Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An empirical investigation of dyadic verbal interaction in the chronic paediatric health care delivery… MacKinnon, Joyce Roberta 1980

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1980_A1 M23_5.pdf [ 7.72MB ]
JSON: 831-1.0076989.json
JSON-LD: 831-1.0076989-ld.json
RDF/XML (Pretty): 831-1.0076989-rdf.xml
RDF/JSON: 831-1.0076989-rdf.json
Turtle: 831-1.0076989-turtle.txt
N-Triples: 831-1.0076989-rdf-ntriples.txt
Original Record: 831-1.0076989-source.json
Full Text

Full Text

AN EMPIRICAL INVESTIGATION OF DYADIC VERBAL INTERACTION IN THE CHRONIC PAEDIATRIC HEALTH CARE DELIVERY SYSTEM by JOYCE ROBERTA MACKINNON B.O.T., McGill University., 1966 M.A., Western Michigan University, 1975 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF. DOCTOR OF PHILOSOPHY ^ . J THE FACULTY OF GRADUATE STUDIES (Interdisciplinary Studies) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1980 (c) Joyce Roberta MacKinnon, 1980 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Joyce R. MacKinnon Department nf Graduate S t u d i e s The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date A p r i l 10, 1980 11. ABSTRACT The primary objective of this study was to analyze dyadic verbal interactions and to determine whether they were associated with the roles of the participants. These "interactions occurred in the chronic paedia-t r i c health care delivery:system between parents of handicapped children and physicians and between those same parents and other members of the health team. Additional objectives of this research included testing of the r e l i a b i l i t y of the Sequential Analysis of Verbal Interaction (SAVI) instrument and its u t i l i t y in the health care system. Clinical data were obtained during regularly scheduled appointments in the form of audio-taped interviews using 37 parent-professional 'and 37 parent-paraprofessional dyads. A six-minute sample was selected from each interview tape, coded at three-second intervals, transcribed into the class of communication behaviour and analyzed. Subsequent to data collection, r e l i a b i l i t y and u t i l i t y of the SAVI instrument were examined and determined to be appropriate for this study. The major finding of the study was that parents of handicapped children used different verbal messages and behaviours than professionals and paraprofessionals. The communication pattern for a l l three groups, using Agazarian.'s (1968) model was cross-purpose. Under a proposed model, adapted from Agazarian's, the parents' pattern of communication was considered to approximate the problem-solving pattern more closely. The conclusion drawn from this exploratory study was that very l i t t l e of a personal nature was occurring in interpersonal communication, which in turn hindered the development of a problem-solving pattern of communication. i i i . An important direction for further research would be the testing of the predominance of the cross-purpose pattern of communication using a larger and more homogenous sample of professionals and paraprofessionals. Dr. :Varice F. Mitchell Thesis Supervisor iv. CONTENTS Page ABSTRACT i i LIST OF TABLES. vi LIST OF FIGURES v i i i ACKNOWLEDGEMENTS ix CHAPTER I - INTRODUCTION 1 The Concept of Role 3 Communication 8 The Health Care Delivery System 14 CHAPTER II - RESEARCH FRAMEWORK 23 Conceptual Model 24 The SAVI System 26 Validity 32 Development of Hypotheses 35 CHAPTER III - METHOD. . 42 Pilot Study 42 Subjects and Setting 44 Procedure 47 Methods of Data Analysis 53 CHAPTER IV - RELIABILITY AND GENERAL CHARACTERISTICS OF THE DATA 56 Reliability 56 Characteristics of the Data 68 CHAPTER V - RESULTS OF HYPOTHESES TESTING 74 Hypotheses 1-20 75 V. Page Comparisons of Theoretical and Observed Verbal Profiles of Groups 85 CHAPTER VI - DISCUSSION AND IMPLICATIONS 93 Discussion 93 Implications 108 Conclusions of the Study 113 Future Research 116 BIBLIOGRAPHY 119 APPENDICES A. SAVI DEFINITIONS 128 B. SAMPLE FORMS 136 C. CODING SYMBOLS 138 D. TALLY SHEET 139 E. COMPUTATION FOR THEORETICAL MODELS 140 F. INDIVIDUAL ANALYSES 143 G. PERCENTAGES OF VERBAL MESSAGES AND BEHAVIOURS USED BY PARENTS, PROFESSIONALS AND PARAPROFESSIONALS 146 v i . LIST OF TABLES Table Page 1. Demographic Characteristics of Parents, Professionals and Paraprofessi onal s 45 2. Intra-rater Reliability 59 3. Inter-rater Reliability - Coders in Geographic Proximity 61 4. Inter-rater Reliability - Coders Geographically Separate 62 5. Summary of Coding Errors 64 6. Total Coded Three-second Units of a Six-minute Sample of Verbal Interaction Between 37 Parent-Professional and 37 Parent-Paraprofessional Dyads 69 7. SAVI Category Frequency for Each Group 71 8. Number, of Avoidant Behaviours Used by Parents and Professionals 75 9. Number of Avoidant Behaviours Used by Parents and Paraprofessionals 75 10. Number of Approach Behaviours Used by Parents and Professionals 76 11. Number of Approach Behaviours Used by Parents and Paraprofessionals • 77 12. Number of Contingent and Approach Behaviours Used by Parents Within Factual Messages 77 13. Number of Contingent and Approach Behaviours Used by Professionals Within Factual Messages 78 14. Number of Contingent and Approach Behaviours Used by Paraprofessionals Within Factual Messages 79 15. Number of Factual and Orienting Messages Used by Parents 79 16. Number of Factual and Orienting Messages Used by Professionals 80 17. Number of Factual and Orienting Messages Used by Paraprofessionals 80 v i i . Table Page 18. Number of Factual and Personal Messages Used by Parent-Professional Dyads 81 19. Number of Factual and Personal Messages Used by Parent-Paraprofessional Dyads. . . . 82 20. Number of Contingent and Approach Behaviours Used by Parent-Professional Dyads Within Personal Messages . . . 82 21. Number of Contingent and Approach Behaviours Used by Parent-Paraprofessional Dyads Within Personal Messages . 83 22. Spearman Rank Correlation Coefficients Between Theoretical and Observed Problem-Solving Patterns of Communication and Levels of Education 86 23. Spearman Rank Correlation Coefficients Between Observed Patterns of Communication of Dyads of the Same Sex With Dyads of the Opposite Sex 87 24. Spearman Rank Correlation Coefficients Between Theoretical and Observed Patterns of Problem-Solving Communication and Appointments 88 25. Spearman Rank Correlation Coefficients Between Theoretical and Observed Problem-Solving Patterns of Communication and Frequency of Interaction of Dyads. . . 89 26. Spearman Rank Correlation Coefficients Between Theoretical and Observed Problem-Solving Patterns of Communication and Acquaintance of the Dyads 90 27. Spearman Rank Correlation Coefficients .Between The Three Theoretical Patterns With the Observed Pattern of Communication for Each Group 91 28. Spearman Rank Correlation Coefficients Between Theoretical and Observed Patterns of Communication With Different Types of Paraprofessionals 96 29. Spearman Rank Correlation Coefficients Between Theoretical and Observed Patterns of Cross-Purpose and Problem-Solving Communication From Other Researchers' Studies 97 vi i i . LIST OF FIGURES Figure Page 1. Conceptual Model 24 2. SAVI Nine Square Grid 29 3. SAVI Theoretical Communication Patterns 31 4. SAVI and Proposed Theoretical Communication Patterns. . . 105 ix. ACKNOWLEDGEMENTS The completion of any major effort requires reliance on others for guidance, support and stimulation. My f i r s t acknowledgement is directed towards my Committee Chairman, Dr. Vance F. Mitchell, who cl a r i f i e d the path to my goal, alleviated my anxieties, and guided me with wisdom and sensitivity during the past four years. My Guidance Committee: Dr. Merle Ace, Dr. Bryan Clarke, and Dr. Brenda Morrison deserve special tribute as they gave so generously of their time and shared their own invaluable and unique expertise. The privilege of having been guided by a truly interdisciplinary team has been a deeply meaningful experience. The Dean of Graduate Studies and his Associates have supported this interdisciplinary endeavour for which I am most grateful. I also wish to acknowledge the assistance of the university examiners, Dr. Ronald Taylor and Dr. Roy Turner for.their contribution and efforts on my behalf. Thanks are directed to Dr. Thomas Mallinson, who introduced me to the SAVI instrument and to Dr. Yvonne Agazarian and Dr. Anita Simon, developers of SAVI, who showed interest and gave support in this research. In the process of a dissertation, there are numerous tasks that need careful attention and assistance. Bio-medical communication's staff assisted with equipment selection and tape duplication. Professionally trained coder, Maria Showalter-DiEgidio coded the data enthusiastically and competently. Programmer, Ronald Sizto played a major role by inter-acting with the computer for me. Colleen Colclough typed the final draft with accuracy and speed. X. Two practising physicians deserve special recognition. Dr. Kenneth Turnbull, whose sensitivity to interpersonal communication, c l a r i t y of thought and knowledge of the health care delivery system, enriched my learning. Dr. Geoffrey Robinson helped in the selection of the paediatric centres and acted as a liaison person for me. Without the help of two very special friends,"the path to my goal would have been, at times, impassable. For the past two years, Lynn Meredith generously volunteered her time in typing verbal transcripts and edited versions of each chapter scrupulously and competently. Margo Buchanan edited the manuscript, shared my concerns and cared for the outcome. I am also grateful to physicians, paraprofessionals, and the many parents of handicapped children who trusted my intrusion enough to share very personal and confidential interviews. Last, but not least, I wish to thank my entire family for their support and confidence in me. This manuscript is dedicated to my father, who never saw i t to completion, and to my mother. 1. CHAPTER I INTRODUCTION A comprehensive paediatric health care delivery system includes out-patient clinics where chronically handicapped children receive treatment and where progress is monitored. Here parents can meet and engage in dialogue with the professionals and various paraprofessionals involved in the care of their child. Many of the ideas for the research reported here resulted from 12 years of experience working with handicapped children and their parents both in Canada and the United States. Inadequacies in the humanistic and communicative aspects of care were confirmed as an area of concern by many professionals, paraprofessionals and parents over the years. The major objective of this research lay in systematically;exploring patterns of verbal interaction which occurred when parents communicated with professionals and paraprofessionals. Through the analysis of tape recorded interviews, i t was hoped that insights would be gained regarding the patterns of communication between individuals. Other objectives of this study included testing the r e l i a b i l i t y of the Sequential Analysis of Verbal Interaction (SAVI) instrument, reported in chapter four, and examining its u t i l i t y in chronic paediatric health care settings as dis-cussed in chapter six. Two key areas emphasized in this research were role and communication. The concept of role is complex. Role definitions and theories were dis-cussed and those appropriate to this study were extracted for elabora-tion. The role of parent of a handicapped child, of professional and of paraprofessional was reviewed since i t represented the major independent conceptual variable under examination. Communication was divided into 2. factors affecting organizational communication and interpersonal communica-tion. Dyadic verbal interaction was the dependent conceptual variable measured by the SAVI instrument. The health care delivery system also represented an important aspect of this study. The verbal interaction which occurred within the chronic paediatric health care delivery system might carry implications for improvements within that special type of organizational system. 3. The Concept of Role In the present study the roles of parents of handicapped children, professionals and paraprofessionals are compared as the incumbents engage in verbal interaction. Differences, too, are observed between the commun-ication profiles of participants who had been socialized into various role behaviours. Thus a brief review of the literature on role defin-itions, role theory and roles is considered appropriate. Role Definition and Theory Many authors have attempted to define role with much obfuscation arising because the term means different things to proponents within various disciplines. The following writers have found l i t t l e agreement on a definition of role (Bates, 1956; Berlo, 1960; Calhoun, 1973; Goffman, 1959; Levinson, 1958; Linton, 1936; and Parsons, 1951). For example, Linton (1936) differentiated between role and status. Status, he elabor-ated, is a collection of rights and duties while role represents the dyna-mic aspects of a status. Levinson, (1958) defined role as an aspect of organizational physiology which involves function, adaptation and process. Calhoun (1973) simplified the definition of role by stating i t is a study of interaction. For the purposes of this study role is defined as the interactive aspects of a collection of rights and obligations. Similar to writers on role definitions, proponents of role theory hold different tenets (Bertrand, 1972; Biddle and Thomas, 1966; Braga, 1972; Ekel, 1969; Gordon, 1966; Levinson, 1958; Sarbin, 1954; Simon, 1976; Thomas,. 1966;.and Thomas_.and Biddle, 1966). Sarbin (1954) extracted from the basic theory the notion of interaction and role but to i t added a second type of interaction, that of self. His theory aims at attending to the 4. structures within the organism and the environment and their interaction. From a global perspective, he noted, "contemporary role theory regards human conduct as the product of the interaction of self and role" (p. 223). Levinson (1958) provided yet another theoretical approach when he demonstrated an interest in the individual within organizations. A member in an organization may have as great an impact on the structure of an organ-ization as i t imposes on the individual. For this reason the author developed what he refers to as organizationally given role demands and personal role definitions. Forces within an organization influence indi-viduals to adapt while forces within individuals lead them to select cer-tain forms of adaptation over others (Levinson, 1958). As Levinson (1958) stated: "In short, every individual and every sociocultural form contains within i t s e l f the seeds of it s own destruction - or its own reconstruc-tion" (p. 179). Roles Roles are important to consider because they represent a major focus of this study. A comparison is drawn between roles of parents, profes-sionals and paraprofessionals and their resultant communication profiles. Roles are a complex composition of idealistic and re a l i s t i c behaviours as perceived by oneself and others (Berlo, 1960) and are displayed through individuals' overt conduct in confirming and justifying their roles (Sarbin, 1968). Individual and Organizational Roles Perlman (1974) stated that vital roles are an integral part of the personality, c r i t i c a l to the well-being of the individual, while forming the underpinning of other valued l i f e roles. Individuals, according to 5. Ruddock (1969), receive personal satisfaction by seeking out roles'which are accepted by others. Temporary or permanent changes in roles must be accepted and integrated by the individual and approved by significant others (Kolb, 1967). The range of roles open to an:individual varies depend-ing on resources - money, age, appearance, speech and education (Ruddock, 1969). Individuals' perceived roles and their roles as defined by society have a significant effect on how they act and others react (Kast and Rosenzweig, 1974). Role behaviours are guided by past perceptions and experiences (Laing, Phi 11ipson and Lee, 1966). The feelings and emotional reactions of individuals are more important than their role or theoretical orientations (Rogers, 1969) while interpersonal role relationships are based on emotions more frequently than logic (Selye, 1974). According to Simon (1976), the roles of individuals in an organiza-tion are shaped by their goal identification with,and their location in, the organization. In organizations the roles of individuals in authority positions have declined because of decentralization, the challenge of pro-fessionals, and an increased emphasis on communication (Leavitt, 1972). When an individual outside an organization seeks service from an organization, communication strains result from dissimilarities in role expectations (Lennard and Bernstein, 1966), differences in status (Carkhuff, 1969), a threatening situation (Thomas, 1971), divergent personalities (Traux and Mitchell, 1971) and the experience of the helper (Bergin, 1971). The Role of Parents of Handicapped Children Since many family functions have been taken over by large bureau-cratic organizations, norms governing the roles of parents are not clear (Mott, 1965) and they become more ambiguous with the advent of a handi-6. capped child (Perlman, 1974). Ross (1972) found that the mother's roles are nurturer and protector. A father's primary role commences, he stated, when the child reaches school age and he becomes the child's planner and provider for the future. With a defective child, he noted, the mother frequently turns her attention to the child and away from her husband. Therapy frequently serves, accord-ing to Ross, to cement the mother-child bond and to dilute the father-child and husband-wife bonds. The Role of Professionals According to Gartner (1976), medicine, law and theology are the established professions. Applied Research Associates (1971) in Canada stated they were unsuccessful in defining a professional. I l l i c h (1977) de-scribed a professional as: "dominant, authoritative, monopolistic, legal-ized" (p. 19). A car mechanic is a professional according to his descrip-tion. The cloistered environment of medicine, according to Thorne (1973), creates a strong loyalty and commitment to the guild, which covertly teaches a condescending attitude even toward other professions with egali-tarian ideologies. The roles of physicians are changing and in the future they will need to: continue to change (Poynter, 1971); become self-evaluative in their interpersonal s k i l l s (Matarazzo, 1971); be involved in research and in helping people to adjust to disease processes (Crichton, 1970); and refer patients to other and new disciplines (Evans, 1975; Evans, 1979; Lalonde, 1974; Lewis, 1969; Rice, 1975; and Stoekle and Zola, 1972). 7. The Role of Paraprofessionals Paraprofessionals differ from professionals in that they are not quite peers though they voice similar concerns (Hughes, 1973). Their education is less intense, of lower quality and less theoretical (Applied Research Associates, 1971). In addition, paraprofessionals lack autonomy but have a responsibility to both the client and the employer (Gartner, 1976); and developed from the i n i t i a t i v e of young people, not from professional associations (Applied Research Associates, 1971). The paraprofessionals in their roles can serve as buffers between the physician and the consumer (Foulkes, 1974), or can assume many a c t i v i -ties carried out by physicians (Evans, 1979; Lalonde, 1974; and Rice, 1975). In summary, the concept role has been defined, elements of role theory extracted and the roles of parents, professionals and paraprofess-ionals have been elaborated. 8. Communication It will be recalled that roles and communication are the two major concerns of this research. Communication is the major focus now under discussion. The resulting communication profiles of the participants, as measured by the Sequential Analysis of Verbal Interaction (SAVI) system represent the dependent variable under investigation. The professionals and paraprofessionals in this study were interacting within an organiza-tional structure while the parents interacted outside this structure. Individuals engaging in a communication process, whether in organizational communication or interpersonal interactions are affected by many factors. Some of these factors are briefly outlined in the following review. Communication Process Communication has been defined by Rogers and Agarwala-Rogers (1976) as a process that involves the sharing of information between two or more people. According to Schreivogel (1972) communication is an action process taking place in time and space, influenced by cultural and personal, exper-ience and shaped by simultaneous events. Writers who have defined commun-ication concur that i t involves an internal process on the part of the individual (Boulding, 1970; Fabun, 1974; Miller, 1967; Schreivogel, 1972; Weick, 1969; and Westley and MacLean, 1970). Problems in communication erupt, according to Goffman (1959), when dissonance exists between the cognitive and affective components of a message. The expressiveness of individuals may embody two different kinds of activity - the expression individuals "give" and the expression they "give off". The former expression contains more of the factual content of the message whereas the latter is loaded with factual but also affective 9. . connotations of the message (Goffman). The more credible the communicator, according to March and Simon (1970), the more readily the message is accepted as valid. Similarly, the credibility of the communicator is important in the eyes of the recipient for a threat to be effective (Watzlawick, 1976). People most resistant to change, according to Bauer (1969), are most attentive to the message. He noted that our social model of the process of communication is morally asymmetrical and f o r t i f i e s the model of a one-way exploitative process of communication. Generally, he added, the commun-ication process must be regarded as a transaction - each gives in order to get. There are certain f a c i l i t a t i n g and inhibiting factors which affect the accuracy of communication. Hovland, Jam's and Kelley (1964) reported research evidence which indicates that communication is affected by the communicator's intentions, trustworthiness and expertise. Asch (1970) discovered that people can be persuaded to agree verbally by group pressure but cannot be coerced to act against their judgements. Partridge (1970) stated that talking to people in a linguistic style they comprehend creates a more positive relationship. The author called this using "correct punctuation". Simon (1976), while emphasizing the role of formal communication, stressed the more important role of informal communication. According to him, factors determining whether individuals become involved in informal communication depend on: personal motives, attitudes, a b i l i t y to influence and the position of authority. Effectiveness of the communication, according to Thayer (1968), must include knowledge about the individual's personal/values since they are c r i t i c a l to accepting or rejecting the communication. We must, he added, 10. develop not only better ways of "communicating-to" but also better ways of being "communicated with" (p. 39). The communication process occurring within an organization is vital to its functioning. Organizational communication, therefore, is discussed in the following section. Organizational Communication Organizational communication is concerned with communication as i t takes place within the organization and in its environment. According to Rogers and Agarwala-Rogers (1976) organizational communication differs, then, from other types of communication in that i t is structured and geared toward some formality and predictability. Gibson, Ivancevich and Donnelly (1973) indicated that the structure of the communication network affects the accuracy of the messages, the task performance of the group, and the satisfaction of group members. Some of the factors they suggest which may influence communication nega-tively or positively include: hierarchical position, personality and per-ception, leadership, the division of labor, financial status, information exchange within the organization, and the grapevine of informal communica-tion. The guidelines they recommended to afford effective communication include utilization of feedback, use of multiple channels and the u t i l i t y of simple language. Katz and Kahn (1970) focused on the organization's reactions to communication overload. In an organization, communication acts as controlled and coordinated stimuli linking the decision centres in a synchronized pattern (Scott, 1975). Simon (1976) referred to the importance of transmittal of informa-tion not only to the decision centres but also from these centres to other parts of the organization. 