . . . . . . . . . . . . . . . 54 Table 1. LIST OF TABLES Percentage of Total Maneuvers per Session Scored as Moving Toward, Moving Away, and Moving Against Made by Subjects (Males and Females Combined) in the Acquaintance, Friends, and Knowing Dyads. . . . . . . . . . . . . . . Mean Number and Percent of Responses per Session Made by Groups on the Movement Scale . Major Sex x Groups x Session Analysis of Variance for Moving Toward Maneuvers . . . . . Sex x Groups x Session Analysis of Variance for Movement Scale--Moving Away Maneuvers. . . Sex x Groups x Session Analysis of Variance for Movement Scale-—Moving Against Maneuvers . Mean Level of G-E-W Scale Ratings. . . . . . . Sex x Groups x Sessions Analysis of Variance for G-E-W Scale--Genuineness . . . . . . . . . Sex x Groups x Session Analysis of Variance for G_E_W scale-_Empathy o o o o o o o o o o a Sex x Groups x Session Analysis of Variance for G-E-W scale--Warmtho o o e a o o o o o o 0 vi Page 27 28 29 30 32 35 36 38 39 LIST OF APPENDICES Appendix Page A. The Movement Scale. . . . . . . . . . . . . . 59 B. Genuineness-Warmth-Empathy Scales . . . . . . 63 C. Interpersonal Check List. . . . . . . . . . . 75 D. Demographic Questionnaire . . . . . . . . . . 78 E. Rating Scale. . . . . . . . . . . . . . . . . 79 F. Instructions. . . . . . . . . . . . . . . . . 80 G. Correlation Coefficients of Observers l and 2 for Movement Scale and G-E—W Scale by Sessions. . . . . . . . . . . . . . . . . . 83 H. Intercorrelation Coefficients Between Genuineness, Warmth and Empathy . . . . . . 84 I. Correlation Coefficients Between Subscales of Movement Scale and the G-E—W Scales by Sessions . . . . . . . . . . . . . . . . 85 CHAPTER I INTRODUCTION Brief Statement of Problem The present study has a three-fold purpose: (1) to investigate the nature of the acquaintance process; (2) to characterize different types of dyadic relationships in terms of the strategies or maneuvers employed by the members as well as selected personality characteristics believed to be important determinants of dyadic relationships; and (3) to generate research-based hypotheses for later studies on psychotherapeutic dyads. Experimentally contrived dyads were used which were instructed to define their relationship as (1) getting acquaintanced, or (2) becoming friends, or (3) to know all they could about the other. These dyads were then observed for the strategies they employed designed to Move Toward, Move Away, or Move Against the other member, and on the levels of Genuineness, Warmth, and Empathy dis- played by dyad members under these conditions. Rationale This research developed out of an interest in the psychotherapy relationship and a desire to characterize it. Many researchers have reached the conclusion that there is 1 a strong relationship between the acquaintance process and psychotherapy, and ". . . a study of the relationship‘ in psychotherapy [can] be viewed as a special case of this highly significant general area of psychological re— search interpersonal relations." (Stollak, 33 31., Psy- chotherapy Research, 1966, p. 504.) It followed then that the first step was to investigate the nature of dyadic re- lationships of various types in order that certain important variables be defined and thus generate increasingly more meaningful hypotheses about the therapeutic dyad. It seemed that dyadic interactions have at least two main determi— nants: (l) the strategies or maneuvers that an individual brings to a situation and which characterize his usual ap— proach to a relationship, and (2) the strategies that are deemed appropriate within the context of the situation. Recent theorists and researchers have considered these de- terminants in the past but too often one has been chosen to the exclusion of the other. The present study attempts to bring the strategies brought by the person together with the situational strategies. Previous Research on Dyadic Interactions Much of the previous research on personality vari- ables has been an attempt to posit a unifying principle based on underlying needs to account for mate selection or friendship choices. The main proponent of a single factor theory of complementarity of needs has been Winch (1952) ". . . each individual seeks within his or her field of eligibles for that person who gives the greatest promise of providing him or her with maximum need gratification." Winch hypothesized that maximum need gratification occurs when two peOple have different and complementary need pat- terns rather than similar ones. On the other hand, there has been equal support for the single factor similarity hypothesis (Izard, 1960) as well as several studies which do not support either notion (e.g. Bowerman and Day, 1956). Newcomb (1961) took a different approach when he suggested that interpersonal attraction is based on the perceived similarity regarding common objects of importance. PeOple respond more positively to those who share their views. However, Izard (1963) concluded that Newcomb's findings point to an immature relationship while mature individuals seek others who have different characteristic than their own. Apparently mature individuals do not need friends who reflect their own personality characteristics in order to enhance their own security operations. They have somehow achieved this through other means, thereby permitting their acquaintanceships to proceed into more productive areas. In a study of friendship choices of adolescent girls in a training school, Jennings (1950) found that both similarity and differences between dyad members or friendships were important, but even more so ". . . to promote a friendship a person must not only be able to provide rewards for others but must also be willing.‘ The most chosen girls were found to possess more capacity of identifying with others i.e., were more empathic. Such factors as similarity and empathy have also emerged from studies of effective psychotherapy generalized to interpersonal relations. Fiedler (1950), in a study of "ideal" therapeutic relationships hypothesized that the therapy relationship may be only a variation of good inter- personal relationships in general. A similar conclusion was reached by Schofield (1964) who stressed the importance of the acceptance of the patient by the therapist. Simi— larities between the therapist and patient have been inves- tigated (McNair, Callahan, and Lor, 1962; Fiedler, 1950; Lesser, 1961; and Snyder, 1961) with the conclusion by Levinson (1961) that some similarities may facilitate good relationships and therapeutic progress while others may be sources of impasses. Many research studies growing out of the client centered approach point to a few essential attributes of the therapist for effective psychotherapy. Rogers (1957) in an article giving what he believes to be the necessary and sufficient conditions for change, specifies empathy, warmth, and congruence on the part of the therapist. He further states that a therapeutic relationship is not unique unto itself but is a concentrated prolonged relationship which encapsulates "constructive qualities which often exist in part in other relationships" (p. 81). Research evidence supporting this comes from extensive studies reported by Truax and Carkhuff (1966) where they repeatedly found that the level of accurate empathy, warmth, and genuineness dem- onstrated by the therapist was directly related to positive changes in the patient. Finally, in a recent review of critical arguments concerning psychotherapy Berenson and Carkhuff (1967) concluded that the variables that prove to be significant contributors to the outcomes in therapy also apply to other kinds of interpersonal situations. They regard that such qualities as positive regard, genuineness, flexibility and openness are conditions of facilitative human encounters and living in general. From several diverse sources then investigations in the fields of friendship, interpersonal relations and psychotherapy all point to the central importance of such factors as openness, genuineness, warmth and empathy in relationships. Although much of the research evidence stem- ming from psychotherapeutic studies generalize to all inter- personal relations, this hypothesis has not yet been tested. Should the generalization of the central importance of the three factors be borne out in other types of relationships, it would provide additional evidence for the notion of a continuum of relationships. Research Concerning Situational and Process Determinants of Maneuvers In the past decade there has been increasing trend away from static factors such as traits toward an ever changing field approach. Originating in cybernetics and leading to the present game theories, such terms as strat- egies, maneuvers, tactics, systems, outcomes, and cost are being used more and more frequently to describe behavior. Relationships are viewed as a process in which both the members and the process change with each maneuver. These maneuvers however, are not random but are organized in greater or less degrees of complex strategies. In one of the most relevant accounts Haley (1963) employes a communication model to assert that when one studies a two person system, he enters the field of communication and must describe the system in terms which apply to the exchange of communicative behavior between two people. When people meet and begin to form a relationship they select certain types of behavior from a wide range