Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Observation of communication behavior : the development of a research method for use in health care organizations McGill, Mary Elizabeth 1976

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

Download

Media
831-UBC_1976_A1 M23.pdf [ 9.71MB ]
Metadata
JSON: 831-1.0093937.json
JSON-LD: 831-1.0093937-ld.json
RDF/XML (Pretty): 831-1.0093937-rdf.xml
RDF/JSON: 831-1.0093937-rdf.json
Turtle: 831-1.0093937-turtle.txt
N-Triples: 831-1.0093937-rdf-ntriples.txt
Original Record: 831-1.0093937-source.json
Full Text
831-1.0093937-fulltext.txt
Citation
831-1.0093937.ris

Full Text

2ISERVATI0N_0F_CgMMUOQATI for ase i n liMLTH_CARE_ORGANIZilIOaS by Mary Elizabeth McGill Bachelor of Science in Rehabilitation, U.B.C., 1965 Submitted i n p a r t i a l fulfilment of the reguireaents for the degree of DOCTOR OF PHILOSOPHY in THE FACULTY OF COMMERCE AND BUSINESS ADMINISTRATION He accept t h i s thesis as conforming to the required standard: THE UNIVERSITY OF BRITISH COLUMBIA August, 1976 ($) Mary Elizabeth McGill, 1976 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Mary Elizabeth McGlll Department o f Commerce and. Business Administration The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date August 9, 1976 ABSTRACT The objective of the Dissertation was to validate Sobert Bales' Interaction Process Analysis, and to adapt the method for analysis of s t a f f and patient communications on a hospital ward. a comprehensive review of the group, organizational, and health care l i t e r a t u r e emphasized the lack of instruments for observing actual communication behavior i n t h i s organizational context. The study therefore focused on r e f i n i n g the research method, rather than on analyzing s p e c i f i c interaction patterns. A four-month f i e l d study was conducted i n a 300-bed r e h a b i l i t a t i o n and extended care hospital i n Southern Saskatchewan. Two wards were selected for study—an adult physical r e h a b i l i t a t i o n unit and a chronic s e t t i n g . Subjects included a l l patients on the two wards, t h e i r f a m i l i e s , and a l l s t a f f involved i n d i r e c t or i n d i r e c t provision of care, A team of two non-participant observers coded dyadic communication behavior in-process on the wards. Hypotheses were tested regarding the r e l a t i v e iaportance of organizational position and context in explaining the variance of the interaction p r o f i l e s . Support was also sought for the construct and face v a l i d i t y of the instrument. The Dissertation included d e t a i l s of the atetnod and the t r a i n i n g required to enable other researchers to employ the technique. Procedures for providing feedback to the subjects were also delineated, as active p a r t i c i p a t i o n was considered c r u c i a l to the project's success. P a r t i a l support was obtained for two aspects of the construct v a l i d i t y of Interaction Process A n a l y s i s — t h e inclusivenass of the categorization and the independence of the category frequencies. Evidence of face v a l i d i t y was also given. A multivariate analysis of variance demonstrated that the position dyad was the most s i g n i f i c a n t explanatory variable for the interaction p r o f i l e , but the particular observer was also an important f a c t o r . The impact of organizational context varied with tae dimension-Alternative explanations were given for the observer e f f e c t , and directions explored for future research. The study concluded that Bales* Interaction Process Analysis has a d e f i n i t e contribution to make to tne study of behavior in health care organizations. Vance F. M i t c h e l l , Thesis Supervisor i v TABLE OP CONTENTS CHAPTER I-A. INTRODUCTION B. RATIONALE C. BACKGROUND D. OBJECTIVE E. LITERATURE BEVIES 1. Role Theory 2. The Communication Process 3. Communication Studies i n Health Care 4. Bales' Interaction Process Analysis F. DEVELOPMENT OF HYPOTHESES CHAPTER II A. LOCATION 44 B. SDBJECTS 46 C. RESEARCH MODEL 47 D. STUDY DESIGN 51 E. METHOD 66 F. DATA PREPARATION 72 1 2 6 8 9 9 14 17 23 42 CHAPTER_III A. PRELIMINARY ASSESSMENT OF CATEGORY USE 79 B. RELIABILITY 81 C. TESTS OF THE HYPOTHESES 85 D. HOSPITAL FEEDBACK 99 CHAPTER I? A. DISCUSSION 104 1. Theory 104 2- Method 106 3. The Research Model 115 4. U t i l i t y of the Method 116 B. FUTURE RESEARCH DIRECTIONS 119 C. CONCLUSIONS 121 BIBLIOGRAPHY 124 APPENDICES A. . Results of P i l o t Study 135 B. Staff and Patient Codes 138 C. Submission to SSHC Management Committee 142 D. Presentation to Department Heads 151 E. Staff and Patient Orientation Sheet 156 F. Communication P r o f i l e s Presented to 158 Management Committee G- Feedback Information Sheet 170 H. F i n a l Report on Communications 172 Research Project v i LIST OF TABLES I STABILITY OF SUBJECTS1 INTERACTION BEHAVIOR 28 BY CATEGORY II BALES' NEW IPA—ESTIMATED NORMS AND 39 DIRECTIONAL INDICATORS III STRENGTH OF PHI VALUES FOR CATEGORY USE 80 BY OBSERVER IV VARIMAX ROT&TED FACTOR MATRIX—BOTH HARDS 87 V VARIMAX ROTATED FACTOR MATRIX—SECOND WEST 87 VI VARIMAX ROTATED FACTOR MATRIX—STATION FIVE 88 VII OBLIQUE FACTOR STRUCTURE—BOTH WARDS 88 VIII OBLIQUE FACTOR STRUCTURE—SECOND WEST 89 IX OBLIQUE FACTOR STRUCTURE—STATION FIVE 89 X EIGENVALUE, MAJOR VARIABLE, AND PERCENTAGE OF 90 VARIANCE EXPLAINED FOR EACH FACTOR XI NUMBER AND DIRECTION OF CO-LOADINGS FOR 92 CATEGORY 6 XII ABSOLUTE AND RELATIVE FREQUENCIES OF 93 CATEGORY 0 BY WEEK XIII MANOVA—SIGNIFICANT F-STATISTICS FOR WEEK 95 XIV MANOVA—SIGNIFICANT F-STAIISTICS FOR WARD 95 XV MANOVA—SIGNIFICANT F-STATISTICS FOR 96 DAY, OBSERVER, TIME PERIOD, LOCUS XVI MANOVA—SIGNIFICANT F-STATISTICS FOR 97 DAY, OBSERVER, DYAD v i i LIST OF FIGURES 1. A MODEL OF THE ORGANIZATIONAL POSITION 12 2. BALES' SYSTEM OF CATEGORIES USED IH OBSERVATION 25 AND THEIR MAJOR RELATIONS 3. BASCANA HOSPITAL 45 4. RESEARCH MODEL 47 5. TIMING OF FIELD STUDY 52 6. . SAMPLE PROFILE FROM HOSPITAL FEEDBACK 78 7. PHASES OF RESEARCH AND TARGET DATES FOR 146 COMPLETION v i i i ACKNOWLEDGMENTS No man i s an i s l a n d . No man stands a l o n e . I t i s hard t o know where t o begin or end when making acknowledgments f o r a D o c t o r a l D i s s e r t a t i o n . So many people's l i v e s have touched mine along the way--to mention them a l l i s i m p o s s i b l e . However, as I have moved along my ca r e e r path, c e r t a i n i n d i v i d u a l s have provided t h a t " e x t r a " impetus which should be r e c o g n i z e d a t t h i s time. Dr, Mary Pack encouraged me t o enter the p r o f e s s i o n s of O c c u p a t i o n a l and P h y s i c a l Therapy. Dr. Brock F a h r n i and the f a c u l t y o f the D.B.C. , School of R e h a b i l i t a t i o n Medicine s t i m u l a t e d c o n s i d e r a t i o n of the t o t a l p a t i e n t and questioned e x i s t i n g p a t t e r n s of c a r e . , Dr. George Szasz broadened my h o r i z o n s t o other dimensions o f the h e a l t h care team, and Robert McDermit c h a n n e l l e d t h i s i n t e r e s t i n t o communication r e s e a r c h . Between 1967 and 1972, teamwork with s t a f f and p a t i e n t s a t Hascana H o s p i t a l caused me t o r e t h i n k my c a r e e r o b j e c t i v e s , and s t i m u l a t e d my r e t u r n to u n i v e r s i t y i n search of a broader understanding of o r g a n i z a t i o n a l behavior and h e a l t h care r e s e a r c h . The past f o u r years of study a t U.B.C. were c r e a t i v e because of the v a r i e t y of pe r s o n a l c o n t a c t s , p a r t i c u l a r l y the f a c u l t y and graduate s t u d e n t s i n the O r g a n i z a t i o n a l Behavior D i v i s i o n of the F a c u l t y o f Commerce and Business A d m i n i s t r a t i o n , and members of the i x Department of Health Care and Epidemiology of the Faculty of Medicine. Through t h e i r contributions, new pathways have been explored and a multitude of questions raised f o r the future. Dissertations require many kinds of support. . Throughout the development and execution of the t h e s i s , my Committee—Dr. Vance M i t c h e l l , chairman. Dr. Carl Sarndal, and Dr., Dean Dyeno of the Faculty of Commerce, Dr. Anne Crichton from Health Care and Epidemiology, and Dr., Tom Mallinson, of the Department of Communication Studies at Simon Fraser University—have a l l played a v i t a l r o l e . For the past two years, I have been fortunate to receive f i n a n c i a l support from a Canada Council Doctoral Fellowship. Dr. Fred S i l l e r , of the U.B.C. Faculty of Commerce, f a c i l i t a t e d a Department of Labor student grant to a s s i s t i n the f i e l d study, and Dr. . Jack Boan, from the University of Regina Department of Economics, obtained computer funding for interim data analysis. Excellent research assistants are a rare coaaodity. I have been fortunate to be associated with two of the f i n e s t — M a r y Brown for the p i l o t study, and Peter Bowman for the Dissertation research. Both made invaluable contributions to the research program. The s t a f f and patients at Hascana Hospital i n the Summer of 1975 deserve s p e c i a l mention. l e worked together as a team to tackle a shared concern for the delivery of integrated patient care. X A Doctoral program cannot be undertaken alone. I t requires constant encouragement. , Without the support of parents and f r i e n d s , I would not have attempted to follow t h i s path. And without their caring presence,the t e r r a i n would have been impassable. We have shared Baany moments, good and bad, as we have traversed our paths together. Hy new family have provided the motivation when the going was the steepest. I f t h i s research i s to be dedicated to anyone i t must be to them, for they are the future. The climb up the mountain has been envigorating. , From t h i s viewpoint the horizon i s broad and the ranges beyond are f u l l of promise for e x c i t i n g discoveries. I t i s my sincerest wish that those whose past contributions have made the present a r e a l i t y w i l l be able to share with me i n exploring the future t e r r a i n . Betty McGill Vancouver, B r i t i s h Columbia June, 1976 1 CHAPTER_I A. INTRODUCTION! Development and validation of a method f o r observing and analyzing communication behavior of s t a f f and patients on a hospital ward are the objectives of t h i s d i s s e r t a t i o n . The study begins with an examination of the problems of funct i o n a l s p e c i a l i z a t i o n and coordination facing today's health care organizations. Communication i s one aethod of achieving horizontal coordination between members of the treatment team, but instruments are not avail a b l e for observing actual i n t e r a c t i o n behavior i n the organizational s e t t i n g . Their potential contribution to an understanding of coordination i n health care organizations i s recognized, and thus the rati o n a l e for the the s i s i s developed. Four key areas are emphasized i n the l i t e r a t u r e r e v i e w — r o l e theory, the communication process, communication studies i n health care, and observational methods. Sobert Bales* Interact4Qn_Process Analysis (IPA) i s seen as the most promising approach to studying communication behavior i n health care organizations. However, t h e o r e t i c a l validation and methodological refinements of t h i s instrument are reguired before s p e c i f i c i n t e r a c t i o n patterns can be interpreted. 2 Eight hypotheses concerning the construct v a l i d i t y of IPA and i t s use on the hospital ward are generated from issues raised i n the l i t e r a t u r e and the r e s u l t s of a p i l o t study conducted by the researcher. Design of the present research follows l o g i c a l l y from the foregoing analysis. Location and sample for the four-month f i e l d study are described, the research model delineated, and variables operationally defined. Developmental aspects of the research are discussed before the f i n a l data c o l l e c t i o n i s described. After summarizing the r e s u l t s of the study, an outline i s given of the process for feedback to the p a r t i c i p a t i n g organization. The functional u t i l i t y of the method i s thereby assessed. In discussing the findings, i a p l i c a t i o n s for communication theory, observational studies, and organizational research are also examined. . The investigator's long-term goal i s to develop a set of methods which could be used by researchers and s t a f f i n health care organizations to observe t h e i r communication behavior and analyze i t s e f f i c i e n c y and effectiveness. B. , RATIONALE: "As medical practise has become more specialized, i t has become more impersonal .... I t has also become more interdependent .... Since s p e c i a l i z a t i o n creates i n t e r -dependence, i t creates the need for coordination (Thompson, 1961, p. 48,50)." 3 One of the major dilemmas confronting organizations i n the 1970's i s the need for functional s p e c i a l i z a t i o n as opposed to coordination. This problem was discussed by many early management t h e o r i s t s . , Fayol (1949, p* 20) recognized that s p e c i a l i z a t i o n of functions was the natural r e s u l t of the d i v i s i o n of work duties advocated by Taylor. Gulick (1937, p. 6) emphasized that the subdivision of work led inevitably to the necessity for coordination. Recent analyses by Lawrence and Lorsch (1967) and H a l l (1972) support the interdependence of these two dimensions. Organizations have several alternative mechanisms fo r achieving coordination. V e r t i c a l coordination, as discussed by flooney(1937) and Gulick (1937), i s generally handled either through s t r u c t u r a l approaches such as organizational design or through the use of formal authority and power. Horizontal coordination i s achieved by the development and exchange of ideas, frequently referred to as "communication**. According to Roberts, O'Reilly, Bretton, and Porter (1974), "Communication seems to t i e organizations together. Consequently, the r e l a t i o n s h i p of various facets of communication to other processes such as coordination and decision-making appears to us to have high pay-offs i n terms of understanding organizational behavior (p. 520) ." However, "communication" as an organizational process has received l i t t l e attention from theorists and empirical researchers (Porter & Roberts, 1972). 4 Because of the exponential growth of medical knowledge during t h i s century, providers of d i r e c t or i n d i r e c t patient care have been forced to s p e c i a l i z e . Health care organizational structures r e f l e c t t h i s development. Total units such as treatment centres or hospitals are frequently separated according to age of patient (pediatric, adolescent, adult, g e r i a t r i c ) , l e v e l of care (acute, r e h a b i l i t a t i o n , a c t i v a t i o n , chronic), and/or disease c l a s s i f i c a t i o n (e.g., a r t h r i t i s , p sychiatric, cancer). Within a l l of the above types of i n s t i t u t i o n s , s p e c i a l i z a t i o n occurs by function (nursing, housekeeping, medical s t a f f ) . Functional s p e c i a l i z a t i o n i s thus a key dimension of t o d a y s health care organizations. Such s p e c i a l i z a t i o n , however, necessitates coordination. The "team approach to patient care" has evolved as a method f o r coordinating these highly s p e c i a l i z e d d i s c i p l i n e s . Considerable disagreement exists over conceptual and operational d e f i n i t i o n s of the "health team". Hany authors have assumed i t to be an organizational objective; a number have attempted to operationalize the concept; a few believe that they have succeeded 1. S i g n i f i c a n t l y , a recent study by Crichton (1975) found that s t a f f in r e l a t i v e l y well-defined units were unable to 1 For a thorough review of t h i s l i t e r a t u r e , see Warner's (1975) annotated bibliography of health care teamwork (p. 1-119). 5 est a b l i s h the boundaries of t h e i r team. Hessen's (1958) analysis of "Hospital ideology and communication between ward personnel" determined that health personnel l i m i t e d t h e i r volume of inter-departmental i n t e r a c t i o n . They channeled the majority of t h e i r communications within occupational l i n e s . Georgopoulos and Hann (1962) reported s i m i l a r findings i n t h e i r study of perceived patterns of information exchange in community general hospitals. D i f f e r i n g behavior patterns of health s p e c i a l i s t s have been hypothesized to create barriers to coordination of patient care, but few empirical attempts have been made to explore t h i s dimension. Researchers have studied such organizational variables as climate and leadership s t y l e s as fierceived by members. Less attention has been paid to developing methods for recording actual behavior i n s i t u . Dubin (1962), i n "Business behavior behaviorally- viewed", presented many of the early studies which explored t h i s dimension. He stated, " I f behavioral science means what i t says, then behavior and the si t u a t i o n s of behavior must be the f i r s t l e v e l of analysis.... Perhaps i t would be productive of immediate and very substantial advances i f good behavioral s c i e n t i s t s rediscovered behavior as t h e i r c e n t r a l datum (p. 11,51)." Kerlinger (1966, p. .503-525), and S e l l t i z , Jahoda, Deutsch, and Cook (1959, p.199-234), have provided a foundation f o r students of the observational method. However, i n spite of the a v a i l a b i l i t y of such references, few recent empirical studies have used observational methods 6 to study organizations. This may be because such investigations often take the researcher into methodologically muddy water. The large number of uncontrollable variables i n the organizational context makes behavioral studies prone to systematic error variance (Thorndike, 1949, p. 75). Measurement of communication behavior i s a p a r t i c u l a r l y challenging task because of the absence of a t h e o r e t i c a l foundation on which to bui l d empirical investigations. The lack of instrumentation can be partly attributed to the dynamic aspects of the process {Roberts S 0*Reilly, 1974b). The amorphous nature of the concept of communication i s also a contributing f a c t o r (Roberts & 0*Reilly, 1974a). In summary, to understand the process of coordinating specialized departments i n health care organizations, more comprehensive and r e l i a b l e technigues for measuring fkctuaj. communication behavior of s t a f f and patiants must be developed. Only i n t h i s way can subjects* perceptions of inte r a c t i o n be combined with th e i r observed behavior patterns to give a comprehensive picture of the t o t a l organizational communication system. C. BACKGROUND: Between 1967 and 1969 the researcher conducted a p i l o t study of "Staff communication i n a r e h a b i l i t a t i o n h o s p i t a l . 7 and i t s r e l a t i o n to integration of patient care" (HcGill, 1968) at Hascana Hospital, a 300-bed r e h a b i l i t a t i o n and chronic-care f a c i l i t y i n Eegina, Saskatchewan. The f i r s t year was spent attempting to develop a method for content coding of verbal i n t e r a c t i o n s . During t h i s period, i t became increasingly apparent that the c r i t i c a l behavioral dimension was how s t a f f and patients interacted, rather than what they said, since the content of t h e i r communication was primarily determined by the nature of the problem at hand. E f f o r t s were then focused on developing a set of categories for "how" people communicated. As the r e s u l t of an extensive l i t e r a t u r e search and dialogue with other researchers, the decision was made that Robert Bales' Interaction Process Analysis (IPA) (1950a) was the most appropriate instrument for the s i t u a t i o n . , There were several reasons f o r t h i s choice. The f i r s t was the desire to use an e x i s t i n g instrument which was at l e a s t p a r t i a l l y validated rather than to develop a new one. The second was the decision to monitor the process of communication rather than i t s content. This narrowed the alt e r n a t i v e s to three—Flanders' (1970) Interaction Analysis, Chappie's (1949) Interaction Chronograph, and Bales' (1950a) Interaction Process Analysis. Flanders' was eliminated because the categories would have required extensive modification for use i n a hospital s e t t i n g . Of the other two, Bales' categorization was considered to have a more comprehensive t h e o r e t i c a l foundation. 8 Bales 1 method, which was developed for the small group laboratory, reguired modification of data c o l l e c t i o n procedures before i t could be used i n the organizational context. Appropriate changes were made, and a p i l o t study was attempted on a l l s i x hospital sards. Data from two wards were not useable because of the age of patient (children) or the informality of the ward atmosphere (veterans). Analysis of interaction patterns on the other four wards indicated that the modified fora of IPA could i s o l a t e communication p r o f i l e s f o r s p e c i f i c occupational groups at a p a r t i c u l a r ward l o c a t i o n . Findings also offered some support for the construct v a l i d i t y of the instrument (see Appendix A). However, because of d i f f i c u l t i e s i n computerizing the p i l o t study data, further v a l i d a t i o n was reguired before the instrument could be comprehensively assessed. D. OBJECTIVE^. The objective of the research was to determine whether Interaction Process Analysis, a general set of communication process categories, can give r e l i a b l e and v a l i d i n t e r a c t i o n p r o f i l e s across a broad spectrum of s t a f f and patient positions i n a health care organization. , This goal was to be accomplished by: 9 (a) r e f i n i n g the method of data c o l l e c t i o n for use on a hospital (b) assessing the construct . v a l i d i t y of Bales* communication categorization. (c) developing and documenting a procedure for t r a i n i n g health care personnel to use the instrument. (d) examining the u t i l i t y of the data for health care administrators and p r a c t i t i o n e r s . E- LITERATURE SEVIIII: The present research evolved to f i l l gaps i n the t h e o r e t i c a l and empirical l i t e r a t u r e on organizational communication. Since an integrated approach to the problem of measurement did not e x i s t , four areas were examined which had a relevant contribution to make: role theory, the communication process, communication studies i n health care, and Bales' Interaction Process analysis. The f i r s t three are discussed b r i e f l y ; emphasis i s given to the fourth, as i t provided the major methodological foundation f o r the study. 