11. In small group communication Bales (1973) stressed the importance of role and status in determining states of equilibrium. Similarly, :Blau and Scott (1973) noted when status differences are present they distort the error-correcting power of social interaction (reluctance to c r i t i c i z e those in high status positions). Wessen (1972) found that communication in a hospital ward is channelled along occupational lines where differences in education and socio-economic backgrounds decrease communication. Examining physicians' and surgeons' communication, Hage (1974) discovered that physicans emphasize communica-tion and recognize the need for socialization and feedback whereas surgeons do not as i t impinges on their authority and status. Formal communication does not provide for the active ingredients of socialization and feedback since i t is usually of an instructional nature (Hage). He added that a l l organizations do have socialization and feedback but to varying degrees. An appropriate conclusion of this section is presented by Huse and Bowditen's (1973) statement that feedback, perception and interaction can not only help communication within the organization but also aid the organization as a system for social exchange and interchange. Interpersonal Interaction Interaction is a two-way process of exchange, whereas communication may involve only one exchange (Huse and Bowditch, 1973). Interaction fre-quently involves risk-taking and creating the desired impression (Goffman, 1959), or the "faulty interactant" who makes the participants feel uncom-fortable (Goffman, 1957; and Gross and Stone, 1973). Interaction tends to oscillate between problem-solving attempts and social-emotional responses (Bales, 1972). A number of factors may affect interpersonal interaction. Bales 12. (1970) pointed out that people are frequently more concerned about the con-tent of what they are saying than the process of interaction which is intuitively understood by most listeners. Similarly, Schreivogel (1972) noted that people must mean what they say; their messages and gestures must agree or the meaning is lost. Experimentally, i t has been shown by Homans (1950) that as the frequency of interaction between individuals increases, their activities become similar. Individuals of equal rank interact more frequently than those from unequal ranks and higher ranking individuals in i t i a t e more interactions than those from lower ranks (Homans). Equalization of power is c r i t i c a l to problem-solving interactions ( F i l l e y , 1975). Interaction may be affected by what takes place within individuals. Schreivogel (1972) stated that communication begins in the interior of individuals while absorbing the world within themselves and subjecting i t to their intellect and emotions. This interior world, he added, is com-posed of culture, environment, personal likes and dislikes. Similarly, Selye (1974) described man's interactions as being guided much more fre-quently by emotion than logic. Man, he stated, tends to use logic later, i f at a l l , to rationalize the emotional act and to pursue his course more efficiently. Interpersonal balance, according to Foa (1966), refers to equal communication between two persons, a significant feature of inter-personal interaction. Self identity, Laing, Phi 11ipson and Lee (1966) stated, is the individual's perception of self plus the perception one believes others hold and is an important dimension of interpersonal inter-action. L i t t l e research has continued in the area of self-identify, according to Cook (1971), because there has been no attempt to improve the method of research. Problems in interaction can occur when the norms of reciprocity are ignored and one's behaviour is contingent upon the 13. other's behaviour (Wilmot, 1975). External factors affecting interaction include such elements as eye contact and space, distance, time and posture. According to Argyle and Dean (1973), eye contact serves in a variety of functions of which the most important is feedback, together with signalling that the channel is open. Eye contact, they noted, may serve as an indicator of approach or avoid-ance as i t relates to distance. Sommer (1972) studied distance and inter-action and reported situations where"interaction is discouraged, such as in a library where people choose distant seats. This, he compared to s i t -uations where interaction is encouraged as in a cafeteria where people tend to s i t across from one another. According to Giffin and Patton (1971), interacting over time tends to train both persons to respond in specified ways thus making the interaction more stable. Postures, of which there are approximately 30 different ones in American culture, may affect inter-actions (Scheflen, 1972). These postures vary within subgroups in i n s t i -tutions, social classes, geographic regions and personality types and they vary according to age, gender, status, position and health (Scheflen, 1972). These general and specific factors, then, encourage or discourage inter-action. In summary, internal and external factors affecting the communication process in general and organizational and interpersonal interaction specifically were identified and discussed in this section. 14. The Health Care Delivery System This investigation focused in part on the verbal patterns of communi-cation within two dyads - parents of handicapped children with professionals, and those same parents with paraprofessionals. The health care delivery system is related to this study by virtue of the verbal interactions taking place within this system which in turn might carry implications for the delivery of health care services to children. Organizations and Systems According to Simon (1976), the term organization refers to the "com-plex pattern of communication and relationships in a group of human beings" (p. x v i i ) . During its functioning an organization: 1) moulds and develops personal qualities and habits; 2) provides for the execution of authority and influence; and 3) structures communication, thereby determining the environments in which decision are made (Simon, 1976). In this study the term health care encompasses medical care in addi-tion to health education, prevention, maintenance and rehabilitation ser-vices. Medical care refers to care given by physicians (professionals) while hospital or sickness care involves the management of acute disease or injuries by professionals or paraprofessionals. The delivery systems are the assembly of agencies approved by government to provide care and service. Chronic paediatric health care delivery systems include out-patient agencies providing primarily the rehabilitation aspects of health care to physically and neurologically handicapped children. Health care organizations are subsumed under the rubric of the health care system. Systems and organizations are similar, according to Scott 15. (1975), because both are concerned with growth and s t a b i l i t y , individuals and their interactions. Their differences, he noted, relate to their focus, which for a system is concern for every level (human and non-human) of function whereas an organization attends primarily to the human level of function. Jaco (1972) described the evolution of health care as unrelated, in-dependent and divergent programs which survived and developed cooperative relationships that emerged as health care systems. Frequently, voluntary associations and religious sects developed, organized and maintained a multitude of health care services which augmented similar services pro-vided by governmental and public hospitals at the federal, state, county and city levels. Health care service was only a century old while hospi-tals had seen three centuries of "unplanned development" thereby reflecting the dominant interest of acute care and the relative lack of concern for health care (McKeown, 1975). Hospitals might indeed be one of the reasons for the problems in our health care system because in the l a t t e r 1 s develop-ment, attempts were made to simulate hospitals with a l l their attendant attitudes, power, values and continuing deficiencies in the humanistic, interactive and communicative aspects of care (Gaudry, 1975). Issues Related to the Health Care Delivery System Problems in the health care f i e l d have been attributed to a variety of sources including: 1) the power of physicians, 2) lack of funds, 3) lack of focus on preventative medicine and poor maintenance of services, and 4) inadequate communication within the organization and with clients. Power of Physicians Foulkes (1974) stated that the principal power rested with the medical profession because government had not assumed control or leadership in the 16. health care system. According to Lalonde (1974), the Canadian physicians' power evolved because of their role in the control of infectious and para-s i t i c diseases, advances in surgery, lowered infant mortality and develop-ment of new drugs. Medicine, according to 111ich (1977), obtained the power to. dictate what constitutes a health need and therefore became a corporate agency with a mandate to test the whole population and identify the group of potential patients. Turnbull (1979) viewed the current power vested in physicians as a function of: their expertise, their comprehension of health care problems, their i n i t i a t i v e in seeking solutions, and the influence of politicians. He questioned whether average citizens could understand the complex health perspective sufficiently to alter their l i f e styles and to set their goals high enough to maintain an improved l i f e style. The politicians, other professionals and the general public lack the overall expertise to cope with these problems. At the same time, he stated that many physicians view their input into medical objectives, methods and standards as being eroded by increased government interference. Physicians have been accused of increasing their t e r r i t o r i a l boundar-ies, resulting in a greater span of control. According to I l l i c h (1977), as the profession increased its span of control, results became counter-productive - the more medicine, the more illnesses. Foulkes (1974) and Gaudry (1975) opposed this view saying that physicians devoted their exper-tise to acute care patients thus neglecting and resisting the development of preventative and chronic care, which was subsequently initiated by other non-medical specialists. As an avenue for reducing medical dominance, Foulkes (1974) prescribed decentralization of service through regionalization incorporating consumer 17. participation and greater utilization of paraprofessionals. Similarly, Lalonde (1974) advocated a multidisciplinary team approach to care that would dilute the power of the doctor and provide service at a reduced cost. Turnbull (1979) maintained that: physician-directed nurse practi-tioners can significantly cut the costs of routine care; acute care is best managed by physicians; and financing and economics may be handled by accountants and business administrators with direct input from physicians. The c r i t i c a l remarks made by politicians concerning the power of physicians seem unfounded. Their power, according to Evans (1979) and Lalonde (1974), can be regulated by provincial governments and, according to Zola (1977), by agencies such as the World Health Organization which in 1948 gave the medical profession f i r s t claim over the label of illness and anything related to i t . Lack of Funds Gaudry (1975) attributed our impoverished Canadian health care delivery system to financial decisions pointing out that 1% of the budget for health care was spent on research and development; h% was spent on the delivery of health care research; and the remaining 98^% went primarily toward meeting the direct cost of "sickness care" delivery. In his view, governments are not spending enough on health care maintenance and preventa-tive measures. "Ounces are proffered where tons are needed" (p. 210). According to the Foulkes' report of 1974, the Government of British Columbia must now be prepared to lead, re-organize, and innovate i f patterns of health care are to be improved and costs controlled. Recently, Evans (1979) advocated cutbacks in new technologies (e.g., body scanning, computerized brain and foetal monitoring) since they un-necessarily raise the costs of health care. According to Turnbull (1979), 18. however,, these non-invasive potentially life-saving technologies can fre-quently identify and in many cases aid in the prevention or the therapy of central nervous system damage. The short term savings made possible by reduced technology utilization seem counterproductive to the long term costs of treatment or institutionalization for handicapped children and adults. Evans (1979) believed that part of the health care funding problem resulted from the presence of too many practising doctors who could be re-placed by less costly personnel. The Canadian Medical Association (1979), however, has been concerned about the shortage of doctors in certain areas and the numbers of Canadian trained doctors emigrating to the United States (1 in 500). During 1979 there seemed to be a continued tightening of monies in health care. In the course of the present study this investigator had the opportunity to explore in an informal way some of the problems of health care deemed important by those most closely involved. Certainly the p r i -mary problem seen by physicians and their a l l i e d colleagues was a lack of funds. With more monies, problems could be alleviated such as: attract-ing persons with higher qualifications and expertise, providing more con-temporary equipment, and providing better f a c i l i t i e s for care and treat-ment. Anxieties were expressed over two proposed hospitals directly con-cerned with the prevention and i n i t i a l care of handicapped'children. Severe budgetary cutbacks were making these f a c i l i t i e s seem inadequate to many health care workers even prior to their completion. Lack of Focus on Prevention and Maintenance of Service Lalonde (1974) stressed the need for a change in the Canadian health care value system. He saw a need to raise the level of "care" to the same 19. level of importance as "cure" before sufficient attention would be paid to the needs of people with chronic or intractible illness. Lalonde pointed out that of the 10 major causes of death in 1900, six were due to infections whereas in 1970 only one of the major causes of death was an infection. Because our health care system is based on curing the sick, he elaborated, treatment of the chronically i l l is not very satisfying for practitioners. Treatment involves long term care with the goal of arresting the progress of a disability rather than curing i t . Turnbull (1979) stated that chronic care delivery has been slow to develop due to scarcity of monies but noted that inroads are being made in British Columbia. Long Term Care (1979) now provides home care or in-residential care ($6.50 per day, regardless of income) for a l l ages from young adults to the elderly. I l l i c h (1977), in his diatribe against medical practices, proposed making individuals responsible for their own health but he did not suggest the medium for removing the structures, barriers and controls of the pre-sent system. Turnbull (1979) noted that efforts to alter people's l i f e styles or have them assume responsibility for their own health have not been very encouraging. He cited,for example, the public's increased knowledge that overeating can lead to obesity and that smoking may lead to lung cancer seems to have had l i t t l e effect in changing habits. Robinson (1979) stated that some preventative medicine has stemmed from the combined efforts of medicine and po l i t i c s . He cited the example that seat belt legislation in British Columbia was inspired by the Provincial Medical Association and put into effect by the Provincial Government. Preventative medicine in this instance had to resort to punitive measures (fines), un-fortunately, to realize its effectiveness (Turnbull, 1979). 20. Inadequate Communication A number of observers have cr i t i c i z e d the adequacy of communication be-tween professionals and with clients. Foulkes (1974) attacked the medical profession for f a i l i n g to accept the reality that much of health is depen-dent upon the quality of one's relationship with others in the environment. According to Gaudry (1975), inadequate communication exists at the system's level, is at the root of many health care issues and therefore cannot be resolved by funding alone. A formal policy, he stated, or some means of sharing knowledge and research findings needs to be instituted throughout the health care system. Applied Research Associates (1971) in Canada stressed the need for fostering inter-professional communication and action. They noted that professionalization was sought by those seeking status, income and autonomy thereby narrowing the focus and creating "no trespass-ing" signs between professionals. Through interviews the researchers found professionals rated their f i e l d as preeminent in a l l respects and appeared unable to determine their relationship with other professionals and the general community. Strauss (1972) identified the major problem in communication as the imposition of middle class values on the poor and a lack of communication shown towards them. Another author, Straus (1972), stated that communica-tion was both understudied and underrated - not only "communication with and about patients, but communication between members of the medical team as well" (p. 259). Turnbull (1979) contended that communication between physicians, while constrained by time does occur regularly during in-service education, medical rounds, seminars and peer review. The physician-client interaction is frequently hampered, however, by insufficient time which can be one of 21. the major factors resulting in inadequate communication. The foregoing discussion of issues in the health care'delivery system highlights some of its complexities. The intertwining of systems at a l l levels, each of which is closeted in invisible boundaries, poses a well-nigh impossible task in ferreting out the pol i t i c a l from the medical re-sponsibilities of health care. Problems exist in differentiating health care from medical care or medical from hospital care, and professional from paraprofessional. The terms themselves are used interchangeably but with very different meanings implied. The complexity of the health care delivery system predictably results in communication problems between and within sub-systems. In addition, lack of funds may trigger power struggles within the system leading to the exclusion of clients or professionals whose positions are not securely entrenched. It is suspected personnel within this system become "navel gazers" since their own domain is a l l they have time, energy, interest or expertise in protecting - a natural phenomenon in our changing, complex and autonomous society. The advice of Gaudry ( 1 9 7 5 ) is well taken when he recommended every-one must strive to make the: health care delivery system more equitable, more efficient and more rational. Otherwise, reforms  will be attempted by ideologues and medical  i l l i t e r a t e s (p. 207). Summary In this chapter an exploration was undertaken of the two major areas of roles and communication and the related area of the health care delivery system. Roles of parents, professionals and paraprofessionals were re-viewed. Factors affecting the communication process, organizational commun-ication and interpersonal interaction were identified and discussed. 22. Issues related to the health care delivery system were examined. They in-cluded the power of physicians, lack of funds, lack of focus on preven-tion and maintenance of service, and inadequate communication. 23. CHAPTER II RESEARCH FRAMEWORK Introduction In this chapter a research framework is presented including the variables, the instrument and the hypotheses used in this study. First, a conceptual model is developed wherein the major conceptual variables under examination are graphically displayed. Next, the theory underlying the testing instrument, the Sequential Analysis of Verbal Interaction (SAVI) is described. Following this, the validity of the SAVI system is discussed. Finally, the major hypotheses to be tested are presented. Definitions Some terms found in the following conceptual model and throughout the paper are used with specific meanings intended and are defined below. Role: This term refers to the interactive aspects of a collection of rights and obligations of an individual. Roles of the three following groups are defined. Professional: In this study professional refers only to physicians. Paraprofessional: Paraprofessionals are "not quite peers" of the pro-fessional "although they voice common concerns" (Hughes, 1973, p. 11). They include: occupational therapists, physiotherapists, speech patholo-gists, social workers, psychologists, special educators, and community health nurses. Parents: Parents refer to those persons who are the natural, foster or adoptive parents of a chronically handicapped child. Chronically Handicapped Child: This term refers to an individual ranging 24. in age from birth to 18 years who has sustained an injury or impairment lasting longer than six months in any or a l l of the following domains: physical, mental or emotional. Conceptual Model The model for this research consists of conceptual independent, dependent and intervening variables as depicted in Figure 1. FIGURE 1 CONCEPTUAL MODEL Independent Variable Intervening Variables Dependent Variable Role Education Experience Frequency of Interaction Length of Acquaintance Type of Appoint-ment Age Sex Verbal Interaction Pattern (SAVI) The conceptual independent variable refers to the role of parents, professionals and paraprofessionals. An individual's role behaviour results from the articulation of personal, organizational and environmen-tal roles as discussed in chapter one. These roles, according to Newcomb (1949), tend to be highly persistent. The role attributes of a mother are proposed to be nurturer and protector while those of a father are planner and provider. Professionals evolve from a medical or predominantly 25. "curing" model whereas most paraprofessionals develop from a health or primarily "caring" model, thus individuals in different roles were expected to demonstrate different components of verbal behaviours and messages. The conceptual dependent variable, the SAVI system, consisted of 40 categories of verbal messages and behaviours. In this study the 40 cate-gories were collapsed into a three-by-three matrix to enable three types of verbal behaviours (avoidant, contingent and approach) and three types of verbal messages (personal, factual and orienting) to be examined. The conceptual intervening variables included: education, sex, experience, age, reason for the v i s i t , frequency of interaction and the length of acquaintance of the interactors. Despite similar levels of education and different ideologies, no significant differences in communication patterns were found by Cox (1976) between professionals and clients during verbal interaction. It was sus-pected by this investigator that this finding occurred because of the similarity in education, not the divergent ideologies as proposed by Cox. Because professionals' and paraprofessionals 1 roles are typically f i l l e d by males and females respectively and mothers typically attend c l i n i c s , sex of the interactors could be a factor in the communication pattern. Mothers, i t might be expected, would find a common bond with other females rather than with males since communication through familiar roles is easier (Newcomb, 1949). When c l i n i c v i s i t s are regular a l l participants have prior knowledge of typical procedures. On the other hand, a consultation or i n i t i a l v i s i t might be anticipated to raise the anxiety level of the participants be-cause of theiunknown nature of the v i s i t and thereby affect the communica-tion. If the frequency of interaction and the length of acquaintance is extended over time i t could result in more personal and approach levels 26. of communication than during the f i r s t encounter. Having described the conceptual model, the next step is to examine the SAVI system. The SAVI System The fundamental question posed in this study concerns the nature of the communication between parents of handicapped children and professionals on the one hand, and between those parents and paraprofessionals on the other. In order to study these communication profiles the Sequential Analysis of Verbal Interaction (SAVI) system was selected. This system in its conceptual form represents a communication model and in its opera-tional form provides a systematic approach to the analysis of inter-actional data. According to the authors, Simon and Agazarian (1967), "All verbal behavior is defined as problem-solving behavior which either approaches or avoids the problem of communication" (p. 17). These approaching and avoid-ing behaviours may be viewed as driving and restraining forces in solving what and how to communicate. Every act conveys a message and every message has two interdependent components - what people are talking about, called "factual" information, and how people are relating to each other, "personal" information. Factual information is viewed as raw material for task achievement while the personal portion is the raw material for the development of interpersonal relations. Vagueness in either of the components of a message leads to ambiguity within the message and the communication process and is labelled dissonance. When one component is missing, the receiver has to second guess what i t would have been i f i t were included. Because both factual and personal information can be relayed at both cognizant and non-cogni-zant levels, dissonance can arise in different ways and in different com-binations. The typical response to dissonance in communication, however, is to increase the effort to convey factual and avoid personal information For effective communication to occur the problems of both What and How to communicate must be resolved (Agazarian, 1968). An important factor in communication is the relevance of the informa-tion to some purpose or goal and whether the goal is explicit or implicit. The problem of communication (the how) must be resolved before any infor-mation (the what) can be used toward a goal or purpose. An explicit goal occurs when people deliberately and consciously direct their behaviour toward a certain purpose. An implicit goal may be private or public, known or unknown. When people behave as i f they are pursuing one aim but claim they are seeking another, the behaving is the implicit goal whereas the claiming refers to the explicit goal. When these two goals are in different directions, the behaviour will relate to the implicit because i t takes priority over the explicit goal (Agazarian, 1968). Problems in communication revolve around three concerns, according to Agazarian (1968): 1) maintenance issues (building interpersonal rela-tions by sharing personal information; 2) procedural issues (decision making, management and structuring); and 3) goal issues (problem-solving or explicit goal achievement). These three concerns are priorized and un-less the problems of communication are resolved, people cannot effectively work on problems. Thus, maintenance work must be done before other kinds of work can proceed. For effective problem-solving to occur, i t is not sufficient that people communicate effectively. They must also convey information that relates to the problem they have come together to solve -the explicit goal. 28. Using SAVI,three kinds of verbal behaviour are identified which alter the potential for the transfer of information: 1. Avoidance behaviors make i t more likely that information they convey will be misunderstood, distorted or lost in that they move away from solving the problem of HOW to communicate and increase stress arising from unsolved problems. 2. Approach behaviors make i t more likely the information they contain will be transferred accurately and will move towards solving the problem of HOW to communicate. 3. Contingent behaviors are neutral verbal behaviors which make no contribution ito reducing dissonance in communica-tion. Whether the information they convey is used for problem-solving is contingent on the respondent's reaction. (Agazarian and Simon, 1976) The system further divides each of the verbal behaviors into Personal, Factual and Orienting messages. Descriptions of these messages are as follows: 1. Personal messages are always interpersonal. 2. Factual messages contain information that"is basic to data input and data processing. • 3. Orienting messages orient the direction of communication and carry the information necessary to judge, evaluate, and direct the phases of the problem-solving process. (Agazarian and Simon, 1976). The SAVI categories of verbal behaviours and messages are content free; that is they do not take any specific subject matter into account (Agazarian, 1968). Appendix A contains the operational definitions of the SAVI categories and sub-categories. These categories, when combined, form the nine SAVI theoretical classes of verbal behaviour as illustrated in Figure 2. These nine classes are as follows: 1. Personal-Avoidance Messages are directly related to interpersonal stress. 