of potential behavior. By this selection of certain behaviors they arrive at a mutual definition of their relationship. From the initial meeting then, each message that they exchange either rein- forces this definition or indicates a shift to a new kind of relationship. The attempt to define the relationship is viewed by Haley as a struggle for control of the defi: nition of that relationship, rather than control of the other person. In a similar vein Szasz (1961) emphasized the game theory of life. He postulated that life consists of a series of games with various levels and kinds of rules. PeOple learn certain games which they must play according to cer- tain rules, strive toward the goal of the game, and cannot change any one part of the rules without also changing other parts. His notions are somewhat more encompassing than those of Haley's but he would not argue with the latter's emphasis on the definition of a relationship as an essential factor, as are the rules of a game. Game theory and communication analysis are also used by Thibaut and Kelley (1959) to approach dyadic interactions in terms of learning theory, economics, and sociology. Their basic assumption is that every "individual voluntarily enters and stays in any relationship only as long as it is adequately satisfactory in terms of his rewards and costs." (p. 47). Withinthe set of persons available for interaction, there may be a variety of different reasons for choosing one person over another, depending upon the rewards from each. "In some cases the rewards depend on similarity (as in value support) and in others the rewards depend on dif- ferences which are in some manner complementary. The pre— diction of interpersonal relationships should, in our opin- ion, always take account of reward and cost considerations . . ." (p.47). Specifically referring to early interactions, they emphasize that each person explores the matrix of possible outcomes. If the initial outcomes compare favorably with alternative relationships, then there is motivation for continued approaches. Conversely, if the initial outcomes fall below the expectations of alternative relations there is decreasing motivation for continued approaches. Haley, Szasz, and Thibaut and Kelley converge in their thinking in their agreement on consistent strategies of maneuvers to define a relationship. These maneuvers can be subsumed under the rubric of three main types: those designed to further the relationship, those that one member employes to draw away from the relationship, and those ma- neuvers to confront or to change the definition of the re- lationship by hostile means.1 A consideration of the above points to several im- portant variables. First, certain qualities or character- istics of the individual are influential in the future course of a relationship. While such characteristics as needs and attitudes have been postulated as crucial variables, some 1It is interesting to note that similar types of maneuvers have been arrived at by Horney, (K. Horney, Our Inner Conflicts, W. W. Norton & Co., 1945) from her cofisid- eration of neurotic conflict. Faced with an insecure and hostile attitude toward the world, the individual has three modes of operation Open to him. He can try to get people on his side, try to dominate and fight them, or try to ig- nore them. Further, in her discussion of these modes her views come in close correspondance with Haley's discussion of strategies and the struggle for control of the relation- ship definition. 9 of the most convincing and clearest evidence comes from the work of Truax and his group. He found that the character- istics of genuiness, warmth and empathy were of great im- portance in psychotherapy dyads. If, as many have suggested, the psychotherapy relationship is a special case of dyadic relationships in general, might it not be expected that these same variables of acceptance of the other, sensitiv- ity to his feelings, and congruence within one's self also have meaning in determining the outcomes of a relationship? For this reason then, these three variables were chosen to be investigated. It further seemed important to consider the situ- ation in which the relationship to be defined occurs. The situation surrounding the relationship suggests certain maneuvers as appropriate or inappropriate. This is only a suggestion however as the person defining the relation- ship also defines the appropriate maneuvers. Thus, for example, a shop foreman would not address the owner by his first name unless given permission to do so. These sugges- tions are often rules of convention or socialized patterns of behavior which (1) exist in conjunction with the char- acteristics the person brings to the situation; and