1. Role_Theory! The term " r o l e " has d i f f e r e n t connotations across the s o c i a l sciences (Gordon, 1966). I t s psychological usage i s conceptually broader—a role i s a set of behaviors exhibited by a l l incumbents of a p a r t i c u l a r s o c i a l position--e.g. the " r o l e " of the nurse. Sociologists generally view a role as one subset o f consistent norms within a s o c i a l p o s i t i o n . 10 Role thus governs particular types of behavior of the incumbent—e.g. the nurse i n the " r o l e " of supervisor. The l a t t e r , narrower d e f i n i t i o n i s used throughout t h i s presentation. Bates (1955-6) c l a r i f i e s the concepts " p o s i t i o n " , " r o l e " , and "norm": "1. Position: A location i n a s o c i a l structure which i s associated with a set of s o c i a l norms. 2. Role: A part of a s o c i a l position consisting of a more or les s integrated or related sub-set of s o c i a l norms which i s distinguishable from other sets of norms forming the same position. 3. Norm: A patterned or commonly held behavior expectation. A learned response, held i n common by members of a group (p. 314)." He maintains that within a given culture there are a limited number of roles which are combined i n various ways for di f f e r e n t s o c i a l positions. For example, the £ole of "mother" could be shared by persons i n many positions: "nurse", " s o c i a l worker", and "housekeeper". The E d i t i o n of "nurse" could contain a number of roles: "mother", "subordinate", "supervisor", "woman", "cleric", " f r i e n d " , "coordinator", "nurturer", "medicine dispenser", etc. The d e f i n i t i o n of " p o s i t i o n " i s c l a r i f i e d by Biddle and Thomas (1966). They describe i t as, "...a c o l l e c t i v e l y recognized category of persons for whom the basis for such d i f f e r e n t i a t i o n i s their common at t r i b u t e , t h e i r common behavior, or the common reactions of others toward them (p. 29)." The concept of " r o l e , " i n the narrower sense discussed above, i s not as e a s i l y understood. D. J. Levinson (1959) i d e n t i f i e d three of i t s uses: 11 (a) organizationally given role demands—role expectations of others. (b) personal r o l e d e f i n i t i o n — r o l e expectations which the incumbent places upon himself. (c) r o l e b e havior—actual performance. I t would appear that a unifying concept of " r o l e " i s not available to p a r a l l e l "position". One of the hallmarks of a p a r t i c u l a r s o c i a l position i s that i t i s e a s i l y i d e n t i f i a b l e . In contrast, a s p e c i f i c role may be d i f f i c u l t to ascertain. The investigator must distinguish whether reference to a r o l e means self-expectations, others' expectations, or overt behavior. , For example, a s o c i a l worker i n the r o l e of "supervisor" may . have certa i n expectations about appropriate in t e r a c t i o n behavior. These expectations may be guite d i f f e r e n t from those of s o c i a l work aides, the director of s o c i a l services, patients, and f a m i l i e s , which, i n turn, can d i f f e r from each other. Actual role behavior may r e f l e c t a combination of a l l of the above expectations, as well as personality and s i t u a t i o n a l influences. f i a f l £ e _ l depicts McGill's (1975) aodel of the organizational position " s t a f f nurse". A personality core i s added to Bates' conception of the position-role r e l a t i o n s h i p . Between the personality and the set of role expectations of the position a dotted l i n e has been drawn to denote a permeable interface allowing for (and, indeed, necessitating) t h e i r i n t e r a c t i o n . 12 FIGORE_J[ A MODEL OF THE ORGANIZATIONAL POSITION McGill suggests that f o r each recognizable s t a f f or patient position there i s a "position-set" analogous to Merton's (1966) description of "status-set" as a "..•complex of d i s t i n c t positions assigned to ind i v i d u a l s both within and among s o c i a l systems (p. 7 4 ) . " Social workers on a hospital ward, for example, would have stable patterns of inter a c t i o n with those i n t h e i r position-set ( s t a f f nurse, physiotherapist, occupational therapist) but not with those outside i t ( e l e c t r i c i a n , dietary aide). Returning to Figure,,,1, boundary . relationships between the position and the organizational environment can be depicted by the outer l i n e of the e l l i p s e . I t , too, should be drawn with a broken l i n e to indicate permeability. The length of the interface between a position and a p a r t i c u l a r s o c i a l environment can be considered proportional to the amount of r e c i p r o c i t y which the incumbent has with others i n that position-set. For example, the outer boundary f o r a nurse, who i s i n contact with many persons on the ward during an eight-hour s h i f t , would be longer than that for the hospital administrator, who v i s i t s the ward only occasionally to converse with the head nurse. 13 Gouldner (1960) examined r e c i p r o c i t y i n exchange relationships within a dyad, concluding that power can be balanced or unbalanced i n the short-run, but i n the long-run w i l l reach equilibrium, While he r e s t r i c t e d his analysis to complementary ro l e s such as father/son or supervisor/subordinate, r e c i p r o c i t y also can be examined between organizational positions such as physiotherapist/ pharmacist. Thus i t may be postulated that within a position-set, stable r e c i p r o c i t y relationships can be i d e n t i f i e d , Reciprocities can also be examined using the model i n ZiaM£e_l« When two positions are placed i n organizational proximity and the incumbents are reguired to i n t e r a c t , r e c i p r o c a l role expectations may complement, supplement, or c o n f l i c t with each other. For example, given the positions " s t a f f nurse" and "nursing attendant," the role expectations f o r supervisor/subordinate would generally be complementary. Those for medication-dispenser/bed-makar would supplement each other. Areas of potential c o n f l i c t could be created i f incumbents of both positions were expected to coordinate other s t a f f or nurture the patient. Given the above complexities, distinguishing between role expectations and role behavior in an organizational context i s a complex task. I n i t i a l analyses of r e c i p r o c a l communication behavior can therefore be more f r u i t f u l l y undertaken at the l e v e l of the " c o l l e c t i v e l y recognized" s o c i a l position. 14 2• The Communication Process: The communication l i t e r a t u r e has been reviewed on three l e v e l s of a n a l y s i s — t h e underlying structure of conversations, and studies of the organization and of the group. Most discussions have focussed on i n t r a - i n d i v i d u a l processes—experience, encoding, decoding, perception and i n t e r p r e t a t i o n . In contrast, very l i t t l e attention has been paid to the i n t e r - i n d i v i d u a l dimension of information transmission. Individual communication theory, which r e l i e s heavily on the findings of developmental and abnormal psychology (e.g. Lewin, Sogers), was of tangential relevance and was not reviewed i n depth. Since the study's objective was to develop a consistent, r e l i a b l e , and useful instrument, i n c l u s i o n of ambiguous non-verbal in t e r a c t i o n was not considered to be appropriate. Literature dealing with t h i s dimension was therefore only b r i e f l y examined (Swenson, 1973, P. 81-114). a) The structure of.conversations. Anthropologists have developed the ethnographic method fo r comparative study of cultures. In t h i s research technique, the organized practises of everyday s o c i a l l i f e are meticulously observed and analyzed i n an objective manner. Soc i o l o g i s t s have refined the technique f o r application to smaller a n a l y t i c a l units, including the interpersonal communication s e t t i n g . Texts on ethnography of communication edited by Sudnow 15 (1972) and Turner (1974) shed considerable l i g h t on how conversations are structured. Such issues as how "openings", "sequencing", and "talking to" versus " t a l k i n g f o r " are accomplished are p a r t i c u l a r l y relevant as they provide a framework into which the communication process can be f i t t e d . , fc>) The organization. Much of the l i t e r a t u r e on organizational communication (e.g. Thayer, 1961, 1968) has a strongly p r e s c r i p t i v e bias. Few o r i g i n a l t h e o r e t i c a l contributions have been made. Roberts et a l . (1974b) decry the lack of t h e o r e t i c a l models for integrating communication with other variables. Their multi-level approach to the d e f i n i t i o n of organizational communication (interpersonal, within-organizational, organizational-environmental) i s a f i r s t step i n t h i s d i r e c t i o n . Gibb's (1972a, 1972b) Trust Openness Realization of Interdependence theory (TORI) i s another attempt to bridge t h i s gap. Likert (1961), Miles (1975), and many other organization t h e o r i s t s use communication pr i n c i p l e s as a part of the i r analyses of organizational climate and behavior, but have not developed o r i g i n a l conceptual frameworks. Hhile recent empirical studies of communication by O'Reilly and Roberts (1974; Roberts & O'Reilly, 1974a) focused on perceived communication, these researchers emphasized the incompleteness of such data. The need for research programs to describe organizational communication 16 behavior i s recognized, but such research has generally been li m i t e d to tabulating the freguency of verbal or written acts. One exception i s Rackham and Honey's (1972) discussion of behavioral evaluation of t r a i n i n g i n two organizations. They considered Bales' approach but rejected i t as not providing the desired type of data for t r a i n i n g . S i g n i f i c a n t progress has also been made i n the educational sphere. Several f i e l d studies of communication processes were reported at the American Psychological Association's 1961 "Symposium on conceptual frameworks fo r analysis of classroom s o c i a l i n t e r a c t i o n . " Two of these are of p a r t i c u l a r relevance. Lewis and Newell (1962) observed classroom behaviors using fourteen categories which seemed to be an extension of Bales' dimensions. They focused on two major c l a s s i f i c a t i o n s of communicator—sender and receiver. The 1962 a r t i c l e was described as a "preliminary report", but no l a t e r references have been located. Flanders (1962, 1970) developed an observation scheme for analyzing teacher-pupil verbal communication i n a classroom. While his ten i n t e r a c t i o n categories are s p e c i f i c to the educational setting^ his method for t r a i n i n g observers has been well documented. He studied i n t e r a c t i o n between singular and grouped s o c i a l positions ("teacher" and " p u p i l " r e s p e c t i v e l y ) , but t h i s combination does not appear to have detracted from the u t i l i t y of the t o o l . This i s an 17 important finding since many other organizations have both types of positions within a single position-set. Amidon and Hough's (1967) c o l l e c t i o n of papers on Flanders' method brought together the experiences of a number of researchers who have attempted to quantify classroom i n t e r a c t i o n behavior. Many of the methodological issues considered i n t h e i r text (e.g. research design, observer t r a i n i n g , r e l i a b i l i t y , applications for inservice t r a i n i n g programs) are relevant to other organizational contexts. c) The group. The major thrust of investi g a t i o n of communication behavior has occurred at t h i s l e v e l . McGrath and Altman's (1966) Small group._research provided an excellent "synthesis and c r i t i q u e of the f i e l d " . Both Chappie's (1949) Interaction Chronograph and Bales' (1950a) Interaction Process Analysis were developed f o r use i n the small group laboratory. 3. Communication,.Studies In Health Care 1: Two texts p a r t i c u l a r l y related to the concerns of the research were The community general h o s p i t a l by Georgopoulos and Mann (1962) and The psychiatric h o s p i t a l as 4 Two computerized l i t e r a t u r e searches of this t o p i c have been conducted by the researcher, covering a l l a r t i c l e s c i t e d i n Index.Medicus from 1950 to the present. 18 a small ..society by C a u d i l l (1958). Georgopoulos and Mann analyzed the structure and function of ten general hospitals. Their questionnaire, which included perceived communication networks, was administered to the board of trustees, medical s t a f f , administrators, non-medical administrative department heads, nurses, and X-Ray and laboratory technicians. Interaction of non-supervisory nurses was also examined i n depth. The authors 1 exclusive reliance on perceptual data l i m i t s i t s usefulness for observational studies. C a u d i l l explored the s o c i a l system of the p s y c h i a t r i c h o s p i t a l . He obtained his data through camouflaged observation on the ward, followed by open recording of patient behavior using the ethnographic method of data c o l l e c t i o n . This approach has considerable promise f o r research i n health care organizations as a main or an adjunctive source of data. A second section of C a u d i l l * s study examined communication i n administrative and c l i n i c a l s t a f f conferences using a modified form of IPA. One of his major findings was the e f f e c t of status and role (used i n the broader sense) on the amount and type of p a r t i c i p a t i o n i n the conferences. He also analyzed the effects of c e r t a i n int e r a c t i o n processes upon the t o t a l h o spital system. lessen (1958) studied the frequency of verbal information exchanges between hospital s t a f f . Systematic observation techniques were used on two wards, together with 19 formal and informal interviews with seventy-five doctors, nurses, d i e t i c i a n s , laboratory technicians, and non-professional ward personnel. He r e s t r i c t e d his data to absolute frequency of in t e r a c t i o n s . Stevens (1967) applied Wessen's method to a health education complex. She found an increase i n l a t e r a l communication which she attributed to the unit's unique organizational structure. Lum (1970) assessed i n t e r a c t i o n patterns of nursing personnel i n a large private general hospital i a Honolulu. She developed a questionnaire which measured the frequency, content, perceived importance, and satisf a c t o r i n e s s of s e l f -and o t h e r - i n i t i a t e d contacts. Wilkinson (1973) examined in t e r a c t i o n patterns and communication channels using a sociometric analysis of a series of guestions and an interview format. No d i r e c t behavioral coding was dons. The focus was on i n t e r a c t i o n within a t o t a l i n s t i t u t i o n . Werner's (1974) description of "Teaching medical students i n t e r a c t i o n a l s k i l l s " , while inte r e s t i n g from a methodological standpoint, used content rather than process as the basic phenomenon for investigation. Nursing researchers have developed several systems f o r coding the behavior of students or professionals inte r a c t i n g with patients. Topf (1969) formulated a Communication S k i l l s Checklist to assess the i n t e r a c t i o n s k i l l s of nursing students. The instrument assesses e f f e c t i v e and i n e f f e c t i v e behaviors i n 20 eight categories, r e l y i n g heavily on raters* perceptions. Johnson (1964) examined the interaction between eighteen nursing students and t h e i r patients i n the psychiatric s e t t i n g . Each student interacted with one patient. The observation instrument had ten content and process categories. No reasons were given f o r t h e i r s e l e c t i o n . Combining content and process dimensions i n one instrument i s considered by the present researcher to be methodologically unwise. Johnson also mentioned that, "The pressure of the observer i n the room with the student-therapist and the patient did a l t e r the i n t e r a c t i o n (p. 341)." This e f f e c t , which was evidently not anticipated, should be controlled and/or minimized i n such research studies through extensive orientation and f a m i l i a r i z a t i o n programs for subjects. Diers and other researchers at the Yale University School of Nursing (Diers & Leonard, 1966) used objectives of nursing care to develop a category system for studying nurse-patient i n t e r a c t i o n . , In Nurse Orientation System (NOSY) a conversation may be oriented towards an object, patient, nurse, or other person. Each d i r e c t i o n i s sub-categorized according to the primary emphasis on f e e l i n g , thinking, or doing. In McBride's (1967) study of the effects of three nursing approaches to patients i n pain, NOSY was used to assess how well the independent variable "nurse's approach to patient" was operationalized i n the experiment (see also 21 Diers, Schmidt, McBride, S Davis, 1972). Diers and Schmidt (1968) assessed the loss of data through t r a n s c r i b i n g tape-recorded conversations. Their major source of error was the transcribing secretary, who had l i t t l e f a m i l i a r i t y with the context from which the data were c o l l e c t e d . This i s not a problem i n studies which code interactions in-process. Nevertheless, i t emphasizes the need for coders to understand certain aspects of the organization under investigation. . A program of i n t e r a c t i o n research was i n i t i a t e d by Hess (1969) through the Center for the Study of Medical Education at the University of I l l i n o i s . He modified S i t h a l l ' s method for studying teacher-pupil i n t e r a c t i o n , developing eleven behavioral categories (Heifer & Hess, 1970, p. 328; see also N. C. Scott, Donnelly, Gallagher, & Hess, 1973, p. 175 for a s l i g h t modification). The unit of recorded behavior was "an uninterrupted, apparently purposeful action by the student (Hess, 1969, p. 935)." This i s less precise than Bales' d e f i n i t i o n of an "act", and requires additional i n t e r p r e t a t i o n by the observer. In Hess* i n i t i a l study, patient interviews by medical students were videotaped. They were l a t e r coded by three physicians who received four hours of t r a i n i n g i n r a t i n g . A guestionnaire designed to gather s i m i l a r information was also completed by the raters, and i n t e r - and i n t r a - r a t e r r e l i a b i l i t y was assessed. The in t e r a c t i o n analysis scores were found to be more r e l i a b l e than the questionnaire 22 r e s u l t s i n determining e f f e c t i v e versus i n e f f e c t i v e communication behaviors. Heifer (1970; Heifer & Hess, 1970) modified Hess* instrument, comparing interviews by senior and freshmen medical students, fhree actors played the role of the mother of a seriously i l l c h i l d . Five trained observers coded the resultant videotapes. Heifer's use of a quasi-experimental design, with actors trained to exhibit s p e c i f i c communication behaviors, holds considerable p o s s i b i l i t i e s for other researchers. H. C. Scott et a l . (1973) used Hess* approach to evaluate a course i n physician-patient re l a t i o n s h i p s f o r medical students. Seventy-six simulated interviews were videotaped and then coded by two raters with a r e l i a b i l i t y of over .85. These were compared with ratings by ten expert judges of the effectiveness of students* interview behavior. The f i n a l studies which were noted were those conducted by Howland and h i s associates. Early a r t i c l e s by Howland (1963a, 1963b, 1963c; Howland S McDowell, 1964) outlined the development of a general Hospital Systems Model for studying health care behavior. These t h e o r e t i c a l references explored p o s s i b i l i t i e s for examining the "nurse-patient-physician t r i a d " (1963a, p. 229) and discussed the need f o r behavioral data on i n t e r a c t i o n processes (1964, p. 295). A recent a r t i c l e by Daubenmire and King (1973) discussed the t h e o r e t i c a l aspects of the research. Howland (1966) gave b r i e f d e t a i l s of the gathering of technological data by two 23 observers i n an operating theatre, but no information about the observation of int e r a c t i o n s . They are now conducting a research program to test the model using observational data gathered by extensive videotaping i n patient rooms. Future reports on t h i s research should be followed to determine the categorization system and the methodological findings on videotaping. In summary, four researchers—Topf, Johnson, Diers, and Hess—have developed t h e i r own coding systems for studying the process of nurse-patient i n t e r a c t i o n . In each, categories were designed to measure s p e c i f i c types of desired nursing behavior, rather than general i n t e r a c t i o n patterns across a broad spectrum of s t a f f and patient positions. As such, the above nursing categories are not appropriate for analysis of i n t e r - p o s i t i o n behavior. 