2. Factual-Avoidance Messages.sidetrack the information flow into fantasy or comic r e l i e f which are disruptive at the time but in the 29. FIGURE 2 SAVI NINE S Q U A R E GRID u z < 9 o > P E R S O N A L M E S S A G E Attacking 1 Attacking Self Defensive Complaining & Blaming Sarcastic Jokes F A C T U A L M E S S A G E Story-telling Small Talk Reminiscing Funny Stories O R I E N T I N G M E S S A G Obscuring Thinking Out Loud Being Obscure Oughtitudes z U J O z r — Z 8 Personal Introduction Informing Personal History Credentials Personal Position Personal Questions Facts and Figures General Information Factual Questions -General Questions Orienting Opinion, Speculation Commanding & Proposing Encouraging & Discouraging Yes-Buts X u < O cc Q _ < Interpersonal Relating Summarizing & Respondir 8 Personal Answer Personal Sharing Personal Support Taking a Stand Affectionate Jokes Factual Answer General Answer Clarifying Issues Paraphrasing Summarizing Interactive Building Building on Ideas Work Jokes (c) Agazarian & Simon, 1974 30. long run may serve a useful purpose. 3. Orienting-Avoidance Messages deflect the information flow into abstractions and condensations that defy mutual understanding. 4. Personal-Contingent Messages contain information that may very well be used as factual information. 5. Factual-Contingent Messages perform the function of putting data in the bank. Their use is contingent on whether people process the data or not. 6. Orienting-Contingent Messages contain information which may orient the direction of communication in terms of time, in terms of levels of abstraction and in terms of sequence of content. 7. Personal-Approach Messages are interpersonal and preserve the person in the communication process from being dehumanized. 8. Factual-Approach Messages perform the function of processing infor-mation by summary, clar i f i c a t i o n or answers. 9. Orienting-Approach Messages contain the kind of information neces-sary for consensus. These messages can only exist when people are interacting at the factual and interpersonal levels. (Agazarian and Simon, 1976) If most of the talk classified into the categories in Figure 2 accumu-lates in the upper l e f t four squares, a fight is probably in process. If most of the talk, however, f a l l s in the lower right four squares, the most probable activity is work. Cocktail party talk would appear predomi-nantly in squares four and two, while monologues disguised as dialogues would appear in squares five and six, and intimacy in squares four and seven. How people talk to each other (what kinds of verbal behaviours they use) will yield different patterns of distribution within the nine classes of behaviour. These have been labelled by Agazarian (1968) as three dis-tinct patterns - defensiveness, problem-solving and cross-purpose (see Figure 3). A high frequency in one of these classes implies a distinctly different group climate from that found when the frequency is high in a different class. FIGURE 3 SAVI THEORETICAL COMMUNICATION PATTERNS Defensive Pattern of Communication Personal Factual Orienting Avoidance H M H Contingent L H M Approach L M L Problem-solving Pattern of Communication Personal Factual Orienting Avoidance L L L Contingent M H H Approach M M M Cross-purpose Pattern of Communicati Personal Factual Orienting Avoidance L M L Contingent L H H Approach L M L Relative Frequences: L = Low («7%) M = Medium {> 7 < 20%) H = High (> 20%) 32. If the frequency is high in the avoidance classes of behaviour, the overall pattern indicates defensive communication. It may be inferred that even i f goal relevant information has been conveyed, i t was probably processed in the defensive pattern. In problem-solving, however, there is a greater probability that approach behaviours are involved and would carry information related to the explicit goal. The cross-purpose pattern has the greatest frequency of talk in the contingent area. It represents a heavy input of information without much evidence of that information being processed or used. Interpersonal relationships in the cross-purpose pattern are not maintained and may, in fact, be ignored (Agazarian, 1968). In summary, the SAVI system was selected to test the.hypotheses for the following reasons: 1) the general kinds of verbal behaviours, types of messages and patterns of talk which can be extracted from the theoreti-cal framework; 2) the specificity and appropriateness of the categories; and 3) the content free nature of the system. Having described the SAVI system, the validity of the instrument will now be examined. The r e l i a b i l i t y of the * instrument was tested and is reported in chapter four. Validity Validation in general appears to depend on the researcher's needs at the time, and the degree of testing to which the instrument has been exposed. Evidence for the validity of the SAVI system is presented in this section. Having reviewed most of the "classic" articles on validity (American Psychological Association, 1967; American Psychological Association, 1975; Anastasi, 1969; Bechtoldt, 1959; Campbell, 1960; Campbell and Fiske, 1959; Campbell, 1976; Cronbach and Meehl, 1955; Kaplan, Bush and 33. Berry, 1976; Kerlinger, 1966; Loevinger, 1967; Nunally, 1967; Nunally and Durham, 1975; Royce, 1977; and Tenopyr, 1977), i t seemed that the greatest problem associated with the SAVI system would be amassing empirical evi-dence attesting to the existence of the system's theoretical constructs. The literature suggests that high r e l i a b i l i t y gives credence to the validity of the constructs. All other types of validity feed into and may be utilized as evidence that the constructs do exist and act as hypoth-esized. Simon and Agazarian (1967) stated that the SAVI system appears to have face validity but needs construct and predictive validation, category by category (p. 16). Empirical evidence has accumulated since, which gives support to predictive and construct validity as indicated by results of the following studies. All studies examined reported both intra and inter-rater r e l i a b i l i t i e s of approximately 80%. The r e l i a b i l i t y portions of these studies appear to have been performed under stringent conditions and appear to provide some evidence that the categories exist and must be relatively discrete since people can discriminate between them. The studies were extensive: three were doctoral dissertations (Browne, 1977; Cox, 1976; and Harvey, 1976) and the other was an exploratory study by one of the developers of SAVI (Agazarian, 1968). The fact that the SAVI system was developed with a strong base involv-ing communication, systems, and behavioral theories also helps to support the construct validity of the instrument. Further, the system's theoreti-cal framework appears logically deductive and has been found to support a number of hypotheses. Cox (1976) found support for her three hypotheses when she used an adult psychiatric population in interaction with thera-pists representing three models of treatment. Browne (1977), using fre-34. quentv.and infrequent attenders at a Family Practice in interaction with nurse practitioners and physicians, reported support for her hypotheses. Harvey (1976) found clear trends in verbal behaviours when the messages (personal, factual and orienting) were partitioned from the verbal behaviours (avoidant, contingent and approach) across the l i f e of a group. Agazarian (1968) investigated the relationship between the set of SAVI categories and two outside content variables of verbal behaviour: target-ness, when the verbal behaviours were on target with the main goal of the group, and convergence, when the message was on the same topic. Agazarian (1968) postulated there would be a dynamic relationship between the parti-cular type of verbal behaviour and the content of the information i t con-veyed. When a group pattern of cross-purpose talk was examined a relation-ship between targetness and ten SAVI categories and convergence and nine SAVI categories was found. The group pattern, she explained, represents , an intervening variable that influences the transfer of information. This determination of SAVI category relationship with independent c r i t e r i a of content is a further step towards validation of the SAVI instrument. In summary, the results of previous researchers' findings appeared to provide sufficient evidence of the validity of the SAVI system to permit its use in this research. Pilot testing, however, will provide additional evidence. The next stage involves the development of the hypotheses. 35. Development of Hypotheses Role was selected as the major independent conceptual variable because of i t s importance and relative s t a b i l i t y over time (Bowman, 1949; and Newcomb, 1949). Filley (1975) found that people adopt roles comfortable to them and prefer to interact-with others whose role behaviour is predict-able.- When people attempt to change their roles they are frequently punished by their friends because of the unpredictability of their new behaviour, even though the change is for the better ( F i l l e y ) . Newcomb's (1949) notion that people communicate better through familiar rather than unfamiliar roles was supported by McGill's (1976) finding. She found that the major independent variable influencing the communication profile is the dyad and that communication tends to follow occupational lines. Paraprofessionals tend to be in marginal or linkage positions where i t might be expected that their role behaviour would strive to appease the demands of both physicians and parents. On the other hand, physicians might be expected to be less concerned about acceptance and therefore more willing to take risks because of their positions of authority and power. In support of this expectation, Cox (1976), in her study of psychiatrists and doctoral level psychologists and social workers, found that psychia-t r i s t s use more extreme behaviours (approach and avoidance) than either of the other two groups. Since the primary purpose of professionals and paraprofessionals is to provide help they might be expected to demonstrate more approach behaviours than would parents. Browne (1977) reported that nurse practi-tioners and physicians demonstrate more approach behaviours than clients regardless of the client's behaviour. She found, also, that unfamiliarity with a client creates more avoidant than approach responses from health 36. professionals. Clients, she noted, are the greatest users of avoidant behaviours. From this finding i t was expected that parents, out of self defensiveness and fear, would show more avoidant behaviours (fight.or, flight) than either professionals or paraprofessionals. Paraprofessionals, because of their marginal role and orientation to a "coping" model, would tend to engage in other than avoidant behaviours; and professionals, because of their ultimate power and their orientation to a "curing" model, might afford the risks of engaging in avoidant and approach behaviours. Filley (1975) stated that trusting (approach) cues evoke trusting behaviour and distrusting (avoidant) cues e l i c i t distrusting behaviours. Browne (1977) found that patients' approach and avoidant behaviours relate to professionals' approach and avoidant behaviours. In this study i t was expected professionals and paraprofessionals would use approach behaviours more frequently than parents. Thus the following hypotheses were proposed regarding verbal messages and behaviours displayed by parents, professionals and paraprofessionals. Hypothesis 1 Parents use a greater proportion of avoidant behaviours than professionals. Hypothesis 2 Parents use a greater proportion of avoidant behaviours than paraprofessionals. Hypothesis 3 Professionals use a greater proportion of approach behaviours than parents. Hypothesis 4 Paraprofessionals use a greater proportion of approach behaviours than parents. 37. Hypothesis 5 Parents use a greater proportion of contingent than approach behaviours when only factual messages are examined. Hypothesis 6 Professionals use a greater proportion of contingent than approach behaviours when only factual messages are examined. Hypothesis 7 Paraprofessionals use a greater proportion of contingent than approach behaviours when only factual messages are examined. Hypothesis 8 Parents use a greater proportion of factual than orienting messages. Hypothesis 9 Professionals use a greater proportion of factual than orienting messages. Hypothesis 10 Paraprofessionals use a greater proportion of factual than orient-ing messages. Hypothesis 11 Parent-professional dyads use a greater proportion of factual than personal messages. Hypothesis 12 Parent-paraprofessional dyads use a greater proportion of factual than personal messages. Hypothesis 13 Parent-professional dyads use a greater proportion of contingent than approach behaviours when only personal messages are considered. Hypothesis 14 Parent-paraprofessional dyads use a greater proportion of contin-gent than approach behaviours when only personal messages are.con-sidered. Another group of hypotheses were proposed which considered the years of experience each of the participants had accrued in their respective role 38. as parent, professional or paraprofessional. It was expected that as years of experience increased so would confidence in their role and therefore, the use of approach behaviours. Younger participants when compared to older participants might be expected to act more defensively about guard-ing, protecting and validating their role, and thus use fewer approach behaviours. Hypotheses 15 to 20 are stated as follows: Hypothesis 15 For parents there is a positive relationship between years of experience as a parent and percentage of approach behaviours used. Hypothesis 16 For professionals there i s J a positive relationship between years of experience as a professional and percentage of approach behaviours used. Hypothesis 17 For paraprofessionals there is a positive relationship between years of experience as a paraprofessional and percentage of approach behaviours used. Hypothesis 18 For parents there is a positive relationship between age and percentage of approach behaviours used. Hypothesis 19 For professionals there is a positive relationship between age and percentage of approach behaviours used. Hypothesis 20 For paraprofessionals there is a positive relationship between age and percentage of approach behaviours used. Another area of investigation'involved the examination of three theoretical patterns of communication proposed by Agazarian (1968). These three patterns of defensiveness, problem-solving and cross-purpose talk were discussed earlier and are displayed in Figure 3 (p. 31). For problem-solving activity to occur, time must be spent developing interpersonal 39. relationships and approach behaviours must be used. Problem-solving activity turns into cross-purpose talk when both interactors are talking and agree-ing but neither is listening. Thus, in cross-purpose talk the communication is devoid of the humanizing elements of personal and approach behaviours. According to Filley (1975), problem solvers share a l l relevant informa-tion and direct their energies towards defeating the problem. Frequently, well intended helpers impose their own' values and therefore "may be dictating what the other party should want rather than providing what the other party does want" (Fill e y , 1975, p. 27). Health professionals can so easily f a l l into this trap of dictating what the parents should need rather than consulting with them on what they do need. Often, parents have goals for their child while professionals and paraprofessionals have formed different goals for the same child. Thus, problem-solving communi-cation is replaced or never occurs because cross-purpose talk takes pre-cedence. In cross-purpose talk much information is given by both participants but l i t t l e of i t is used. It seemed c r i t i c a l to determine whether cross-purpose talk takes precedence over problem-solving talk and to examine when the problem-solving pattern is most active. Both patterns were examined, therefore, with emphasis on investigating the problem-solving pattern. Finally, the following expectations were examined with respect to the intervening variables of education, sex, type of appointment, frequency of interaction, and length of acquaintance. Education As the level of education of parents and paraprofessionals in-creases so does the correlation between the theoretical and observed problem-solving patterns of communication. 40. Dyads of the same sex show a higher correlation between their observed patterns of communication than dyads of the opposite sex. Appointment Regular vi s i t s result in a higher correlation between the observed and theoretical problem-solving patterns of communication than i n i t i a l v i s i t s . Frequency of Interaction As the frequency of interaction between dyads increases so does the correlation between the observed and theoretical problem-solving patterns of communication. Length of Acquaintance As the length of acquaintance increases so does the correlation between the observed and theoretical problem-solving patterns of communication. Role and Pattern Parents', professionals' and paraprofessionals' observed patterns of communication correlate higher with the cross-purpose theoretical pattern than with the problem-solving or defensive theoretical patterns. Problem-solving The paraprofessionals 1 observed pattern of communication correlates higher with the problem-solving theoretical pattern of communica-tion than that of either the parents or the professionals. Summary A conceptual model was explained depicting the relationship thought to exist among the major phenomena observed in the present study. In the model, the role behaviour of parents, professionals and paraprofessionals was the independent conceptual variable. The SAVI system with its result-ant category frequencies of verbal messages and behaviours which could be transposed into a communication profile became the dependent variable. A number of intervening variables believed to influence the dependent variable were delineated. The theory of the SAVI system was described and its validity discussed. The variables in the conceptual model were the major input for the development of the hypotheses which were presented and elaborated. 42. CHAPTER III METHOD In this chapter the method of data collection is described. Informa-tion is provided regarding pilot testing, subjects and setting, data collection and processing procedures, and the stati s t i c a l techniques utilized. Prior to contact with any of the population to be sampled the projected study was approved by the University of British Columbia's Ethics Committee. Different f a c i l i t i e s were involved to obtain an adequate sample size of professionals, paraprofessionals, and parents who had children with similar types of chronic disabilities treated in the greater Vancouver area. Dr. 6. Robinson, professor in paediatrics and a practising paedia-trician, helped to identify the four f a c i l i t i e s from which data were obtained. Of these four, one was an assessment centre; another was an ambulatory c l i n i c in a paediatric hospital; and the remaining two were out-patient paediatric treatment f a c i l i t i e s . Pilot Study At the outset a pilot study was conducted: to test the adequacy of the equipment to be used in data collection; to examine the appropriate-ness of parental consent and participant demographic forms developed for the study (Appendix B); to test the i n t e l l i g i b i l i t y of the explanation to be given to the various participants; to ensure that an adequate degree (approximately 70% agreement) of intra-rater r e l i a b i l i t y could be established; and to examine construct validity of SAVI. The data were collected by audiotape recordings of c l i n i c a l inter-views between professionals and parents of handicapped children, and between 43. paraprofessionals and parents of handicapped children. Tape recorders appropriate to this research were selected after testing them in a variety of room situations with background noise typical of that found in c l i n i c and treatment situations. The demographic form was tested on parents, paraprofessionals and professionals. Since the form resulted in problems of interpretation the researcher was present to interpret when necessary and to check that each was accurately completed. During the pilot test i t became apparent that the briefer the explanation presented to participants, the better. However, questions raised by the participants were answered. Basically the type of information given to the parents was as follows: I am doing a research project at this Centre. The purpose of my research is to examine the way doctors and other health professionals communicate with parents of handicapped children. Dr. and Ms. (Mr./Dr.) have agreed to participate. It involves tape recording your session with and then with . Will you help? SAVI Check Subsequent to the testing of equipment, the forms and explanation were used in two real situations involving parents and professionals. The two tape recordings obtained from these sessions provided the data from which talk samples were selected. These sample data were coded according to the SAVI system and then recoded six weeks later. The r e l i a -b i l i t y between the f i r s t and second coding was 72%. Since the researcher is not a trained coder this figure was deemed sufficiently high to ensure the appropriateness of the SAVI system to the research situation. Trained individuals, however, whose r e l i a b i l i t y could be expected to be higher were used to code the data of the main study. During the pilot study the SAVI categories seemed exclusive and sufficiently well defined for the 4 4 . coder to make coding decisions. Because the constructs underlying the SAVI instrument did appear to exist and act as anticipated, construct validity was considered adequate to proceed with the research plan. Subjects and Setting Selection Once consent was obtained from the paediatric centres involved, each individual professional's (physician) agreement to participate was sought.. All physicians contacted agreed to participate in the research. A c l i n i c schedule, then, was obtained and contact was made with one of the para-professionals involved with the parent of the handicapped child. The time most frequently selected was that just prior to, or immediately following the parent's appointment with the physician to ensure that the researcher obtained both tapes and that some degree of random assignment was included. Once professional and paraprofessional agreement was obtained the researcher made contact with the parents on the day they arrived for their c l i n i c , therapy or assessment appointment. The researcher explained to each parent the purpose of the study and obtained their written consent. Some basic demographic data were collected on a l l participants. This in-formation was gathered during the usual waiting period prior to the appoint-ment. Examples of demographic and parent consent forms may be found in Appendix B. Demographic data characteristics of the participants are included in Table 1. Parents A total of 3 7 parents interacted with both a professional and a para-professional , thus making 74 dyadic interactions. TABLE 1 DEMOGRAPHIC CHARACTERISTICS OF PARENTS, PROFESSIONALS AND PARAPROFESSIONALS Parent Professional Paraprofessional Number 37 7 26 Age - (range) 22-71 32-63 25-56 - (mean) 33 47 35 Sex (males) 2 6 2 (females) 35 1 24 Education Elementary 11 Secondary 14 Training 8 Diploma 2 8 Bachelor's 2 10 Master's degree 6 Ph.D. degree 2 Medical Doctor 7 Experience (in years) Range 1-48 1-43 1-25 Mean 9 19 10 Appointment Regular 35 17 18 Special 39 20 19 Acquaintance First v i s i t 45 21 24 Less than 1 year 17 8 9 More than 1 year 12 8 4 Frequency of Interaction First Visit 45 21 24 1-2 times/year 10 8 2 3-12 times'/year 12 8 4 Every week or more 7 0 7 Parents = 37 (32 natural, 3 foster and 2 adoptive) Professionals Seven professionals interacted with the 37 parents. Four of these professionals were taped with six different parents, the f i f t h with seven, the sixth with four, and the seventh with two different parents. Taping the professionals on different occasions with different parents provided an opportunity to observe changes in the professionals' communication pro-f i l e s over situations and time. The seven professionals included: four paediatricians, an orthopaedic surgeon, and two physiatrists. A total of 37 professional-parent interactions were taped. Paraprofessionals Twenty-six paraprofessionals were included in the study. There were three community health nurses, five social workers, eleven physiothera-pists, three occupational therapists, one educational consultant, one psychologist, and two speech pathologists. Five paraprofessionals expressed concern about participating. Three of these refused and two agreed with reservation. All were occupational therapists. These paraprofessionals were not from one f a c i l i t y but spread throughout the four organizations involved. It must be pointed out that the researcher is an occupational thera-pist. This similarity in background may have posed a greater threat to these individuals, fearing the exposure of inadequacies. Background similarity, on the other hand, may have permitted them to express their lack of interest more openly. It is believed, however, that the former is more indicative of the true situation since none of the people in the population was known to the researcher previous to the data collection. Setting The actual taping session took place in the office, c l i n i c or treat-ment room typically used by the professional or paraprofessional. This 47. posed some problems since i t meant taping, on occasion, in large treatment rooms where other activity was in progress. The tape recorder was placed in open view. The researcher was not in the room during taping. This decision was made following the pilot study as the researcher believed her presence was intrusive. Remaining outside the room also relieved the researcher of the pressures of immediate feedback as to the quality of the participants' interpersonal s k i l l s or in providing tips as to how to improve their communication. Tape Rejection For this study i t was necessary to have two tapes on each parent -a tape of the parent with the professional and a companion tape of the parent with the paraprofessional. Five tapes had to be discarded because companion tapes were not obtained. In one instance the professional turned off the tape recorder because the child was crying and forgot to restart the machine. There were three occasions when the parents were taped with professionals, but because of the inappropriateness of the child's referral to that centre, paraprofessionals' interview with the parent was can-celled. In another instance a change of schedule resulted in the parent being interviewed by the paraprofessional but not by the professional. No tapes were discarded because of breakage. Procedure It will be recalled that 37 professional-parent and 37 paraprofessional-parent interviews, varying in length from eight minutes to an hour and a half, were collected. A total of six minutes of conversation per tape was used for data analysis. Two, three-minute segments were selected, one from the middle of the f i r s t half and the other from the middle of the 48. second half of each tape. The total amount of conversation was determined by noting the numbers on the tape recorder's digital counter when conversa-tion began and ceased. Once the two mid-points on the tape were computed mathematically, a stop watch and the digital counter were utilized to determine the one and a half minutes to the right of these points. To deter-mine the one and a half minutes to the l e f t of these points, the same number of digits were used to estimate the point and then a stop watch was again used to ensure a f u l l minute and a half" had elapsed. The digital counter could not be used alone because of speed variation of digital counters. A typescript of each six-minute sample of tape was prepared and three-second intervals were marked off. The selection of the six-minute sample from each tape recording was based on the findings of Agazarian, 1968; Browne, 1977; and Harvey, 1976, each of whom talked about the appropriate-ness of using this time frame for samples. In fact, Browne demonstrated she could use a three-minute sample with interviews of greater than 10 minutes in length with a high degree of accuracy (p. 179). The typescripts and tapes were air expressed to a professionally trained coder of the SAVI system. This trained coder was utilized to elim-inate researcher bias. The coder was not aware of the hypotheses being tested. The necessary 60 hours for the investigator to learn SAVI coding were not deemed feasible in terms of time or costs. SAVI Coding Following the coding of a l l the data from the typescript into one of the 45 SAVI categories, the codes (Appendix C) were transferred to a t a l l y sheet (Appendix D). There was a t a l l y sheet for each participant to allow the individual's portion of the dyadic interaction to be examined separately. The frequencies from the t a l l y sheets were then transferred to a Fortran 49. coding form together with the individual's demographic information in pre-paration for keypunching. Verification of the data was performed with less than one percent of errors. These errors were rectified. Finally, the keypunched data were placed onto a computer f i l e . The following is an excerpt from a three-minute sample of verbal interaction between a parent and a professional. /..../ = three seconds or a change in category. SAVI Code Column: Dr: Pt: Dr: Dr: Pt: Dr: Would you describe her convulsion 1 a l i t t l e Mrs. _ Well/ 2 Was i t a real?