which (2) may be more or less flexible depending on how the rela— tionship is defined by the participants. The latter point led to the presentation of varying relationship definitions to the subjects in the present study. It was hypothesized that there would be a relationship 10 between the characteristics of genuineness, warmth, and empathy which an individual brings to a relationship and the situation and definition of the relationship--in this case, dyads instructed to define their relationship as acquaintances, friends, or to know the other dyad member. As acquaintances are thought to have less invested in the relationship and have more freedom to like or dislike each other, it was expected that they would allow for a greater number of maneuvers to terminate or express dissatisfaction with the other. In the present study hostile or confronting maneuvers are termed Moving Against (the other) maneuvers. Acquaintances may also be expected to display lower levels of Warmth and Empathy than friends or knowing dyads, steming from their low involvement with each other, the transitory nature of the relationship definition, and the freedom for. Moving Against maneuvers. As this study artifically or experimentally manip- ulated the relationship definitions e.g., to become friends, these subjects are expected to be less genuine in their encounter than either acquaintance or knowing dyads. Dyads instructed to become friends must make some efforts toward mutually satisfying rewards in the relationship. Hence, the hypothesis of higher levels of Warmth and Empathy as characteristics facilitating rewards. At the same time, in order for these dyads to be friends, the subjects must demonstrate their friendship in some fashion. Friends show 11 interest in each other, are at least somewhat attentive to what the other is saying, and work to resolve differences. These maneuvers are subsumed in the present study under Moving Toward (the other) maneuvers. Dyads in which the relationship is defined as learning all they can about the other person would not be expected to include angry or hostile maneuvers as appropri- ate to the situation. On the contrary, high levels of Warmth, Empathy, and perhaps Genuineness would appear to be characteristics which facilitate the relationship as it is defined. At the same time, to reveal oneself or to have an "open" relationship is often highly anxiety provoking and threatening. They are therefore expected to engage in revealing maneuvers only in conjunction with a high fre- quency of tactics designed to pull away from or to avoid the other in the relationship. Maneuvers such as these to avoid the relationship are termed Moving Away (from the other) in the present study. A final note concerning what the individual brings to a relationship must be considered. The early studies of Winch, Izard and others emphasized the needs of the in- dividual. In addition to needs, socio-economic factors have been suggested by Winch (1952) and others as influ- encing choice of partners, contacts, etc. It seemed impor— tant then to obtain some measure of gross personality char- acteristics and background information as possible 12 intervening variables in the relationships to be formed. A personality test Specifically oriented toward interper— sonal behavior, the Interpersonal Checklist, (ICL) was developed by LaForge (1963) and refined by Leary (1957). This test is summarized by two dimensions, Lov and Dom. The former ranges from warm, accepting behaviors to ingra— tiation, overly accepting and complient behaviors. The other dimension, Dom, ranges from independent, decisive, rational behaviors to harsh, critical and rigid maneuvers. As the Lov dimension has warmth as a component and behavior which involves reaching out in relationships, it was hypothesized that there would be a positive relation- ship between the Lov scores for subjects and their levels of Warmth and Empathy, as well as frequency of Moving To- ward maneuvers. Conversely, rigid, critical, unaccepting people i.e., high on Dom scale, are expected to display lower levels of Warmth and Empathy, and more frequent Mov- ing Against tactics. Hypotheses I. The general hypothesis with respect to the acquaint- ance process and the levels of empathy, warmth and genuine- ness is as follows: That individuals have characteristic styles or tactics which are employed in the process of getting acquainted which are characterized by Moving Toward, Moving Away or Moving Against and that these are correlated with levels of Warmth, Genuineness, and Empathy. 