4. Balesy Interacti^gn^Process Analysis; In 1950, Robert F. Bales and his associates at the Harvard Laboratory of Social Relations reported the development of a method for investigating communication behavior i n the small groups laboratory (Bales, 1950a) based on Parsons' Pattern Variables. Interaction Process Analysis (IPA) employed a set of twelve mutually exclusive categories of communicative behavior which, when placed i n a who-to-whom matrix by a trained observer, enabled the researcher to construct " i n t e r a c t i o n p r o f i l e s " f o r indi v i d u a l s and groups. A l l face-to-face verbal and overt 24 non-verbal interactions were recorded. Bales (1950a) stated that, "The present set of categories i s meant to be a general purpose framework for observation which can be used to obtain a series of standard indices regarding the structure and dynamics of i n t e r a c t i o n i n any small group....,The set of categories i s aeant to be completely i n c l u s i v e i n the sense that every act which can be observed can be c l a s s i f i e d in one p o s i t i v e l y defined category (p. 33,35)." The twelve categories focus on how the s o c i a l actor communicates. They are arranged i n two hierarchies which are combined into a si n g l e , mirror-image l i s t . Neutral emotional content i s i n the middle, and strong p o s i t i v e and negative a f f e c t at the top and bottom, respectively. The categories are paired, working outwards from the c e n t r a l task problems of communication and evaluation to the peripheral emotional problems of tension reduction and reintegration as depicted i n Figure 2. While t h i s instrument has been used extensively i n small group research for over twenty-five years, none of the studies covered by the present review gives s t a t i s t i c a l support for the construct v a l i d i t y of Bales* theory of communication. P a r t i c u l a r areas of concern are the inclusiveness of the categorization, the independence of each dimension, and the interdependence of the category pairs. Each requires further construct validation. In Bales* o r i g i n a l research he used two to seven "trained observers" behind a one-way screen, each having an electro-mechanical Interaction Recorder with a continuously moving paper tape. For each communication "act" (simple 25 FIGURE_2 liLISi_SYSTEM_OF_CATEG^RIES USED IN_QBSSHVATIQN AND THEIfi MAJOB RELATIONS S o c l a l -fhot loaal Area i pos i t ive Task Arcai Keutral Soel&l-tfcotlonal Area i j, J \ \ Shows s o l i d a r i t y , r a i s e s o t h e r ' s s t a t u s , C l v c s help, riwi'.zvii 2 Sh',wr, tension r»l»*=e. Jokes, laughs, shows s a t i s f a c t i o n i Agrees, chows passive acceptance, undr.rstar.ds, concurs, e a s p l l o s i * Clves s w r t s t ' o n , d i r e c t i o n , l = p l y i n 3 autonomy l o r ether i C l v c s o r ' - i l o ^ , e v a l u a t i o n , a n a l y s i s , expresses f e e l i n g , w l s h i 6 C l v r n o r i e n t a t i o n , i n f o r n a t l o n , repeats " c l a r i f i e s , c o n i l r a s i ? Asks f o r o r i e n t a t i o n , ' i n f o r c a t l o a r e p e t i t i o n , c o n f i r m a t i o n i 8 Asfcr, f o r o p i n i o n , e v a l u a t i o n , t r . a l y s l s , expression of f e e l l n g i 9 Asks f o r s t v-entlon, d i r e c t i o n , posslbl.. vj»ya of a c t i o n t ' 10 M s f . T f o j , shows pa.-.sivo r o j e c t l o n , f o r m a l i t y , v l t h o l d s help n Chows t e n s i o n , asks; f o r help, w i t h d r a w out of f i e l d i 12 Shows ant.T-onlsn, d e f l a t e s o t h e r ' s s t a t u s , defends or a s s e r t s s e l f i KETl 6 ftrofcle*s of Corxnunlcatlon i froMer . s of Evaluation C JroVloria of Control A froVlcns of Decision • fttoltlcns of Tension Hoducllon * — » 1 a b e d e f J f . TroV *na of Re i n t e g r a t i o n A P o s l t i v o Konotlcns B Attempted Answers C Cuostlons D Nc^atlva Reactions (fxoa B i l os , 1950-a, p. 9) \ s u b j e c t - p r e d i c a t e c o m b i n a t i o n o r an o v e r t n o a - v e r b a l a c t i o n ) t h e sender and r e c e i v e r s e r e r e c o r d e d under the a p p r o p r i a t e c a t e g o r y . Because t h e c a t e g o r y f r e q u e n c i e s d i f f e r e d w i d e l y i n magnitude, t h o s e a t t h e extremes ( i . e . C a t e g o r i e s 1 . and 26 12) were given p r i o r i t y over those i n the centre. This was done to i n f l a t e the numbers i n the less freguent categories to a s t a t i s t i c a l l y s i g n i f i c a n t l e v e l , a practise which i s considered methodologically unsound. Although Bales outlined an extensive training program for observers, he gave i n s u f f i c i e n t coding d e t a i l s for others seeking to use the technigue. Procedures employed by most other researchers have not been reported. A notable exception i s the thorough i n s t r u c t i o n a l manual for IPA research using typed t r a n s c r i p t s which was developed by Mishler and Waxier (1968). Considerable modifications to t h i s manual are reguired for in-process scoring. Training i n IPA i s thus an area which reguires further attention. One methodological concern which has been very thoroughly covered i n the l i t e r a t u r e i s that of r e l i a b i l i t y . Bales devoted an entire chapter of his o r i g i n a l text (1950a) to "Appraising observer r e l i a b i l i t y . " An exhaustive process was proposed f o r assessing both i n t r a - and i n t e r - r a t e r r e l i a b i l i t y . He discussed three key problems of r e l i a b i l i t y as those of " u n i t i z i n g , " "categorizing," and " a t t r i b u t i n g " (p. 101). The f i r s t refers to dividing a conversation into separate communication acts, the second to placing each act i n a s p e c i f i c category, and the t h i r d to i d e n t i f y i n g the sender and receiver of the message. The f i r s t two are of p a r t i c u l a r concern during in-process monitoring of dyadic communication. 27 For i n t r a - r a t e r r e l i a b i l i t y (test-retest),- Bales (1953) reported c o e f f i c i e n t s of .65 to .98 for the twelve categories (Pearson's r ) . The mean for the twelve was .92, with the following values under .85 : Category 4(.81); Category 8 (.65); Category 9 (.83); Category 10 (.70).. For i n t e r - r a t e r r e l i a b i l i t y , Heinicke and Bales (1953) compared two sets of laboratory studies at Northwestern and Harvard. Each used a s l i g h t l y d i f f e r e n t set of categories. The main difference was that in the Northwestern groups scoring was done by recording one category for each speaker's t o t a l utterance, rather than one per "act" as done at Harvard. R e l i a b i l i t y was determined for two trained observers on the basis of gross t o t a l s for each category, and thus did not give any i n d i c a t i o n of act-by-act congruence. The r e s u l t s showed higher r e l i a b i l i t y f or the Northwestern observers, using the larger unit of analysis. Those for the Bales technigue ranged from .90 (Category 2) to .74 (Categories 9 and 7), with a median of .86. Bales (1953) also examined the consistency of subject behavior for two separate sets of studies. The r e s u l t s are fa r from encouraging, i n spi t e of the statement that, "...the general picture of the correlations indicates that a positive relationship e x i s t s between the behavior of the same subject from one time to another i n a l l the categories...(which) i s s u f f i c i e n t l y good to give the researcher some confidence (p. 568)." Table_I duplicates h i s findings f o r one of the sets. 28 TABLE_I STABILITI_0F_SJIBJICTS1_IN Kimber of pextlcl jvintB (ti) 3 5 5 4 5 Buabrr of sesclona 2 2 4 4 Category 1 .w -.1* .53 .26 .13 2 -.62 .67 .9* -.15 .50 .93 .91 .70 •?5 .73 * .26 .79 .83 .84 -.13 5 .8? .91 .50 C .80 .79 .90 -.01 .76 7 .97 .39 .86 • .01 .33 e .92 .27 .99 .96 .19 9 ».03 .01 .33 .97 .61 10 -.53 .68 .69 .97 11 .0? .85 .78 .09 .23 12 .00 -.25 .96 .00 1.00 ToUl .99 .96 .85 .92 • .56 • (from tales, 1953. r. able 2-) While a l l categories except 1, 4 and 11 had at l e a s t one s t a b i l i t y c o e f f i c i e n t above .90, there was no consistent pattern of high and low c o e f f i c i e n t s across a l l groups. The differences i n the p r o f i l e s of s t a b i l i t y and the negative cor r e l a t i o n s (one as high as -.62) should e i t h e r be •accounted for or minimized i f the u t i l i t y of the instrument i s to be demonstrated. , No allowance was made i a the Bales s t a b i l i t y study for the e f f e c t s of task, time, and other factors which may have led to di f f e r e n t interaction patterns during subsequent meetings of the same group 1. 1 Methods whereby the present study handled t h i s s t a b i l i t y problem are discussed below, page 41. 29 Psathas (1961) compared Bales 1 technique of in-process scoring with the use of written and taped protocols. In the l a t t e r , the observer focused on non-verbal behavior, r e l y i n g on the tape t r a n s c r i p t s for verbal content. Psathas compared the number of in-process acts per t i a e period to the number of "pooled" (verbal plus non-verbal) acts, and found that in-process scoring missed 23% of the pooled t o t a l . When he corrected the pooled scores for differences i n number of acts, he found that the freguency d i s t r i b u t i o n s of acts by category corresponded very c l o s e l y except f o r Categories 10 through 12. These three categories are a l l i n the negative socio-emotional area (10, disagrees; 11, shows tension; 12, shows antagonism). In-process scoring was higher on Category 10, while the pooled method was higher for.Categories 11 and 12. Waxier and Mishler (1966) reported the most c r i t i c a l review of IPA r e l i a b i l i t y problems to date. Their extensive analysis of marginal differences versus act-by-act comparisons i s an important one for t h i s research. When 10,910 acts were scored using both approaches, two raters disagreed on the categorization of 38.1%. Over h a l f of these disagreements were due to incorrect categorization of categories 3 into 5, 5 into 6, 10 into 5, and 10 into 6. Thus, of one hundred and thirty-two possible types of disagreement, four contributed most of the errors. 30 In an early a r t i c l e . Bales (1951) summarized some of the " S t a t i s t i c a l problems of small group research", i d e n t i f y i n g four d i f f i c u l t i e s with scoring: (a) the unit of analysis i s not s u f f i c i e n t l y defined that a l l observers can come up with the same number of acts. (b) there i s no way of obtaining a "correct" answer i n an absolute sense, since a l l scoring involves human judgment. (c) events cannot be repeated for recount. (d) the speed with which the in t e r a c t i o n moves precludes recording each and every act (p. 314). A l l of these concerns are extremely relevant. Numbers one, two, and four can be managed by thorough observer t r a i n i n g with documentation of i n t e r - r a t e r r e l i a b i l i t y on an act-by-act basis. Number three can be resolved i n the laboratory by using videotaping, but presents major problems i n the organizational s e t t i n g . Bales proceded to out l i n e four ways of analyzing observations: (a) Interaction p r o f i l e s - based on t o t a l meeting, sub-periods and i n d i v i d u a l s . These can be tabulated as raw scores or as percentages of the t o t a l i n t e r a c t i o n . (b) Sequencing u iof acts - each act i s recorded twice, as the f i r s t and second number of a pair. (c) Phases - e.g. div i d i n g a meeting into three egual parts. (d) Matrices - the d i s t r i b u t i o n of p a r t i c i p a t i o n between members, placing the t o t a l number of acts i n a who-to-whom matrix. Senders can also be arranged in rank order of p a r t i c i p a t i o n freguency (p. 315). Most early researchers using IPA emphasized the l a t t e r two approaches. Recent work i n the f i e l d has r e l i e d more on 31 the analysis of p r o f i l e s . There has been very l i t t l e research done on sequencing.. This may be because of the arduous nature of the analysis, where every act must be examined twice—once as the f i r s t act of a sequential pair, and once as the second. Bales postulated that sequential acts were interdependent. By dealing with a process of i n t e r a c t i o n , he reasoned that the frequency of acts f a l l i n g into any c e l l would a f f e c t those i n adjacent and paired c e l l s . This would mean that standard t e s t s of s t a t i s t i c a l s ignificance such as Chi-square could not be used. No evidence was found i n the l i t e r a t u r e to support the assumption of interdependence of the category pairs. Turning from issues of r e l i a b i l i t y and s t a t i s t i c s to those of methodology, studies were examined which took IPA out of the small groups laboratory and into the f i e l d . O'Eourke (1963) observed the same three-person family groups at home and i n the laboratory. His basic premise was that "The quality of in t e r a c t i o n i n s o c i a l groups can be expected to vary as the s i t u a t i o n a l contexts of the interactions and the structures of the groups (p. 422).'* He used a non-participant observer present i n the room i n both settings, but did not discuss the impact of the observer (an issue of v i t a l concern to organizational researchers). A shortened form of IPA with only three categories was employed. No in d i c a t i o n s of r e l i a b i l i t y were given. His finding that the balance between socio-eaotional and 32 instrumental interactions was contingent upon the f a m i l i a r i t y of the environment was considered p a r t i c u l a r l y relevant. The major focus of the review of f i e l d studies was on the use of IPA i n health care settings. C a u d i l l (1958) was the e a r l i e s t researcher to report such a study., His work has been discussed above. Korsch and Negrete (1972) observed physician-parent communication during diagnostic v i s i t s to a large p e d i a t r i c c l i n i c , and were able to r e l a t e mutual d i s s a t i s f a c t i o n with the v i s i t to c e r t a i n types of i n t e r a c t i o n p r o f i l e s (Francis, Korsch, S Morris, 1969; Freemon, Megrete, Davis, & Korsch, 1971; Korsch, Gozzi, & Francis, 1968). M. S. Davis (1968) used IPA to code 154 patient interviews by fourth year medical students or attending physicians. Coders worked from tape recordings; d e t a i l s of t r a i n i n g or r e l i a b i l i t y were not given (except that r e l i a b i l i t y averaged .85). In addition to the construction of i n t e r a c t i o n p r o f i l e s and Bales* indices, Davis did an extensive factor analysis of the doctor-to-patient data. P r o f i l e s of the twelve i n t e r a c t i o n categories for both physician and patient were correlated with each other and with themselves, forming a twenty-four by twenty-four matrix. From t h i s , ten factors were extracted which represented d i f f e r e n t types of patient-doctor communication behavior. The study then related these interaction factors to compliance by the patient, finding some patterns which 33 helped to explain noncompliance. Conant (1965) used IPA to "examine the development and nature of the ro l e relationships of public health nurses and patients i n home v i s i t s (p. 304)." Twelve public health nurses made a t o t a l of forty-eight home v i s i t s to twenty-four antepartal patients. The interactions were tape recorded and then scored by two persons trained i n IPA. The instrument was chosen because of the apparent appropriateness of Bales 1 task/emotional problem-solving model. Observed behaviors were related to perceived s a t i s f a c t i o n , as obtained by separate interviews with nurses and patients. Conant devoted considerable time to examining the li m i t a t i o n s of IPA. The exclusion of in t e r a c t i o n content from the coding scheme was discussed. She f e l t that t h i s was a handicap to be overcome.. Diers and Leonard (1966), i n considering the dual objective of assessing communication process versus content, stated, " I t i s possible to combine both kinds of categories i n one system, but for the sake of the mental health of the researcher, i t i s probably better to confine any one system to either content or process. I f both kinds of categories are included i n one system, the coder and researcher are faced with deciding whether there i s an important difference between, for instance, a question that i s 'supportive' and a statement that i s ' unsupportive • (p. .226)." The researcher agrees with Diers, believing that i t i s better to do an accurate evaluation of in t e r a c t i o n processes, rather than attempt to use a scheme which combines process and content. 34 Conant also raised questions of i n t r a - and i a t e r - r a t e r r e l i a b i l i t y . She found that Bales' categories were not always mutually exclusive. "An act i s not necessarily either completely task or emotional i n nature., I t may contain elements of both and have several purposes and e f f e c t s (p. 308).*' This finding, together with her observation that Categories 3 and 11 were r e l a t i v e l y undifferentiated between patient and nurse behaviors, has considerable importance f o r the construct v a l i d i t y of the instrument. Hhereas Conant saw these findings as l i m i t a t i o n s , an alternative approach would be to build on Bales' conceptual framework, modifying i t where appropriate in the l i g h t of additional empirical evidence. Conant found that the intervening variable of patients' race emerged i n an "ex post facto analysis." I t i s important to consider such intervening variables, as they have an important contribution to make i n understanding the t o t a l i n t e r a c t i o n process. Conant's f i n a l c r i t i g u e of IPA was re l a t e d to the scoring of negative socio-emotional categories. She stated, "In the nursing v i s i t s that were analyzed the showing of tension (category 11) sometimes seemed to be related to the s i t u a t i o n or the topic under discussion rather than to the r e l a t i o n s h i p of the participants (p. 309)." She seems to have ignored Bales' requirement that acts r e f e r r i n g to the outer s i t u a t i o n be scored as actor-to-X rather than to the co-communicator (Bales, 1951a, p.,185). This finding regarding Category 11 could also be important to the instrument's construct v a l i d i t y . , 35 Mishler ana Waxier (1966, 1968) studied the relati o n s h i p between patterns of family i n t e r a c t i o n and the presence of schizophrenia. They used family discussions of questionnaire feedback and the method of revealed differences as the basic source of data. Sessions were tape recorded and transcribed before being subjected to eleven types of coding procedures (see 1966, p. 69 for a summary of the codes). These varied i n complexity from a simple who-to-whom categorization to Bales 1 IPA. While the authors raised many questions about the r e l i a b i l i t y of IPA, they nevertheless made extensive use of these data i n inte r p r e t i n g t h e i r findings. In summary, M. S. Davis, Korsch, and Mishler and Waxier a l l employed IPA extensively i n physician-patient interviews, and made constructive use of i t as an evaluative t o o l . Conant was more c r i t i c a l , but her i n t e r e s t was focused on obtaining s p e c i f i c nursing care data rather than on understanding the t o t a l process. A l l of the studies reviewed above which used IPA i n health care organizations found that stable i n t e r a c t i o n p r o f i l e s could be measured for role or position groups i n a health care organization 1. They therefore support the 1 Whereas these studies examined communication between one health profession and a single type of patient, the present research looked at the inte r a c t i o n network of a larger unit--the hospital ward. Many s t a f f groups were included. Patients also d i f f e r e d i n type and extent of d i s a b i l i t y . . 36 premise that each s o c i a l position within the organization has a consistent set of interaction p r o f i l e s . Another focus of the l i t e r a t u r e review was on studies which used IPA as a t o o l to further an understanding of other s o c i a l phenomena. F. G. Scott (1962) observed three-generation f a m i l i e s i n the laboratory. She used a modification of Bales* scoring to determine how patterns of support between any two family members were affected by group size and family position. Her use of "family position" i s a departure from the general emphasis on the analysis of i n d i v i d u a l behavior. It p a r a l l e l s a focus on the organizational position. Hanheim (1960) investigated intergroup i n t e r a c t i o n , focusing on the r e l a t i o n s h i p between the amount of intergroup d i f f e r e n t i a t i o n and c o n f l i c t . His study design involved two t r i a d s which were experimentally manipulated to create differences i n perceived status (high/low) and leadership (assigned/emergent). The two groups communicated with each other via written notes which were l a t e r analyzed using IPA. Manheim employed Bales' Index of Expressive Malintegrative Behavior (EMB), which i s calculated by taking the number of negative socio-emotional acts as a percent of a l l positive and negative socio-emotional acts.„ Landsberger's (1955) analysis of the "Mediation of labour-management disputes" found some evidence of the predictive v a l i d i t y of IPA . He stated. 37 "...ultimate success of the session could be p a r t i a l l y predicted from the parties' state of mind when they embarked upon the session; the more h o s t i l e t h e i r expressed fe e l i n g s , the le s s l i k e l i h o o d of success (p. 558)." His assumption of eguality of parties i s generally not present in the health team. According to Pellegrino (1970), "...the physician i s the most potent force i n decision making, not only in regard to care of i n d i v i d u a l patients but i n regard to the setting of p o l i c i e s , goals and commitments of the ho s p i t a l (p. 303)." However, some teams appear to be working towards a more e g a l i t a r i a n form of decision-making within t h e i r group meetings or conferences. Landsberger 1s approach might therefore be used to compare teams perceived by t h e i r members to be e g a l i t a r i a n with those perceived to be authoritarian. Greater predictive v a l i d i t y would be expected with the former group, since each member would see an opportunity to influence the outcome of the conference. In Bales' l a t e s t book, Personality And Interpersonal Behavior (1970), he expands the twelve categories, plotting each on three dimensions of the values structure of the i n d i v i d u a l . He conceptualizes these three as personality dimensions which a f f e c t the individual's i n t e r a c t i o n p r o f i l e . They were derived from a study of sixty subjects who: "...met i n twelve groups of f i v e f o r a series of f i v e two-hour meetings, each time with a somewhat d i f f e r e n t task. There were two interviewing and inference tasks, one somewhat formalized s e l f - a n a l y t i c task, and two value-dilemma tasks based on cases. The data f o r a l l f i v e meetings were pooled for each i n d i v i d u a l (p. 391)." Extensive factor analyses were performed on f i v e types 38 of variables: (a) personality t r a i t s , measured i n advance by written t e s t s . (b) observation of overt behavior, using IPA. (c) content c l a s s i f i c a t i o n of value statements made during the groups. (d) ratings of the i n d i v i d u a l by other group members., (e) guesses by the i n d i v i d u a l of the ratings he would receive from others (p. 391-392)., The three personality dimensions were derived from these data. The f i r s t , "upward/downward" (H/D), r e f e r s to the tendency on the part of the i n d i v i d u a l towards dominant (0) versus submissive (D) i n t e r a c t i o n behavior. This i s conceptualized as a pure leadership f a c t o r . The second, "forward/backward" (F/B), i s concerned with whether the i n d i v i d u a l 1 s communication behavior contributes to (F) or retards (B) the accomplishment of the task at hand. The t h i r d dimension, "positive/negative" (B/U), describes the expression of a f f e c t or f e e l i n g . The o r i g i n a l twelve IPA categories given i n Figure^2. page 25, were revised to create a new categorization, which was hypothesized to be d i r e c t i o n a l l y consistent along the three dimensions. The only major change involved Categories 2 (tension release) and 11(shows tension), although most of the others were c l a r i f i e d and re-named. Bales then developed d i r e c t i o n a l l abels for each category, based on frequencies above or below the median range of i n t e r a c t i o n s , Horms for the medians were i n f e r r e d from predicted changes between the old and the new schemata but have not yet been 39 empirically v e r i f i e d . Table_II gives the suggested inte r p r e t a t i o n for acts i n i t i a t e d . TABLE_II BALES1_NEW_IPA--BST^ Catecory of Aelo Initialed If lovi «»» Mndiua Ranfio If Hlf;hi I. Scoas Friendly R 2 . 