/' Pt: It was a grand mal state. The ambulance took her i n . / ^ They had to put her in oxygen and that. They said, to them i t looked like she was 5 having a stroke./ She was going to c the right a l l the time./ I really didn't realize that...that was a seizure. I just...didn't think she'd take anymore./'7 They didn't know what i t was and doctors in town didn't/^ know./^ Was she?/ Yeah./11 10 She had been standing at 12 the table with her Dad and I./ Unhum./13 1. Command - Com. 2. Fragment - F 3. Factual Question - FQ 4. Facts & Figures - FF 5 . Facts & Figures - FF 6. Facts & Figures - FF 7 . Opinion - Op 8 . Facts & Figures - FF 9. Facts & Figures - FF 10. Factual Question - FQ 11. Factual Answer - FA 12. Facts & Figures - FF 13. Encouragment - ENC 50. SAVI Code Columns: Pt Dr Pt Dr Pt And then she went into this real big/ 15 Was she frothing from the mouth?/ No./16 Was she incontinent?/ 14 17 Dr Pt Dr Pt 20 Yes. She wasn't frothing from the •j Q mouth but there was just/ nothing there./^ Was she like shaking or/ Yes./ 2 1 or was she s t i f f like that/' She was s t i f f mostly until we got her ,22 to the ambulance and half way to town and then she started to shake/ ,24 Quiet/ Dr: One side more than the other/ 25 Pt: She was just a l l pulled to the right and sort of/ 2^ ?7 No i s e r 7 They said i t must have been some-thing to do with the l e f t side of 28 her brain/ Noise/" Noise/ 3 0 Dr: And following the seizure, how long did i t last or do you remember or 31 32 did/ anyone write i t down?/ 14. Facts & Figures - FF 15. Factual Question - FQ 16. Factual Answer - FA 17. Factual Question - FQ 18. Factual Answer - FA 19. Factual Answer - FA 20. Factual Question - FQ 21. Factual Answer -• F/ 22. Factual Question - FQ 23. Factual Answer - FA 24. Quiet - Q 25. Factual Question - FQ 26. Factual Answer - FA 27. Noise -28. Facts & Figures - FF 29. Noise - N 30. Noise - N 31. Factual Question - FQ 32. Factual Question - FQ 51. SAVI Code Columns Pt: No. It lasted a l l the way to town and probably lasted about 15-20 33 minutes/ before, until the /34 ambulance did get to the house. Dr: All that time she was like that?/ Pt: She was just. She quit breathing is what she did. She wasn't shaking al l that time. She quit breathing/ and I kept shaking her to make her 37 go - ah, ah (mother gasps)/ just until the ambulance got there./ 39 38 Dr: Unhum./ Pt: and then they gave her oxygen./ 40 Dr: After she was in hospital, did they 41 notice any changes in terms/ of any weakness on one side or 42 the other or can you remember?/ Pt: No. Well i t was/ 4 3 44 Noise/** Dr: She was put on phenobarb and since Pt Dr Pt then she's been on that medication?/ Yes./ 4 6 Except i t ' s been increased/4'' She's never had another seizure .,48 since./ 45 Dr: ok/ 49 33. Factual Answer - FA 34. Factual Answer - FA 35. Factual Question - FQ 36. Factual Answer - FA 37. Factual Answer - FA 38. Factual Answer - FA 39. Encouragement -ENC 40. Factual Answer - FA 41. Factual Question - FQ 42. Factual Question -FQ 43. Factual Answer - FA 44. Noise - N 45. Factual Question - FQ 46. Factual Answer - FA 47. Facts & Figures - FF 48. Facts & Figures - FF 49. Encouragement - ENC 52. Noise/ Noise/ 50 51 Now, when did you notice the spots? 52 Did you notice them/ ah, anytime 53 coming on?/ Pt: About last year when she was in hospital but I just thought i t was, 54 maybe, another rash/ cause i f she got reactions on the rest of her 55 body that was a rash alright./ 56 Dr: Ok./ Noise/ 57 Pt: She's had that since she was real t i n y . / 5 8 SAVI Code Columns: 50. Noise - N 51. Noise - N 52. Factual Question - FQ 53. Factual Question - FQ 54. Factual Answer - FA 55. Factual Answer - FA 56. Encouragement - ENC 57. Noise - N 58. Facts & Figures - FF In :the above example the lack of personal messages being displayed is apparent. At no time does the professional ask the mother how she f e l t about the ordeal nor does she lend any personal support to the mother. This example does display a very high content of task oriented verbal behaviour. There is almost a constant barrage of questions on the part of the professional and a persistent attempt on the part of the parent to provide factual answers. During this brief encounter i t is obvious this is a fact-finding mission. This sample of verbal behaviour also displays the constant use of some categories to the exclusion of the majority of the 45 categories available. The professional is asking questions and the parent is answering and pre-senting facts and figures. This is typical of many of the tapes, although a greater variety of categories of verbal behaviour is shown on some. It is important to remember that the code for any statement was determined by simultaneously listening to the audiotape while scoring the transcript. The tone of the voice is essential in determining the essence of the message. Methods of Data Analysis Since a variety of sta t i s t i c a l tests were performed they are outlined here under the six levels of analyses which were conducted. The f i r s t two levels contained general descriptions and global findings from the data. The analyses in levels three, four and five were selected to test specific hypotheses. The sixth level consisted of an examination of the patterns of verbal interaction displayed by the three different groups. The sequential nature of the observations resulted in no one observation being completely independent from another. The amount of auto correlation, how-ever, should be relatively insignificant. In the third, fourth, f i f t h and sixth levels of analysis, therefore, the independence of the observa-tions was assumed even though this is not s t r i c t l y true. In the fourth level, where correlated proportions were tested, the correlation resulted from the fact that the sum of proportions was equal to"1. This correla-tion could be substantial in comparison with the auto correlation pre-viously mentioned. First Level The frequencies in each of the nine SAVI areas were examined for each group of participants. Second Level The 45 categories within the nine areas were examined for each group. 54. Third Level For hypotheses 1 to 4 a test of significance of the difference between two independent proportions was performed (Ferguson, 1976, p. 173). Fourth Level For hypotheses 4-14 the significance of the difference between two correlated proportions was calculated (McNemar, 1947). Fifth Level Hypotheses 15-20 were tested by use of the Pearson product moment correlation. Sixth Level This final level of analysis required several steps and was developed to examine verbal interaction patterns for both groups and individuals. The following steps were necessary to compute this analysis (an example of these steps is provided in Appendix E). 1. The three theoretical patterns of verbal communication of defensive-ness, cross-purpose and problem-solving, (Appendix E) are presented by Simon and Agazarian (1967). For the three patterns the authors proposed relative levels of low, medium, and high for each of the nine cells compris-ing the individual patterns. These relative levels were not attributed a range of raw frequency or percentage values thus making them impossible to test. Agazarian (1968) did present one case (p. 71.), however, where the observed frequencies were converted to percentages and reported. She explained how the observed findings from this one instance related to the cross-purpose theoretical pattern. What was missing, then, was a range of values for the theoretical levels of low, medium, and high so that compari-sons with other observed levels could be related to one of the three theore-tic a l patterns of communication. After examining the information provided by Agazarian (.1968). and the raw data presented by Browne [1977) and Cox (1976), some structure was added to make objective comparisons possible between the observed data and theoretical patterns. The structure con-sisted of assigning the following percentage ranges to the relative levels: Low < 7%; 7% < Medium < 20%; and High > 20%. 2. The frequency distributions of the observed data were converted to percentages so that the nine cells in each theoretical pattern totalled 100%. The percentages for each, cell were changed into relative levels of low, medium, and high. 3. The nine cells of observed data were compared to the nine theoretical cells for one of the three theoretical patterns being examined. 4 A rank ordering, from one to nine for the observed and theoretical patterns, was performed. When ties occurred, each was assigned the average of the ranks which would have been assigned had no ties occurred. Therefore, the correlation was corrected for ties (Hays, 1973, p. 791). Once the values for each set of observed and theoretical data were determined a Spearman rank correlation coefficient was calculated. Since the hypotheses predicted the direction of the relationships, one-tail tests were deemed appropriate. The .05 level of confidence was used as the basis for rejecting the null hypothesis. Summary In this chapter the results of the pilot testing were reported. The settings in which data were collected for the major study and subject-selection procedure were explained. The procedure for selecting a sample from the tape recordings was delineated and the six levels of data analysis were discussed. 56. CHAPTER IV RELIABILITY AND GENERAL CHARACTERISTICS OF THE DATA The present study did not provide an opportunity for further investi-gation bf construct validity. It was possible, however, to examine the r e l i a b i l i t y of the SAVI instrument which was an objective of this research. In this chapter, then, the r e l i a b i l i t y results are discussed in addition to the general characteristics of the data. Reliabi1ity After reviewing other researchers' r e l i a b i l i t y studies i t was decided to examine typical coding errors and observe inter-rater r e l i a b i l i t y when coders had not jointly used a practice tape to coordinate their coding s k i l l s . The results of these findings are discussed. Reliability Studies Reviewed Reliability studies have been documented by both the developers of the Sequential Analysis of Verbal Interaction (SAVI) system and other users of the instrument. According to Agazarian and Simon (1976), 80% r e l i a b i l i t y , as assessed by the percentage of category agreements between coders, was considered acceptable due to the complexity of the system. They stated that hearing the overall pattern of communication was more important than getting every single verbal behaviour correct. Simon and Agazarian (1967) reported an intra-rater r e l i a b i l i t y of 86% after a six-week interval and an inter-rater of 91%. Cox (1976) established an intra-rater r e l i a b i l i t y of 82% (range 77 to 93%) and an inter-rater r e l i a b i l i t y , using the investigator and another coder, of 80% 57. (range 77 to 83%). In 5731 instances of verbal behaviour Browne (1977) found an intra-rater r e l i a b i l i t y of 95% and an inter-rater r e l i a b i l i t y of 84%. Using three coders on five-minute samples, Harvey (1976) found r e l i a b i l i t y was 82% (range 76 to 87%). The Process In this report r e l i a b i l i t y studies were carried out using both intra and inter-coder r e l i a b i l i t y on 14 of the total 74 tapes. Because only seven professionals were interviewed, each,with different parents, i t was decided to have each professional represented. One tape from each pro-fessional was randomly selected, thus giving seven tapes of different pro-fessional-parent verbal interactions. The companion tape of the parent with a paraprofessional provided the other paraprofessional-parent verbal interactions for r e l i a b i l i t y testing. Intra-rater r e l i a b i l i t y studies were accomplished by having the pro-fessionally trained coder of the SAVI system code the 14 tapes. These same tapes were recoded six weeks later, as suggested by Agazarian (1968), to ensure that memory was not a confounding factor. Inter-rater r e l i a b i l i t y was established under two different conditions. Under the f i r s t condition, the coders were together when coding was per-formed and under the second condition these coders' results were compared to the results obtained by the professionally trained coder who lived in another city. Thus, r e l i a b i l i t y was examined when people were in geographic proximity and when they were geographically separate. The four coders used under geographic proximity conditions did not discuss or talk during coding but presumably could discuss problems between tapes and thereby improve their s k i l l . In addition, these coders practised together on a practice tape prior to the coding of the data in order to 58. sensitize themselves both to coding problems which might arise and to the SAVI system. The f i f t h coder, who lived in another city, did not have the benefit of practising or talking with the other four coders. The four coders consisted of the two developers of the SAVI instru-ment and two trainees. These four coders were divided so that only two coders coded each tape. The two coders, then, worked in varying combina-tions. These combinations involved: the two developers coded five tapes, the two trainees coded one tape, and a developer and a trainee coded the remaining nine tapes. The inter-rater agreement scores ( r e l i a b i l i t y ) were then compared to the scores of the professionally trained coder who had no contact with the other coders, nor any knowledge that her coding results would be compared to the other four coders. Each category label for every verbal act was compared and the percen-tage of errors in the total number of codes made was calculated. The percentage of errors was then subtracted from 100% to obtain the rater r e l i a b i l i t y . This was the method utilized by Agazarian (1979) because i t represented the most conservative and rigorous form of calculating r e l i a -b i l i t y ( i .e., by comparing every item rather than the total number of codes). In addition, r e l i a b i l i t y was tested at the category level instead of the area level of the SAVI system. For each item, this represented a choice among 45 possible categories as compared to only nine areas; thus coding and testing were rigorous. Results of Reliability Testing The overall intra-rater r e l i a b i l i t y , as shown in Table 2, averaged 90% for the i n i t i a l coding by the professionally trained coder. This r e l i a b i l i t y figure was based on 1875 coding decisions made on the 14 tapes. TABLE 2 INTRA-RATER RELIABILITY Tape Total Codes Errors % Error Reliability 10 124 5 4 .03 95.96 11 137 6 4.38 95.60 36 153 14 9.15 90.80 37 126 10 7.94 92.06 42 176 20 11.36 88.63 43 175 28 16.00 84.00 48 143 4 2.79 97.20 49 133 15 11.28 88.72 64 122 14 11 .48 88.52 65 128 14 10.94 89.06 68 102 15 14.71 85.29 69 100 5 5.00 95.00 72 146 19 13.01 86.99 73 n o 12 10.91 89.09 Tapes Total Error % Error Reliability 14 1875 181 9.50 90.50 60. The r e l i a b i l i t y scores ranged from 84-97% agreement. Inter-rater r e l i a b i l i t y scores, as shown in Table 3, using four coders averaged an overall r e l i a b i l i t y agreement of 87% with a range of r e l i a b i l i t y scores from 72-94%. Inter-rater agreement scores with coders geographically separate, as displayed in Table 4, resulted in lower r e l i a b i l i t y scores as expected. The inter-rater agreement averaged 79% with a range of r e l i a b i l i t y scores from 70-86%. By selecting the seven highest and seven lowest r e l i a b i l i t y scores and examining the coders involved in those scores, individual differences can be noted in Tables 3 and 4. For example, when examining coder number five's performance, the following differences were found. These seemed to depend on coding partners and the individual tapes involved. When coder number five coded with just one other person, as shown in Table 3, the performance ranked within the lowest seven r e l i a b i l i t y scores for all five tapes coded. This performance did change, however, when another , coder was included (see Table 4). Then, two scores in the highest and three scores in the lowest end of the r e l i a b i l i t y scores were found. While individual differences in a b i l i t i e s to code data using SAVI do exist, consideration should be given to the d i f f i c u l t y level of interpret-ing the tapes. Some tapes posed greater coding problems than others. Di f f i c u l t tapes tended to include those where the speakers changed cate-gories frequently or where there were disruptive background noises, rapid speech, ethnic accents and colloquial language. To identify tapes which seemed more d i f f i c u l t to code, the six lowest r e l i a b i l i t y scores were ex-tracted from each of Tables 2, 3 and 4. When the same tape number was found to belong to the lowest six scores in two out of the three tables i t was considered a problem tape for the coders. Tape number 68 was TABLE 3 INTER-RATER RELIABILITY CODERS IN GEOGRAPHIC PROXIMITY Tape Coder* Total Codes Error % Error Reliabil ity 10 3,4 124 10 8.06 91.94 + 11 3,4 137 16 11.68 88.32 + 36 4,5 153 21 13.73 86.27 -37 2,3 126 8 6.35 93.65 + 42 3,4 176 13 7.39 92.61 + 43 3,5 175 25 14.28 85.71 -48 • 2,5 143 21 14.69 85.31 -49 3,5 133 21 15.78 84.21 -64 3,4 122 10 8.20 91.80 + 65 2,3 128 11 8.59 91.41 + 68 2,5 102 29 28.43 71.51 -69 2,3 100 13 13.00 87.00 -72 2,3 146 34 23.29 76.71 -73 2,3 n o 7 6.36 93.63 + Tapes 14 Total 1875 Error 239 % Error 12.85 Reliability 87.14 + = Seven highest scores - = Seven lowest scores * Coders: 1 = Professionaly trained coder 2 = Developer #1 of SAVI 3 = Developer #2 of SAVI 4 = Trainee #1 of SAVI 5 = Trainee #2 of SAVI 62. TABLE 4 INTER-RATER RELIABILITY CODERS GEOGRAPHICALLY SEPARATE Tape Coders* Total Codes Error % Error Reliabil ity 10 1,3,4 124 18 14.52 85.48 + 11 1,3,4 137 26 18.98 81.02 + 36 1,4,5 153 38 24.84 75.16 -37 1,2,3, 126 21 16.67 83.33 + 42 1,3,4 176 35 19.89 80.11 + 43 1,3,5 175 38 21.71 78.28 -48 1,2,5 143 20 13.99 86.01 + 49 1,3,5 133 24 18.05 81.95 + 64 1,3,4 122 17 13.93 86.06 + 65 1,2,3 128 33 25.78 74.22 -68 1,2,5 102 31 30.39 69.61 -69 1,2,3 100 24 24.00 76.00 -72 1,2,3 146 32 21.91 78.08 -73 1,2,3 110 30 27.27 72.73 -Tape 14 Total 1875 Error 387 % Error 20.85 Reliability 79.14 + = Seven highest scores - = Seven lowest scores * ;= Coders: 1 = Professionally trained coder 2 = Developer #T of SAVI 3 = Developer #2 of SAVI 4 = Trainee #1 of SAVI 5 = Trainee #2 of SAVI 63. identified as a d i f f i c u l t tape since i t appeared within the lowest six r e l i a b i l i t y scores in a l l three tables whereas tape numbers 43 and 72 were found in two of the three tables. The interesting finding is that coder number five was involved in coding two of these problem tapes (68 and 43) out of the total of five tapes this individual coded. It must be stated, therefore, that caution must be exercised in judging an individual coder's performance. Analysis of Coding Errors The term error refers to a discrepancy or disagreement between coders on a coding decision. Coding errors were aggregated and analyzed by each tape under the following r e l i a b i l i t y situations: 1) the professionally trained coder coded a six-minute sample from each of the 14 tapes and recoded them six weeks later; and 2) the four-person coder team practised on a practice tape, following which the 14 tapes were each coded by one pair of coders. Comparisons were also made between situations one and two. A coding decision was made every three seconds, or on category or person changes,thus resulting in a mean number of approximately 120 codes per six minutes of talk. Despite the fact that six-minute samples had been selected from each tape, there was a discrepancy in the total number of codes for each tape. This discrepancy resulted from the speed of speech of the interactors. The total number of coding errors per tape (see Table 5) was consis-tently higher under the more rigorous coding situation (coder #3) with the exception of tape 72 where minimal differences were found. In coder situation number one (professionally trained coder) there were only five instances (tapes 37, 42, 43, 64 and 65) out of a total of 42 occasions 64. TABLE 5 SUMMARY OF CODING ERRORS TAPE TOTAL CODES CODER Z ERRORS TYPE OF ERROR COMMON ERROR SAMPLE ERROR COMMENT TAPE TOTAL COOES CODER r ERRORS TYPE OF ERROR COMMON ERROR SAMPLE ERROR COMMENT TAPE TOTAL CODES CODER £ ERRORS TYPE OF ERROR COMMON ERROR SAMPLE ERROR COMMENT to 124 1 5 i A 0 c CI = FF(4 ) FF=5 48 143 1 • 4 0 A 4 C Ad/0m^2 Para=2 Sum=2 69 100 2 1 3 6 A 7 C Add/0m=5 0P = 5 FA=3 sea t t e r e d 69 100 . 3 24 20 A FA=0P(4) FA=10 0P = 10 FF=7 s c a t t e r e d 10 124 2 10 9 A FA= T0L(4 ) T0L=4 1 C FA=6 48 143 2 . 21 12 A 9 C FA=C la r (5 ) FA=14 C l a r - 6 sea t t e r e d 4 C 10 124 3 18 15 A 3 C FF=CLAR(6) FF=12 C la r=9 FF = 4 FQ=5 48 143 3 20 19 A PA=0P(3) PA= 11 s c a t t e r e d 72 146 1 19 15 A - FF = 3 s c a t t e r e d 11 137 1 6 2 A 4 C Add/0m=2 FF=2 FA=2 s c a t t e r e d 1 C P.Pos=5 0P=8 4 C 0P=3 72 146 2 34 16 A Spec=OP(10) 0P = 23 Spec=10 Add/0m=7 s c a t t e r e d 11 137 2 16 9 A 7 C FA=FF(4) FA=4 FF=5 P.Pos=6 49 133 1 15 10 A 5 C Prop=FF(4) Prop=4 FF=6 s c a t t e r e d 18 C 49 133 2 21 12 A Prop=0P(4) 0P=IO 72 146 3 32 27 A 0P=AJ(4) 0P=11 s c a t t e r e d 11 137 3 26 20 A 6 C FF=FA(4) P . P o s = U FF=8 Op=5 PA=6 9 C Prop=Sum(4) Prop=9 Sum=4 FF = 5 5 C AJ=5 FF=7 FA=3 49 133 3 24 17 A 0P=Prop(3) OP=10 s c a t t e r e d 73 110 1 12 4 A Add/0m=5 FF=4 36 153 1 14 6 A 8 C CI=FA(4) CI=4 PA=8 Add/0m=4 7 C F A = P . H i s t ( 3 ) Prop=10 . FF=8 8 C 0P=4 Spec=4 36 153 2 21 16 A 5 C FF=Para(3) FF= i 1 Para=4 FA=4 w ide s c a t t e r e d 64 122 1 14 8 A 6 C Add/0m=4 P.P0S=4 0P=2 s c a t t e r e d 73 110 2 7 4 A 3 C FF=0P(4) FF = 4 0P=5 64 122 2 10 7 A FF=FQ(4) FF=5 73 n o 3 30 16 A FF=0P(15) 0P=29 36 153 3 38 30 A 8 C F F = C l a r ( 7 ) F F - F A ( 7 ) FF=20 FA=12 C l a r - 7 3 C 0P=Prop(4) 0P=5 FQ=5 Prop=4 14 C FF=15 867 To ta l E r r o r s 37 126 1 10 6 A 4 C Add/0m=3 FF=3 FA=3 s c a t t e r e d 64 122 3 17 12 A 5 C 0P=FF(4) 0P=11 FF=7 s c a t t e r e d 3) 126 2 8 8 A 0 C FA=7 FF=2 s c a t t e r e d 65 128 1 14 10 A 4 C - FF = 7 CI = 5 s c a t t e r e d 0P=2 65 128 2 " ' 11 3 A FF=FQ(4) FF = 5 * Note 37 126 3 21 18 A 3 C FF=0p(8) FF= 14 0p=9 FA=6 s c a t t e r e d 8 C 0P=Prop(4) 0P=5 FQ=5 Prop=4 42 176 1 20 10 A 10 C 0P=FF(4) 0P=7 FF=7 Add/0m=4 s c a t t e r e d 65 128 3 33 20 A 13 C FF=FA(12) FA=0P(4) FF= 16 FA=20 OP-10 42 176 2 13 11 A 0P=FF(4) 0P=9 2 C FF = 6 68 102 1 15 9 A - Add/0m=6 FF = 5 0P=3 C 1 = 3 42 176 3 35 35 A 0 C C ! = FF(7) FF = 2I FA=12 CI = 7 s c a t t e r e d 6 C 68 102 2 29 11 A 18 C T0L=Frag(8) F r a g = U T0L=9 0P=8 .FF=7 s c a t t e r e d 43 175 1 28 25 A 3 C FF=Spec(4) FF = 13 Spec=6 s c a t t e r e d 43 175 2 25 23 A FF=0P(9) FF=12 s c a t t e r e d 2 C 0P=17 68 102 3 31 19 A 12 C FF=0P(6) FF=18 0P=15 s c a t t e r e d 43 175 3 38 31 A 7 C B/C=FF(10) FF= 19 s c a t t e r e d 0P=18 8 / O 1 0 69 100 1 5 1 A 4 C FF = 3 0P=3 65. Note. Tape - The 14 tapes randomly selected for r e l i a b i l i t y studies. Even numbers are professionals with parents. Odd numbers are para-professionals with parents. Total Codes = The total number of coding decisions per tape. Coder = The three different coding situations 1 = the professionally trained coder coding the 15 tapes after a six-week lapse of time. 2 = the 4 coders in Philadelphia (2 developers of the SAVI system and 2 trainees of the SAVI system). Two coders coded each.of the 14 tapes. This was the group in geographic proximity. . 