13 From this general hypothesis the following specific hypotheses were derived: II. A. There are a positive intercorrelation among. Empathy, Warmth, and Genuineness. There are negative intercorrelation between the levels of Empathy, and Warmth and the frequency of Moving Away and Moving Against behaviors. The general hypothesis regarding the relationship between "defined" relationships and levels of accurate em— pathy, warmth and genuineness is as follows: the levels of Genuineness, Warmth, and Empathy vary dependent on the type of dyadic relationship that is defined. III. A. Dyads instructed to become friends or to know the other person display higher levels of Warmth and Empathy than dyads instructed to become acquainted. Dyads instructed to become friends display lower levels of Genuineness than those instructed to become acquainted or to know the other person. The general hypothesis regarding the relationship between the acquaintance process and type of defined rela- tionship was as follows: Moving Toward, Moving Away, and Moving Against tactics occurring in a dyad vary with the type of relationship that is defined. A. Dyads instructed to get to really know the other person show more Moving Away tactics 14 than dyads instructed to become friends or to get acquainted. B. Dyads instructed to be friends engage in sig- nificantly more Moving Toward tactics than the other two groups of dyads. C. Dyads instructed to become acquainted display a greater number of Moving Against behaviors than the other two groups. IV. The general hypothesis regarding the relationship between sex of the subject and levels of Empathy, Warmth, and Genuineness is as follows: There is an interaction between the sex of the subject and the levels of Empathy, Warmth, and Genuineness. A. Females display higher levels of Empathy and Warmth initially but over time, or sessions, there is no difference in the levels between males and females. B. Males manifest significantly higher levels of Genuineness initially but over the sessions there is no difference between males and females. V. The general hypothesis regarding the relationship of personality variables to Empathy, Warmth, and Genuineness and to the acquaintance process is as follows: Certain gross personality dimensions are positively correlated with levels of Empathy, Warmth, and Genuineness and to Moving Toward, Away, and Against behaviors. 15 There is a positive relationship between the Lov score for subjects and the levels of Warmth and Empathy, as well as for the frequency of Moving Toward maneuvers. There is a positive relationship between the Dom score and Moving Against, and a negative relationship with the levels of Warmth, and Empathy. CHAPTER II METHOD Overview The basic conditions of the study involved selecting subjects from college students who were not previously ac- quainted, pairing males with females, and then randomly assigning the dyads to one of three types of relationships (to get acquainted, to be friends, or to really know the other). Each dyad met for nine half hour sessions which were observed and tape recorded. Each dyad member was Ob— served and rated on the over-all levels of Genuineness, Warmth, and Empathy displayed in a session as well as the frequency of Moving Toward, Moving Away and Moving Against maneuvers. Following the ninth session each subject com- pleted the Interpersonal Checklist as well as a rating sheet on his impressions Of the experiment and a demographic questionnaire. Subjects Subjects were unmarried Michigan State University students between the ages Of 18 and 22 enrolled in an in- troductory psychology course. Participation in psychology experiments was Optional but contributed extra-credit l6 17 points toward the course grade. Subjects were told that the experiment had to do with how people got acquainted. NO preselection Of subjects occurred except that they be unmarried and must sign the sheet Opposite a stranger of the Opposite sex. Thirty males and thirty females were selected, paired into male-female dyads and then the thirty dyads were randomly assigned to one of the three instruction groups yielding ten dyads in each condition. Materials The Movement Scale This scale was devised for the present study and consisted of fifteen significant common verbal and non- verbal interactions. The interactions were chosen tO be descriptive Of the three categories e.g., leaning forward, moving closer to the other was scored as Moving Toward; monosyllabic replies to questions about the self or to feelings were included as Moving Away; and such tactics as shaking a finger at the other were labeled Moving Against. (See Appendix A for complete scale.) An attempt was made to keep the scale on the level of observable data as much as possible in order to reduce the number of inferences to be made by the observers. Observers were trained to crite— rion prior to the study by Observing dyads similar to those to be used in the present study. At least 70% agreement between raters for each category was achieved, with a range of 70% to 89% agreement. 