6 - 4 , 8 P 2 . Dramatizes DP ' UB . 3 . Agrees NB 8 . 0 - 1 3 . 6 pj. *. Clves Kusccctlon DB 3 . 0 - 7 . 0 OF 5. Clvcs Opinion B 1 5 . 0 - 2 2 . 7 r C Gives Information U 2 0 . 7 - 3 1 . 2 D 7. Asks for lnforuation KJ 4 . 0 - 7 . 2 OP 8. Asks for Opinion N 2 . 0 - 3 . 9 P ' 9. Asks for Suceostion UB . 6 - 1 . 4 DP 10. Disagrees P 3 . 1 - 5 . 3 B ». Shows Tension OF 3 . 4 - 6 . 0 DB 1 2 . Seems Unfriendly P 2 . 4 - 4 . 4 H -• (froo Riles, 1970, p. 96) ** . These norms are ostlnated by A long process of inference, described ln Appendix 4 , tete3 lower than the Medium Kango shown are classified I>ow| r a t e 3 higher than the Kcdlua Panes shown aro classified High, For example, i f an ind i v i d u a l * s frequency i n Category 1 i s below the median range i t r e f l e c t s a negative (N) tendency. Conversely, a high proportion of acts i n Category 1 r e f l e c t s a positive (P) value orientation. In discussing his new formulation of the theory. Bales (1970) stated, " I t should be recognized that six t y subjects i n twelve groups i s a very small number for such a mountain of data analysis and so many hypotheses., No claims are made for the r e l i a b i l i t y or representativeness of the findings. These problems must be dealt with i n the future. The present study, for a l l i t s complication, i s only exploratory (p. 391)." 40 In spite of the above caveats, plus the fact that the d i r e c t i o n a l hypotheses are based on personality dimensions which may or may not be applicable at the p o s i t i o n a l l e v e l . Bales 1 new formulation i s considered preferable to the old one for data c o l l e c t i o n on the hospital ward 1. 5. ,. Summary: The following conclusions emerge from the review of the l i t e r a t u r e on Interaction Process Analysis (IPA): (a) No s t a t i s t i c a l v a l i d a t i o n has been found f o r the t h e o r e t i c a l foundations of IPA. (b) Acceptable l e v e l s of i n t r a - and i n t e r - r a t e r r e l i a b i l i t y can be achieved for scoring of category t o t a l s . (c) Act-by-act comparison, while giving i n d i c a t i o n s of r e l i a b i l i t y , must be convincingly demonstrated i n any IPA study. (d) I t i s e s s e n t i a l to establish s t a b i l i t y of subject in t e r a c t i o n behavior, or to account f o r a major portion of i t s variance through s i t u a t i o n a l moderators. (e) Although IPA has been taken out of the laboratory, no studies were found which measured the impact of t h i s methodological change. (f) IPA does have tentative indications of predictive v a l i d i t y , helping the researcher to understand other dimensions of organizational behavior. 1 Both categorizations were assessed during the t r a i n i n g period of the present study, and the decision made that the new one was more suitable for t h i s organizational context. Support for t h i s conclusion i s given i n Chapter . ?II. 41 The present research has been designed to consider the above fa c t o r s . The construct v a l i d i t y of Interaction Process Analysis was examined. Act-by-act r e l i a b i l i t y was documented, as was the t r a i n i n g process whereby t h i s was achieved. Bales' s t a b i l i t y problem was handled i n two ways. F i r s t l y , a l l interactions for a particular position dyad were combined, disregarding the i n d i v i d u a l except when there was only one position incumbent on that ward. Thus, i n d i v i d u a l variations due to short-term intra-personal factors such as fatigue, mood-swings, etc., were eliminated. By changing a systematic source of error variance i n t o a random one, i t was suggested that greater p r o f i l e s t a b i l i t y could be achieved. Secondly, the project confronted the s t a b i l i t y issue through the measurement of intervening contextual variables. These were hypothesized to moderate the r e l a t i o n s h i p between the communication dyad and i t s i n t e r a c t i o n p r o f i l e . Support was given for taking the method out of the laboratory into various organizational locations. Questions of experimenter a r t i f a c t , which were c r i t i c a l to the research design, were considered i n d e t a i l . F i n a l l y , the functional u t i l i t y of IPA was a v i t a l focus of attention. The method should be s u f f i c i e n t l y powerful for broader investigations of organizational behavior i f i t i s to make a s i g n i f i c a n t contribution to the d i s c i p l i n e . 42 F. DEVELOPMENT C^_HYPOTHESESi The researcher's o r i g i n a l intention i n formulating Hypothesis 1 was to test two aspects of Bales* communication theory—the o v e r a l l category independence, plus the interdependence of the category pairs. H1, as i t was o r i g i n a l l y stated, did not r e f l e c t t h i s intention: H1 The paired Categories ( i . e . 1-12, 2-11, etc.) are interdependent. I t was therefore reformulated as follows: Hl-a The paired Categories ( i . e . 1-12, 2-11, etc.) are sequentially interdependent. H1-.b The r e l a t i v e frequencies of the twelve categories are independent., Hypothesis 2 was suggested by data from the p i l o t study (see appendix A). where Categories 2 (tension release), 6 (gives information), and 11 (shows tension) tended to occur i n conjunction with others. They were postulated to be adjunctive communicative behaviors which worked i n concert .with the others to c l a r i f y (Category 6) and modify (Categories 2 and 11) the t o t a l i n t e r a c t i o n sequence. Conant's (1965) finding that Category 11 was r e l a t i v e l y undifferentiated between patient and nurse behaviors also supported t h i s hypothesis. The formal statement of Hypothesis 2 was: H2 Three of the Categories (2, 6, and 11) are more strongly correlated with the others than are the remaining nine. The t h i r d hypothesis also evolved from the need to 43 contribute to the construct validation of Bales* theory of communication through an assessment of the inclusiveness of the categorization: H3 Bales* twelve interaction Categories are i n c l u s i v e of a l l types of verbal communication behavior. The remaining f i v e hypotheses were designed to test the research model described i n d e t a i l i n Chaj3ter_.il- Together, they assessed s t a b i l i t y of subject behavior (H4), the impact of s i t u a t i o n a l moderators (H5, H6), the e f f e c t of various organizational l o c i (H7), and the guestion of experimenter a r t i f a c t (H8). H4 A major independent variable which influences the communication p r o f i l e on a hospital ward i s the position dyad. H5 Task factors have a weaker, but s t i l l s i g n i f i c a n t , e f f e c t on the int e r a c t i o n p r o f i l e s . H6 The contextual factors of day, week and time do not have a s i g n i f i c a n t e f f e c t , when controlled for other intervening variables. H7 The loc a t i o n of i n t e r a c t i v e behavior ( i . e . ward, locus) has a strongly s i g n i f i c a n t mediating e f f e c t on the p r o f i l e s of a l l dyads. H8 The effect of a s p e c i f i c observer does not produce variation i n communication p r o f i l e s . Having developed the research hypotheses, d e t a i l s of the study method w i l l now be discussed. 44 CHAPTER II A. LOCATION: In order to test the above hypotheses, a four-month f i e l d study was conducted at Hascana Hospital, a d i v i s i o n of the South Saskatchewan Hospital Centre, i n Regina, Saskatchewan. This was the s i t e of the p i l o t study. The researcher had been employed by that hospital f o r f i v e years as a physiotherapist and researcher, and had gained acceptance by the hospital s t a f f . Twenty-three percent (144/614) of the s t a f f who were employed during the p i l o t project were s t i l l at Hascana when the present study began, including sixteen of the twenty department heads and assistant d i r e c t o r s of nursing, plus the head of the medical s t a f f . Considerable i n t e r e s t i n the outcome of the p i l o t project was expressed by ward and treatment s t a f f when the researcher returned to work i n the hospital physiotherapy department i n the summer of 1974. The two wards selected for the study were "Second Best" (extended care) and "Station Five" (adult physical r e h a b i l i t a t i o n ) . Each ward was a functional subsystem within the h o s p i t a l , with commonly accepted geographic boundaries and a single "nursing s t a t i o n " where patient care a c t i v i t i e s were coordinated. Figure 3 gives a map of the hospital and the two study wards. FIGURE 3 WASCANA HOSPITAL SCCOAIO wfsT h: KVVWWWWWWWI _W\\\\N 1 LOCUS tnn*T A O O M fl*TM«OOMS 46 B. SOBJECTS! The study population was operationally defined as follows: 1. Patient: a person who resides on one of the above wards during the study. Patients v i s i t i n g from other areas of the hospital were excluded. Maximum patient population was sixty-one on Second West, and thirty-three on Station Five. 2. S t a f f s a l l community and consultant physicians treating patients; a l l persons employed by the hospital who provide d i r e c t or i n d i r e c t (e.g. housekeeping) services to patients; i n d i v i d u a l s employed by other organizations who part i c i p a t e i n d i r e c t patient care. Numbers i n each s t a f f c l a s s i f i c a t i o n are given i n Appendix B. 3. Volunteer; a person not employed by the h o s p i t a l or another agency, but who works with patients under the d i r e c t i o n of the coordinator of volunteer services. Approximately twenty volunteers were active during the study. 4* Family: immediate r e l a t i v e s of a patient who take r e s p o n s i b i l i t y f o r his/her well-being. This number varied according to p a r t i c u l a r patients on the wards. Thus, a l l persons involved i n d i r e c t or i n d i r e c t care of patients on a p a r t i c u l a r ward were considered potential subjects, whether or not they belonged to a recognized "team". Contributions of patients and families were also monitored. 47 C. RESJAJCH_MODELi. The r e l a t i o n s h i p between the independent, dependent, and intervening variables i s depicted i n Figure-4. FIGURE,4 RESEARCH MODEL INDEPENDENT INTERVENING DEPENDENT POSITION Environmental Methodological TaBk INTERACTION DYAD Context Artifact Factors PROFILE For purposes of categorizing the independent variable, the following d e f i n i t i o n s were accepted: 1. Position:, "a c o l l e c t i v e l y recognized category of persons for whom the basis for such d i f f e r e n t i a t i o n i s t h e i r common at t r i b u t e , their common behavior, or the common reactions of others toward them (Biddle & Thomas, 1966, p. 29) ." 2- £atient_Positionj. a sub-categorization of patients, based on the common reactions of s t a f f towards them regarding length of hospital stay, physical mobility, and communication d i s a b i l i t y . 3. Sjtaff_Positionj. a c o l l e c t i v e l y recognized category of s t a f f whose common attribute i s a s p e c i f i c job t i t l e (e.g. "housekeeping aide", "director of housekeeping"). 4. Staff .Department;, a small group of persons from d i f f e r e n t positions i n a single hospital department (e.g. Maintenance) who perceive that on the wards they a l l perform 48 s i m i l a r communication functions with s t a f f and with patients. For example, the one e l e c t r i c i a n , one plumber, one carpenter, etc., were a l l c l a s s i f i e d as "maintenance". 5- _?Y.<_di a pair of communicators, i d e n t i f i e d by position (e.g. " s o c i a l worker-physiatrist"). The dependent variable, communication, was r e s t r i c t e d to face-to-face verbal and overt non-verbal transmission of information from one person to another on the ward. I t was operationalized as the Interaction P r o f i l e , a percentaged graph of the freguency of Bales' twelve new communication Categories. For each dyad, two p r o f i l e s were generated—e.g. nurse-doctor and doctor-nurse. Intervening variables were of three types. The f i r s t was labeled "environmental context". "Week", "day", and "time" of observation were hypothesized to have d i f f e r i n g e f f e c t s on the in t e r a c t i o n p r o f i l e . "Locus" of i n t e r a c t i o n ( i . e . where i t occurred on the ward) was another contextual variable. The two l o c i used i n the p i l o t study had a strong moderating e f f e c t . One of those, the "nursing s t a t i o n " , was included i n the present research. The other, a "ward walk", was subdivided i n t o "patient rooms", "cor r i d o r s " , and "charting room". The intervening variable "ward" had a strongly s i g n i f i c a n t e f f e c t i n the p i l o t study. The present research was ca r r i e d out on two types of ward, making Nagi's (1965) d i s t i n c t i o n between "impairment" and " d i s a b i l i t y " relevant. 49 Patients on Second Best a l l had a long-standing " d i s a b i l i t y " or "pattern of behavior that evolves i n s i t u a t i o n s of long-term or continued impairments that are associated with functional l i m i t a t i o n s (p. 103)." In contrast. Station f i v e patients exhibited a range of behaviors from early "impairment" through recognition of "functional l i m i t a t i o n s " to long-standing " d i s a b i l i t y " . A second group of intervening variables was that produced by methodological a r t i f a c t . The only one guantified by the study was the p a r t i c u l a r "observer" who was recording. The t h i r d type of intervening variable, "task f a c t o r s " , referred to the task being accomplished during the communication. The l i t e r a t u r e suggests that task parameters have an impact upon the i n t e r a c t i o n p r o f i l e i n the small group s e t t i n g , but a suitable c l a s s i f i c a t i o n scheme fo r t h i s dimension i s not available. Therefore, a general task categorization was to have been developed during the study. The basis which was to have been used f o r t h i s c l a s s i f i c a t i o n was Dubin*s (1958) c l a s s i f i c a t i o n of four basic behavior systems, which he postulated were c h a r a c t e r i s t i c of every j o b — i . e . , technological (job-specific task a c t i v i t i e s ) , formal (behaviors governed by organizational r u l e s ) , non-formal (the actual way i n which the job gets done), and informal (direct voluntary interpersonal r e l a t i o n s h i p s ) . 50 The above model, while using the terns "independent" and "dependent" variables, did not test the assumptions of caus a l i t y i m p l i c i t i n t h i s usage. The d i r e c t i o n of the re l a t i o n s h i p seemed to be a l o g i c a l o n e — i t i s not as plausible to postulate that interaction p r o f i l e s cause a p a r t i c u l a r dyad to i n t e r a c t , p a r t i c u l a r l y i n the h o s p i t a l s e t t i n g where much of the int e r a c t i o n i s necessitated by the task of patient care—-but t h i s could not be tested i n the present study. The researcher recognizes that communication on a hospital ward does not occur i n i s o l a t i o n - - i t i s influenced by a host of e x t r i n s i c and i n t r i n s i c f a c t o rs. However, the objective of t h i s study was to validate a method, not to generalize the r e s u l t s . Therefore, only one subset of contextual moderators was examined. The decision to l i m i t the scope of the research model i s e s s e n t i a l in the i n i t i a l stages of any research program—the dependent variable must be capable of being accurately measured before i t s in t e r a c t i o n with other dimensions can be comprehensively -assessed. 51 D. STUDY_DESIGN: 1. Research Assistance: Because documentation of the t r a i n i n g required to use the technique was a major objective of the study, a f u l l - t i m e research assistant was employed for the f i e l d study, A twenty-two year old male was selected who had recently graduated i n Psychology, He had a sound background i n research, including ethnography, but no experience i n hospitals (as patient, s t a f f or v i s i t o r ) . This choice of a person with no previous knowledge of the health care - system was deliberately made i n order to balance the researcher 1s -extensive involvement and reduce any bias which might thereby be introduced. 2- Development of_Methodj. A scenario of the f i e l d study, including preparatory meetings and follow-up sessions with the hospital, i s given i n Figure 5. Because many of the methodological d e t a i l s were developed during the f i e l d study, each issue i s examined separately i n t h i s section before summarizing the method used for the f i n a l d ata-collection. 52 FIGURE 5 TIMING OF FIELD STUDY l_l l _ _ . 2 _ L 7 . i _ ~ p o s s i b i l i t i e s explored for conducting the f i e l d study at Hascana Hospital 0__.o_J§£_.___l_2_ - presentation made to the management committee of the SSHC November 12 f 1974 - hos p i t a l commitment obtained January._20-22_._1.975 - preliminary meetings held with hospital administration _______7 - ward s t a f f oriented to study May 20-30 - t r a i n i n g of research assistant begun; s t a f f orientation completed; study explained to a l l patients June 2-30 - coding system developed f o r s t a f f and patients; part-time ward observation commenced; new s t a f f and patients oriented as necessary July 2-15 - f u l l - t i m e ward observation undertaken; time periods f i n a l i z e d July 15-16 - frequency study completed to f i n a l i z e l o c i July 21 - August 3 - Bales' old vs. new formulations evaluated August5-7 - "shakedown" of f i n a l method completed August 9-22 - f i n a l data co l l e c t e d Aug„25- .September 4 - wind-down of study December 9-19 - f i r s t stage of feedback to h o s p i t a l undertaken Fe___a_y_1.0__1.__l 9 76 - second stage of feedback completed a) Preliminary contact with ho s p i t a l . P o s s i b i l i t i e s f o r conducting the f i e l d study at Wascana Hospital were f i r s t discussed with the assistant executive director of the South Saskatchewan Hospital Centre (SSHC) i n the Summer of 1974. A "Summary and implications" of the study (see Appendix C) 53 was prepared f o r submission to the October meeting of the SSHC management committee. after permission was received for Hascana Hospital's p a r t i c i p a t i o n , the researcher met with the administrator, the head of the medical s t a f f , the director and four assistant d i r e c t o r s of nursing, and a l l sixteen c l i n i c a l and service department heads for three days i n January, 1975. The research instrument was explained, and the type of data which the hospital could expect to receive was discussed. Given the choice of two wards, they selected Second West and Station Five i n order to see the contrast between a custodial care and an active treatment ward. A summary of the meetings was distributed to a l l p a r t i c i p a n t s (see Appendix D). The process used to obtain organizational commitment was s i m i l a r to H. Levinson's (1972) approach to obtaining consent f o r diagnostic intervention (p. ,14). P a r t i c u l a r attention was paid to e t h i c a l considerations surrounding the issue of "informed consent" (Argyris, 1968; Ring, 1967). Freedom of exclusion or withdrawal, p a r t i c u l a r l y for persons i n singular positions ( i . e . the only incumbent on that ward), was made e x p l i c i t . Persons i n singular positions who chose to participate were to receive the i n i t i a l feedback on t h e i r i n t e r a c t i o n p r o f i l e s . , I t would be t h e i r prerogative to decide the extent of d i s t r i b u t i o n of that data, a l l patients were to be given the option of exclusion from the data-collection. 