3 = coders 1+2. Sum of Errors = The total number of errors made per tape Type of Error Common Error Sample Error Whether the error was an Area (Class) or a Category type: A = Area or Class; C = Category. The most common error discovered in that tape. The kinds of category confusion which existed. A sampling of the top 1, 2 or 3 errors and their frequency of occurrence. Area Errors FF = Facts and Figures 5 301 TOL = Thinking out loud 3 13 FA = Factual Answer 8 106 Clar Clarification of information 8 41 P. Pos. = Personal Position 4 37 OP = Opinion 6 276 P.A. = Personal Answer 7 25 ADD/Om = Additions & Omissions - 13 Para = Paraphrasing 8 6 Spec = Speculation 6 20 B/C = Blaming & Complaining 1 23 Sum = Summarizing 8 6 Prop Proposal 6 31 FQ = Factual Question 5 15 Frag = Fragment 6 11 AJ = Affectionate Joke 7 5 P. Shar = Personal Sharing 7 5 66. (14 tapes x 3 situations) where more coding errors occurred than under coding situation two (coders with practice sessions together). Out of the total of 42 instances, there were 33 instances where class errors were greater than category errors, eight instances where they were less and one instance when both class and category errors remained the same. Class errors were more common and are more serious than category errors because they represent errors which cross the boundary into another SAVI area, and thus alter the meaning attached to the communication. Category errors indicate errors which occur within the confines of one of the nine SAVI areas, thus do not alter the meaning since a category is a subset of a class. For example, should a coder label a three-second segment of talk as "clarifying information" instead of the correct code "facts and figures", the error is of a class type. This is a class type error since "c l a r i f y -ing information" belongs to the factual-approach area whereas "facts and figures" represent the factual-contingent area. On the other hand, i f the coder confused a "proposal" for "speculation" these are both cate-gories within the factual-orienting area. According to the developers of the instrument, Agazarian and Simon (1976), the fact that an error was made was not as c r i t i c a l as the fact that the coders would code i t con-sistently on the next occasion. The professionally trained coder (#1) had a higher frequency of class errors than category errors. The discrepancy between the number of class and category errors was not of a great magnitude, with the exception of tapes 43 and 72 which have already been identified as problem tapes. This is mentioned since the professionally trained coder coded a l l the tapes for this study and therefore her performance is of greater interest. The common errors were those where categories were confused frequently with other categories. This is indicated by the number in the brackets in 67 the summary of coding errors (Table 5). In examining the column of common errors, 18 common errors were found out of a total of 40 categories used. These 18 different category errors represented eight area..errors out of a total of nine SAVI areas. The class of contingent-orienting behaviours, where the greatest number of errors (5) occurred, also contains the greatest density of categories. The class with no errors was factual-avoidance which only contains three categories. The sample errors (see Table 5) were identified by selecting two to four of the categories which had the highest frequency of errors on each tape. When there was a wide discrepancy in the frequency of errors on any one tape after the two, three or four categories were selected, the sample error identification was discounted. By examining the sample errors, i t became readily apparent that cate-gories of facts and figures, opinions, and factual answers were the cate-gories resulting in the greatest overall number of errors. Facts and figures constituted 35% of coding errors and opinions resulted in 32%, while factual answers represented 12% of the errors (see note to Table 5, p. 65). These three categories represented 79% of the total errors made, apparently presenting a major problem to the coders. Since they repre-sented such a large portion of the errors, the implication arises that these categories were not exclusive but overlapped with other categories. This lack of specificity in the operational definition of these three categories should be refined. The remaining 21% of errors were spread across 15 categories, each constituting from .6% to %5 of the errors, thus probably representing random type errors. Facts and figures, opinions and factual answers were categories which were responsible for a large percentage of the errors but they were also 68. frequently utilized - 17%, 13% and 13% respectively. These three categories represented the classes of factual-contingent, factual-approach, and orient-ing-contingentwhere 94% of the verbal activity occurred. In summary, both the lack of specificity of the operational definitions and the high u t i l i t y of these three categories were factors in the errors found. Of these two factors the definitional problem is believed to be the greater since other categories were utilized as frequently ( i . e . , encouragement and factual questions) but did not create a similar magnitude of error. It appears, therefore, that refinement of the operational definitions of these three categories, in addition to the other 15, would substantially improve the u t i l i t y of the SAVI instrument. In 25 out of 42 instances the errors showed a wide range of scatter throughout the categories, similar to .that which would be expected in ran-dom type errors (see Table 5, p. 64). In other instances the errors seemed to be focused on specific categories. Characteristics of the Data It will be recalled that the data for the present study were obtained from tape recordings of 37 parent-professional and 37 parent-paraprofessional interviews. The parent-professional dyads showed 2117 coded units of verbal behaviours for parents as compared to 2461 for the professionals. Examining the parent-paraprofessional dyads, there were 2439 units of coded verbal behaviour from parents and 2328 from the paraprofessionals (see Table 6). Parents had a greater number of utterances coded as avoidance and approach behaviours than did professionals and paraprofessionals. They TABLE 6 TOTAL CODED THREE-SECOND UNITS OF A SIX-MINUTE SAMPLE OF VERBAL INTERACTION BETWEEN 37 PARENT-PROFESSIONAL AND 37 PARENT-PARAPROFESSIONAL DYADS 37 Parents 37 Professionals P F 0 Total P F 0 Total Av 62 02 10 74 16 01 12 29 C 53 487 508 1048 96 1117 1015 2228 Ap 23 972 00 995 06 198 00 204 138 1461 518 2117 118 1316 1027 2461 37 Parents 37 Paraprofessionals P F 0 Total P F 0 Total Av 72 05 04 81 00 00 06 06 C 85 566 634 1285 103 958 1089 2150 Ap 53 1019 01 1073 09 162 01 172 210 1590 639 2439 112 1120 1096 2328 Av = Avoidance Behaviour C = Contingent Behaviour Ap = Approach Behaviour P = Personal Message F = Factual Message 0 = Orienting Message 7Q. also used personal and factual types of messages slightly more often. Professionals and paraprofessionals, on the other hand, had a greater number of instances coded as contingent behaviours. In addition, they utilized orienting kinds of messages more frequently. Thus, i t may be stated parents showed a preference for avoidance and approach behaviours and personal and factual messages whereas professionalsand paraprofessionals tended towards contingent behaviours and orienting messages. Parents incorporated a variety of messages and behaviours into their talk while professionals and paraprofessionals demonstrated a more stereotyped selection of messages and behaviour. When the frequencies for each of the SAVI categories were examined for each group, some difference in their usage could be identified. Table 7 shows the five most frequently used categories by a l l groups com-bined, in order of greatest frequency: facts and figures (#16), encourage-ment (#25), factual answers (#39), opinions (#20), and factual questions (#18). These categories covered three different areas: factual-contingent, factual-approach and orienting-contingent. Five categories were not used at a l l by any of the participants. They were: reminiscing (#6), oughti-tudes (#10), credentials (#12), building on ideas (#44) and taking-a-stand (#37). The areas involved in these unused categories included: personal-avoidance, factual-avoidance, orienting-avoidance, personal-con-tingent, orienting-approach, and-personal-approach. By referring to Table 7,differences can be found between parents, professionals, and paraprofessionals and their use of various categories. The parents when interacting with both the professionals and the parapro-fessionals utilized the category of factual answers most frequently. The professionals and paraprofessionals had a similarly high preference for factual questions (18). Both these categories had considerably larger 71. TABLE 7 SAVI CATEGORY FREQUENCY FOR EACH GROUP Category # Parent Profess-ional Parent Parapro-fessional Total Attack 1 2 13 2 0 17 Self-Defensiveness 2 34 1 22 0 57 Blame/Complain 3 26 1 48 0 75 Sarcastic Joke 4 0 1 0 0 1 Gossip 5 1 0 2 0 3 Reminiscing 6 0 0 0 0 0 -Funny Story 7 1 1 3 0 5 Think-out-loud 8 10 8 4 5 27 Obscuring 9 0 4 0 1 5 Oughtitudes 10 0 0 0 0 0 -Personal History 11 0 1 1 4 6 Credential s 12 0 0 0 0 0-Personal Position 13 48 81 71 70 270 Per. Question 14 0 5 0 16 21 Ritual 15 5 9 13 13 40 Facts &. Figures 16 409 350 516 289 1564 + General Inform. 17 1 40 0 28 69 Fact. Question 18 76 702* 50 607* 1435 + Gen. Question 19 1 25 0 34 60 Opinion 20 200 385 315 358 1258 + Interpretation 21 9 0 8 13 30 Speculation 22 15 34 26 39 114 Assumption 23 0 4 3 3 10 Projection 24 0 0 1 0 1 Encouragement 25 260 393 260 502 1415 + Discouragement 26 7 15 7 1 30 Yes/But 27 2 4 2 3 11 Command 28 8 40 3 48 99 Proposal 29 7 140 9 122 278 Quiet 30 267 0 230 0 497 Noise 31 431 0 485 0 916 Laughter 32 58 0 95 0 153 Fragments 33 41 0 58 0 99 Personal Answer 34 22 0 37 0 59 Personal Sharing 35 0 0 14 0 14 Personal Support 36 0 2 0 5 7 Take-a-stand 37 0 0 0 0 0 -Affectionate Joke 38 1 4 2 4 11 Factual Answer 39 818* 84 749* 81 1732 + General Answer 40 54 24 139 13 230 Clarification 41 95 6 122 6 229 Paraphrasing 42 5 72 7 58 142 Summarizing 43 0 12 2 4 18 Build-on-ideas 44 0 0 0 0 0 -Work Joke 45 0 0 1 1 2 2914 2461 3307 2328 11010 + = most used category; - = least used category; * = most used category within group. 72. frequencies than any of the other categories, thus making the interview a very task oriented activity of constant factual questioning and factual answering. The professionals and paraprofessionals utilized the neutral area of factual-contingent primarily, while the parents used the factual-approach area. The categories used least frequently within each group will not be discussed since there are too many unused categories for each group but these are displayed in Table 7. It is interesting to note several cate-gories where one group used a category whereas another group avoided i t . Parents used categories 2 and 3 of self-defensiveness, and blaming and complaining but the other two groups did not. The professionals and para-professionals used factual and general questions (18 and 19) whereas parents rarely seemed to engage in this behaviour. The professionals and paraprofessionals used proposals (29) to a much greater extent than the parents. Categories 30-33 (noise, silence, laughter and fragments) were elim-inated from most of the analysis because of the d i f f i c u l t y in determining who created the silence; frequently the noise or laughter was made by a third person. Fragments were isolated utterances which were too brief to code. Frequently both interactors would begin to talk at once making neither i n t e l l i g i b l e to the listener. Since these primarily meaningless bits of data posed problems greater than their u t i l i t y , the decision was made to t a l l y them and place them under the parents' data to allow their inclusion in the overall data base but removal for hypothesis test-ing. Noise, (31) which had a high frequency of occurrence, was created by children, outside construction and t r a f f i c , phone calls and interrup-73. tions by others. Personal answers and personal sharing (categories 34 and 35) came from parents exclusively. Neither professional nor paraprofessional en-gaged in this verbal behaviour. Parents provided answers and c l a r i f i e d information (39, 40 and 41) while the professionals and paraprofessionals engaged more frequently in paraphrasing information (42). None of the three groups used the most valued categories of building-on-ideas and participating in work-jokes (44 and 45). In conclusion, areas incorporating categories 16-19, 20-33 and 39-43 were used most frequently. These areas included: factual-contingent, orienting-contingent and factual-approach. Parents utilized the factual-approach while professionals and paraprofessionals used the former two areas. These three areas of most usage represented 93.5% of the verbal behaviour when a l l three groups were combined. Summary In summary, r e l i a b i l i t y was examined over three different situations involving five different coders. The type of category error made most frequently occurred outside the area to which that category belonged. The implications of this were discussed. The three most used categories were identified and elaborated. Having concluded that the coding r e l i a b i l i t y reached an acceptable level, the characteristics of the data were dis-cussed. 74. CHAPTER V RESULTS OF HYPOTHESES TESTING The results of hypotheses' testing of the present study are presented in this chapter. First the results of testing hypotheses 1 - 4 are given. They involve independent proportions and the groups' usage of verbal behaviours and messages. These are followed by the results obtained for hypotheses 5 - 1 5 concerning verbal behaviours and messages with corre-lated proportions. Following this,the results of tests of hypotheses 15 -20 are presented, u t i l i z i n g Pearson product moment correlations of age and experience with the percent of approach behaviours for the groups. Finally, the verbal interaction patterns (defensiveness, cross-purpose and problem-solving) are examined by comparing the observed with the theo-retical patterns. The levels of confidence accepted for hypothesis testing are indicated throughout this report by the following scheme: * £ <--..05 ** £ < .01 n.s. not significant WHen stating the hypotheses, the alternate form only of each hypothesis is presented. Throughout the test the following abbreviations will be used: Ha refers to the alternate form of the hypothesis. Hypotheses 1 - 14 will be tested using a one t a i l test where: * Rvalue of 1.65 (probability < .05) ** _z value of 2.33 (probability < .01) 75. Hypothesis 1 Ha: Parents use a greater proportion of avoidance behaviours than professionals. TABLE 8 NUMBER OF AVOIDANT BEHAVIOURS USED BY PARENTS AND PROFESSIONALS Avoidant Behaviours Total Behaviours z score Parents 74 2117 4.6 ** Professional s 29 2461 ** p < .01 As predicted, the parents use avoidant verbal behaviours significantly more often (Table 8) than the professionals, z = 4.6, _p_ < .01. Hypothesis 2 Ha: Parents use a greater proportion of avoidant behaviours than paraprofessionals. TABLE 9 NUMBER OF AVOIDANT BEHAVIOURS USED BY PARENTS AND PARAPROFESSIONALS Avoidant Total Behaviours Behaviours z score Parents 81 2439 10 0 ** Paraprofessionals 6 2328 ** p < .01 76. As predicted, the parents use avoidant verbal behaviours to a signi-ficantly greater degree (Table 9) than the paraprofessionals, z = 10.0, £ < .01. Hypothesis 3 Ha: Professionals use a greater proportion of approach behaviours than parents. TABLE 10 NUMBER OF APPROACH BEHAVIOURS USED BY PARENTS AND PROFESSIONALS Approach Total Behaviours Behaviours z score Professionals 204 2461 =•_ ^g'yy Parents 995 2117 Results are significantly at variance with the hypothesis, z = -29.77 (Table 10). Accordingly, the hypothesis that professionals use a greater proportion of approach behaviours than parents must be rejected. Hypothesis 4 Ha: Paraprofessionals use a greater proportion of approach behaviours than parents. 77. TABLE 11 NUMBER OF APPROACH BEHAVIOURS USED BY PARENTS AND PARAPROFESSIONALS Approach Total Behaviours Behaviours z score Paraprofessionals 172 2328 Parents 1073 2439 •30.5 Results are significantly at variance with the hypothesis, z_ = -30.5 (Table 11). Accordingly, the hypothesis that paraprofessionals use a greater proportion of approach behaviours than parents must be rejected. Hypotheses 5 - 1 4 are those involving proportions which are not in-dependent. Hypothesis 5 Ha: Parents use a greater proportion of contingent than approach behaviours, when only factual messages are examined. TABLE 12 NUMBER OF CONTINGENT AND APPROACH BEHAVIOURS USED BY PARENTS WITHIN FACTUAL MESSAGES Parents Factual Messages z score Contingent 1053 ^ ^ Approach 1991 78. Results are significantly at variance with the hypothesis, z_ = -17.36, (Table 12). Accordingly, the hypothesis that parents use a greater propor-tion of contingent compared to approach' behaviours, when only factual messages are considered, must be rejected. Hypothesis 6 Ha: Professionals use a greater proportion of contingent than approach behaviours,when only factual messages are examined. TABLE 13 NUMBER OF CONTINGENT AND APPROACH BEHAVIOURS USED BY PROFESSIONALS WITHIN FACTUAL MESSAGES Professionals' Behaviours Factual Messages z s c o r e Contingent 1117 35.56 ** Approach 198 ** p < .01 As predicted, professionals use contingent verbal behaviours s i g n i f i -cantly more often (Table 13) than approach behaviours, when only factual messages are considered, z_ = 35.56, £ < .01. Hypothesis 7 Ha: Paraprofessionals use a greater proportion of contingent than approach'behaviours,, when only factual messages are examined. 79. TABLE 14 NUMBER OF CONTINGENT AND APPROACH BEHAVIOURS USED BY PARAPROFESSIONALS WITHIN FACTUAL MESSAGES Paraprofes Behaviours Contingent Approach ** p < .01 As predicted, paraprofessionals use contingent verbal behaviours to a significantly greater extent (Table 14) than approach verbal behaviours, when only factual messages are considered, z_ = 34.72, £ < .01. Hypothesis 8 Ha: Parents use a greater proportion of factual than orienting messages. sionals Factual Messages 958 162 z score 34.72 ** TABLE 15 NUMBER OF FACTUAL AND ORIENTING MESSAGES USED BY PARENTS Parents 1 Number of Messages Messages z score Factual 3051 33.62 ** Orienting 1157 ** p < .01 As predicted, parents :use factual verbal messages significantly more often (Table 15) than orienting messages, z_ = 33.62, £ < .01. 80. Hypothesis 9 Ha: Professionals use a greater proportion of factual than orienting messages. TABLE 16 NUMBER OF FACTUAL AND ORIENTING MESSAGES USED BY PROFESSIONALS Professionals 1 Number of Messages Messages z score Factual 1316 5.85 ** Orienting 1027 ** p < .01 As predicted, professionals use factual verbal messages significantly more often (Table 16) than orienting messages, z_ = 5.85, £ < .01. Hypothesis 10 Ha: Paraprofessionals use a greater proportion of factual than orienting messages. TABLE 17 NUMBER OF FACTUAL AND ORIENTING MESSAGES USED BY PARAPROFESSIONALS Paraprofessionals' Number of Messages Messages Factual 1120 Orienting 1096 z score 0.94 n.s. n.s. 81 There was no significant difference (Table 17) between the proporti of factual messages as opposed to orienting messages used by parapro-fessionals, z = 0.94, not significant. Accordingly, hypothesis 10 must be rejected. Hypothesis 11 Ha: Parent-professional dyads use a greater proportion of factual than personal messages. TABLE 18 NUMBER OF FACTUAL AND PERSONAL MESSAGES USED BY PARENT-PROFESSIONAL DYADS Parent-Professional Number of Messages Messages z score Factual 2777 85.38 ** Personal 256 ** p < .01 As predicted, the parent-professional dyads use factual verbal messages significantly more often (Table 18) than they do personal messages, z_ = 85.38, £ < .01. Hypothesis. 12 . Ha: Parent-paraprofessional dyads use a greater proportion of factual than personal messages. 82. TABLE 19 NUMBER OF FACTUAL AND PERSONAL MESSAGES USED BY PARENT-PARAPROFESSIONAL DYADS Parent-Paraprofessional Number of Messages Messages Factual 2710 Personal 322 ** p < .01 As predicted, the parent-paraprofessional dyads use factual verbal messages significantly more often (Table 19) than they do personal messages, _z = 68.73, £ < .01. Hypothesis 13 Ha: Parent-professional dyads use a greater proportion of contingent than approach behaviours, when only personal messages are considered. z score 68.73 ** TABLE 20 NUMBER OF CONTINGENT AND APPROACH BEHAVIOURS USED BY PARENT-PROFESSIONAL DYADS WITHIN PERSONAL MESSAGES Parent-Professional Number of Behaviours Personal Messages z score Contingent 149 12.37 ** Approach 29 ** p < .01 83. As predicted, the parent-professional dyads use contingent verbal behaviours significantly more often (Table 20) than approach behaviours, when only personal messages are considered, z_ = 12.37, £ < .01. Hypothesis 14 Ha: Parent-paraprofessional dyads use a greater proportion of contingent than approach verbal behaviours, when only personal messages are considered. TABLE 21 NUMBER OF CONTINGENT AND APPROACH BEHAVIOURS USED BY PARENT-PARAPROFESSIONAL DYADS WITHIN PERSONAL MESSAGES Parent-Paraprofessional Number of Behaviours Personal Messages z score Contingent 188 g 13 * * Approach 62 ** p < .01 As predicted, the parent-paraprofessional dyads use contingent verbal behaviours to a significantly greater extent (Table 21) than approach behaviours, when only personal messages are considered, z_ = 9.13, £ < .01. In hypotheses 15-20 age and experience were correlated with the percentage of approach behaviours used. The Pearson product moment corre-lation was the sta t i s t i c used. 84. Hypothesis 15 Ha: There is a positive correlation for parents between years of experience as a parent and percentage of approach behaviours used. There is no significant correlation between parents' years of exper-ience as a parent and percentage of approach behaviours used, r_ ••=- -.17, £ = .14, n.s. Accordingly, hypothesis 15 must be rejected. Hypothesis 16 Ha: There is a positive correlation for professionals between their years of experience as a professional and the percentage of approach behaviours used. There is no significant correlation between professionals' years of experience as a professional and percentage of approach behaviours used, r_ = -.17, £ = .33, n.s. Accordingly, hypothesis 16 must be rejected. Hypothesis 17 Ha: There is a positive correlation for paraprofessionals between the years of experience as paraprofessionals and percentage of approach behaviours used. There is no significant correlation between paraprofessionals' years of experience as paraprofessionals and percentage of approach behaviours used, r. = -.08, £ = .62, n.s. Accordingly, hypothesis 17 must be rejected. Hypothesis 18 Ha: There is a positive correlation for parents between age and percentage of approach behaviours used. 85. There is no significant correlation between parents' age and percen-tage of approach behaviours used, _r = -.12, £ = .30, n.s. Accordingly, hypothesis 18 must be rejected. Hypothesis 19 Ha: There is a positive correlation for professionals between age and the percentage of approach behaviours used. There is no significant correlation between professionals' age and percentage of approach behaviours used, r_ = -.19, £ = .25, n.s. Accord-ingly, hypothesis 19 must be rejected. Hypothesis 20 Ha: There is a positive correlation for paraprofessionals between age and percentage of approach behaviours used. There is no significant correlation between paraprofessionals' age and percentage of approach behaviours used, r_ = -.10, £ = .57, n.s. Accordingly, hypothesis 20 must be rejected. Comparisons of Theoretical and Observed Verbal Profiles of Groups The researcher questioned whether levels of education, sex of inter-actors, types of appointments, frequency of interaction, length of acquaintance, and interactors' roles altered patterns of communication. In addition, i t was questioned whether the cross-purpose pattern of communication would change as participants interacted more frequently or knew one another over a longer period of time. The three theoretical patterns of communication, i t will be recalled, are problem-solving, cross-purpose and defensiveness. The following section addresses these concerns. The analysis used here involved a number of steps which were outlined in chapter three with an example provided in Appendix E. The Spearman rank correlation coefficient (r ) was the s t a t i s t i c used. Levels of Education It was expected that: As the level of education of parents and paraprofessionals increases so does the correlation between the theoretical and observed problem-solving patterns of communication. TABLE 22 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PROBLEM-SOLVING PATTERNS OF COMMUNICATION AND LEVELS OF EDUCATION (rs) Parents (N = 37) - Problem-solving 1. Elementary (N = 11) .79 2. High School (N = 14) .79 3. High School plus (N =8) .79 4. Diploma (N = 2) .79 5. Bachelor's degree (N = 2) .79 Paraprofessionals (N = 24) 1. Diploma (N = 8) .83 2. Bachelor's degree (N = 10) .83 3. Master's degree (N = 4) .83 4. Ph.D./Ed.D degree (N = 2) .83 87. The correlation between the theoretical and observed problem-solving pattern of communication and levels of education remained constant (Table 22) for both parents (r = .79) and paraprofessionals (r = .83). The paraprofessionals as a group did show a slightly higher correlation with the problem-solving pattern of communication than did the parents. The conclusion, then, was that problem-solving patterns did not increase through higher levels of education for parents or paraprofessionals. Sex of Dyads It was expected that: Parent-professional and parent-paraprofessional dyads of the same sex show a higher correla-tion between their observed patterns of communication than dyads of the opposite sex. TABLE 23 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN OBSERVED PATTERNS OF COMMUNICATION OF DYADS OF THE SAME SEX WITH DYADS OF THE OPPOSITE SEX Same Sex Opposite Sex Parent-Professional Dyads and .95 .90 Parent-Paraprofessional Dyads In Table 23 i t can be seen that a slightly higher correlation (r^ = .95) was obtained by dyads of the same sex as compared to dyads of the opposite sex (r^ = .90). Thus, i t would appear that as expected, communication patterns with the same sex were slightly more similar than patterns with the opposite sex. Kind of Appointment It was questioned whether: Regular vi s i t s result in a higher correlation between the observed and theoretical problem-solving pattern of communication than i n i t i a l v i s i t s . TABLE 24 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PATTERNS OF PROBLEM-SOLVING COMMUNICATION AND APPOINTMENTS Problem-solving Regular Apointment .83 Initial Appointment .79 As questioned, regular appointments did result, but only slightly, (Table 24) in a higher correlation with the problem-solving pattern of communication (r = .83) than did those with an i n i t i a l appointment —s (r* = -79). Frequency of Interaction of Dyads It was expected: As the frequency of interaction between each dyad increases so does the correlation between the observed and theoretical problem-solving patterns of communication. 89. TABLE 25 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PROBLEM-SOLVING PATTERNS OF COMMUNICATION AND FREQUENCY OF INTERACTION OF DYADS Problem-solving Second V i s i t to More First V i s i t Than Once a Week Parent-Professional 7 Q 0 Dyads • / y - 8 i Parent-Paraprofessional 7 Q Q 0 Dyads ' / y As expected, there was a slightly higher correlation (Table 25) of the problem-solving pattern of communication with dyads who met more than once (r^ = .83) than with those who interacted on only one occasion (is = -79). Acquaintance of Dyads It was expected: As the length of acquaintance of dyads increases so does the correlation between the observed and theoretical problem-solving patterns of communication. 90. TABLE 26 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PROBLEM-SOLVING PATTERNS OF COMMUNICATION AND ACQUAINTANCE OF THE-DYADS Parent-Professional Dyads Parent-Paraprofessional Dyads Dyads Combined Problem-solving One Occasion ,79 ,79 ,79 Two Occasions -Over Three Years .83 .83 .83 As expected (Table 26), the problem-solving pattern of communication correlated slightly higher when the acquaintance of the dyads was extended over more than one occasion (r^ = .83) as compared to only one occasion ( r s = .79). Role and Pattern of Communication It was expected: Parents', professionals' and paraprofessionals' observed patterns of communication show a higher correlation with the cross-purpose theoretical than with the problem-solving or defensive theoretical patterns. 