18 G—E—W Scales These scales were develOped by Truax and Carkhuff (1967) for use in psychotherapy. Genuineness, a 5—point scale ranged from Level 1 where the subject presents a facade to Level 5 where he is freely himself and is not playing a role. Warmth, also a 5—point scale, ranged from Level 1 indicating evaluation, conditional liking or dis- approval, tO Level 5 where the person warmly accepts and cares for the other person. Empathy, a 9-point scale, ranges from Level 1 indicating that the subject is preoc— cupied, disinterested, and misunderstands the other to higher levels Of empathy indicating that he is "with" the other, sensitive to him, and can communicate this. (See Appendix B for complete scales.) The raters for the G-E-W Scales were trained to criterion by listening to tapes of dyads similar to those used in the present study. The scoring was discussed until the raters achieved at least 70% agreement on the levels of each variable. ICL LaForge's Interpersonal Checklist was administered to each subject to describe himself and his partner. In this test subjects are presented with a list Of adjectives and asked to check each one that he feels applies to him (or to his partner). These adjectives yield two summary 19 scores, LOV and Dom, representing cross points on a matrix of eight personality dimensions. Lov scores range from warm, acceptance Of people to ingretiation and over- complience. Dom ranges from independent, decisive, rational behavior to harsh, critical, and rigid behavior. Thus both scales have an intensity factor built in. (See Appendix C for complete test.) Demographic Questionnaire This was a short questionnaire designed for the present study to provide information on the background of the subjects. It asked such questions as size of hometown, occupation of parents, college major, extra-curriculum activities, social activities, etc. (See Appendix D.) Rating Scale This short rating scale was also devised for the present study for use in post hoc analysis. Subjects were asked to locate themselves and their partner on a five- point scale of such factors as warm—cold, sincere-not sin- cere. They also located themselves on such dimensions as liking their partner, whether they would like to continue the relationship, and how well they felt they knew their partner. (See Appendix E for complete scale.) 20 Procedure Observers All raters or Observers used in the study were undergraduate students who participated as part Of an indi— vidual study course. Eight raters were used, four trained on the Movement Scale and four on the G-E-W Scales. Each subject was instructed to appear at a given place and time. The two subjects comprising a dyad were present at the same time. NO attempt was made to introduce them except to identify them as partners. Each member of the dyad was handed an instruction sheet (see Appendix F) which told that they were to participate in an experiment which related to the manner in which people interact or relate to each other, in addition to specifying their particular group instructions. For the remainder Of the experiment at the beginning Of each session the subjects were handed the instruction sheet to read again and reminded what their task was, e.g., to become friends, etc. The dyad was then taken to an interview room fur- nished with three chairs and a desk or table. There was a tape recorder placed on the desk and each subject wore a neck micrOphone. The experimenter, or an assistant, turned on the recorder and then left the room. At the end Of a half hour the session was terminated and arrange- ments made for the next meeting of the particular dyad. 21 Each dyad met together twice in a period of seven days. This procedure was repeated for each Of the 30 dyads until they had all met for a total of nine such half hour sessions. At the end of the ninth session, each subject was requested to arrange for another meeting, this time to complete the ICL, the demographic questionnaire, and the rating scale. Observational Measures The Movement Scale was employed to score the dyadic interaction in terms Of Moving Toward, Moving Away, or Moving Against maneuvers. Of the nine half hour sessions that each dyad met together, the lst, 3rd, 5th, 7th, and 9th were selected as sessions to be scored. The first session was not used in the data analysis and served as a "Warm-up" due to the presence Of the Observers in the exper— imental setting. Two Observers were seated in the room with the dyad during the scoring interviews1 and recorded a score for the most significant maneuvers