54 b) Staff orientation. At the beginning of the f i e l d study a one-half hour orientation was given to a l l s t a f f with possible involvement on the two wards., Each person was provided with a summary sheet, plus a copy of Bales 1 categorization (see Afipendix E. plus F i g u r e ^ , page 25). Host sessions were conducted i n d i v i d u a l l y or i n small groups. Several larger meetings were held when department heads wanted to give a l l s t a f f an overview of the project (e.g. Housekeeping, Physiotherapy). When a l l persons on s t a f f had been informed, newcomers were briefed during the regular hospital orientation for new s t a f f or by i n d i v i d u a l appointments. A policy was established that people's communications would not be monitored u n t i l they had received f u l l d e t a i l s of the project and had given permission f o r i n c l u s i o n . Over the course of the study, three s t a f f members elected to be excluded for various reasons, i n d i c a t i n g that t h i s was a viable a l t e r n a t i v e . Staff i n conversation with a patient were reguested to ask the researcher to leave i f they considered i t advisable. , c) Patient orientation. Explanations to patients were t a i l o r e d to t h e i r l e v e l of understanding. Some patients with normal communication a b i l i t y received a l l iaformation given to s t a f f , including the two handouts.. Others (e.g. with brain damage) were given a s i m p l i f i e d explanation. A number of patients on the chronic care ward were oriented through a s t a f f member interpreting i n t h e i r 55 f i r s t language (e.g. German, Ukrainian). Patients with normal receptive a b i l i t y but severe expressive impairment often reguired several discussions before having the opportunity to have a l l t h e i r questions answered. Only one patient requested to be excluded, and he, after two weeks of being omitted, questioned why he was being " l e f t out." Uhen told that t h i s was at h i s own request, he said exasperatedly, "Oh, that was th,en!", and demanded to be observed from that point on. One other patient with paranoid tendencies was excluded at the suggestion of the s t a f f . New patients did not receive the orientation u n t i l t h e i r second day on the ward, and were not observed u n t i l a f t e r t h i s was completed. Host f a m i l i e s were given an explanation of the study by one of the observers. Hhere t h i s was not possible, patients were asked to relay the information to t h e i r f a m i l i e s . d) Training. A major emphasis was placed on t r a i n i n g during the study. The f i r s t step i n the r e t r a i n i n g of the researcher and the i n i t i a l f a m i l i a r i z a t i o n of the research assistant was the reading of Bales 1 Interaction Process Analysis (1950a), with emphasis on the Appendix, "Definitions of the categories (p.,177-195)," Tapes of radio interviews and "open-line" shows were then used, together with extensive discussion, to e s t a b l i s h consensus on how a p a r t i c u l a r act should be coded. This 56 method of t r a i n i n g was helpful for improving categorizing a b i l i t y , but not for u n i t i z i n g ( i . e . dividing conversations into separate a c t s ) . Mishler and Waxier*s (1968) "Interaction Code Book" was useful as a s t a r t i n g point for the l a t t e r . The observers then moved to an intensive one-week study of English grammar handbooks i n order to be able to guickly i d e n t i f y simple versus complex sentences, clauses versus phrases, etc. At the end of t h i s period, decisions about u n i t i z i n g were made to f a c i l i t a t e in-process scoring (as opposed to Waxier and Hishler's focus on tape-transcipts) . The observational method focused on verbal behavior. Only those non-verbal i n t e r a c t i o n s which were overt and e a s i l y c l a s s i f i a b l e into a p a r t i c u l a r category (e.g. smile, anger) were recorded. The impact of other non-verbal aspects on the reception of the verbal message i s recognized—e.g. support, modification, or negation of the manifest content. However, considerable i n t e r p r e t a t i o n i s generally reguired to assign such behavior to one of the IPA categories. This i s time-consuming and a poten t i a l source of observer expectancy bias. Therefore only simple non-verbal acts were categorized. During the t r a i n i n g period, a c a r e f u l comparison was made between Bales' old and new categorizations. The old one was used for the f i r s t half of the tra i n i n g period, and the new one for the second half. 57 The f i n a l s e l e ction was made i n favor of the new one fo r two reasons. The f i r s t was that the observers found the d i r e c t i o n a l i n d i c a t o r s very helpful i n categorizing. For example, when there was doubt whether an act belonged i n Category 4 (gives suggestion) or 5 (gives opinion), the decision could be made on the basis of ascendance (0), Category 4, versus forward task movement (F), Category 5. The second reason f o r selecting the new version was that the modified Category 2 (dramatizing) appeared to be a more meaningful type of i n t e r a c t i o n behavior on the h o s p i t a l ward than was the old Category 2 (tension release). Because Category 2 included a s i g n i f i c a n t proportion of dramatizing about "out-group" a c t i v i t i e s ( i . e . not related to the ward), these acts were coded separately as Category "M". This method was chosen over Bales* "sender^-to-X" approach because i t preserved the i d e n t i t y of the co-communicator. I t was also a useful d i s t i n c t i o n during the feedback to the hospital s t a f f . After the new categorization had been selected, the sections on "Definitions of the i n t e r a c t i o n categories" (p. 99-135) and Appendix 4, "Revision of the i n t e r a c t i o n categories" (p. 471-491) of Bales* (1970) _fersona_iJ:y_and interpersonal behavior were studied exhaustively. Each observer made notes on key points. These were discussed and c o l l a t e d . The r e s u l t i n g twelve-page summary i n point form was re-read every evening during the f i n a l three weeks of data c o l l e c t i o n to ensure r e g u l a r i t y of coding. 58 e) R e l i a b i l i t y * The major avenue f o r assessing t h i s dimension was the use of the above-mentioned radio tapes. These swere selected because the variety of two-party conversations resembled the types of interchange being monitored on the wards. The h o s p i t a l refused to allow any taping of conversations on the premises, making r e l i a b i l i t y checks on data i d e n t i c a l to the study material impossible. Taping had been used b r i e f l y i n the p i l o t study to e s t a b l i s h i n t r a - r a t e r r e l i a b i l i t y . Although extensive explanations and assurances had been provided to s t a f f , t h i s taping had had a very upsetting e f f e c t on communication patterns, reducing t r u s t i n the research team f o r several weeks. Using the radio-taped conversations, a aaximum of eighty "acts" was coded simultaneously by both observers without i n t e r r u p t i o n . This number was chosen because a majority of conversations on the ward were of l e s s than s i x t y acts* duration. Therefore, the eighty-act c r i t e r i o n was considered s u f f i c i e n t l y rigorous. The tape was replayed act-by-act and consensus reached on the "correct" unit and category. I t was repeated a t h i r d time, with freguent interruptions, i n order that the observers could check t h e i r o r i g i n a l coding and determine the number of "correct" acts, and the number of errors i n categorizing and u n i t i z i n g (both "missed" acts and "extra" ones). A rough r e l i a b i l i t y check was then made for each observer by expressing the number of correct acts as a percent of the t o t a l number of acts. Errors were also subdivided into those of categorizing 59 versus u n i t i z i n g . This procedure provided the observers with feedback on t h e i r types of mistakes, and thus guided the learning process. During the f i n a l three weeks of data c o l l e c t i o n , at le a s t two eighty-act r e l i a b i l i t y checks were made at the end of each day's observations to monitor the observers' minimum r e l i a b i l i t y . One week after the conclusion of data c o l l e c t i o n , six conversations from the beginning, middle, and end of the three weeks were recoded to es t a b l i s h i n t r a - r a t e r r e l i a b i l i t y . f) Staff position coding. A three-digit code was developed for a l l s t a f f positions and departments. The f i r s t d i g i t was generally either a " I " or a "0", depending on whether the position was singular ( i . e . °1" incumbent per ward) or grouped ("0") . The l a s t two d i g i t s were then allocated according to position or department. Three grouped codes were also used, each beginning with a "2"; 260 = group of s t a f f 261 = group of patients 262 = group of s t a f f and patients In addition, when a s t a f f group consisted e n t i r e l y of one position, the l a s t two d i g i t s used were those of that position (e.g. a group of s t a f f nurses was coded "200"). Hherever possible, the nursing d i r e c t o r s , department heads, and other singular positions were coded with the l a s t 60 two d i g i t s of other department members (e.g. housekeeping aides were coded " 0 1 0 M ; the dire c t o r of housekeeping was " 1 1 0 " } . Appendix B gives the complete s t a f f code. g) Patient position coding., Three factors were considered to influence the communications sent to and received by p a t i e n t s — l e n g t h of stay on that ward, physical mobility, and communication impairment. Thus each patient was assigned a t h r e e - d i g i t code, one number f o r each factor (see Ap.pendix_B) . The actual coding procedure varied with the dimension. £i=Sath_of_Stay. was calculated for each patient on the f i r s t day of every week. Mobility for patients on Second West was coded by the researchers according to t h e i r observed physical mobility over the previous week. On Station Five, t h i s was done i n consultation with nursing s t a f f . Patients on both wards remained i n the same mobility c l a s s i f i c a t i o n unless t h e i r l e v e l changed for more than one day.. For example, a patient of code "4" (moves f r e e l y about the ward but cannot transfer from bed to wheelchair independently) who developed pneumonia and was confined to bed was recoded to a " 0 " ; a bedridden patient of code «Q*» who l e f t her room once per week to make a telephone c a l l was not recoded. I n i t i a l l y the research team worked with the two assistant directors of nursing to code each p a t i e n t 1 s communication impairment . This c l a s s i f i c a t i o n was then discussed with ward s t a f f , who disagreed with a number of 61 the p a t i e n t s ' codes. Because the c r i t i c a l determinant of s t a f f - p a t i e n t communication was c o n s i d e r e d to be the s t a f f ' s £__£___!__ °f p a t i e n t a b i l i t y , i t was decided that p a t i e n t c l a s s i f i c a t i o n on t h i s dimension should be done by s t a f f working with them. Nursing was chosen as the c l a s s i f y i n g department because of c o n s i s t e n t c o n t a c t with a l l p a t i e n t s and a v a i l a b i l i t y f o r c o d i n g . Hethods d i f f e r e d on the two wards: __£2__«_il__: Once per week f o r s i x weeks p r i o r to the f i n a l data c o l l e c t i o n , every s t a f f nurse and n u r s i n g attendant working with a group o f p a t i e n t s c l a s s i f i e d each p a t i e n t a c c o r d i n g t o r e c e p t i v e and e x p r e s s i v e impairment. T h i s was intended to be an i n d i v i d u a l assessment, but sometimes c o n s u l t a t i o n d i d o c c u r . At the end of the s i x weeks, a l l p e r c e p t i o n s were t a b u l a t e d , the r e c e p t i v e and e x p r e s s i v e modes obt a i n e d f o r each p a t i e n t , and a communication "code™ assigned-The p a t i e n t r e t a i n e d t h i s code d u r i n g the f i n a l d a t a - c o l l e c t i o n u n l e s s communication a b i l i t y changed markedly.. For example, a p a t i e n t who had a " s t r o k e " and c o u l d no l o n g e r speak would be recoded. S t a t i o n F i v e : Because the communication a b i l i t y of many p a t i e n t s on the r e h a b i l i t a t i o n ward was changing r a p i d l y , the above method of c l a s s i f i c a t i o n c o u l d not be used. I n s t e a d , a t the beginning of every week ( i n c l u d i n g those of the f i n a l data c o l l e c t i o n ) a l l n u r s i n g s t a f f on the ward met f o r f i f t e e n to twenty 62 minutes, discussed the patients, and reached a consensus on their communication impairment. These values were then used for the remainder of the week, except i n cases of a marked change as mentioned above. New patients were coded the day after t h e i r a r r i v a l by the admitting and/or senior nurse. While i t i s recognized that t h i s difference i n methods between the two wards i s not i d e a l , i t was one of the exigencies which the researcher was forced to accept. Staff on the two wards were accustomed to functioning d i f f e r e n t l y , and since they were being reguested to take time from t h e i r busy schedule to do the weekly coding, they needed f l e x i b i l i t y i n t h e i r method of accomplishing i t most e f f i c i e n t l y and e f f e c t i v e l y . h) Day _and t i m e s e l e c t i o n . Since only minimal s t a f f were maintained on weekends, the f i v e weekdays (Monday through Friday) were selected for intensive study. During the early weeks of observation i t was determined that peak periods of i n t e r - d i s c i p l i n a r y communication occurred between 0800 and 1430 hours. Three one-and-one-half-hour time periods within t h i s span were chosen to provide optimum coverage—0800 to 0930, 1030 to 1200, and 1300 to 1430 hours. This s e l e c t i o n of day and time excluded communication between nursing s t a f f during and immediately a f t e r the change of s h i f t , and also conversations between nurses and 63 patients or f a m i l i e s in evenings and on weekends, but i n t e r - d i s c i p l i n a r y interaction was given p r i o r i t y . At the outset of the research, nursing administration requested feedback on s h i f t changes, as they f e l t that communication at t h i s time was poor. , To meet t h i s request, four days following the study were used to code a l l three s h i f t changes on both wards. These data were tabulated separately and presented a f t e r the second feedback session. . 1) Choice_of l o c i , This was accomplished through t r i a l of a number of s i t e s on the wards during the f i r s t s i x weeks of observation, plus an intensive 2-day study of communication frequency at various locations. Decisions made were: (i) nursing st a t i o n - the issue was whether to separate the inner Dedications/service area from the outer desk. I t was decided to keep t h i s as one locus rather than two, as frequency i n both areas was often interdependent, A conversation would begin i n one part and then move to the other, or might be conducted with one person i n each, ( i i ) charting room - freguency here varied markedly with the time and day, as the room was also used for s t a f f conferences and patient c l i n i c s . During a week, s u f f i c i e n t communication occurred to warrant i t s i n c l u s i o n as a locus. A further pragmatic consideration was that the i n c l u s i o n of one lower-freguency locus gave 64 the observer the opportunity to s i t f o r f i v e minutes out of every t h i r t y and record in a more l e i s u r e l y f a s h i o n — a welcome respite from the continuous coding while standing or walking at the other l o c i , ( i i i ) c o r r i d o £ S - reasons for subdividing d i f f e r e d on the two wards: Second Best was H-shaped, with patient rooms down the outer sides of both limbs of the "H," and the nursing s t a t i o n , charting room, and elevator at the crosspiece (see Figure 3, page 4 5 ) . Patients were divided into four "teams". The two on the South side had mostly bedridden patients with low coamunication s k i l l s . The two teams on the North side had patients on a l l communication and mobility l e v e l s . From ethnographic impressions i n the early weeks of observation, i t seemed that patterns of care on the North side d i f f e r e d from those on the South, and therefore the decision was made to separate c o r r i d o r s into "North" and "South." Although one could walk the length of either corridor i n 4 5 seconds, the s t a r t of conversations at one end were missed i f the observer was at the other or had his hack turned at the time. Station Five was more compact, L-shaped, with the nursing s t a t i o n and charting room at the junction of the two limbs (see Figure 3 ) . Therefore, the observer i n one corridor could not see i f conversations were occurring i n the other. Because of t h i s , c o r r i d o r s were subdivided into "North" and "East" for t h i s ward. 65 (iv) rooms - these were divided accordiag to t h e i r adjacent c o r r i d o r s . (v) solarium - t h i s was excluded because i t was rarely used during the day except for overflow of conferences or c l i n i c s from the charting room. (vi) service rooms^bathSP.QI§<.,§tc. - conversations i n these areas were included with "corridors." To improve r e l i a b i l i t y , rules were Bade specifying inclusion/exclusion of conversations between l o c i , at the boundaries of the ward, etc. While s t a f f and patients were aware of the boundaries on each ward, they seldom made an obvious e f f o r t to t a l k outside the area. j) Week. In the o r i g i n a l research design, i t was planned that complete data would be c o l l e c t e d from the two wards for a two-week period. , Observers would alternate wards every day. a f t e r a t r i a l of t h i s system, i t was recognized that each locus was being covered for only f i v e minutes out of every t h i r t y . Infrequent or b r i e f communicators (e.g. physician, occupational therapist, or family) could come, communicate, and leave during the twenty-five minutes that the observer was at other l o c i . Thus, much i n t e r - d i s c i p l i n a r y communication was missed. In order to r e c t i f y t h i s , i t was decided to have two observers on the ward at the same time, but spaced three l o c i apart. This meant that coverage was provided f o r a p a r t i c u l a r locus f i v e minutes out of f i f t e e n , rather than out of t h i r t y . 66 Consequently, the variable "week" could not be f u l l y assessed, since one ward was covered f o r one week, and then both observers moved to the other ward. k) Task .design. One of the intervening variables proposed f o r the research was "task design." While the researcher recognizes that t h i s may be a s i g n i f i c a n t moderator, i t could not be tested during the study., The observers had reached t h e i r maximum recording capacity without t h i s additional variable. The decision was therefore made to leave exploration of t h i s variable to a l a t e r date, when fewer contextual factors were being monitored. E. METHOD! The f i n a l method for data c o l l e c t i o n evolved from the foregoing decisions (see Figure 5, page 52) . A three-day "shakedown" of the procedure was conducted on Station Five, August 5 through 7. The only add i t i o n a l problem encountered was that of exact synchronization of both observers' watches with the ward clock. The only solution to t h i s was the use of stop watches. Held i n the palm of the l e f t hand, and only c l i c k e d at the beginning and end of each ninety-minute period, they were no more obtrusive than the ordinary watches which had previously been used, and were not a concern for s t a f f or patients. . 6 7 The study was designed to spend the week of August 11 on Second West and the week of August 18 on Station Five, However, the observers switched wards for one day each week for two reasons. F i r s t , a regular conference was being held on Station Five on Thursday, August 14, but not the following Thursday because of vacations. Second, since a possible general s t a f f shutdown was imminent due to secondary picketing by an external party involved i n a labor dispute, the ward switch was made i n order to have one day's f i n a l data on Station Five f o r comparison with the shakedown data. Thus, each week's data were composed of Monday, Tuesday, Wednesday and Friday on one ward, and Thursday on the other. Each week's observation began on Sunday, with orientation of new s t a f f , patients, and f a m i l i e s . The patient c l a s s i f i c a t i o n was also revised, and one or two hours' practise observation was done to f a m i l i a r i z e new subjects. Extensive r e l i a b i l i t y checks were also made and f i n e r points of coding were discussed. Observations from Monday through Friday were conducted during the previously discussed time periods (0800 to 0930, 1030 to 1200 and 1300 to 1430 hours). A time sampling of behaviors was employed, with four minutes and t h i r t y seconds spent at each locus, and t h i r t y seconds allowed f o r moving to the next one. Data were recorded d i r e c t l y onto eighty-coluaa Fortran coding forms. The intervening variables were entered i n 68 columns one through sixteen before the ninety-minute observation period began. Then, when a conversation started, i n t e r a c t i o n was immediately recorded as w_o speaks to whom (using the position codes) and how (an act-by-act Bales categorization). The process was similar to that of simultaneous t r a n s l a t i o n . For example, the seguence: DOCTOfi TO NORSE : Good morning. How i s Mrs. Smith? Do yon think we can stop the I.V.? NORSE TO DOCTOR : She had a good night, but she's r e s t l e s s today. would be recorded as follows: 040 : 189-000 : 56-In the above example, the doctor (040) greeted, ("1M) and asked for an opinion ("8") and a suggestion (,,9")« The nurse (000) did not return the greeting, gave an opinion ("5") and information {"6 M). This could be a breakdown i n communication. The sequencing aspects of in t e r a c t i o n were not examined i n the data analysis. However, i f the phenomenon was consistent, the p r o f i l e for doctor-nurse would show a higher percentage of Categories 1, 8, and 9. The nurse-doctor p r o f i l e would have a d e f i c i t of Category 1, and peaks for 5 and 6. Such p r o f i l e comparisons could then be followed by random or s e l e c t i v e examination of seguences. Two l i n e s of the coding sheet were used f o r each conversation—one f o r each person's utterances—moving from l e f t to ri g h t as the dialogue progressed. For example, i f a nurse was ta l k i n g to a nursing attendant, the coding might 69 look l i k e t h i s : 000 : 167 36 66444 41-001 : 65122 3 3 1-In t h i s case, the conversation began with three acts by the nurse COOO") , followed by f i v e by the nursing attendant ("001"), then two by the nurse, etc. When no communications were occurring at a p a r t i c u l a r locus, the observers used the back of the coding sheet to record ethnographic impressions. Loci were rotated systematically i n the following order: 0 = nursing station 2 = S. corridors (Second West) or N. corridors (Station Five) 3 = S. ,rooms (Second West) or N. rooms (Station Five) 1 = charting room 4 = N. corridors (Second West) or E. corridors (Station Five) 5 = N. rooms (Second West) or E. rooms (Station Five) Each day observers began at successive l o c i (e.g. Monday at "0" and "1," Tuesday at n2n and "4") i n order to obtain as comprehensive coverage as possible. Because only f i v e days' observations were completed per ward, l o c i "3" and "5" only reached the s t a r t i n g position once per week, whereas the others did so twice (once for each observer). 