91 . TABLE 27 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THE THREE THEORETICAL PATTERNS WITH THE OBSERVED PATTERN OF COMMUNICATION FOR EACH GROUP Defensive Cross-Purpose Problem-Sol ving Professionals .48 .91 .83 Paraprofessionals .51 .85 .83 Parents .44 .89 .79 As expected, professionals', paraprofessionals' and parents' patterns of communication (Table 27) did correlate more highly but only slightly, with cross-purpose (.91, .85 and .89 respectively) than with problem-solving (.83, .83, and .79 respectively) or defensive (.48, .51 and .44 respectively) patterns of communication. Problem-Solving Pattern It was expected: The paraprofessionals' observed pattern of communication shows a higher correlation with the problem-solving theoretical pattern of communication than that of either the parents or the professionals. By referring back to Table 27 i t can be seen that there was no difference between paraprofessionals'and professionals' utilization of the problem-solving pattern of communication (r = .83 for both groups). As —s was expected, parents did u t i l i z e the problem-solving pattern of communi-cation less (r^ = .79). This expectation, then, was only partially f u l -f i l l e d . The differences in the r (Spearman rank correlation coefficients) —s would be expected to be small since the relative frequencies attributed 92. to each cell are so similar, in addition to the small numbers (9). It will be recalled that the relative frequencies for five of the nine cells of cross-purpose and problem-solving patterns are identical. Summary The results df hypotheses testing and comparisons between the theo-retical and observed patterns of verbal communication were s t a t i s t i c a l l y analyzed and reported. 9 3 . CHAPTER VI DISCUSSION AND IMPLICATIONS Several topics are elaborated in this chapter. First, the hypotheses are discussed in the light of the findings. Second, the u t i l i t y of the SAVI system is evaluated. Third, the study's a) contribution to role theory and SAVI theory, b) implications for educational practice and c) specific implications for health care systems are examined. Fourth, some major conclusions of the study are formulated. Finally, some avenues for future research are proposed. Discussion During the i n i t i a l stages of data collection the majority of the interviews involved people who were interacting for the f i r s t time. This was a factor which could not be altered. The researcher, therefore, obtained data from an additional group of six paraprofessionals, each in interaction with three different parents. Each of these dyads had worked together on a minimum of six different occasions. These data provided 18 additional parent-paraprofessional interactions. Since these additional parent-paraprofessional dyads knew one another, they were expected to show differing profiles, coming from a more relaxed base. The interesting finding, however, is that their verbal messages and behaviours and patterns of communication, for the most part, did not differ significantly from the original group. Standing as an additional group, their data were not used in testing the hypotheses and are reported here only when their behaviour deviated from that of the original group. 9 4 . Patterns of Communication The single most interesting finding was that the cross-purpose pattern of communication predominated over problem-solving and defensive patterns for a l l groups. Cross-purpose communication, as the label.implies, is low in building on the ideas of others, in personal and humanizing ele-ments, and in jokes relating to work. These negative elements of cross-purpose talk do not preclude its being a task-oriented behaviour - inform-mation is given but not utilized. The suggestion might be made that professionals and paraprofessionals indulge in cross-purpose interaction patterns in an attempt to maintain integrity and self-preservation when confronted with insolvable problems. The use of this pattern could be a form of deviant non-involvement which decreases the risk of becoming entangled in the parents' emotional web of demands. The cross-purpose pattern might be a form of compliant verbal exchange whereby professionals and paraprofessionals go through the motions of helping to solve problems while rationalizing that parents' problems are primarily social, not medical, and thus outside their j u r i s -diction. Parents, being faced with cross-purpose communication from both professionals and paraprofessionals, could become engulfed and therefore also assume that pattern of verbal behaviour. While the predominance of the cross-purpose pattern was hypothesized and confirmed for the three groups, this researcher then questioned the extent of the dominance of cross-purpose talk and the existence of problem-solving communication. The prevalence of these two patterns of talk was investigated by examining each individual's profile and each paraprofessional's discipline in addition to studying data from other researchers' studies. 95. Results of individual parents', professionals' and paraprofessionals' communication profiles are found in Appendix F 1-3. Eighty-three percent of the individuals from a l l three groups used cross-purpose talk. In the additional group where participants knew one another, 92% of the interactions were also of this type. The cross-purpose pattern also predominated when paraprofessionals were separated into their seven disciplines (Table 28). The data collected by Agazarian (1968), Browne (1977), Cox (1976) and Harvey (1976) were examined to determine the typical pattern of communication. Agazarian's, Browne',s and a. portion of Cox's data showed a cross-purpose pattern. In each of the situations (Table 29) where medically oriented people were involved (Cox and Browne) the cross-purpose pattern dominated. Since cross-purpose talk dominated on both the individual and group levels in this study and occurred in other studies, i t should be questioned whether this is the prevailing talk in daily interactions. Problem-solv-ing might be an ideal people strive for, but cross-purpose talk may represent the majority of interaction since i t carries task achievement elements though devoid of interpersonal and approach exchanges. Conversa-tions, perhaps, predominate in monologues disguised as dialogues. Another interesting finding, contrary to expectation, was that para-professionals did not engage in problem-solving communication more fre-quently than parents or professionals. According to F i l l e y (1975), problem-solving consists of high quality and mutually acceptable solutions where facts and feelings are reported and feedback given without judge-ment or accusation. Since less than 5% of the paraprofessionals' total 96. TABLE 28 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PATTERNS OF COMMUNICATION WITH DIFFERENT TYPES OF PARAPROFESSIONALS Type of Problem- Cross- Total Paraprofessional solving Purpose Codes Community Health Nursing .83 .89 307 Social Work .83 .89 633 Speech Pathology .83 .89 787 Occupational Therapy .83 .89 789 Physiotherapy .83 .90 1087 Psychology .83 .89 49 Education Consulting .60 .70 16 Note: Education Consultant's verbal behaviour unrepresentative because of extremely low frequency of total codes. TABLE 29 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PATTERNS OF CROSS-PURPOSE AND PROBLEM-SOLVING COMMUNICATION FROM OTHER RESEARCHERS' STUDIES 1. Agazarian (1968) - data from page 71 Situation - Group of Academic Specialists Cross-purpose .92 Problem-solving .62 2. Harvey (1976) - Data from Appendix G (Tables Gl, G2, G3) Situation - Small GrouD of Graduate Students and Teachers Table Gl -Cross-purpose .64 Problem-solving .85 Table G2 -Cross-purpose .59 Problem-solving .85 Table G3 -Cross-purpose .71 Problem-solving .83 3. Browne (1977) - Data from Table 5, page 84 Situation - Health Professionals with clients Cross-purpose .86 Problem-solving .75 4. Cox (1976) - Data from pages 237, 238 and 239 Situation - Medical, Behavioral, and Systems Models Medical Individual Group Group Cross-purpose .89 .36 .78 Problem-solving .79 .15 .47 Behavioral Cross-purpose .76 .76 Problem-solving .68 .81 Systems Cross-purpose ' — .76 Problem-solving — .81 98. interactions involved personal messages,and less than 8% included approach behaviours, the criticism that feelings were rarely encouraged and feed-back rarely given must be levelled in their direction. Shave (1979) listed the essential elements for effective communication as: t e l l i n g , asking, listening, observing, understanding and convincing. Since 89% of paraprofessionals 1 exchanges with parents involved their giving opinions and asking questions, l i t t l e time remained for the other elements of effective communication, particularly personal concerns and approach behaviours. The parents were l e f t primarily with the responsibility for answering the numerous questions posed. If the interviewing climate is not set up to encourage mutual exchange of information and discussion, then the real problems are never exposed, thus never resolved (Agazarian, 1968; F i l l e y , 1975; Knowles, 1970; Korsch et a l . , 1968; Korsch and Negrete, 1972; Mambert, 1971; and Miller, 1967). Verbal Behaviours and Messages Parents are responsible for their handicapped child, and may naturally exhibit reactions of greater polarity and possess a greater vested interest in their child's problem than do professionals and paraprofessionals. Parents reacted, therefore, to the interview with greater extremes of avoidance and approach verbal behaviours than the other two groups. Possibly, professionals and paraprofessionals did not incorporate approach behaviours into their interactions, as anticipated, fearing to become too emotionally involved and thus lose their a b i l i t y to be objective. Similarly, they did not u t i l i z e avoidant behaviours since ethically and morally they are service oriented people who have entered this chronic care f i e l d in 99. order to help handicapped children and their families. It might be sus-pected that educators taught professionals and paraprofessionals to be empathic or selected them because they had the potential to be empathic but gave limited guidance in this c r i t i c a l area. When the number of personal messages was compared to factual messages, al l groups used the latter more frequently. Personal information would, in many instances, be restricted and held in abeyance while factual in-formation predominated because an interview is considered a formal type of communication. It was noted though, when personal messages were initiated, they did tend to recur in that interview., Paraprofessionals only failed to conform to the prediction that factual messages would be used more often than orienting messages. Since orient-ing messages are considered those which direct or orient the conversation, paraprofessionals may be demonstrating leadership qualities of either a democratic or autocratic nature. Paraprofessionals appeared to follow the democratic model since they spent 46% of their time giving encouragement, a democratic behaviour, whereas only 2% of their time was spent giving commands, a behaviour more typical of an autocratic leader. The verbal behaviours and messages may be summarized by noting: parents used approach and avoidant behaviours more frequently than the other two groups; paraprofessionals used both factual and orienting messages equally whereas factual messages were preferred by the other two groups; and a l l groups utilized more factual than personal messages. Demographic Variables A number of demographic variables were believed to influence the verbal interactions. Since interactors of similar roles communicate more 100. easily than interactors from dissimilar roles i t was predicted participants of the same sex might communicate with greater similarity than participants of the opposite sex. Although dyads of the opposite sex might result in more exciting and novel interactions, which have the potential for problem-solving communication, their verbal behaviours might not be as predictable. This unpredictability of verbal behaviour might be related to a dissimilarity in the communication pattern with dyads of the opposite sex as was the finding in this study. Problem-solving behaviours are probably not only learned in the edu-cational system and in experiential contacts with people, but surface when individuals feel comfortable and confident in a situation. Thus the finding, that the problem-solving pattern was not associated with higher levels of education,may be more related to past experiential con-tacts of the individual and feelings of self-confidence than to levels of education. It will be recalled that problem-solving behaviours for both dyads increased slightly as the number of vis i t s and the length of the acquain-tance increased. Also, regular appointments correlated more highly with problem-solving than i n i t i a l appointments. Seemingly, as individuals became more familiar with each other, they could devote more energy to solving the problem confronting them than to predicting each others' behaviours. Once this anticipatory stage was over, i t appeared communica-tors could concentrate more on the content of the communication and less on the process of how they should communicate. It was interesting to note that.for each group no predicted signi-101. ficant relationships were found between age or experience and approach behaviours. It was expected that as individuals increased in maturity and experience, they would feel more confident in u t i l i z i n g approach behaviours. The flaw in this prediction may be that neither approach behaviours nor confidence are related to age or experience. Approach behaviours and confidence, however, may be related to each other. Only further research could determine whether a relationship exists between the level of confidence and the frequency of approach behaviours. In summary, professionals and paraprofessionals tended to remain in verbally neutral territory engaging only in factual messages, leaving problem-solving, personal messages and approach behaviours primarily to the recipients of care, the parents. The individual participants from the three groups and their demo-graphic backgrounds, verbal messages and behaviours were not the only factors affecting the results; the SAVI system i t s e l f played an important role in these findings. Thus, the u t i l i t y of the SAVI instrument will be discussed in the subsequent section. U t i l i t y of the SAVI System It will be recalled that a principal objective of the research reported here was an assessment of the u t i l i t y of the SAVI system for studies of communication in paediatric health care delivery settings. In this section the strengths and weaknesses of the system found in the present study will be discussed. In addition, changes will be proposed to two of 102. the three theoretical patterns of communication, cross-purpose and problem-solving. Strengths of the SAVI System The SAVI instrument was selected over other instruments because the underlying theory stressed the problem-solving aspects of communication ' which fitted most closely with this researcher's objective. This was to explore the differences, i f any, found in the problem-solving aspects of dyadic communication when different role enactors communicated. The SAVI category system provided a detailed measure of individual or group communication patterns which could then be collapsed into nine major areas of verbal behaviour. The use of typescripts acted as a mode of standardizing the data for the coders and aided in maintaining intra-and inter-rater r e l i a b i l i t y at a high level. Listening to the entire tape prior to coding the six-minute sample ensured that the dialogue was judged in context. Problem-solving, cross-purpose and defensive theoretical patterns, when compared to observed patterns, provided a novel and insightful way of analyzing the data. The patterns of communication formed by the dis-tribution of verbal behaviours in each.of the nine areas could be identified. The SAVI system proved versatile in a variety of situations and analyses. All utterances were codable into categories. The data could be examined from raw frequencies as was done in this study or from a sequential analysis of each speaker, depending on the needs of the investi-gator. With this system data could be combined so that profiles of large groups, dyads or an individual could be examined. Research could be developed around groups, organizations or a single case study of an indi-103. victual, such as a lecturer. The SAVI system placed minimal time demands on the participants and so enhanced their cooperation. Time is an important consideration in health care research, being expensive, and professionals, paraprofessionals, medical directors and administrators may not receive direct rewards or benefits for their time or cooperation in research studies. Another positive feature of SAVI related to the utilization of tape recordings thereby minimizing the interaction effects between researcher and participants. Weaknesses of the SAVI System The primary problem, which was discovered during intra and inter-rater r e l i a b i l i t y studies, revolved around the precision of the operational definitions of three of the 40 categories this sample used. These problem categories were facts and figures, opinions, and factual answers. The other weakness was the insistence of the instrument's developers (Simon and Agazarian, 1967) that coders must be in geographic proximity to allow use of a practice tape, discuss the type of data, fine tune themselves into the system and adjust their coding procedures to f i t with other coders. It was believed that i f coders had to be in close prox-imity, the instrument was seriously restricted as a research tool. There-fore, inter-rater r e l i a b i l i t y studies were carried out with raters both geographically separated and in close proximity. The average r e l i a b i l i t y for coders who were geographically separate was 79% as compared to 87% for coders in geographic proximity. Thus, an acceptable level of r e l i a -bility, with geographically separated coders is certainly possible and should be encouraged. 104. Proposed Adaptiqns_to the SAVI System The theoretical models of problem-solving and cross-purpose patterns of communication proposed by Agazarian (1968) are very similar to each other. Of the nine areas, five had identical relative frequencies in Agazarian's model (see Figure 4, p. 105). Because of these similarities, statistical and practical differences between the models are d i f f i c u l t to identify. According to this researcher's logic, problem-solving should contain a balance between asking questions and informing on the one hand, and answering questions and summarizing on the other. The areas of factual-contingent and factual-approach, therefore, should hold identical relative frequencies. Agazarian's (1968) model designated the factual-contingent area, a high relative frequency, and the factual-approach area, a medium relative frequency. It is proposed that since both these areas are used extensively, they both should be designated a high relative fre-quency. The adapted problem-solving model is on the right of Figure 4. In the cross-purpose model, the developers have given both the factual-contingent and orienting-contingent areas a high relative fre-quency. This researcher believes that individuals using cross-purpose talk would not be as concerned with shifting or orienting the direction of the conversation but would be more interested in presenting facts and figures, informing, te l l i n g funny stories, and engaging in small talk. It is proposed, therefore, that the contingent-orienting area be desig-nated a medium relative frequency. The cross-purpose adapted model is dis-played in Figure 4. When the relative frequencies of the newly proposed theoretical models of cross-purpose and problem-solving patterns are compared to each FIGURE 4 SAVI THEORETICAL PROPOSED THEORETICAL COMMUNICATION PATTERNS COMMUNICATION PATTERNS Problem-solving Problem-solving P F 0 P F 0 Av. L* L L* Av. L* L • L* C. M H* H* C. M H* H Ap. M M* M Ap. M H M Cross-•purpose Cross--purpose P F 0 P F 0 Av. L* M L* Av. L* M L* C. L H* H* C. L H* M Ap. L M* L Ap. L M L * indicates identical relative frequencies between the problem-solv and cross-purpose models. Av. = Avoidant Behaviour C = Contingent Behaviour Ap. = Approach Behaviours P = Personal Messages F = Factual Messages 0 = Orienting Messages Relative Frequencies: L = Low ( < 7%) M = Medium ( > 7 < 20%) H = High ( > 20%) 106. other, there are only three areas containing identical relative fre-quencies (Figure 4). These proposed models were tested on individuals whose verbal patterns had already been identified as problem-solving. The results of the proposed theoretical models correlated with the observed data are found bracketed in Appendix F 1-3. In a few cases the proposed theoretical models decreased the differences in correlations between problem-solving and cross-purpose patterns thus making them more d i f f i c u l t to differentiate. In most instances, however, the proposed theoretical models increased the differences in correlations between these two patterns, thus making each pattern more distinctive and easier to identify. During hypotheses testing i t was found that parents used personal messages and approach behaviours with greater frequency than professionals and paraprofessionals. Since personal messages and approach behaviours are c r i t i c a l to the problem-solving pattern, the finding that the parents' pattern of communication was not of a problem-solving type was disturbing. The proposed models were tested using the three groups - parents, professionals and paraprofessionals. The findings for each group were as follows: Problem-solving Cross-purpose Parents .89 .86 Professionals .88 .87 Paraprofessionals .74 .80 These findings, then, show that parents as a group utilized the problem-solving pattern of communication. Paraprofessionals remained with a cross-purpose pattern of communication as their predominant type of talk. Professionals, because of the marginality of their factual-approach 107. behaviours (8.03% while the medium range is > 1 < 20%), moved into the problem-solving pattern (Appendix G). Had they used 1.04% fewer factual-approach behaviours they would have demonstrated identical patterns to the parapro-fessionals. Similarly, had the paraprofessionals used .04% more factual-approach behaviours, they too would have fallen into the problem-solving pattern. Considering other relevant data in conjunction with the margin-al ity of the factual-approach area, professionals and paraprofessionals logically f i t closer to cross-purpose talk than to problem-solving. Of the three groups and nine areas, the paraprofessionals 1 and professionals' verbal behaviours in the factual-approach area showed the only instances of marginality. Thus, the proposed theoretical models represent a more logical and better statistical f i t than Agazarian's models when marginality and other relevant findings are considered. In summary, the u t i l i t y of the SAVI instrument was examined and dis-cussed. Adaptations to Agazarian's (1968) cross-purpose and problem-solving theoretical patterns of communication were proposed, t r i a l tested and discussed. The implications for theory and practice are elaborated in the following section. 108. Imp!ications Theoretical implications for role theory and SAVI theory are dis-cussed. Practical implications are put forth for medical and parapro-fessional educators and continuing educators in the Health Sciences. The implications these findings have for the chronic paediatric health care delivery system are stressed. Implications for Role Theory and SAVI Theory The underlying theory supporting this thesis was that the role of the participant would be related to the type of talk that would take place in an interview situation. This finding seems to have been substantiated. The role into which one has been socialized did seem to be related to one's verbal communication pattern. Certainly those within.the organization (professionals and paraprofessionals) communicated differently from those outside the organization (the parents). During this study i t was possible to test the SAVI theoretical models of problem-solving, cross-purpose and defensiveness and to devise a way of examining them s t a t i s t i c a l l y (see Appendix E). It will be recalled that five of the nine areas of the SAVI theoreti-cal models of problem-solving -and cross-purpose patterns have identical rela-tive frequencies. Thus, adapted models of cross-purpose and problem-solving were proposed and tested. These adapted theoretical models are believed to contribute to the SAVI theory. More extensive testing of these models, than was possible in this study, is suggested. 