in a thirty second interval for each member. The Observers were in- structed that each subject must be scored on at least one of the three categories Of the Movement Scale and no more 1Due to a shortage of Observation rooms with one- way mirrors it was necessary to have the Observers present. They were instructed tO be as unobstrusive as possible but it is likely that their presence had some undetermined effect on the S's. 22 than two categories for each thirty second interval. As the scoring intervals were thirty second periods, it was felt that a subject might use two different and equally important maneuvers in this time but to allow three types would dilute the significance of the maneuver for scoring purposes. The same category could not be scored twice in one scoring period. Thus the maximum score for a subject during one twenty minute session would be 80 (two different significant maneuvers for each time period), the maximum score for any one category was 40 (one score in that cate- gory for each time period), and the minimum over all score for a subject as 40 (one score in any category for each time period). Although each session was scheduled as a thirty minute period, it was necessary to allow for coats to be removed, etc. at the beginning and end Of each session. Consequently, the actual scoring was based on a twenty min— ute period beginning when E left the room and the tape re- corder turned on, and concluding twenty minutes later, re— .gardless of any continuation for a few moments past the twenty minutes. The G-E-W Scales were employed by independent raters who judged each subject in the dyad at the end of the same five scoring sessions as were used with the Movement Scale. Dyad members received only one score for a session consid- ered as a whole for each Of the dimensions Of the 23 G-E—W Scales. The ratings were made on the basis Of the tape recordings in an effort to keep the ratings as inde— pendent as possible as well as tO keep such "halo" effects as appearance, etc. to a minimum. All raters were informed as to the session number of the particular dyadic interac- tion to be scored but were unaware of the group to which the dyad belonged. Each subject then, received two inde- pendent scores every thirty seconds on the Movement Scale and two independent ratings for the interaction during the entire interview being scored on the G—E-W Scales, based on tape recordings. Subject Scales The ICL was administered following the ninth ses- sion. Each subject was asked to complete the ICL on him- self and on how he perceived his partner. This provided Lov and Dom summary scores for each subject and his partner. The Rating Scale was administered following the ICL to each subject as was the Demographic Questionnaire. At this time subjects were also given the Opportunity to ask any questions they wished concerning the experiment. Methods of Analysis The analysis was performed in two major divisions. First, each of the variables on the Movement Scale was an- alyzed individually in a 2 x 3 x 4 (sex x groups x sessions) 24 analysis of variance design with repeated measures on the last factor (sessions). The analyses Of variance for the G-E-W Scales, also performed for each individual factor, were 2 x 3 x 3 designs (sex x groups x sessions). It was not possible to use the data from Session 5 due to mechan- ical failure Of the recording machines, leaving Sessions 3, 7 and 9 for analysis. The second major portion of the analysis was per- formed by a correlation matrix in which the inter-correlations of selected variables was Obtained. The Lov-Dom scores for each subject were also entered in the matrix in addition to the G-E—W scores and the Movement Scale results. For both the scales measuring the dependent variables (the Movement Scale and the G-E-W Scales) inter-rater reli- ability estimates were obtained by the Pearson product— moment correlation coefficient. As two sets Of raters were used for each scale, it was necessary to compute two sep- arate reliability coefficients, one for each set. CHAPTER III RESULTS Inter-Rater Reliability for Movement Scale and G-E-W Scales A subject received a score on each dependent vari— able by two independent judges, yielding, therefOre, two scores on each variable. A Pearson product-moment corre- lation coefficient was computed for each variable between the two judges. These correlations ranged from r=.69 between the two Observers on Warmth, session 3, to r=.98 between the two Observers on Moving Against, session 3. (See Appendix 6 for complete table of reliability coeffi— cients.) The average reliability coefficients for each trial computed by a Fisher 8 transformation were sufficiently high to deem appropriate the use of the