70 Depending on the locus, observers were sometimes reguired to c i r c u l a t e i n order to remain accessible to potential conversations. As discussed above, i n the charting room they sat i n a c e n t r a l location where a l l conversations could be heard. In the nursing s t a t i o n , they positioned themselves cent r a l l y and could move immediately to monitor a conversation when i t began. At both these s i t e s i t was possible to code two conversations simultaneously unless one had a very rapid pace. Others were ignored u n t i l the i n i t i a l one was complete. In co r r i d o r s , the observers placed themselves c e n t r a l l y when no conversation was occurring, and moved to communications as they commenced. Monitoring i n rooms reguired continuous walking from room to room i n order to provide egual coverage. Because other conversations could not be seen, observers switched to a new room when a conversation terminated or reached a duration of one minute, whichever occurred f i r s t , i n order to avoid overrepresentation of one pair of communicators. Because of the obvious presence of the research team, ca r e f u l consideration was given to guestions of a r t i f a c t and control (Rosenthal and Rosnow, 1969). The observers had a d e f i n i t e impact on the ward communication structure i n the early weeks of the project. However, after two to three weeks of part-time observation, the s t a f f became accustomed to the presence of the observer. Webb, Campbell, Schwartz, and Sechrest's (1966) discussion of Dnobtrusiye measures was 71 c a r e f u l l y studied for ways to reduce the r e a c t i v i t y of the method, but t h e i r suggestions were not p a r t i c u l a r l y applicable- Instead, the researcher elected to follow Argyris' (1968) approach to minimizing the "Unintended conseguences of rigorous research" by building i n motivation fac t o r s f o r the subject and the organization, and by developing meaningful p a r t i c i p a t i o n . Throughout the study, data sheets were open to examination by subjects. Staff and patients soon r e a l i z e d that nothing which they said could be traced to them as i n d i v i d u a l s , since only coded positions and i n t e r a c t i o n s were recorded. This further reduced apprehension, and thus the observer's impact. On Sundays, when ward a c t i v i t i e s were minimal, observers were available to give further reassurance and explanations. Having the observer record constantly, noting ethnographic impressions i n the conversation breaks, was another means of reducing the obtrusiveness of the method. I t was f e l t that subjects were les s l i k e l y to notice someone who writes continuously than one who does so i n t e r m i t t e n t l y . I t had been suggested that a combination of audio-tape, video-tape and/or interviews be used after the major data c o l l e c t i o n to assess subjects' perceptions of the obtrusiveness of the method 1. This was not possible because 1 This suggestion was made at a workshop of the Organizational Behavior Division of the U.B.C. Faculty of Commerce. 72 of time l i m i t a t i o n s . However, informal discussions with s t a f f and patients during and a f t e r the study indicated that they became guite comfortable communicating i n the presence of observers except when discussing t h e i r home s i t u a t i o n s and personal l i v e s . A r e l a t i v e l y high freguency of negative socio-emotional acts during the month of July indicated that subjects were able to communicate negatively when the occasion arose. In discussions with s t a f f , a t h i r d factor which they mentioned as contributing to open communication was the promise of feedback on t h e i r i n t e r a c t i o n patterns., During orientation sessions, the researcher explained that the data which s t a f f would receive could only be as accurate as t h e i r input of data to the study. In other words, the onus was placed on the s t a f f to communicate as "normally" as possible i f they wished to receive accurate, helpful feedback. F. DA___P___AH_TIO__ A l l data from the developmental three weeks ( " 1 n , " 2 " , and " 3 " ) and the f i n a l three weeks ( " 5 " , " 6 " , and "7") were keypunched, v i s u a l l y examined for errors, corrected, and then reorganized by a s p e c i a l computer program 1. In t h i s manner, the seguential material was transferred to a form 1 This program was developed by L.. Chen. 73 suita b l e f o r further analysis, condensing 13,684 cases of raw data to 10,654. These data were further v e r i f i e d before being compiled into two major f i l e s (one for weeks 1 to 3; the other f o r weeks 5 to 7) using the Statistical„_ac_ages f o r _the_spcial sciences. Version 6.02 (Nie, H u l l , Jenkins, Steinbrenner, & Bent, 1975). Considerable time was reguired for the editing process i n order to eliminate not only keypunching errors missed i n the i n i t i a l v i s u a l e d i t i n g , but also l o g i c a l observer errors (such as miscoding sender/receiver "005" as "003 M for one day). Ethnographic data provided considerable assistance for finding the l a t t e r errors. As discussed above (page 30), Bales outlined four ways of analyzing in t e r a c t i o n d a t a — p r o f i l e s , sequencing of acts, phases, and matrices. In the present study a who-to-whom position matrix was used. For each c e l l i n the matrix, two int e r a c t i o n p r o f i l e s were determined—one for the sender, and one for the receiver. Percentaged p r o f i l e s were calculated (the freguency of each category converted to a percent of the t o t a l interactions) to standardize the data. This created some concern f o r i p s a t i v i t y (Hicks, 1970), since there i s a weak mathematical interdependence of the twelve percentages—i.e. when one increases, another must decrease. However, indica t i o n s from the p i l o t study data were that t h i s interdependence would be minimal. Matrices and p r o f i l e s were tabulated for each observation period, and the moderating s i t u a t i o n a l variables noted. 74 Complete sequencing of acts was not examined because of the magnitude of the data-handling problem. However, t h i s information was available from the raw data to answer s p e c i f i c questions during the feedback process. For example, i f the i n t e r a c t i o n p r o f i l e s for a p a r t i c u l a r dyad showed a pote n t i a l source of communication inconsistency, a random or selected sample of th e i r interactions could be analyzed sequentially. When Bales and Strodtbeck (1951) examined the phases of problem-solving groups, they defined phases as " q u a l i t a t i v e l y d i f f e r e n t sub-periods within a t o t a l continuous period of interaction i n which a group proceeds from i n i t i a t i o n to completion of a problem involving group decision (p. 485)-" They found that f o r a pr;oble.m-soiivjnq group, movement occurred over time from problems of communicaton through evaluation to control (see key to £iaM£e_2, page 25). The group concurrently had an increase i n the r e l a t i v e freguency of both positive and negative reactions. Phases were not assessed i n the present research f o r two reasons. The f i r s t was that ongoing communication on a hospital ward i s not i d e n t i c a l to a "continuous period of i n t e r a c t i o n " i n which a problem i s i d e n t i f i e d , i t s ramifications explored, and a solution reached. More often the problem i s solved through a number of discrete interactions over an extended period of time. The second reason was that, as discussed above (page 37 ), a "group 75 decision" i s r a r e l y , i f ever, made i n the medical context. The two major data sets were grouped according to dyad, and int e r a c t i o n p r o f i l e s calculated for each who-to-whom combination. Those with less than f i v e time periods or le s s than f i f t y acts were eliminated (unless they completed a dyad pair i n which the other half met these c r i t e r i a ) , and the remaining p r o f i l e s were examined f o r consistency over time. These two sets were not d i r e c t l y comparable f o r two reasons. The f i r s t was that weeks 1-3 used Bales 1 old categorization system, and weeks 5-7, the new. , This may have resulted i n major differences i n Category 2, and minor differences in several others. Because of the weak i p s a t i v i t y of the percentaged p r o f i l e s , these changes contributed an unknown variance to the two sets., The second, and more s i g n i f i c a n t a l t e r a t i o n , occurred because of an unanticipated intervening variable which was not being monitored during the study—the weather. Hascana Hospital, a two-storey frame building, had no a i r conditioning (except i n certain offices) and l i t t l e a i r c i r c u l a t i o n because of ward design. Therefore, when temperatures i n the f i r s t week i n July reached 38 degrees Celsius, and humidity was over 7 0 % , wards quickly became unbearable f o r s t a f f and patients (and observers!). Working and l i v i n g under such conditions, tempers f l a r e d guickly and tension l e v e l s rose. Interestingly, the observational instrument proved 76 s u f f i c i e n t l y s e n s i t i v e to p i c k up these changes i n emotional c l i m a t e . When the p r o f i l e s f o r the J u l y data s e t were examined, the percentages of a c t s i n C a t e g o r i e s 11 and 12 were almost double those f o r August, when t e a p e r a t u r e s were lower. T h i s f i n d i n g was c o n s i s t e n t f o r most s t a f f - t o - s t a f f , s t a f f - t o - p a t i e n t , and p a t i e n t - t o - s t a f f i n t e r a c t i o n s . Because o f the mathematical interdependence o f the percentages, the r e l a t i v e f r e g u e n c i e s o f the o t h e r ten c a t e g o r i e s were lowered and t h e two s e t s c o u l d not be d i r e c t l y compared. The importance of the a l t e r a t i o n i n the p r o f i l e s c o r r e s p o n d i n g to f l u c t u a t i o n s i n t h e weather w i l l be d i s c u s s e d i n Chapter IV. Chi-sguare s t a t i s t i c s were c a l c u l a t e d f o r C a t e g o r i e s 1 through 12, and f o r 3 through 10, on f o r t y - s i x of the p r o f i l e s . , Both a n a l y s e s showed a c o n s i s t e n t p a t t e r n a c r o s s the two data s e t s , g i v i n g some i n d i c a t i o n of p r o f i l e s t a b i l i t y over time (greater on the 3-10 a n a l y s i s ) . Two s p l i t - h a l f checks were a l s o done on the major p r o f i l e s "nurse-nurse" and "nurse attendant-nurse a t t e n d a n t " w i t h i n the second s e t , with s i m i l a r r e s u l t s . P r o f i l e s f o r the second data s e t with g r e a t e r than f i v e p e r i o d s and f i f t y a c t s (a t o t a l o f 309—118 f o r Second west and 191 f o r S t a t i o n Five) were p l o t t e d , u s ing another s p e c i a l computer program 1. These formed the b a s i c feedback data f o r the h o s p i t a l . Graphed p r o f i l e s f o r both p a i r s of a 1 T h i s program was w r i t t e n f o r the study by P. Cahoon. 77 dyad (e.g. nurse-doctor, and doctor-nurse) sere reduced i n size and transferred to a singl e sheet of paper. _ i _ _ r _ _ 6 gives an example of the p r o f i l e s f o r one dyad. Sets of p r o f i l e s were compiled for each s t a f f position and s t a f f department p a r t i c i p a t i n g i n the study. A single information sheet was written to accompany the p r o f i l e sets, with i n s t r u c t i o n s for t h e i r i n t e r p r e t a t i o n (see A__e,ndix_G). Department heads received copies of a l l p r o f i l e s of t h e i r s t a f f , accompanied by tables giving the exact values f o r the graphs. The f i n a l three weeks" data were then divided into three days* "shakedown" on Station Five and f i v e days* f i n a l data for each ward. A l l dyads with l e s s than f i v e time periods or l e s s than one hundred acts were eliminated, together with those involving groups ("260", e t c . ) , s e l f ("900"), or patients ("300" through "659"). The regaining s t a f f - t o - s t a f f data were used for the s t a t i s t i c a l t e s t s of the hypotheses. WSCJWq HOSPITAL CCmUNJCATIW PRCFJ1.E PERCENTAGE OF RCTS -H ERCH CATEGORY FOR CATEGCRT VflSCBNfi H0SP1TM. C O W U N I C B T r W PROFILE PERCENTAGE OF ACTS IH EACH CATEGORY FOR SECOND WEST GENERAL PRRCTITIOl'JER PERIODS^ TO B C T S = REGISTERED NURSE 79 CHAPTER i l l A. PRELIMINARY ASSESSMENT OF CATEGORY USE; When the data had been edited and c o l l a t e d . Phi c o e f f i c i e n t s were calculated on every ward/week combination to determine d i f f e r e n t i a l category usage f o r the two observers. Phi i s a measure of association s i m i l a r to Chi-sguare. I t t ests the difference between two categorical ratings but i s independent of the sample siz e (N). I t has a minimum value of - 1 . 0 0 and a maximum of + 1 . 0 0 . The optimum value i s 0 . 0 0 . _________ reports Phi c o e f f i c i e n t s greater than ± 0 . 0 9 , and mean values (X) for the two data sets by ward. For weeks 1 - 3 , ten out of seventy-eight Phi c o e f f i c i e n t s reached a value of ± . 1 0 ; for weeks 5 - 7 , seven out of fift y - t w o were at or beyond t h i s l e v e l . Tests of the significance of Phi ( i . e . the n u l l hypothesis that Phi=0) can be obtained from the knowledge that N times Phi-sguare eguals Chi-sguare. However, because of the large sample size i n t h i s study, the n u l l hypothesis would be rejected i n a l l cases even though Phi i s very small. This would be e n t i r e l y an e f f e c t of the large sample s i z e , and not a r e f l e c t i o n of any strong association. 80 TABLE_III STRENGTH OF PHI VALUES FOB CATEGORY USE-BY OBSERVER CatO Cl C2 cr. C3 C4 C5 C6 C7 C8 C9 CIO C l l C12 1 2W .10 .11 .21 35 .12 - t* • 2W .11 z 65 .10 .20 i) 2W . 12 S5 .11 • .1'* .13 J 123 zv .11 S5 .11 . !5 | 5 (2W) r 35 * 2V .10 6 S5 .1? .10 .12 7 2¥ : i 2 • '-5 S5 • 13 X 56? 2W . i i S5 • NO DATA WERE COLLECTED FOB 2K IN KEEK 5 J . A. Davis (1971) has outlined an ar b i t r a r y set of conventions for describing the strength of the association between two nominal variables as measured by the s t a t i s t i c "Yule*s Q. ** According to h i s c r i t e r i a , Q values between ± .01 to .09 should be interpreted as a "negligible positive or negative association", and those i n the range .10 to .29 as a "low association"., Since Phi i s si m i l a r to Q, i t i s reasonable to use these c r i t e r i a when interpreting the above findings. There was therefore only a ne g l i g i b l e or low association between o v e r a l l category use by the two observers on either ward. 81 / B., RELIABILITY! Hany of the communication studies reviewed have used Pearson R to assess i n t e r - and int r a - r a t e r r e l i a b i l i t y . As discussed by W. A. Scott (1955, p. ,323), t h i s s t a t i s t i c i s only appropriate when dealing with i n t e r v a l or r a t i o data. Since IPA i s a nominal categorization, use of R was considered inappropriate. Consequently, i t was not possible to use standard r e l i a b i l i t y measures such as Kuder Richardson 20. The taped r e l i a b i l i t y data were therefore tested by two alternate s t a t i s t i c s — C o h e n ' s Kappa (percentage act-by-act agreement with chance removed) (1960), and Scott's P i (marginal agreement with chance removed) (1955) . Two approaches were used because a single s t a t i s t i c i s not available f o r handling act-by-act agreement f o r in-process scoring, where errors of u n i t i z i n g are found i n addition to those of categorizing. Cohen's Kappa, used extensively by Waxier and Hishler (1966), handles categorizing errors, but when adapted to account f o r u n i t i z i n g errors i t underestimates the r e l i a b i l i t y . , In both instances, one of Cohen's assumptions—i.e. that there i s no c r i t e r i o n f o r correctness which can be used f o r comparison—was violated by the r e l i a b i l i t y data. Nevertheless, the importance of obtaining a measure of act-by-act r e l i a b i l i t y during the t r a i n i n g stages of the research necessitated the use of Kappa at that time. , During 82 the f i n a l data analysis, an assessment of marginal category t o t a l s was desirable, since the data were to be used i n that form. Pi was therefore used, even though i t did not account for consistent differences between observers. Scott's P i , an index of inter-coder agreement, was designed for use i n survey research "to neasure the r e l i a b i l i t y of c l a s s i f y i n g a large number of responses i n t o nominal scale categories (W. A. Scott, 1955, p. 321)." I t ranges from 0.00 to 1.00, and assumes that the d i s t r i b u t i o n of proportions over categories for the population i s known and i s egual f o r both raters. This assumption was not met by the r e l i a b i l i t y data. P i i s calculated by the formula: I - P± where P0 = observed percent agreement P e •= percent agreement expected on the basis of chance Cohen's Kappa was also designed for s t a t i s t i c a l testing of nominal data, but on an act-by-act basis. I t makes the following assumptions: (a) units are independent (b) judges operate independently (c) categories are independent, mutually exclusive, and exhaustive According to Bales' theory of communication, these three assumptions were met by the data, with the possible contaminating influence of seguential interdependence. 83 This s t a t i s t i c i s used in si t u a t i o n s where there i s no c r i t e r i o n f o r correctness. The formula i s : Po - R l - Pc where P 0 = proportion of units i n which the judges agreed P c = proportion of units for which agreement i s expected by chance I t i s calculated from the actual d i s t r i b u t i o n of assignment of acts to N categories by two observers, resulting i n an N by N matrix. Although the formulae for the two s t a t i s t i c s appear very s i m i l a r , "Pc" and «pe" are defined d i f f e r e n t l y . For P i , "Pe" i s the sguared mean difference between the t o t a l acts assigned to each category by the two raters: P.. * P. a X P* + Pa 1 0 0 where Pfl = __ P^ + • P 6 = C P<-6 +.....+P*6 In contrast, the "Pc" for Kappa i s defined as the proportion of acts expected to be assigned to a particular c e l l by both observers on the basis of chance, given d i f f e r i n g marginal t o t a l s , i . e . the joint p r o b a b i l i t y of the marginal proportions: (P,.)(P..) p where P = P„ = P. 84 Pi s t a t i s t i c s f o r the three weeks of August 3 to 22, i n c l u s i v e , were: August 3 - 7 .91 10 - 15 .73 17 - 22 .71 ov e r a l l r e l i a b i l i t y f o r three weeks was .86 R e l i a b i l i t y using Kappa was calculated i n two ways; 1. categorizing errors only -August 3 - 7 .66 10 - 15 .71 17 - 22 .61 o v e r a l l r e l i a b i l i t y .67 2. categorizing plus u n i t i z i n g errors -August 3 - 7 .58 10 - 15 .57 17 - 22 .48 o v e r a l l r e l i a b i l i t y .56 Given the variation i n the above figures, i t i s d i f f i c u l t to draw conclusions about the r e l i a b i l i t y data. Since the study data were analyzed as percentaged p r o f i l e s of category t o t a l s , the figures f o r P i are the most appropriate to consider at the present stage of analysis. These are within acceptable range, although the deterioration across the three weeks should be noted (and w i l l be discussed i n Chapter IV). The o r i g i n a l objective of using Kappa during the developmental stages of the research to monitor the progress of achieving act-by-act r e l i a b i l i t y could not be met because of the lack of s u i t a b i l i t y of the s t a t i s t i c . This was an 85 unanticipated problem. The study has therefore uncovered the need for future dialogue with s t a t i s t i c i a n s to develop a suitable measure of act-by-act r e l i a b i l i t y f o r in-process scoring of inte r a c t i o n data. Intra-rater r e l i a b i l i t y was determined by P i to be .80 for the researcher, and .86 for the research a s s i s t a n t . An explanation for t h i s difference w i l l be presented below. C. _E_L_S_OF__HE__YP It w i l l be r e c a l l e d that the f i r s t part of Hypothesis 1 was: Hl-a The paired Categories ( i . e . 1-12, 2-11, etc.) are sequentially interdependent. After the data were c o l l a t e d , the researcher r e a l i z e d that Hl-a could only be assessed by sequential analysis of the data. As previously discussed, t h i s i s an extremely time-consuming and arduous task which could not be done i n the present study. This hypothesis was therefore not tested. The next two hypotheses were examined together; Hl-b The r e l a t i v e freguencies of the twelve categories are independent. H2 Three of the Categories (2, 6, and 11) are more strongly correlated with the others than are the remaining nine. 86 These hypotheses sere tested by a series of R-type factor analyses on three subsets of the f i n a l data—each ward aggregated by day, observer and dyad, plus a combined analysis of ward, observer, dyad. In each case, p r i n c i p a l components analysis was used, with i t e r a t i o n s to es t a b l i s h communalities. By performing both orthogonal (varimax) and obligue (direct oblimin) rotations, orthogonality of the variables could be v e r i f i e d . The number of factors was set at twelve, with no minimum Eigenvalue, to determine whether each category would load on a separate factor. T_bles_I______a____I give the varimax rotated factor matrices, which assume orthogonality as conceptualized by Bales. A l l three analyses provided strong support for the orthogonality of Bales' new IPA categories. . Excluding Category 6 (which w i l l be discussed below), there was no i n d i c a t i o n that Bales' new categories tended to load together on the f a c t o r s . Hl-b was thus confirmed i n t h i s analysis. Removing the assumption of orthogonality, the three obligue rotations summarized i n Tables - Vill, - - y i l l - i rand, IX showed a weak interdependence of some categories, but t h i s was minimal. In general, varimax loadings of .10 to .19 tended to increase on the obligue structure to between .20 and .29 when the orthogonality r e s t r i c t i o n was l i f t e d . These values were s t i l l low, r e l a t i v e to the high loading of the single dominant variable. This analysis further supported Hypothesis 1-b. , 87 ________ VARIMAX ROTATED FACTOR MATRIX—BOTH WARDS VARIABLE FACTOR 1 FACTOR 2 FACTOR 3 FACTOR U FACTOR 5 FACTOR 6 FACTOR 7 FACTOR 8 FACTOR 9 FACTOR 10 FACTOR 11 FACTOR 12 P 1 .98 P 2 -.9* P 3 • 95 ?4 .97 P 5 -.22 .21 • 93 P6 -A3 -.1*1 -.36 • 50 P 7 .96 P8 .95 P9 .96 P 10 • 96 P 11 .96 P 12 -.21 -.91 ' -23 (K • kk) ONLY FACTOR LOADINGS > .20 ARE REPORTED _______ VARIMAX ROTATED FACTOR MATRIX—SECOND WEST VARIABLE FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FACTOR 6 FACTOR 7 FACTOR 8 FACTOR 9 f w c r o n 1 0 FACTOR 1 1 FACTOR 1 2 P 1 -.96 P 2 -.97 0 P 3 -.97 P4 .97 P5 • 97 P6 .38 - . U 2 - . 