109. Implications for Educators in Medicine and Health The chronic care f i e l d entails a long-term commitment to treatment for the handicapped child. Expertise and confidence in u t i l i z i n g inter-personal s k i l l s would go a long way to improving the treatment of chroni-cally handicapped children. According to Korsch (1968 and 1972), parents' dissatisfaction with paediatric health care personnel focused on the lack of warmth and personal interest shown to them and their child. The major implication resulting from this study is directed at edu-cators in professional and paraprofessional fields. The health care f i e l d is a practically oriented area and practical experience in interpersonal s k i l l s is equally as c r i t i c a l as theory. More attention could be directed towards providing medical students, residents and paraprofessionals with practical experience, both in interviewing s k i l l s and in monitoring their communication profiles prior to patient contact. Patient simulation could provide students with a setting in which they could work through a range of verbal behaviours, with assistance, prior to being expected to function independently and effectively in their roles. Ideally, medical and health instructors should also be effective communicators and thus pro-vide role models for students to emulate. Modern technology and health care knowledge is continually becoming outdated and replaced but the need for interpersonal communication remains. Focus should be placed, there-fore, not only on technology and knowledge demands, which are so variable, but also on interpersonal communication which encompasses the very founda-tion of human experience. Continuing educators could provide professionals and paraprofessionals with splinter s k i l l s at least to aid them in becoming better communicators. no. Ongoing small group practical sessions are needed where participants can learn and practise in a non-threatening atmosphere, since communication is a very personal thing. According to the results of this study, parents, for the most part, are co-operating in the interaction. They, too, should be given some instruction and practice, perhaps in a role playing situation, as to how they can communicate their problems. Parents appear motivated and should be given an opportunity to learn how to maximize the benefits of their time spent with professionals and paraprofessionals. The implications of this study for health delivery systems are elabor-ated in the following section. Implications for the Chronic Paediatric Health Care Delivery System An emphasis on interpersonal communication studies as an adjunct to "Quality of Health Care" is an additional implication of this research. Because cure is no longer an issue for chronically handicapped children, the quality.of care becomes c r i t i c a l for their parents. One of the four major barriers hindering the improvement and development of the quality of care given to these children is created by inadequate inter-personal communication. Informal and personalized communication is seen as vital in the f i e l d of paediatric chronic care. Knowledge of the family's l i f e style is important in meeting the needs associated with future medical, surgical and social problems and to allow trust and confidence to develop and grow between the interactors. In this study informal and personalized m . communication was found to be used infrequently while formal communication of a more instructional nature prevailed. In chronic care, shared infor-mation through effective interpersonal communication regarding management, treatment and other long-term goals can lessen or curtail the handicap. In this study inadequate communication was associated with some children's handicaps, according to the reports. These handicaps might have been pre-vented had attending physicians 1istened to the parents' early concerns. Mothers reiterated accounts of their suspicions which were later confirmed on the birth of their handicapped child. The additional three barriers related to the quality of the health care delivery system include power, funding and prevention (as discussed in chapter one). These are also believed to be dependent upon effective communication. While some element of power was evidenced in physicians' verbal behaviour, i t was less than that used by paraprofessionals. Both physicians and paraprofessionals assumed leadership roles over parents by engaging in orienting behaviours (42%, 47% and 25% of the time respectively). Both rarely used behaviours typically associated with power, such as attacking, demanding, commanding or obscuring verbal behaviours. They did, however, use power in exercising control over the interaction by their utilization of orienting messages. This verbal exercise of power by physicians and paraprofessionals is associated more highly with the cross-purpose rather than the problem-solving pattern of communication. Funding proposals are constantly suffering cutbacks which appear to be related to a lack of communication. Both givers and recipients of funding feel exploited. Trusting relationships do not 112. appear.? to have developed. One area in which these cutbacks are being suggested is technology which can, for example, identify newborns in distress and often prevent handicapping conditions. Many of the parent-professional dialogues indicated that these new technologies had been available but were not used for their child. Approximately 40% of these children had conditions which occurred at birth and are now largely pre-ventable with new technology and effective communication (listening and shared information). The costs of technology and the effective use of communication do not seem as great as the life-long costs of care and possible institutionalization of these children. Accounts of parents' experiences included lack of information given by physicians which resulted in inadequate communication and in turn created.difficulty and delay in gaining entry into the health care system. Upon entry into the system, parents experienced further frustration when lack of communication failed to prepare them for treatment delays. These were due to waiting l i s t s and lack of trained personnel. Prevention seems limited with inadequate communication prevailing between the experts and the public. Many handicapped children are known to be associated with older, high-risk mothers (Erickson, 1978; Goldberg, Edmonds and Oakley, 1979; Sells and Bennett, 1977). In this study the age of mothers was certainly indicative of an older at-risk population since the mean and median ages were 33 years with a mode of 29 years. The parents did not appear disadvantaged from the point of view of exper-ience and education since the average time spent in the parenting role was nine years and 70% of the parents had completed secondary or higher educa-tion. More effective interpersonal communication and public education are needed with these high risk populations. 113. Summary If cross-purpose talk dominates throughout the delivery of paediatric health care services, then the costs of health care could escalate un-necessarily. This cross-purpose talk could result in making parents more dependent upon the system, since their problems (as they experience them) are rarely discussed, thus never resolved in terms of their concerns for their handicapped child. Not only is the professionals' and parapro-fessionals' quality of care-giving in question, but the effectiveness of delivery of care to parents, in terms of costs, patient outcomes and the incidence of handicapping conditions is also in serious question. In the remaining sections, the conclusions of the study are elaborated and some ideas for future research are put forth. Conclusions of the Study Through an intensive analysis of the data, the following conclusion is inescapable. Professionals and paraprofessionals engaged in very similar verbal patterns of communication - they asked questions and oriented the conversation, thus assuming the leadership role. The parents only infrequently asked questions. Having listened to a l l of the taped interviews, at no time was the legitimacy of the professionals' or paraprofessionals' roles questioned. Rarely did the professional or paraprofessional engage in informal commun-ication and rarely did the parents attempt to break through the formal communication barrier. None of the groups engaged to any significant degree in personal support, personal sharing, jokes or building on each others' ideas. They did not incorporate interpersonal messages in their 114. talk, thus preventing the communication process from becoming humanized. Parents in this study were seeking a specialized kind of treatment for their handicapped child and the assumption they brought to the situa-tion was that help, advice, and solutions would be forthcoming. Because the paraprofessionals' communication patterns were so similar to the pro-fessionals, the question arises as to whether they too are imitating the medical model. This,model, based on a "curing" approach with formal communication, cannot be transplanted into the "caring" mould with its emphasis on informal communication. The conclusion, then, is that the "curing" approach, with its lack of emphasis on feelings, dominates the delivery of health care even in the "caring" situation. Although inroads have been made in that professionals and paraprofessionals are aware of the importance of communication, they have not yet risked shifting from the s t r i c t l y formal and factual to the more informal and helping mode of communication. Parents hope for a dialogue which will help to alleviate some of the problems they encounter in coping with their handicapped child. Instead, they tend to be bombarded with questions accompanied by unsolicited factual information and opinions. Cross-purpose talk emerges and dominates. The findings from supported those of Cox's (1976) study where she found that different ideologies did not alter the communication pattern. This study corroborated Cox's that levels of education did not alter the communication pattern. The finding of a predominant pattern of cross-purpose talk was strengthened by analyzing Cox's (1976) and Browne's (1977) data and discovering the same pattern predominated for medical personnel. These results help to confirm the presence of the 115. cross-purpose pattern of talk as more than a chance occurrence. Limitations of the Study In any study there are limitations and this one is no exception. The nature of the study was primarily exploratory with attention given to the r e l i a b i l i t y and u t i l i t y of the SAVI instrument. The results, there-fore, must be viewed in this light. The findings cannot be generalized out-side the realm of the specific centres from which the data were collected, because the sample size of professionals and paraprofessionals was in-sufficient to permit this. There was the potential limitation of using only a six-minute sample while judging i t to be representative of the entire tape. In future, i t would be worthwhile during pilot testing to code an entire tape in addition to coding a sample to test for representativeness of the selected sample. Professionally trained coders were taught, when in doubt about a code, to code i t with the last code (Agazarian, 1968). This added to the factual and contingent areas since most conversations utilized these two areas. Greater refinement of the operational definitions might alter the current findings. There is the possibility that the theoretical patterns of communication need revisions as proposed. The conclusion regarding the u t i l i t y of the SAVI instrument is that i t permits creativity and is an adaptable and versatile tool with a definite application for the health care f i e l d . It is possible to reach a consistent and high level of r e l i a b i l i t y using this instrument. Empirical investigations similar to this study were not found. It appears this study makes a special contribution in uncovering the predomi-nant communication profile of participants in chronic paediatric health 116. care delivery settings to be one of cross-purpose communication. Further study is required to confirm categorically that cross-purpose talk is the predominant pattern of communication. It is appropriate, how-ever, to discuss briefly in the next section some additional considerations for future research which may plant a seed in the minds of other researchers. Future Research In the research strategy used here, the analytic process was applied to chronic paediatric health care delivery settings. The micro-analytic approach was primarily dictated by the SAVI system methodology. A number of. suggestions using this system are proposed for future research direc-tions . An attempt should be made in a future study to acquire greater numbers of a more unified sample (i . e . , one type of paraprofessional) in order to eliminate some of the diversity of the present sample. With this more unified sample, then, testing for the predominance of the cross-purpose pattern of communication could strengthen or weaken the validity of the present findings. An investigation into other "helper-helpee" relationships within and outside the health f i e l d (lawyer-client, accountant-client, counsellor-client and parent-child) would provide additional evidence. Whether the cross-purpose pattern of communication is the most typical pattern used across professional lines is an important question to answer. The determination of the prevalent pattern of communication - cross-purpose, problem-solving or defensiveness - among and between both pro-117. fessionals and paraprofessionals could substantiate or refute the findings of McGill (1976). She found significant differences in communication behaviours among health professionals of different positional levels using Bales' Interactional Process Analysis. A future study could make a substantial contribution by comparing not only the findings but the Bales and SAVI instruments. The question of whether individuals could change their communication profiles is c r i t i c a l and must be answered with substantiated research results. To research this question, an empirical study involving experi-mental and control groups could be employed. The experimental group could receive training and practical experience in effective communication and in monitoring their own communication. The control group would receive no training or experience. Differences found between the two groups would indicate whether change had taken place. A final direction proposed for future research involves the documen-tation of specific tests on the construct validity of SAVI which would increase confidence in the results. A questionnaire could be constructed involving the nine areas of the SAVI system. Participants could be asked to complete this questionnaire, regarding their own use of messages and behaviours, and the messages and behaviours they perceived the other participant to have used. In a dyadic interaction there would be the behaviour as recorded and analyzed by SAVI coders, and self and other's perception of the kinds of verbal utterances displayed by each individual. The correlations, then, between SAVI coded behaviours and perceived utterances, and between SAVI coded behaviours and reported utterances, would add to the construct validity. 118. Summary In common with many other investigations of interpersonal communica-tion, this study has raised more questions than i t has answered. Explora-tory and tentative though these findings may be, they do provide encourage-ment for further study of the relationships between various roles and communication profiles. Future studies also will seek greater understand-ing of the complexities of different roles and their relative importance in u t i l i z i n g a problem-solving pattern of communication, especially in paediatric health care delivery settings. Finally, efforts must be con-tinued to seek improved ways of providing early practical training in interpersonal s k i l l s for professionals and paraprofessionals. These people must be prepared with superior communication s k i l l s to help parents cope with the multiplicity of problems they encounter when rearing their handicapped child. The Celdic report (1970), a Canadian publication on children with special problems, addressed the problem of communication with parents. It stated: The greatest problem of communication is the illusi o n that i t has been achieved. We are convinced that every professional would benefit from training which would help him to increase his s k i l l in communication (p. 436). 119. BIBLIOGRAPHY Agazarian, Y. A theory of verbal behavior and information transfer. Un-published doctoral dissertation, Temple University, 1968. Agazarian,•• Y. Personal Communication. January 30, 1979. Agazarian, Y. and Simon, A. SAVI Manual. Philadelphia: Unpublished Manual, 1976. (SAVI Communications, 1831 Chestnut St.) American Psychological Association, Division of Industrial-Organizational Psychology. Principles for the validation and use of personnel  selection procedures. Washington: American Psychological Association, 1975. American Psychological Association, Standards for educational and psycholog-ical tests and manuals. In D. Jackson and S. Messicks (Eds.), Problems  in human assessment. New York: McGraw-Hill Book Company, 1967. Anastasi, A. Psychological testing. New York: Macmillan, Third Edition, 1969. Applied Research Associates. Professional education - a policy option. A study prepared for the Commission of Post Secondary Education in Ontario, 1971. Argyle, M. and Dean, J. Eye contact, distances and a f f i l i a t i o n . In M. Argyle (Ed.), Social encounters. Middlesex, England: Penguin Books, 1973. Asch, S.E. Effects of group pressure upon the modification and distortion of judgements. In J.H. Campbell and H.W. Hepler (Eds.), Dimensions  in communication: Readings. Belmont, California: Wadsworth Publishing Company, Inc., 1970. Bales, R.F. Personality and interpersonal behavior. Toronto: Holt, Rinehart & Winston, 1970. Bales, R.F. How people interact in conferences. In J. Lavers and S. Hutcheson (Eds.), Communication in face to face interaction. Baltimore: Penguin Books, 1972. Bales, R.F. The equilibrium problems in small groups. In M. Argyle (Ed.), Social encounters. Middlesex, England: Penguin Books, 1973. Bates, F.L. Position, role and status: A reformulation of concepts. Social Forces, 1956, 34, 313-321. Bauer, R.A. The obstinate audience: The influence process from the point of view of social communication. In W.G. Bennis, K.D. Benne and R. Chin. (Eds.), The planning of change. Toronto: Holt, Rinehart & Winston, Inc., 1969. 120. Bechtoldt, H.P. Construct validity: A critique. American Psychologist, 1959, 14 (10), 619-629. Bergin, A.E. The evaluation of therapeutic outcomes. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of psychotherapy and behavior change: An empirical analysis. Toronto: John Wiley & Sons, Inc., 1971. Berlo, D.K. The process of communication. Toronto: Holt, Rinehart & Winston, 1960. Bertrand, A.L. Social organization: A general systems and role theory  perspective. Philadelphia: F.A. Davis Co., 1972. Biddle, B.J. and Thomas, E.J. Differentiation, specialization, and division of labor. In B.J. Biddle and E.J. Thomas (Eds.), Role theory: Concepts and research. New York: John Wiley & Sons, Inc., 1966. Blau, P.M. and Scott, W.R. Processes of communication in formal organiza-tions. In M. Argyle (Ed.), Social encounters. Middlesex, England: Penguin Books, 1973. Boulding, K. Introduction to "The image". In J.H. Campbell and H.W. Hepler (Eds.), Dimensions in communication: Readings. Belmont, California: Wadsworth Publishing Co., 1970. Bowman, C.C. Role playing and the development of insight. Social Forces, 1949, 28, 195-199. Braga, J.L. Role theory, cognitive dissonance theory, and the inter-disciplinary team. Interchange, 1972, 3(4), 69-78. Browne, R.M. Patient and professional interaction and its relationship  to patients' health status and frequent use of health service. Un-published doctoral dissertation, University of Toronto, 1977. Calhoun, C.J. General status: Specific role. New Orleans: Amer. Anthro. Assoc., 1973"! (ERIC Document Reproduction Service No. ED 087 684). Campbell, D.T. Recommendations for APA test standards regarding con-struct, t r a i t , or discriminant validity. American Psychologist, 1960, 15.(8), 546-553. Campbell, D.T. and Fiske, D.W. Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 1959, 56(2) 81-105. ~ Campbell, J.T. Psychometric theory. In M.D. Dunnette (Ed.), Handbook  of industrial and organizational psychology. Chicago: Rand McNally, 1976. Canadian Medical Association. Number of doctors emigrating is up, shortages in certain areas also up - CMA. University Affairs, 1979, 20(7), 11. 121. Carkhuff, R.R. Helping and human relations: Practice and research. Volume II. Toronto: Holt, Rinehart & Winston, Inc., 1969. The CELDIC Report. One mi 11 ion children. A National study of Canadian children with emotional and learning disorders. Toronto: Leonard Crainford for The Commission on Emotional and Learning Disorders in Children, 1970. Cook, M. Interpersonal perception. Middlesex, England: Penguin, 1971. Cox, B.A. Communication systems in psychotherapy: An empirical investiga- tion into the treatment ideologies of patients and therapists. Un-published doctoral dissertation. Simon Fraser University, 1976. Crichton, M. Five patients: The hospital explained. Toronto: Bantam Books, 1970. Cronbach, L.J. and Meehl, P.E. Construct validity in psychological tests. Psychological Bulletin, 1955, 52, 281-302. Eckel, H. Developing and testing research instruments for role analysis  of educational administrators emphasizing devices to study relation- ships between variable definitions of educational administrative  a c t i v i t i e s , style, and norms and role conflict and ambiguity. Final Report. 1969. (ERIC Document Reproduction Service No. ED 043 133). Erickson, D.J. Down syndrome, paternal age, maternal age and birth order. Annals of Human Genetics, 1978, 41_, 289-298. Evans, J.R. Conflict or common cause in the education of the health pro-fessions. In W.C. Gibson (Ed.), Health care, teaching, and research: Prospect and retrospect. Victoria: Morriss Printing Company, 1975. Evans, R.G. Health care: In c r i s i s , but s t i l l among the best. The  Vancouver Sun, August 29, 1979, p. A5. Fabun, D. Communications: The transfer of meaning. Toronto: Collier-Macmillan, 1974. Ferguson, G.A. Statistical analysis in psychology and education. Toronto: McGraw-Hill, 1976. Fi l l e y , A.C. Interpersonal conflict resolution. Glenview, I l l i n o i s : Scott, Foresman & Co., 1975. Foa, U.G. Behavior, norms, and social rewards in a dyad. In B.J. Biddle & E.J. Thomas (Eds.), Role theory: Concepts and Research. New York: John Wiley & Sons, 1966. Foulkes, R.G. Health security for British Columbians Tome one: The politics of change and major philosophical issues. Ottawa: Government Publications, 1974. 122. Gartner, A. The preparation of human service professionals. New York: . Human Sciences Press, 1976. Gaudry, R. Medical research and science policy in Canada. In W.C. Gibson (Ed.), Health care, teaching, and research: Prospect and retrospect. Victoria: Morriss Printing, 1975. Gibson, J.L., Ivancevich, J.M. and Donnelly, J.H. Organizations: Structure, processes, behavior. Georgetown, Ontario: Irwin-Dorsey, 1973. Gi f f i n , E. and Patton, B.R. Fundamentals of interpersonal communication. New York: Harper & Row, 1971. Goffman, E. Alienation from interaction. Human Relations, 1957, 10, 47-60. Goffman, E. The presentation of self in everyday l i f e . New York: Doubleday & Company, 1959. Goldberg, M.F., Edmonds, L.D., and Oakley, G.P. Reducing birth defect risk in advanced maternal age. Journal of American Medical Association, 1979, 242, 2292-2294. Gordon, G. Role theory and illness: A sociological perspective. New Haven: College & University Press, 1966. Gross, E. and Stone, G.P. Embarrassment and the analysis of role require-ment. In M. Argyle (Ed.), Social encounters. Middlesex, England: Penguin Books, 1973. Hage, J. Communication and organizational control: Cybernetics in health  and welfare settings. Toronto: John Wiley & Sons, 1974. Harvey, R.B. A study of communication change in group process seminars  using the Sequential Analysis of Verbal Interaction. Unpublished doctoral dissertation, Bryn Mawr College, 1976. Hays, W.L. Statistics for the social sciences. Toronto: Holt, Rinehart & Winston, Inc., 1973. Homans, G.G. The human group. New York: Harcourt, Brace & World, Inc., 1950. Hovland, C.I., Janis, I.L. and Kelley, H.H. Communication and persuasion: Psychological studies of opinion change. New Haven: Yale University Press, 1964. (Originally published 1953). Hughes, E.C. Introduction. In E.C. Hughes, B. Thorne, A. DeBaggis, A. Gurin & D. Williams (Eds.), Education for the professions of medicine, law, theology, and social welfare. Toronto: McGraw-Hill Book Co., 1973. 1 2 3 . Huse, E.F. and Bowditch, J.L. Behavior in organizations: A systems approach  to managing. Don Mills: Addison-Wesley Publishing Co., 1 9 7 3 . I l l i c h , I. Disabling professions. In I. I l l i c h , I.K. Zola, J. Mcknight, J. Caplan and H. Shaiken (Eds.), Disabling professions. Don Mills: Burns & McEachern, 1 9 7 7 . Jaco, E.G. Society and health care administration. In E.G. Jaco (Ed.), Patients, physicians and illness. New York: The Free Press, 1 9 7 2 . Kaplan, R.M/, Bush, J.W. and Berry, C.C. Health status: Types of validity and the index of well-being. Health Services Research, 1 9 7 6 , 1 1 ( 4 ) , 4 7 8 - 5 0 7 . Kast, F.E. and Rosenzweig, J.E. Organization and management. Toronto: McGraw-Hill Book Co., 1 9 7 4 . Katz, D. and Kahn, R.L. Communication: The Flow of information. In J.H. Campbell and H.W. Hepler (Eds.), Dimensions in communication: Readings. Belmont, California: Wadsworth Publishing Company, 1 9 7 0 . Kerlinger, F.N. Foundations of behavioral research. Toronto: Holt, Rinehart and Winston, Inc., 1 9 6 6 . Knowles, M. The modern practice of adult education: Andragogy versus pedagogy. New York: Association Press, 1 9 7 0 . Kolb, D.A. Behavior change in transition roles. (Working Paper), 1 9 6 7 . (ERIC Document Reproduction Service No. ED 0 1 8 8 0 6 ) . Korsch, B.M., Gozzi, E.K. and Francis, V. Gaps in doctor-patient communi-cation: Doctor-patient interaction and patient satisfaction. Pediatrics, 1 9 6 8 , 4 2 / 5 ) , 8 5 5 - 8 7 1 . Korsch, B.M. and Negrete, V.F. Doctor-patient communication. Scientific  American,1972, 2 2 ( 2 ) , 6 6 - 7 4 . Laing, R.D., Phillipson, H. and Lee, A.R. Interpersonal perception: A  theory and a method of research. New York: Springer Publishing Company, 1 9 6 6 . Lalonde, M. A new perspective on the health of Canadians. Ottawa: Government Publications, 1 9 7 4 . Leavitt, H.J. Managerial psychology (third edition), Chicago: University of Chicago Press, 1 9 7 2 . (Originally published 1 9 5 8 ) . Lennard, H.L. and Bernstein, A. Expectations and behavior in therapy. In B.J. Biddle and E.J. Thomas (Eds.), Role theory: Concepts and  research. New York: John Wiley & Sons, Inc., 1 9 6 6 . Levinson, D.J. Role, personality, and social structure in the organizational setting. Journal of Abnormal and Social Psychology, 1 9 5 8 , 5J3, 1 7 0 - 1 8 0 . 124. Lewis, C.E. The team is in the doctor's bag. In NLN Convention Papers (Ed.), Roles on today's team: Relationships, doctor, administrator, director of nursing. New York: National League for Nursing, 1969. Linton, R. The study of man: An introduction. New York: Appleton-Century, Loevinger, J. Objective tests as instruments of psychological theory. In D.N. Jackson and S. Messick (Eds.), Problems of human assessment. New York: McGraw-Hill, 1967. Long term care in B.C.: A citizen's guide. Social Planning and Review  Council of British Columbia. Victoria: Health Branch Stockroom, January 1979, 1-24. Mambert, W.A. The elements of effective communication. Washington: Acropolis, 1971. March, J.G. and Simon, H.A. Communication. In J.H. Campbell and H.W. Hepler (Eds.), Dimensions in communication: Readings (2nd ed.). Belmont, California: Wadsworth, 1970. Matarazzo, R.G. Research on the teaching and learning of psychothera-peutic s k i l l s . In A.E. Bergin and S.L. Garfield (Eds.), Handbook of  psychotherapy and behavior change: An empirical analysis. Toronto: John Wiley & Sons, 1971. McGill, M.E. Observation of communication behavior: The development of a research method for use in health care organizations. Unpublished doctoral dissertation, University of British Columbia, 1976. McKeown, T. The determinants of human health: Behavior, environment and therapy. In W.C. Gibson (Ed.), Health care, teaching and research: Prospect and retrospect. Victoria: Morriss Printing, 1975. McNemar, Q. Note on the sampling error of the difference between corre-lated proportions or percentages. Psychometrika, 1947, 1_2(2), 153-157. Miller, G.A. The psychology of communication: Seven essays. New York: Basic Books, 1967. Mott, P.E. The organization of society. Englewood C l i f f s : Prentice-Hall, 1965. Newcomb, T.M. Role behaviors in the study of individual personality and of groups. Journal of Personality, 1949, j_8, 273-289. Nunnally, J.C. Psychometric theory. New York: McGraw-Hill, 1967. Nunnally, J.C. and Durham, R.L. Validity, r e l i a b i l i t y , and special problems of measurement in evaluation research. In E.L. Struening and M. Guttentag (Eds.), Handbook of evaluation research, Volume 1. Beverly H i l l s : Sage, 1975. 