mean raw score given by the two raters as the raw data. Thus, all analysis of variance tests and correlation analyses were computed on the basis of the average Of the two scores that a subject received by the judges. The Definition Of the Relationship and the Strategies of Moving Toward, Moving Away, and Movinnggainst One Of the general hypotheses stated that the fre- quency of maneuvers on the Movement Scale varies with the type of relationship defined. An over-all picture Of the frequency Of maneuvers for each group in each session is 25 26 provided by Table 1. Visual inspection indicates that in each group, at least half of the total maneuvers were scored as Moving Toward whereas Moving Away and Against responses accounted for much fewer Of the total maneuvers. This dif- ference is shown more clearly in Table 2 by summarizing Table 1 into groups x type Of maneuver (mean and percent). Looking at each group individually, the percent of Moving Toward maneuvers is clearly greater than either Moving Away or Against for each of the groups during any session. Turning to a more important and meaningful compar- ison, Table 2 indicates some differences in the number of Movement Scale responses by each group. Analysis Of Vari- ance to test this general hypothesis is given in Table 3 for Moving Toward maneuvers. The results suggest that there is a tendency (.05>p<.10) for the groups to differ in the frequency of Moving Toward maneuvers. This tendency reaches statistical significance in Table 4 which shows the Analysis Of Variance results for Moving Away tactics. Here groups clearly differ in the frequency of Moving Away maneuvers (p<.05) but they also differed by session. As the groups x session interaction and the main groups effect were both significant (p<.05) a simple effects test (Winer, p. 311) was performed which indicated that during sessions 7 and 9 the Acquaintance group made significantly more Moving Away responses than the dyads instructed to be friends or to know each other (F=2.34,df 2,108, p<.10 groups for 27 m.b m.m m.m m.HH m.HN H.mH N.ma H.vm m.ow «.mh m.mb m.wm m Goflmmom N.h v.5 w.h m.m v.mH m.¢ m.om H.wm N.Mb o.mm m.mh o.vm h coammom o.m h.m o.w w.w v.5H H.0H ©.ma m.mm m.mh o.wm h.mh o.mo m coammom m.w m.w w.m m.mH m.om N.NH m.mm H.mm O.Hh m.mm m.H> m.om m Qoflmmmm w new: 30cm chHm vofl w new: 30cm mpcnm vos w new: 30cm mOch dos pmsflmmfl msfl>oz mmzm mcfl>oz Oumsoa mcfl>oz .mOme maHBOQM One .mO:0flHm .mocmncflmswom can ca AOOGHQEOU moamfiom One mmamzv muOOnQSm an OOME mm pmqflmmfi mcfl>oz one .mmsm mcH>oz .Onmsoa mcfl>oz mm Omnoom scammmm Mom mHo>5osmz Hmuoe mo omensoouomll.a mqmfia 28 M 1 - _ v.s m.mH 1 H m.mm s new: s.m sm.a m.s so.m H.am om.em mangoes _ . s.m am.a m.ma _ mm.m m m.mm es.sm mecmflum v.m hn.m N.mH mm.n v.mh ho.mm mocwpcflmsvofi w HODESZ 6m HOQEDZ mm HOQEDZ awe: new: new: pmswmmm mcw>oz hmsfi mcfl>oz OHmsoB mcfl>oz mamom DGOEO>OE on“ so mmsouw an own: QOHmmmm Hem mmmcommmm mo unmouom One Hmnadz cmozll.m mqmp<.10 30 TABLE 4.--Sex x Groups x Session Analysis of Variance for Movement Scale--Moving Away Maneuvers Source df SS MS F Between Subjects 59 6,495.06 Sex 1 1.43 1.43 p<.10) to be significant in an increase in Empathy levels. Finally, the Analysis Of variance on levels of Warmth, depicted in Table 9, shows neither the type of rela- tionship nor the number Of session to be significant (p>.10). However, the hypothesis Of a significant difference between sexes does find support from the data (see Table 6). Both Tables 8 and 9 indicate that sex is significant (p<.05) for levels of Warmth and Empathy. Analysis of this main effect by means Of the Newman-Kuels procedure (Winer, p. 310) revealed that females are significantly higher than males in levels Of Empathy and Warmth. The results concerning the G-E-W Scales indicate then that contrary to expectations, the level Of Genuine— ness, Warmth, or Empathy that a subject displays in the relationship does not vary with the type of relationship defined in the present study. There is some tendency TABLE 8.--Sex x Groups x Session Analysis Of Variance for G-E-W Scale--Empathy J -_ v:— - Source df SS MS F Between Subjects 59 85.44 Sex 1 6.24 6.24 4.46** Groups 2 2.80 1.40 1.00 Sex by Groups 2 .57 .28 <1.00 Subj. within Groups 54 75.83 1.40 Within Subjects 120 94.50 Sessions 2 3.86 1.93 2.38* Sex x Sessions 2 .71 .36 <1.00 Groups x Sessions 4 .47 .12 p<.10 **p<.05 39 TABLE 9.-—Sex x Groups x Session Analysis Of Variance for G-E-W Scale--Warmth Source df SS MS F Between Subjects 59 58.03 Sex 1 6.05 6.05 6.80* Groups 2 2.70 1.25 1.15 Sex x Groups 2 1.23 .62 <1.00 Subj. within Groups 54 48.05 .89 Within Subjects 120 79.25 Sessions 2 2.51 1.26 1.88 Sex x Sessions 2 2.71 1.36 2.03 Groups x Sessions 4 1.11 .28