2 3 • 5 0 - . 2 5 - . 2 1 P 7 .97 pe - .96 . 2 2 P9 .99 P 1 0 . 2 1 * . - 9 5 p 1 1 .98 P 1 2 .99 (if - ne) ONLY FACTOR LOADINGS > .20 ARE REPORTED 88 TABLE_VI VABIHAX ROTATED FAG TO g M AT R I S TATIQ N FIVE VARIABLE FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FACTOR 6 FACTCR 7 FACTOR 8 FACTOR 9 FACTOR 10 FACTOR 11 FACTOR 12 P I .86 P2 .97 J J . 7 * P » .79 PS .78 P6 .90 P 7 . 8 5 P 8 .69 P 9 .71 P 10 •37 p i l .66 P 12 •38 (R - 160) ONLY FACTOR LOADINGS > .20 AND REPORTED TABLE VII QBLI0.UE_FACTOR_STROC^ VARIABLE FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FACTOR 6 FACTOR 7 FACTOR 8 FACTOR 9 FACTOR 10 FACTOR 11 FACTOR 12 p 1 • 3 3 . 2 7 -1.00 . 2 2 P 2 .27 - 1 . 0 0 .24 • 3 5 P 3 • 3 3 - 1 . 0 0 . 2 7 P » .28 • 3 5 - .26 - 1 . 0 0 P 5 - . 3 0 • 3 6 - . 3 6 . 2 2 - 1 . 0 0 P6 . 1 . 0 0 .28 . 2 7 • 3 3 - . 2 3 - 4 5 - . 3 0 .28 P 7 .24 1 . 0 0 • 36 P8 . 2 3 - .28 .99 .21 • 3 1 P 9 .31 • 3 6 • 99 P 10 - .99 .28 - .26 P 11 - .24 .21 1 . 0 0 • 3 6 P 12 .28 - 1 . 0 0 3 3 • 3 6 (I - 44) ONLY FACTOR LOADINGS > .20 ARE REPORTED TABLE_VIII OBLIQUE FACTOR STRUCTURE—SECOND BEST V A R I A B L E F A C T O R 1 F A C T O R 2 F A C T O R 3 F A C T O R 4 F A C T O R 5 " A C T O R 6 F A C T O R 7 F A C T O R 8 F A C T O R 9 F A C T O R 10 F A C T O R 11 F A C T O R 12 P 1 .29 -1.00 .22 P2 .25 -1.00 .24 P 3 -1.00 .26 P4 .24 .23 .22 -1.00 P5 .21 .21 -1.00 P6 .26 -1.00 .29 .23 P 7 -1.00 .25 .21 P8 .46 1.00 P 9 1.00 P 10 1.00 -.21 .46 P 11 1.00 - .21 P 12 .21 1.00 (H - 118) ONLY FACTOR LOADINGS > . 2 0 ARE REPORTED TABLE_IX QBLI<20E_FACTO RESTRUCTURE V A R I A B L E ! F A C T O R 1 F A C T O R 2 F A C T O R 3 F A C T O R 4 F A C T O R 5 F A C T O R 6 F A C T O R 7 F A C T O R 8 F A C T O R 9 F A C T O R 10 F A C T O R 11 F A C T O R 12 11 .26 -.88 P2 -1.00 -.28 P 3 -.21 .76 P4 .22 .23 .82 -.23 • P5 -.81 • 31 P6 -95 .28 -.20 -.25 -.26 P 7 .88 P 8 -.70 P9 -.72 P 10 • 38 P 11 .67 P 12 -.49 (N - 160) ONLY FACTOR LOADINGS > .20 ARE REPORTED 90 The c l a r i t y of these findings raised the guestion of whether the factor structure was actually as simple as i t appeared, or could t h i s be a s t a t i s t i c a l a r t i f a c t ? A par t i c u l a r concern was the lack of an obvious order f o r variable loadings on the fa c t o r s . Tabie_X gives the major variable loadings. Eigenvalues, and proportions of the variance explained, f o r each of the three data sets. TABLE .X EIG___AL_____AJO__V FOR EACH FACTOR—WARDS SEPARATE AND COMBINED FACTOR BC EIGENVALUE TK WARES VARIABLE % VARIANCE SE( EIGENVALUE 30ND WEST VARIABLE % VARIANCE ST/ EIGENVALUE ITI0N FIVE VARIABLE % VARIANCE 1 2.17 P9 18.2 1.88 P2 15-7 1.41 re 21.0 2 1 .87 71 15.6 1 . 5 0 P4 12.5 1.19 PI 17.8 3 1.68 P 3 14.1 1.44 F3 12.0 .82 P7 12.2 4 1.19 r? 10.0 1 . 2 7 Pll : 10.6 • 72 P5 ] 0 . 7 5 1 . 1 3 pi 9 . 5 1.08 PI 9.0 .69 P 3 10.2 6 1.04 P10 8 . 7 . 9 9 P7 8 .3 • 57 P9 8.4 7 .89 P3 7.5 . 9 7 16 8.1 . 5 ' * P4 8.1 8 .69 Pll 5-8 .89 P5 7.4 .43 6 . 3 9 • 5 5 I>4 4.6 .74 P12 6 .2 .37 Pll 5-6 10 .42 P12 3 - 5 . 7 3 10 6.1 .18 P12 2 . 7 11 . 3 2 P5 2.6 .50 P10 4 .1 •03 P6 • 5 12 - 0 . 0 0 • 0 . 0 • 0 . 0 0 • 0 . 0 -23 P10 -3.6 When the variable/factor r e l a t i o n s h i p was examined more c l o s e l y , some consistency emerged. Categories 2, 3, 7, 1, 8, and 12 loaded i n t h i s order on the factors of the combined wards. This r e l a t i o n s h i p was also evident on the Second West fac t o r s , except that P7 and P1 were reversed. 91 The Station Five factors had P3 and P1 reversed* but otherwise the order confirmed the other two analyses. Therefore, only Categories 4, 5 # 9, 10, and 11 loaded randomly. The lack of t o t a l consistency i s believed to have occurred because there was no single dominant fac t o r which would account for most of the data variance. In Table X. a l l Eigenvalues are r e l a t i v e l y close i n magnitude. No single factor makes a major contribution to the explained variance. Hypothesis H2, which evolved from the analysis of the p i l o t study data (see Appendix A), was p a r t i a l l y supported by two out of the three varimax factor analyses. Category 6 (gives information) loaded moderately (.30 to -50) on four of the factors, and weakly (.20 to .29) on three others i n the Second west data. Using the combined ward data. Category 6 loaded moderately with Categories 2 and 4, and weakly or moderately with 11 and 12. On the Second West data. Categories 3 and 5 also had a moderate-to-weak contribution to t h e i r factors by Category 6. No such pattern was found i n the varimax analysis f o r Station Five. When the d i r e c t i o n of the varimax loadings was examined, a l l fac t o r s but one (Table^IV, Factor 10) were i n the reverse d i r e c t i o n to the other variable. That i s . Category 6 was negatively associated with Categories 1,2, 3, 4, 11 and 12 (those at the extremes of the categorization hierarchy). Examination of the obligue factor structure 92 revealed a s i m i l a r but accentuated pattern because of the general increase i n the factor loadings. T a b l e X I summarizes the number and d i r e c t i o n of Category 6's co-loadings without regard for t h e i r magnitude. TABLE_XI N0MBER_M2_2IRECTION_OF_C Direction Rotation Cl C2 C3 Ch C5 c? C8 C9 CIO C l l C12 varimax 2 2 1 2 1 2 1 reverse oblique 7 3 2 6 1 3 2 2 if . 2 total 9 5 3" 8 2 3 2 2 6 3 varimax same oblique 1 2 k 2 5 1 1 total 1 2 4 2 6 1 - - 1 In general, Category 6 had a strong negative association with Categories 1, 2, 4, 8, and 11, and a weak negative association with 9, 10, and 12., It had a strong positive association with Category 5 (the c e n t r a l , task dimension of giving opinions). Findings for Categories 3 and 7 were ambiguous. I t therefore appears that, using Bales* new categorization, Category 6 remains an adjunct to most other communication behaviors, rather than a unique dimension i n and of i t s e l f . However, Categories 2 and 11, which under the old system also appeared to be adjunctive, have i n the new one been refined to the point where they emerge as unigue types of i n t e r a c t i o n behavior. 93 The oblique rotations were further examined to determine whether any trends were evident towards factors loading along the d i r e c t i o n a l dimensions hypothesized by Bales to correlate with each of the new categories. No such pattern emerged. Hypothesis 3 was stated as follows: H3 Bales 1 twelve int e r a c t i o n Categories are i n c l u s i v e of a l l types of verbal communication behavior. This hypothesis was tested informally by the number of acts coded into a thirteenth category ("0 , , =not codable). Table XII gives the t o t a l number of acts i n t h i s r e s i d u a l category by week, plus t h e i r contributions to the percentage p r o f i l e s . TABLE XII ABSOLUTE S RELATIVE, FREQUENCY OF CATEGORY ,0-BX MEEK WEEK 1 WEEK 2 WEEK 3 WEEK 5 WEEK 6 WEEK ? CATEGORY 0 - ABSOLUTE FREQUENCY 24 170 40 220 406 470 CATEGORY 0 - PERCENT OF TOTAL .40 % 1.68 % 1.80 % 2.16 % 2.62 % 2.85 % TOTAL ACTS 5.901 10,094 2,223 10,178 15.504 16,511 This hypothesis was not" submitted to a formal test of sig n i f i c a n c e using Chi-sguare f o r the same reasons given above (page 80), but the data appear to support H3. The maximum value for Category 0 i s 2.8% of the t o t a l acts by a l l s t a f f and patients. The s l i g h t but steady increase i n 94 percentages between weeks 1 and 7 l i k e l y r e f l e c t s an observer fatigue f a c t o r . Hypotheses 4, 6, 7, and 8 were tested by a se r i e s of Multivariate Analyses of Variance (MANOVA). The data-handling program used was the University of C a l i f o r n i a , Los Angeles' BMD:X63, a MANOVA program which allows f o r unbalanced design and unegual c e l l freguencies. Because of program l i m i t a t i o n s , only the main e f f e c t s could be obt a i n e d — i n t e r a c t i o n e f f e c t s were grouped with the error term. A series of p a r t i a l l y complete MANOVA's was executed to examine combinations of the following independent and intervening variables: position dyad, week, day, time period, ward, locus, and observer. 1. Week: Although t h i s variable could not be rigorously tested from the f i n a l data, some i n d i c a t i o n of i t s impact could be gained by comparing three days' data from Station Five i n the i n i t i a l "shakedown" week with the same three days from the f i n a l data set. A subset of twenty-one dyads common to both weeks was selected, each with a minimum of f i v e time periods and one hundred acts. Data were then aggregated i n two ways: WEEK by DAY by OBSERVES by LOCUS ("WDOL") WEEK by OBSERVER by DyAD ("WODy") 95 after aggregation, category t o t a l s and percentage p r o f i l e s Were calculated for each combination. A l l l e v e l s of independent and intervening variables were then dummy coded, and t h e i r e f f e c t on the dependent variable p r o f i l e assessed. The sig n i f i c a n c e l e v e l of the F - S t a t i s t i c s f or the main e f f e c t s i s given i n Table..XIII,. TABLE_XIII aANpJA-^SIGNI^FIC^HCE OF F-STATISTICS .FQa^MSIK TOOL VODy Veck .001 .001 Day' < ns Observer .001 .001 Locuo ns Dyad .001 N-70 K-83 2- Wardi Again, a subset of 11 dyads common to both wards during the f i n a l data c o l l e c t i o n was used, with the same minimum freguency c r i t e r i a . Findings are summarized in° Table XIV. TABLE_XIV MAHOVA--SIGNIFICANCE OF,F-STATISTICS FOE-WARD -VdDOL VdODy Yard na .05 Har .005 Observer .001 .01 Locus .025 Dyad .001 K-113 96 3- Day. Observer, Time Period,_Locus: Two analyses were completed for these intervening v a r i a b l e s — o n e f o r each ward—using the minimum c r i t e r i a outlined above. In order to reduce the number of c e l l s i n the analysis, the time of day was reduced to three major "time periods"—0800 to 0930, 1030 to 1200, and 1300 to 1430 hours for findings on these four variables, see Tabi___V. TABLE XV MANOVA—SIGNIFICANCE OF F-STATISTICS FOR ' DAY. OBSERVES, TIME PERIOD. LOCOS DOTL(2tf) D0T_(S5) fey •05 •005 Observer . 0 0 1 .001 i'lme Ferloc ns ns Locus .05 • 05 K-165 N-133 Very s i m i l a r patterns were obtained on the two wards, but "day" was l e s s s i g n i f i c a n t on Second Best. Implications of t h i s w i l l be discussed below. 4. Day, Observer, Dyad: The two most s i g n i f i c a n t variables from the preceding a n a l y s i s — i . e . day and observer—were then combined with dyad. Here an i d e n t i c a l pattern emerged for the two wards (see Table_XVI). 97 TABLE XVI A—SIGNIFICANCE OF F-STATISTICS FOR D0Dy(2W) D0Dy(S5) .025 .025 Oboervor .005 .001 Dyad .001 .001 H-no N-160 One question raised by the above analyses was whether d i f f e r e n t ' patterns of si g n i f i c a n c e might emerge f o r grouped roles versus singular ones. Two separate MANOVA's were performed f o r each ward—one on a l l freguent p r o f i l e s where both members of the dyad were grouped positions, and one where either or both members were singular. Results were complicated by d i f f e r i n g d i s t r i b u t i o n s of positions on the two wards. On Station Five, a majority of the positions were singular; on Second Hest, most were grouped. Allowing for the d i f f e r e n t N's involved, patterns of s i g n i f i c a n c e did not appear to d i f f e r markedly i n the two types of an a l y s i s , with one exception. On both wards, locus reached the p < .001 l e v e l f o r singular r o l e s , and was only s i g n i f i c a n t between p < .10 and p < .05 for grouped roles. Summarizing the findings relevant to Hypotheses 4 through 8, the following assessments can be made: H4 The major independent variable which influences the communication p r o f i l e i n a hospital ward i s the p o s i t i o n dyad. This was strongly supported (p < .001). i n a l l analyses. 98 H5 Task fact o r s have a weaker, but s t i l l s i g n i f i c a n t , e f f e c t on the i n t e r a c t i o n p r o f i l e s . This could not be tested, for reasons stated above. H6 The contextual factors of day, week and time do not have a s i g n i f i c a n t e f f e c t , when controlled for other intervening variables. . Results varied according to the par t i c u l a r variable. Day was not s i g n i f i c a n t i n the HDOL analysis, when only three days' data were used. Apart from that, i t was generally s i g n i f i c a n t between p < .05 and p < .01. This e f f e c t was accentuated on Station Five, where - events consistently varied from day to day because of weekly c l i n i c s and team conferences, preparation for weekend leaves, admissions, discharges, etc. Week was highly s i g n i f i c a n t . . This unanticipated e f f e c t i s discussed below under "observer." Time period was consistently not s i g n i f i c a n t . H7 The lo c a t i o n of i n t e r a c t i v e behavior (i.e..ward, locus) has a strongly s i g n i f i c a n t mediating e f f e c t on the p r o f i l e s of a l l dyads. This prediction was p a r t i a l l y upheld. Hard was i n c o n c l u s i v e — i t was s i g n i f i c a n t beyond the p < .05 l e v e l when the data were analyzed one way; the other way, i t was not s i g n i f i c a n t . Locus, with the exception of the "week" analysis (using fewer dyads and days), was s i g n i f i c a n t at between p < .05 and p < .025, supporting t h i s part of the hypothesis. The observed difference between singular and grouped positions 99 i s understandable, since several of the single incumbents worked mainly at the nursing station (e.g. senior nurse, ward clerk) . H8 The e f f e c t of a s p e c i f i c observer does not produce variation i n communication p r o f i l e s . This hypothesis was d e f i n i t e l y refuted. Possible explanations f o r t h i s finding are discussed i n Chapter _ I V . D. HOSPITAL FEEDBACK: As part of the researcher's commitment to the h o s p i t a l , i t was agreed that objective information on communication patterns could be provided to s t a f f , department heads, and administrators. The ho s p i t a l was aware from the outset that the method was not s u f f i c i e n t l y refined to separate "good" from "bad" communication, except as related to general communication p r i n c i p l e s . What could be provided was a mirror which would r e f l e c t to the participants t h e i r communication patterns. The process of i n i t i a l feedback which the subjects and researcher agreed upon was as follows: basic l e v e l s t a f f would receive r e s u l t s before t h e i r department heads, and the l a t t e r , i n turn, before the administration. Persons i n singular positions would meet i n d i v i d u a l l y with the researcher, and would be given the master copy of t h e i r p r o f i l e s . 100 A s e r i e s of one-half hour meetings was held for small groups of s t a f f during the f i r s t week of the December feedback- Each s t a f f participant received copies of the information and patient coding sheets, plus a l l graphs f o r her/his p o s i t i o n . , In some cases, t h i s was only one or two dyads; in others, i t ranged up to twenty-nine pairs of p r o f i l e s . The information sheet was thoroughly discussed, focusing on understanding the layout of the p r o f i l e s and how to i n t e r p r e t them., It was emphasized that the researcher's role was not to assess whether discrepancies i n the paired p r o f i l e s were "good" or "bad". The onus was placed on incumbents of the two positions to examine t h i s dimension. Staff kept t h e i r copies of the material. They were encouraged to go over the p r o f i l e s on their own, or preferably with a group of t h e i r co-workers, and to examine discrepancies and discuss possible causes and solutions. They were t o l d that the researcher would return i n four to six weeks to a s s i s t the hospital to formulate a plan of action using the communication data. In the second week of the December feedback, the researcher met i n d i v i d u a l l y with each department head. They f i r s t received the same basic presentation as t h e i r s t a f f . In most instances, t h i s was followed by a more detailed discussion of guestions raised by the p r o f i l e s . At the end of the f i r s t feedback period, the researcher held a two-hour meeting with three members of the management committee of the South Saskatchewan Hospital Centre, plus 101 the administrative coordinator of iascana Hospital. After reviewing the method of data c o l l e c t i o n , a selected set of p r o f i l e s was presented which, from the researcher's point of view, might have future administrative implications (see Appendix F). Since the o r i g i n a l mandate was only to c o l l e c t the data and present i t to the s t a f f and administration, guidance was sought from the committee on the d e s i r a b i l i t y and format of future input. They reguested that the researcher e l i c i t as many suggestions as possible from the s t a f f during the next feedback v i s i t (keeping i n mind budget stringencies), and involve the soon-to-be-established Staff Development Department (formerly Nursing Inservice Education) as heavily as possible. Staff Development could then take over r e s p o n s i b i l i t y f o r implementing any action a r i s i n g out of the study., The second two-week feedback session was postponed u n t i l February at the reguest of the hospital administrator. I t began slowly, as department heads had just learned of a change in administrator (the t h i r d i n three months) and impending cuts i n s t a f f positions. Because of the general unrest throughout the h o s p i t a l , emphasis was s h i f t e d to maximizing the effectiveness and e f f i c i e n c y of e x i s t i n g s t a f f and programs rather than developing new areas. 102 The researcher and the director of s t a f f development met with the heads and/or supervisors of the following departments: Dietary, Housekeeping, Medical Staff (General Pra c t i t i o n e r s and Consultants), fifursing (both wards), Occupational Therapy, Physiotherapy, Pharmacy, Psychology, S o c i a l Work, and Speech and Hearing. At each neeting, the department members present were asked: (a) What addit i o n a l work had they or t h e i r s t a f f done with the communication feedback since mid-December? (b) Did they see any need for action? I f so, what had they already done? What should be suggested f o r the future? (c) What type of help would they l i k e , either from the researcher or from Staff Development, fo r : - general s t a f f members? - supervisors and/or department heads? Four major inter-department were discussed: (a) low communication by a l l s t a f f with Station Five patients with moderate to severe communication impairment ( i . e . "aphasia") (b) functioning of team conferences on Station Five (c) desire of department heads and/or supervisors for some t r a i n i n g i n how to help s t a f f communicate more e f f e c t i v e l y (d) general s t a f f ' s wish for information on e f f e c t i v e communication. With the exception of number one, these issues were not related to the s p e c i f i c i n t e r a c t i o n p r o f i l e s , but were raised by the department heads as relevant concerns. 103 In addition, a number of intraydepartmental problems were discussed. , These generally focused on the s p e c i f i c communication p r o f i l e s for that department., Examples of the concerns which were raised are: (a) low communication of dietary aides with patients with communication impairment (aphasics) (b) strong communication r o l e of d i e t i c i a n with Station Five patients on an experimental program of therapeutic diets (c) nurse-physician imbalance of Category 1 (seems friendly) (d) low communication of c l i n i c a l pharmacists with a l l patients, e s p e c i a l l y aphasics (e) frequency of negative emotional reactions (tension, unfriendliness) from patient to physiotherapy attendant on Second West Action was taken on most of these concerns. In some cases, the researcher acted as a dire c t resource, conducting classes f o r department heads and s t a f f . In others, only a neutral catalyst was needed to enable the department members to solve t h e i r own problems. Issues which required administrative support and/or decision-making were given addi t i o n a l emphasis i n the f i n a l report to the management committee (see Appendix H for d e t a i l s of concerns r a i s e d , conclusions reached, and action taken during the two weeks). 104 CHAPTER IV A. DISCUSSION! The findings of the study can be divided i n t o four areas of concern—theory, method, the research model, and the u t i l i t y of the method. 1. Theory: The construct v a l i d i t y of any empirical measure must f i r s t be established before consideration i s given to such factors as r e l i a b i l i t y or u t i l i t y . Bales* o r i g i n a l premise that the categories are orthogonal was strongly supported by the factor analyses performed on the present and p i l o t study data., Seguential interdependence of the category pairs could not be assessed. Findings concerning Category 6 (gives information) merit further i n t e r p r e t a t i o n . I t appears that Categories 6 and 5 (gives opinion) are frequently intermingled i n a conversational seguence. However, Category 6 i s negatively associated with the expression of positive and negative a f f e c t (Categories 1 and 12), giving suggestions or leadership (Category 4), and seeking opinions (Category 8). Thus information-giving tends to be done in i s o l a t i o n from these other f i v e dimensions. Likewise, when the f i v e other communication behaviors are exhibited, the amount of 105 information provided i s reduced. While the d i r e c t i o n a l indicators of the new categories were hel p f u l to the observers i n deciding where a p a r t i c u l a r act should be placed i n the i n t e r a c t i o n a l space, evidence of d i r e c t i o n a l loadings was not found i n the factor analyses. Since the new categorization i s s t i l l i n the embryonic stage of development, considerable construct v a l i d a t i o n i s necessary to determine i f these dimensions influence actual behavxor, or whether they are more relevant to perceptual data. The present research guestions the v a l i d i t y of the d i r e c t i o n a l hypotheses for i n t e r p r e t i n g i n t e r a c t i o n behavior at the l e v e l of the organizational position., Other comparative and contrasting studies are necessary to further examine t h i s issue. The study data also showed strong indications of face v a l i d i t y . Empirical variations i n i n t e r a c t i o n p r o f i l e s r e f l e c t e d " r e a l " events occurring during the study which were described by the ethnographic data., When the weather was hot, and the ward atmosphere was described as "tense", the p r o f i l e s had a higher proportion of negative socio-emotional acts. The s t a t i s t i c a l s ignificance of "day" on Station Five i s associated with the observation that d i f f e r e n t events happened on that ward on s p e c i f i c days., On Second West, where there was l i t t l e observable difference between one day and the next (including weekends), the variable "day" had only a minimal impact on the p r o f i l e s . Ho strong patterns of v a r i a t i o n were noted 106 ethnographically within days across the time periods spanned by the study, and time was not a s i g n i f i c a n t factor i n the analysis of variance. The duties of three of the main singular communicators (ward clerk, senior nurse, and assistant d i r e c t o r of nursing) r e s t r i c t e d many of the i r communications to the nursing s t a t i o n . Comparisons of singular versus grouped p r o f i l e s found that "locus" was highly s i g n i f i c a n t for the singular cases. The above findings, i f substantiated by future research, could make a s i g n i f i c a n t contribution to the development of a "theory of organizational communication." 