125. Parsons, T. The social system. New York: The Free Press of Glencoe, 1951. Partridge, E. Punctuation. In J.H. Campbell and H.W. Hepler (Eds.), Dimensions in communication: Readings. Belmont: Wadsworth, 1970. Perlman, H.H. Persona: Social role and personality. Chicago: The University of Chicago Press, 1974. Poynter, N. Medicine and man. Middlesex, England: Penguin Books, 1973. Rice, D. Education for tomorrow's family physicians. In W.C. Gibson (Ed.), Health care, teaching and research: Prospect and retrospect. Victoria: Morriss Printing, 1975. Robinson, G.C. Personal communication, Vancouver, British Columbia, July 3, 1979. Rogers, CR. The characteristics of a helping relationship. In W.G. Bennis, K.D. Benne and R. Chin (Eds.), The planning of change. Toronto: Holt, Rinehart & Winston, 1969. Rogers, E.M. and Agarwala-Rogers, R. Communication in organizations. New York: The Free Press, 1976. Ross, A.O. The exceptional child in the family. New York: Grune & Stratton, 1972.. Royce, J.R. On the construct validity of open-field measures. Psychological  Bulletin, 1977, 84(6), 1098-li06. Ruddock, R. Roles and relationships. New York: Humanities Press, 1969. Sarbin, T.R. Role theory. In G. Lindsey (Ed.), Handbook of social psychology (Volume I). Reading, Massachusetts: Addison-Wesley, 1954. Sarbin, T.R. Role: Psychological aspects. In D.L. S i l l s (Ed.), Inter- national encyclopedia, of the social sciences (volume 13). Toronto: The Macmillan & The Free Press, 1968. Scheflen, A.E. The significance of posture in communication systems. In J. Laver and S. Hutcheson (Eds.), Communication in face to face inter- action. Baltimore: Penguin Books, 1972. Schreivogel, P.A. Communication in c r i s i s . New York: Thomas Nelson, 1972. Scott, W.G. An overview of the f i e l d organization theory. In H.L. Tosi (Ed.), Theories of organization. Chicago: St. Clair Press, 1975. Sells, C.J. and Bennett, F.C. Prevention of mental retardation: The role of medicine. American Journal of Mental Deficiency, 1977, 82(2), 117-129. 126. Selye, H. Stress without distress. Scarborough: American Library of Canada, 1974. Shave, G.A. Getting the message. The Financial Post, 1979, July 21, 13. Simon, A. and Agazarian, Y. Sequential analysis of verbal interaction. Philadelphia: Research for Better Schools, 1967. Simon H.A. Administrative behavior: A study of decision-making in adminis- trative organization (3rd edition). New York: The Free Press, 1976 (Originally published 1945). Sommer. R. Further studies of small-group ecology. In J. Laver and S. Hutcheson (Eds.), Communication in face to face interaction. Baltimore: Penguin Books, 1972. Stoeckle, J.D. and Zola, I.K.' After everyone can pay for medical care: Some perspectives on future treatment and practice. In E.G. Jaco (Ed.), Patients, physicians and illness. New York: The Free Press, 1972. Straus, R. Hospital organization from the viewpoint of patient-centered goals. In B.S. Georgopoulos (Ed.), Organization research on health  institutions. Ann Arbor: The University of Michigan, 1972. Strauss, A.L. Medical ghettos. In E.G. Jaco (Ed.), Patients, physicians  and illness. New York: The Free Press, 1972. Tenopyr, M.L. Content-construct confusion. Personnel Psychology, 1977, 30, 47-54. Thayer, L. Communication and communication systems. Homewood, I l l i n o i s : Richard D. Irwin, 1968. Thomas, E.J. Role conceptions, organizational size and community context. In B.J. Biddle and E.J. Thomas (Eds.), Role theory: Concepts and  research. New York: John Wiley & Sons, 1966. Thomas, E.J. and Biddle, B.J. The nature and history of role theory. In B.J. Biddle and E.J. Thomas (Eds.), Role theory: Concepts and Research. New York: John Wiley & Sons, 1966. Thomas, L.E. Supervision and evaluation of paraprofessionals. Denver. Paper: Annual Meeting Rocky Mountain Psychology Association, 1971. (ERIC Document Reproduction Service No. ED 130 174). Thorne, B. Professional education in medicine. In E.C. Hughes, B. Thorne, A. Baggis, A. Gurin and D. Williams (Eds.), Education for the pro- fessions of medicine, law, theology and social welfare. Toronto: McGraw-Hill, 1973. Truax, CB. and Mitchell, K.M. Research on certain therapist interpersonal s k i l l s in relation to process and outcome. In A.E. Bergin and S.L. Garfield (Eds.), Handbook of psychotherapy and behavior change: An  empirical analysis. New York: John Wiley & Sons, 1971. 127. Turnbull, K.W. Personal communication, Vancouver, British Columbia, November 15, 1979. Watzlawick, P. How real is real? New York: Random House, 1976. Weick, K.E. The social psychology of organizing. Don Mills: Addison-Wesley, 1969. Wessen, A.F. Hospital ideology and communication between ward personnel. In E.G. Jaco (Ed.), Patients, physicians and illness. New York: The Free Press, 1972. Westley, B.H. and MacLean, M.S. A conceptual model for communications research. In J.H. Campbell and H.W. Hepler (Eds.), Dimensions in  communication: Readings. Belmont, California: Wadsworth, 1970. Wilmot, W.W. Dyadic communication: A transactional perspective. Don Mills: Addison-Wesley, 1975. Zola, I.K. Healthism and disabling medicalization. In I. I l l i c h , I.K. Zola, J. McKnight, J. Caplan, and H. Shaiken (Eds.), Disabling pro- fessions. Don Mills: Burns & McEachern Ltd., 1977. 128. APPENDIX A S. A. V. I SYSTEM FOR ANALYZING VERBAL INTERACTION (Agazarian & Simon, 1976) DEFINITIONS SQUARE ONE: ATTACK 1.1 Attack Hostile remarks, accusations and questions; swearing, name-calling and labelling. 1.2 Self-Defensive Apologising for self, putting self down, justifying or ex-cusing self. Negative questions. Taking generalizations personally. "The. tn.oa.btz with yoa Ij,..." "Get the. {>... oat ofi my way.' "Vo yoa think y o u ' r e ia.vm.y1" "Idiot." "He.'4 60 pajtanold.' "Hz'6 puAe i>quah.e one.." "1 only meant..." "I'm i>onx.y, I don't mean to InteAAupt." "1 know I'm not vexy good at this, bat..." "1 hope, yoa won't think I'm talking oat oi tu/in..." "I don't know Ifi thli, li> fieJLevant, bat I..." "I don't Aappo-be yoa'd like, to go to the movie tonight?" ["The. tuouble. wlXh people. t& they always take, things peJUonatly."} "1 don't." 1.3 Blaming and Complaining Resentful narratives, implying that people or circumstances are unfair. Scapegoating. 1.4 Sarcastic Jokes Jokes made at the expense of someone else or one's self. Gallows humor. "Ii It wasn't ion. yoa..." "look what yoa made, me do..." "1 had no choice...." "Mo matteA how hand. I tAy..." "EveAy turn they..." "It wasn't my iault..." "A^teA all I've, done...." "If i, not lain.." "I only open my mouth to change, ieet." "Vo yoa think I'm a maihuoom, keeping me. In the dank and ie.ed-Ing me battdhlt." 129. SQUARE TWO: STORY-TELLING 2.1 Gossip Telling tales about others, s ing rumors. 2.2 Reminiscences Light anecdotes of past experiences. 2.3 Funny Stories Anecdotes and jokes told for a laugh. "Have you haand tha latent about X?" "...and than I said to htm... and than ha hold to ma..." "Tha last tima I was In Eun.opa {at tha shona) [at tha movlas)..." "Whan I was a child [young] [manrtad) [youn aga)..." "Ramamban whan wa..." "Hava you haand tha ona about tha salasman who " "A iunny thing happanad to ma on tha way han.a..." SQUARE THREE: OBSCURING 3.1 Thinking Out Loud Talking in a wandering fashion; looking for the right words. 3.2 Being Obscure Intellectual izing: concealing the meaning in jargon, or unfamiliar and esoteric words. "Alphabet soup" "Professionalese" Depersonalized generaliza-tions. 3.3 Oughtitudes Dogmatic value judgments and generalizations which imply a universal "rightness". "It kind ol saams to.. .ah.. .to ba nalatad to.. .ah...pankaps tha tdaa about.. .on. do I moan shall I axpnass thu..." "Ii ona consldans tha constructs o{, ambiguity, nadundancy and con-t/iadictton within tha spactrum oi Inionmation thaony..." "Tilt out a H-11 to apply to tha A.C. oi 0. ion V.V.G. claananca through tha UCVT oiilca." "B.P. 60 ovar 110. Started I.V. at 0600 pnton. to transpont. 10-4." "It usually happans with tliam, and laavas ona coniusad about that, panticulanLy whan you always know it's that way..." "Evarybody ought to know..." "Ona always..." "It Isn't dona..." "Vou should know battan than to..." "Ho ona avan...." Definitions Page 2 130. SQUARE FOUR: PERSONAL INTRODUCTION 4.1 Personal History Information about personal history which has significance for the speaker. 4.2 Credentials Unsolicited information about "who I am" and where I f i t in the social structure: like name, rank, t i t l e , degrees, etc. 4.3 Personal Position Unsolicited "throw-away" state-ments of "where one i s " in rela-tion to the ideas being dis-cussed, -j Superficial likes and dislikes acknowledged as a personal opinion. 4.4 Personal Question Questions asking for information about meaningful personal issues. 4.5 Ritual Stereotypic, polite formalities; the language of social exchange. [VoluntzeAZd) • "I wo4 the youngest oi {ive and wo fie. my bnothen1 i> hand-me.-downt, until I was 11." "I WOA the. only one tn my family who went to collage.." (Volunteered) •• ' "1 went to X UniveAAity and studied..." "I am an hononnxy membex O f j . . . " "I have thh.e.e yean* ofi thatning -In " "I am accJtedited by..." "Now let ma *e.e...." "I'm &ZUUL confuted about that..." "Walt a moment..." "1 like, that." "I love, apple*." "1 adone. the movie*." "A/ie. you angn.y with me.?" "AJie you okay?" "Would you like to talk about what'& bothexlng you?" "Good moaning." "How one you?" "Thank you." "Won't down?" Discrimination between "Personal Position" statements and some of the other kinds of statements with which they are often confused can be made by remembering: 1) they are always unsolicited. If they are in response to a question they are called "General Answer" (Square 5); 2) they are always superficial. When invested with strong personal feeling, they are called "Personal Sharing" (Square 7); 3) they are not evaluative. Likes and dislikes that are expressed as opinion are called "Opinion" (Square 6); 4) they are casual. When given in data language (like: "this room feels cold to me.") they are a descriptive report of a subjective fact, and labelled as such in "Facts and Figures" (Square 5). Questions asking for personal "facts" are called factual or general questions. Definitions Page 3 131 . SQUARE FIVE: INFORMING 5.1 Facts and Figures Statements of factual data 1 ike quantities, citations, refer-ences, dates, times etc.-, Descriptions about the world that are based on observation and can be verified. Subjective descriptions that are reports of personal sensory experience in data langage.. 5.2 General Information Information that has a theoret-ical or research basis. 5.3 Factual Questions Direct, specific questions for facts.o [MolunteeAed] : "TheAe weAe 62 people. In thii expeA-Iment." "It -LA 5:00 by my watch." "OuA AaleA gitaph li> up thli, yeaA compared with la&t yeaA." {V0lu.nte2A.zd):' "The meeting AtaAted at 9 P.M. with aZJL but one peAhon ph.eAe.nt. The ilAbt ten minuter weAe i>pent In joking, then..." "Thli, ieeli cold to me." "That dAeMi, looki blue to me." "1 heaA mui>lc that 6oundi> good to me.." (VolunteeAed) : "EveAy meJ>£>age hoi, a peAAonal and a topic component." "The phaAeA oi gnoup development aAe illght, ilght..." "TheAe ls> a coAAelatlon between Incidents oi meai>leJ> and the. nlAe. and iall oi the Yangtze AlveA." "Vo you have the time to go to the AtoAe. thli aitennoon?" "What It, the capital oi Bolivia?" "How many?" "When?" "WheAe.?" "Why?' "How?" "What?" "Who?" Factual descriptions do not have to be true to be called factual. For example, "grasshoppers have twelve legs" is factual in that its truth or falsity can be determined by an empirical test. Only questions that seek information are called questions. Other kinds of behavior that are disguised as questions are labelled for what they are. For example, a sarcastic question is called "hostile" (Square 1); a rhetorical question is called either an assumption or a speculation (Square 6) or a description (Square 5), etc. Definitions Page 4 132. 5.4 General Questions Open questions inviting ideas.-, "Anyone. el*e. got an idea on thi*?" "How can we impfiovz thu i>yi>tem?" "What do you think we ought to do now?" SQUARE SIX: ORIENTING 6.1 Opinions Subjective opinions (sometimes stated as fact) as in: Opinion based on a subjective judgment. Interpretative opinion. Speculative opinion Assumptions Projections 6.2 Commands Ordering or directing other's behavior, thinking,etc. Orders and directions in the form of a question. "That'A a good pA.opo-itti.on; I think we. ought to implement it." "What John meant i* that they don't have, the an*wen. yet." "You'fie nexdJiy laying you'fie mad." "...then l{ we. can anAange the. outlet*, we may ven.y well top the hale* fie*ult* inom la*t "• "I know we. all want to thank Wt*. X lofi hen. time...." "1 know 'that you won't back me up." "Vo tku." "Vo that." "Stop tt." "Come hen.z." "Vo the iin*t, and then get the mail out." "Will you plea*e, take thi* lettex?" Only questions that seek information are called questions. Other kinds of behavior that are disguised as questions are labelled for what they are. For example, a sarcastic question is called "hostile" (Square 1); a rhetorical question is called either an assumption or a speculation (Square 6) or a description (Square 5), etc. Definitions Page 5 133. 6.3 Proposals Suggestions for orienting ways to work or things for people to do. Propositions in the form of questions. 6.4 Encouragement Praise, agreement, reinforcing opinions, and other expressions that encourage someone to con-tinue saying what they're saying. 6.5 Discouragement Negations, disagreement, dis-couraging opinions and other expressions that discourage people from continuing what they are saying. Discouragements in the guise of a question. 6.6 Yes-But's A token agreement introducing a disagreement or change of subject or change of agenda. "I ptiopoAz that wz ondzn. oun. agenda pntonitizA." "I think wz Ahould A tant doing thiA now." "Vzn.ha.pA iJt could bz donz thiA way." "Why don't you. do i t thiA way?" "Right..." "I agfizz..." "Okay..." "That'A a good Idza..." "Uh huh..." "Good ginZ..." "Go on..." "Vzah, I know what you mzan..." "No." "...I di^agnzz..." "That'A no good." "Vou'nz wftong." "I don't think that would wonk." "Vou don't nzalty mzan that, do you?" "you an.zn't AzhtouA?" "VZA, but you havzn't takzn into account my ^iXAt point." "VZA, but don't you nzally think..." "On thz onz hand, you could Aay that IA AO, but on thz othzn. hand..." SQUARE SEVEN: INTERPERSONAL RELATING 7.1 Personal Answers Responding to a personal, factual or general question with infor-mation about oneself that is obviously personal, important and meaningful. [In fiZAponAz to a quzAtlon) •• "VZA, I am Ifvightznzd at mzzting pzoplz." [In nzAponAz to a quzAtlon) •• "II you nzally want to know, I'm Attll vzty angny with you." Definitions Page 6 134. 7.2 Personal Sharing Unsolicited important and meaning-ful personal information that is obviously invested with personal feeling. 7.3 Personal Support Paraphrasing or reflecting the personally meaningful feeling part of what someone else said (not interpreting "what you think they really meant!"). Remarks that reflect how the listener feels. Supporting a person's right to speak (even i f you disagree) by taking them, and their feelings, seriously. 7.4 Taking a Stand Non-defensive, non-aggressive, clear, neutrally toned, descrip-tive statements about what one will or will not do. 7.5 Affectionate Jokes Warm, interpersonal humor that increases good feelings between people and increases intimacy. (VoluntzzAzd) : "J am vzxy hnlghtznzd at meeting pzoplz." "I'm a^hatd theAe'A Aomzthlng WAong with mo,." "1 {,zzl vzxy good about uA." "1 love you." [In AzAponAz to an.oth.QJi'& input) : "you Izzl that whatzvex you do, nothing will make, a dlfihzxzncz." [In AzAponAz to a peXAon) : "Right now, you fazzl vzxy angAy." "X, you axz tAylng to get hzaxd, then?" "I'd tike to Atand ho A K ' A Atght to get hzaxd, whetheA we. think Ahz'A Alght OA not." "1 do not Intend to buy youA magazlneA ." "My anAwen tA no." "I witl not continue, to do thiA thiA way." [Said fondly) •• "That'A my wlfie, {OA you -- Ahe'tl atwayA take In AtxayA. We have. AO many animalA, the. zoo zatlA uA when thzy Aun AhoAt." SQUARE EIGHT: SUMMARIZING AND RESPONDING 8.1 Factual Answer Direct specific answers to either factual or general questions that give factual or personal data relevant to the question. Responses to questions that require a right or wrong answer. [In AeAponAe to a question) : "VZA." "No." "I'm leaving at 8:30 tonight." "1 Like, thz Aed onz bzttzA." "A AquaAz hoA {QUA, 90° anglzA." "1776." Definitions Page 7 135. 8.2 General Answers General or indirect answers to factual or general questions con-taining information, judgments, opinions, speculations, or theory relevant to the question. 8.3 Clarifying Issues Expanding on an answer to a ques-tion by giving descriptive definitions or information that c l a r i f i e s ambiguities or con-tradictions in the answer.^ 8.4 Paraphrasing Accurately repeating or reflect-ing the content of something that has just been said. 8.5 Summarizing Restatement of what has been said or done within a meeting by making a synthesizing summary of the con-tent of the previous discussion. SQUARE NINE: 9.1 Building on Ideas To build on, or add to, another's ideas (has a creative, escalat-ing quality). [In response to either a factual on general question) •• "Well, on the. one hand, I think If the. South had won the. Civil War, the. economics of cotton production would have. caused..." "I meant John Kennedy, not Bobble.." [As a paraphrase oi what someone has been saying just before, you speak): "You're, saying that no matter how many hours you put In, you can't finish IX." "To Summarize.: John filled us in on the, fiscal end, and suggested a 5% cut; Ge.orge outlined the pro-duct modifications and we are now trying to assess how these two things will Influence the market-ing strategy." "If we follow John's Idea, we could chart the progress..." "... and we could blow the chart up Into a wall poster..." "...then the whole staff could get the picture..." "And we could put It In the canteen." INTERACTIVE BUILDING 9.2 Work Jokes "In-group" jokes, puns and plays "Well, we're okay on chiefs, now on words that are relevant to the where do we get some Indians?" work being done. Ho remain under the label "clarifying issues," the c l a r i f i c a t i o n must be directly related to the answer. Any new information or change of subject is labeled under its contingent label, like "opinion" or "description", etc. Definitions Page 8 136. APPENDIX B  DEMOGRAPHIC FORM Name of child Name of parent N _ F _ A _ Address Sex Dob Education Frequency of interaction with this paraprofessional Frequency of interaction with this physician How long have you known this paraprofessional How long have you known this physician How long have you been a parent Kind of appointment Special Regular Disability of child Age of child Name of paraprofessional Sex Dob Education Time practising role Type of specialty Date of Interview Name of physician Sex Dob Education Time practising role Type of specialty Date of Interview Center involved Clinic involved 137 APPENDIX B  CONSENT FORM I hereby grant permission to Joyce MacKinnon, B.O.T., M.A., to make audiotape recordings of interview/s involving me and to use the audio-tape/s for research purposes. There will be no mention or record of my name made public. Only code numbers will be used. I understand that I am entitled to refuse to have the interview/s audio-taped and i f I am a parent of a handicapped child that-such refusal would in no way affect the treatment my child receives here. Signed Role Signed (Researcher) CODING SYMBOLS: PERSONAL FACTUAL ORIENTING 1. Attack Self Defense Blame or Com-plain Sarcastic Jokes At. SD B/C SJ 2. Gossip Reminiscing Funny Story G Rem FS 3. Thinking Out Loud Obscuring Oughtitudes TOL Obs Ou 4. Personal History Credentials Personal Position Personal Question P.His. CR P.Pos. PQ 5. Facts & Figures General Information Fact Question General Question : FF Gl FQ GQ 6. Opinion Interpretation Speculation Assumption Projection Encouragement/ Agreement Discouragement/ Disagreement Yes/But Commands Proposals Quiet Noise Laughter Fragment Op Int Spec Assump Proj. Enc Dis Y/B Com Prop Q N L F 7. Personal Answers Personal Share Personal Support Taking A Stand Affectionate Joke PA P.Sh. P. Sup. TAS AJ 8. Fact Answer General Answer Clarifying Issues Paraphrasing Summarizing FA GA CI Para Sum 9. Build on Ideas Work Joke B WJ 139. APPENDIX D TALLY SHEET o CQ c OJ S-rC Q-S-<D s •i— > OJ +J OJ </•> CT) •r— +-> OJ S-o CD 00 CD-CO O ro CO s-OJ E =3 OJ (O Dill C •r— T3 O O cc: CL to o 00 S-OJ OJ o E -o O > OJ o o o ro O S-CL Q_ APPENDIX E SAVI THEORETICAL COMMUNICATION PATTERNS Defensive Pattern of Communication Personal Factual Orienting Avoidance H M H Contingent L H M Approach L M L Problem-solving Pattern of Communication Personal Factual Orienting Avoidance L L L Contingent M H H Approach M M M Cross-purpose Pattern of Communication Personal Factual Orienting Avoidance L M L Contingent L H H Approach L M L Relative Frequencies: L = Low ( < 7%) M = Medium ( < 7 < 20%) H = High ( <_ 20%) 141 . APPENDIX E COMPUTATION FOR DETERMINING THE PATTERN OF COMMUNICATION BETWEEN SAVI THEORETICAL AND OBSERVED PATTERNS OF COMMUNICATION 1. Raw Data: Personal Factual Orienting Total Avoidance 22 201 33 256 Contingent 54 739 1485 2278 Approach 0 173 100 273 76 1113 1618 2807 2. Convert to Percent: Personal Factual Orienting Total Avoidance 1% 7% 1% 9% Contingent 2% 26% 53% 81% Approach 0% 6% 4% 10% 3% 39% 58% 100% 3. Apply Rule to Above Percent: P F 0 Low = < 7 A L M L Medium = > 7 < 20 C L H H High = > 20 Ap L L L 4. Compare Observed with Theoretical: Observed Problem-solving Cross-purpose P F 0 P F 0 P F 0 A L M L L L L. L M L C L H H M H H L H H Ap L L L M M M L M L 5. Do a Rank Ordering (1 - 9) with ties: Observed: 1 2 3 4 5 6 7 8 9 L M H 3.5 •7 8.5 (Values) Problem-solving: 1 2 L 2 5 6 M 5 8 9 H 8 (Values) Cross-purpose: 1 2 3 4 L 3 6 7 M 6.5 8 9 H 8.5 (Values) 6. Do Spearman Rank Correlation Coefficient between: Observed Values and Problem-solving Theoretical Values Observed Values and Cross-purpose Theoretical Values 7. Highest Correlation represents the pattern of communication. Source: Dr.Malcolm Grieg, Computing Centre, University of British Columbia APPENDIX F - 1 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL'AND OBSERVED PATTERNS OF CROSS-PURPOSE AND PROBLEM-SOLVING COMMUNICATION OF PARENTS AND PROFESSIONALS Parents Professionals Cross- Problem- Cross- Problem-Purpose Sol ving Purpose Sol ving Tape 10 .55 .45 .88 .83 12 .86 .75 .89 .83 14 .80 .64 .89 .83 16 .70 .60 .89 .83 18 .86 .75 .89 .83 20 .80 (. 63) .85* (.70) .70 .60 22 .86 .75 .89 .83 24 .89 .79 .89 .83 26 .85 .80 .89 .83 28 .55 .44 .55 .47 30 .70 .60 .85 .80 32 .68 .48 .89 .83 34 .73 .55 .80 .64 36 .85 .73 .71 (.63) .83* (.71) 38 .86 .75 .89 .83 40 .90 .83 .80 (.74) .85* (.87) 42 .86 .75 .89 .83 44 .86 .75 .89 .83 46 .85 .73 .80 (.74) .85* (.87) 48 .32 (. 37) .37* (.53) .80 (.74) .85* (.86) 50 .85 .73 .90 .83 52 .80 (. 77) .81* (.90) .89 .79 54 .70 .60 .72 (.76) .80* (.85) 56 .89 .79 .90 .83 58 .85 .73 .74 (.73) .81* (.85) 60 .89 .79 .80 (.74) .85* (.87) 62 .59 (. 51) .68* (.50) .89 .79 64 .70 .60 .85 .82 66 .89 .79 .70 .60 68 .74 (. 67) .81* (.85) .90 .83 70 .89 .83 .89 .83 72 .81 .75 .89 .83 74 .86 .75 .89 .83 76 .92 .81 .90 .83 78 .90 .83 .89 .79 80 .89 .83 .81 .75 82 .60 .41 .71 .50 * indicates problem-solving communication. Numbers in brackets indicate correlation coefficients using adapted theoretical models. APPENDIX F - 2 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PATTERNS OF CROSS-PURPOSE AND PROBLEM-SOLVING COMMUNICATION OF PARENTS AND PARAPROFESSIONALS Parents . Paraprofessionals Cross-Purpose Problem-Solving Cross-Purpose Problem-Sol ving Tape 11 .23 ( .19) .48* (.59) .80 (.74) .83* (.87) 13 .40 ( .33) .58* (.69) .89 .83 15 .68 .55 .89 .83 17 .86 .75 .90 .83 19 .85 .73 .70 .60 21 .55 .47 .85 .80 23 .86 .75 .85 .82 25 .70 .60 .90 .83 27 .89 .79 .85 .82 29 .68 .57 .85 .82 31 .86 .75 .89 .83 33 .68 .57 .89 .83 35 .73 ( 74) .75* (.87) .80 (.74) .85* (.86) 37 .68 .57 .86 .75 39 .86 .75 .89 .83 41 .90 .83 .89 .83 43 .68 .52 .89 .83 45 .85 .73 .89 .83 47 .86 .75 .90 .83 49 .89 .79 .89 .83 51 .80 ( 74) .85* (.87) .89 .79 53 .80 .65 .89 .83 55 .89 .79 .89 .83 57 .86 .75 .70 .60 59 .89 77) .79 .80 (.74) .85* (.87) 61 .80 (. .81* (.90) .90 .83 63 .85 .73 .70 .60 65 .86 .75 .90 .83 67 .90 .83 .92 .80 69 .85 .73 .83 .71 71 .90 56) .83 .57 (.48) .75* (.63) 73 .63 (. .79* (.66) .89 .83 75 .68 (. 72) .81* (.80) .57 (.49) .75* (.63) 77 .89 .79 .89 .79 79 .81 .75 .89 .83 81 .89 .79 .90 .83 83 .63 .53 .89 .83 * indicates problem-solving communication. ( ) indicates correlation coefficients using adapted theoretical models. APPENDIX F - 3 SPEARMAN RANK CORRELATION COEFFICIENTS BETWEEN THEORETICAL AND OBSERVED PATTERNS OF CROSS-PURPOSE AND PROBLEM-SOLVING COMMUNICATION OF THE ADDITIONAL GROUP OF PARENTS AND PARAPROFESSIONALS Parents . Paraprofessional s Cross-Purpose Problem-Sol ving Cross-Purpose Problem-Sol ving Tape 100 .89 .79 .89 .83 101 .90 .83 .80 (.58) . 8 5 * ( .66) 102 .89 .83 .89 .83 103 .90 .83 .89 .83 104 .89 .83 .89 .83 105 .92 .80 .89 .83 106 .89 .83 .89 .83 107 .89 .83 .89 .83 108 .89 .83 .89 .83 109 .60 .59 .85 .82 n o .90 .83 .89 .83 111 .66 (. 51) . 8 0 * ( .52) .89 .83 112 .89 .79 .90 .83 113 .70 .60 .89 .83 114 .86 .75 .89 .83 115 .80 (. 74) . 8 5 * ( .86) .89 .83 116 .90 .83 .90 .83 117 .98 .52 .74 .58 * indicates problem-solving communication. ( ) indicates correlation coefficients using adapted theoretical models. 146. APPENDIX G PERCENTAGES OF VERBAL MESSAGES AND BEHAVIOURS USED BY PARENTS, PROFESSIONALS AND PARAPROFESSIONALS Parents Personal Factual Orienting Avoidance 2.94 0.15 0.32 Contingent 3.03 23.11 25.06 Approach 1.67 43.70 0.02 7.64 66.96 25.40 Professionals Personal Factual Orienting Avoidance 0.65 0.04 0.49 Contingent 3.90 45.39 41.24 Approach 0.24 8.05 0.00 -4.79 53.48 41.73 Paraprofessionals Personal Factual Orienting Avoidance 0.00 0.00 0.26 Contingent 4.42 41.15 46.78 Approach 0.39 6.96 0.04 4.81 48.11 47.08 Relative Frequencies: L = Low ( < 7%) M = Medium ( < 7 < 20%) H = High (<;'20%) PUBLICATIONS: MacKinnon, J.R. "Role behaviours and communication profiles." Paper presented at the National Conference of the Canadian Association of Occupational Therapists, Edmonton, Alberta, June 1978. Division of Continuing Nursing Education. Rehabilitation  of Persons with Chronic Obstructive Lung Disease: An  instructional package for health professionals. School of Nursing, University of British Columbia, July, 1977 (major contributor to this educational package). MacKinnon, J., Sanderson, E., and Buchanan, J. The MacKinnon Splint. Canadian Journal of Occupational Therapy, 1 9 7 5 , ^ 2 ( * 0 , 157 -158 MacKinnon, J. Attitudes of mothers of handicapped children. Canadian Journal of Occupational Therapy. 1 9 7 2 , 3 9 ( 1 ) . 2 5 - 3 3 . 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items