2. Method! Turning from issues of t h e o r e t i c a l relevance, attention should be focused on the methodological concerns examined during the study. The most important finding concerning the research method was that .the_ position :,dyad - plays a dominant-role i n -determining patterns of communication. The basic premise that "each s o c i a l position within the organization has a consistent set of int e r a c t i o n patterns" was upheld. There are several implications a r i s i n g from t h i s finding. The f i r s t i s that the functional position, as defined above, i s an important focus for the study of organizational behavior. This area of concern has been neglected by researchers i n the f i e l d , who have tended to focus on the in d i v i d u a l , the group or the organization. 107 An extensive review of the use of " r o l e " and " p o s i t i o n " i n the organizational l i t e r a t u r e over the past f i v e years (McGill, 1975) has shown that these concepts have considerable potential for bridging the gap between "macro" and "micro" approaches to the study of organizations. D. J. Levinson (1959), i n a seminal a r t i c l e on "Role, personality, and s o c i a l structure i n the organizational s e t t i n g " , suggests that one of the "most a l l u r i n g q u a l i t i e s (of the concept 'role*) i s i t s double reference to the i n d i v i d u a l and to the c o l l e c t i v e matrix.... The organization provides a si n g u l a r l y useful arena f o r the development and application of r o l e theory (p. 1 7 0 ) . H e has thereby issued a dual challenge to Organizational Behaviorists—-to develop an integrative theory of organizational roles, and to apply t h i s framework to empirical analyses of behavior. Three other works re i t e r a t e d Levinson* s challenge. Hickson (1966) recognized the potential u t i l i t y of role theory for organizational analyses, but focused only on role s p e c i f i c i t y . , Katz and Kahn (1966) suggested that role was, "...the major means for l i n k i n g the i n d i v i d u a l and organizational l e v e l s of research and theory; i t i s at once the building block of s o c i a l systems and the summation of the reguirements with which such systems confront t h e i r members as indivi d u a l s (p.,197)." However, they only explored the role-taking episode. The integrative theme then lay dormant u n t i l Lichtman and Hunt (1971) suggested that, "roles do more than l i n k the i n d i v i d u a l and the s o c i a l s t r u c t u r e — t h e y unite them (p. 252) ." 108 Levinson's dual challenge has not been met. , Few systematic attempts have been made to develop role theory as a conceptual l i n k between the i n d i v i d u a l and the organization. Applications of r o l e theory are d i s j o i n t e d , and empirical analyses of these concepts vary greatly i n r i g o r . Future work in t h i s area holds considerable promise, and should be given p r i o r i t y by organizational researchers. The second implication i s that the understanding of organizational behavior i s enriched by the examination of in t e r a c t i v e behavior of dyads as well as that of in d i v i d u a l s . I t was the who-to-whom combination which influenced the inte r a c t i o n s , not the position incumbent i n i s o l a t i o n . A t h i r d implication i s linked to the finding that stable behavior patterns can be i d e n t i f i e d f or the incumbents of p a r t i c u l a r organizational positions., Whereas Bales found l i t t l e s t a b i l i t y for ind i v i d u a l s across groups, the present study demonstrated s t a b i l i t y f o r p o s i t i o n dyads across a time span of eight weeks and a variety of sit u a t i o n s . Given the above, one of the study's s i g n i f i c a n t contributions to the f i e l d of organizational behavior should be to emphasize the wealth of understanding which can be gained by studying organizations at the l e v e l of the functional position. 109 The other major methodological finding of the study was that, while acceptable l e v e l s of i n t r a - and i n t e r - r a t e r r e l i a b i l i t y could be achieved on data unrelated to the type of organization being studied, the p a r t i c u l a r observer had a strongly s i g n i f i c a n t e f f e c t on the ward i n t e r a c t i o n p r o f i l e s . This was not anticipated, and i s a major concern for future use of the method. There are at l e a s t three possible explanations for the consistent si g n i f i c a n c e of the variable "observer". The f i r s t i s the divergent backgrounds of the two observers. The study design purposely employed persons with d i f f e r i n g exposures to t h i s type of organization i n order to minimize any observer "expectancy e f f e c t . " The researcher i s s t i l l of the opinion that the choice of an a s s i s t a n t with no exposure to the health care system was j u s t i f i e d at the time from the point of view of study design and richness of ethnographic data. However, the c r u c i a l guestion now arises whether an observational team can function e f f e c t i v e l y with such divergent backgrounds. This factor i s p a r t i c u l a r l y suspect as a contributor to the "observer e f f e c t " because of the reasonably high r e l i a b i l i t y attained on the "neutral" taped material. One would suspect that the i n t e r p r e t i v e element during " i n process" coding i s stronger than anticipated. Current studies of organizational climate suggest that a p a r t i c u l a r type of organization develops a unigue climate. I t would therefore seem reasonable to postulate that some 110 aspects of that climate are only accessible to " i n s i d e r s " who have experienced i t s impact and have becoae s e n s i t i z e d to i t s s u b t l e t i e s . An "outsider" who has not been s o c i a l i z e d into a p a r t i c u l a r organizational climate may view ongoing behavior from a d i f f e r e n t perspective than an " i n s i d e r . " I t i s important to recognize that neither view of the organization i s correct a p r i o r i . While the above approach to study design has not yet been explored by organizational researchers, i t has been a s i g n i f i c a n t concern i n the f i e l d of organizational consultation f o r many years. In 1959, an e n t i r e issue of the "Journal of S o c i a l Issues" was devoted to the t o p i c of group and organizational consultation. Several of the authors referred to the multiple roles of the consultant (Gibb, 1959; fi. L i p p i t t , 1959; Seashore 6 Van Egmond, 1959). G. L. L i p p i t t (1959) emphasized that, "A consultant r e l a t e s to a c l i e n t group i n terms of i t s previous background and history, and must work within the dimensions of that s i t u a t i o n (p. 50)." According to H. Levinson (1972), "Pure o b j e c t i v i t y or detachment i s impossible i n the behavioral sciences (p. 17)." He maintains that procedures for data c o l l e c t i o n are s i m i l a r , regardless of who i n i t i a t e s the reguest for a s s i s t a n c e — t h e organizational leader or the consultant/researcher. In e i t h e r case the researcher becomes "embedded i n the organizational process (p. 24).'1 The 1975 Academy of Management meetings included a Managerial Consultation Division at which several speakers 111 considered the r e l a t i v e merits of in-house versus external consultants. F i t z g e r a l d (1975) emphasized that each approach had i t s advantages. One of the main points i n favor of the in-house approach was that f a m i l i a r i t y with organizational i n t r i c a c i e s could lead to a more r e a l i s t i c evaluation of the f e a s a b i l i t y of proposed actions. Conversely, the outside consultant was seen to have a more impar t i a l viewpoint, a d e f i n i t e asset i n some s i t u a t i o n s . The organizational change l i t e r a t u r e has also devoted considerable attention to achieving an optimum balance between involvement and detachment i n the consultant-client r e l a t i o n s h i p (Bennis, Benne, S Chin, 1969; Dalton, Lawrence, 6 Greiner, 1970). Argyris' (1969) discussion of the marginal r o l e of the consultant i s p a r t i c u l a r l y germane. He emphasized the behavioral i n e f f i c i e n c i e s created f o r both parties by a discrepancy between the values of the consultant and c l i e n t . Huse (1975) also explored the impact of trust versus that of impersonality (p. 312-313) and suggested that these should be balanced to meet the p a r t i c u l a r needs of the organizational system. Whereas most of the methodological l i t e r a t u r e focuses on the need to control "expectancy bias", perhaps i t i s time f o r organizational researchers to begin to ma____ th i s phenomenon. In other words, i n observational studies the degree of researcher s o c i a l i z a t i o n should be matched to the objectives of the research. If the main focus i s on findings unaffected by organizational climate or milieu, the 112 research team should consist of novices to that p a r t i c u l a r context. I f , on the other hand, the objective i s to study the i n t e r a c t i v e e f f e c t s of a set of variables a i i k i n organizations, then researchers should be exposed to some of the same s o c i a l i z a t i o n as participants. Thus, the composition of the study team should r e f l e c t the study design, rather than impinge upon i t . This approach has been alluded to by Suchman (1967) i n his comparison of evaluative versus noaevaluative or " s c i e n t i f i c " research. However, he has only considered a l t e r a t i o n s i n research design which are reguired for evaluation studies. Desireable c r i t e r i a f o r the researcher have not been mentioned. There i s a strong i m p l i c i t assumption i n t h i s and other discussions of "applied" versus "basic" research that one or the other approach i s somehow "better" i n an absolute sense. This evaluative bias may be hindering the balanced development of organizational studies. I t should be eliminated, and replaced by an integrative approach to research methodology im which a l l studies, whether t h e o r e t i c a l , empirical, applied, or evaluative, would be assessed on the basis of t h e i r contributions to an o v e r a l l understanding of how people behave i n organizations. Recent discussions of "action research" have made a d e f i n i t e contribution to t h i s concern. Huse (19 75) has summarized t h i s synthesis of research and consultation as follows: 113 "...action research i s a c y c l i c a l process which focuses on several main issues: j o i n t collaboration between c l i e n t and change agent, heavy emphasis on data gathering and preliminary diagnosis prior to action planning and implementation, careful evaluation of r e s u l t s before action i s taken, and the development of new behavioral science knowledge which can be applied i n other organizational settings, as opposed to the application of ex i s t i n g behavioral science knowledge (p. 103) ." A second r i v a l hypothesis f o r the observer e f f e c t i s that each observer may have d i f f e r e n t i a l l y selected conversations to code. It became apparent during discussions towards the end of the study that the researcher's hospital background f a c i l i t a t e d coding of conversations i n patient rooms and cubicles during such treatment procedures as bathing and dressing, whereas the research assistant did not f e e l comfortable entering such s i t u a t i o n s . However, since patient data were excluded from the above analysis, the importance of t h i s observer s e l e c t i v i t y cannot be assessed at present. The t h i r d explanation for the impact of the observer i s an i n t e r a c t i o n between "week" and "observer." It w i l l be noted from the r e l i a b i l i t y data that the researcher's own r e l i a b i l i t y declined over the f i n a l three-week period more rapidly than did the research assistant's. This was l i k e l y the e f f e c t of fatigue on judgment and speed. Because of l i m i t e d funding, the researcher was reguired to do the work of a c l e r i c a l assistant i n evenings and on weekends throughout the study. In addition, during the f i n a l two weeks of data c o l l e c t i o n s t a f f c a l l e d upon the researcher to 114 s e t t l e issues not d i r e c t l y connected with the observational study. This meant that " r e s t " breaks between coding periods were v i r t u a l l y nonexistent. One very important caveat for future research which the fatigue explanation underscores i s the importance of adequate c l e r i c a l assistance f o r observational studies. Because of the degree of mental acuity demanded, i t i s p a r t i c u l a r l y important that observers not be overtaxed. At the outset of the present study, i t was recognized that observational studies "often take the researcher into methodologically muddy water (page 6)." The observer e f f e c t discussed above exemplifies the type of problem encountered by t h i s type of research. However, i t i s important that such findings should not prevent future undertaking of s i m i l a r studies. Rather, they should be considered as new types of obstacles to be overcome by c a r e f u l and systematic investigations. The r e l i a b i l i t y findings indicate that P i values of .85 can be achieved within a two-month t r a i n i n g period. Since the t r a i n i n g program was being refined at the same time, i t i s believed that a four- to six-week period would now be s u f f i c i e n t to achieve t h i s l e v e l . However, t h i s estimate of t r a i n i n g time has yet to be v e r i f i e d . Act-by-act assessment of in-process scoring was not possible because of s t a t i s t i c a l l i m i t a t i o n s . I t s importance during the t r a i n i n g process should not be underestimated. Confidence cannot be achieved i n marginal scores during the 11.5 f i n a l data analysis unless the preceding process has been monitored on an act-by-act basis. U n t i l such a s t a t i s t i c i s a v a i l a b l e , the researcher's method of monitoring. the percentage of u n i t i z i n g and categorizing errors, without considering chance, i s the best ava i l a b l e a l t e r n a t i v e . I t i s imperative that i t be superseded by a suitable s t a t i s t i c as soon as possible. Documentation of the t r a i n i n g procedure fo r IPA in-process scoring was one objective of the study. This material has been co l l e c t e d , but additional time i s reguired to produce a succinct t r a i n i n g manual., The need for such a document was re-emphasized by the study, and t h i s work w i l l receive high p r i o r i t y i n the near future. 3. The Research Model: The two most s i g n i f i c a n t e f f e c t s i n the research model (position dyad and observer) have already been discussed. The contextual factor "day" had a greater impact than was anticipated. I t i s encouraging to f i n d a r e l a t i o n s h i p between variation i n i n t e r a c t i o n patterns and regular weekly events. On Station Five, d i f f e r e n t weekly conferences on p a r t i c u l a r days brought d i f f e r e n t communicators to the ward. Behavior i n these meetings, and on the ward before and a f t e r , varied with the persons present and the topics being discussed. This was r e f l e c t e d i n the observed patterns of communication, a further i n d i c a t i o n of the appropriateness of the instrument for t h i s type of analysis. I t seems 116 s u f f i c i e n t l y s e n s i t i v e to i d e n t i f y consistent d a i l y s h i f t s i n the patterns of i n t e r a c t i o n , and therefore may have considerable potential for monitoring s t a f f and patient behavior changes over time. The type of ward was a doubtful determinant of communication behavior i n t h i s study. Further comparisons are necessary to ascertain i f consistent patterns emerge across health care organizations. The e f f e c t of locus on the communication matrix i s moderated by the positions of the communicators. Hhen singular communicators, whose positions r e s t r i c t s i t e s of i n t e r a c t i o n , are excluded, location within the ward plays only a minor role i n influencing behavior patterns. The non-significance of time period i s encouraging methodologically, as i t indicates that the t o o l i s not hypersensitive. The a b i l i t y to d istinguish minor from major moderators i s important i n any instrument for measuring organizational behavior. 4• Utility_Of_The_Method: The study of interpersonal and group phenomena using IPA, while i n vogue i n the 1950's and early *60's, has f a l l e n somewhat int o disfavor because of researchers' disappointment with the findings. . Many c r i t i c i s m s focus on i t s lack of conceptual richness. This was one of the reasons for Bales' addition of d i r e c t i o n a l dimensions to the instrument. However i t can be argued that the lack of 117 i n t e r p r e t i v e power l i e s in the uses made of the t o o l , rather than the categorization i t s e l f . , Perhaps at the i n d i v i d u a l l e v e l of analysis i t i s lacking i n usefulness.„ However, at the p o s i t i o n a l l e v e l i t appears to have considerable i n t e r p r e t i v e power, while retaining s u f f i c i e n t generality to be used across a wide spectrum. I f the above findings are v e r i f i e d i n future studies, the method could play a v i t a l r o l e i n monitoring changes i n patterns of health care delivery. An example of i t s use would be monitoring a l t e r a t i o n s i n the focus of care from chronic to r e h a b i l i t a t i o n , a d i f f i c u l t t r a n s i t i o n facing many i n s t i t u t i o n s today. An objective measure f o r organizational feedback on active versus custodial patterns of care would be of s i g n i f i c a n t assistance to health care administrators and p r a c t i t i o n e r s . Similar use could also be made with organizational development programs, A team of i n t e r n a l and/or external observers could be trained to assess the unit's i n t e r a c t i o n patterns, to diagnose areas of breakdown, and to monitor e f f o r t s to improve coordination of patient care. One of the major methodological drawbacks, however, i s the cost of such research. The method used i n the present study i s both time-consuming and expensive i n terms of the cost of the research team. However, l i t t l e time was demanded of ho s p i t a l s t a f f and patients, an important consideration for research i n health care organizations where the price of s t a f f time i s high because of the degree 118 of pr o f e s s i o n a l i z a t i o n . Further refinements of t r a i n i n g procedures, the development of a suitable s t a t i s t i c f o r assessing in-process act-by-act r e l i a b i l i t y , and streamlining of methods for c o l l a t i n g and analyzing the data should contribute to a reduction i n the expense of such studies. Future research i n the use of the technique should c a r e f u l l y assess the cost factor and attempt to determine the cost/benefit r a t i o (a controversial topic i n the economics of health care). Considerable time was spent during the study developing and implementing the method for feedback to the pa r t i c i p a t i n g h o s p i t a l . As can be seen from the f i n a l , report to the administrator (Appendix H). some of the issues considered i n the two feedback sessions were only peripherally related to the study data. Others, such as the low communication with aphasic patients on Station Five, were a direc t r e s u l t of t h i s input. Most s t a f f reacted p o s i t i v e l y to the i n i t i a l feedback. When the findings differed from th e i r own perceptions, they had s u f f i c i e n t confidence i n the v a l i d i t y of the data to guestion and reassess their own perceptions, and to c r i t i c a l l y examine discrepancies between the two sources of information. Only one group of f i v e stated to the researcher that the findings were not v a l i d because they had not communicated f r e e l y when the observers were present. A l l others said that they had communicated normally, and therefore the data deserved c a r e f u l consideration. 1 1 9 Several s t a f f groups took action as a result of the i n i t i a l feedback (e.g. Physiotherapy, Speech Therapy). Others reguired more time with the researcher to c l a r i f y t h e i r interpretations. They decided on a course of action during the second phase. Only one department made no attempt to use the feedback data. While the patient data have not yet been subjected to extensive multivariate analysis, v i s u a l examination of the pati e n t - t o - s t a f f p r o f i l e s shows considerable v a r i a t i o n across one or more of the three patient dimensions. , For example, the nurse's interaction with patients tended to vary most with t h e i r communication impairment, whereas the occupational therapist's d i f f e r e d mainly by length of stay. From the foregoing, i t can be seen that the study did have value f o r hospital s t a f f at several l e v e l s . The next guestion to be asked i s the resultant impact on patient care. This effectiveness dimension can only be assessed by lon g i t u d i n a l studies, which are strongly recommended for future phases of the research. B, FUTORE RESEARCH DIRECTIONS: During the i n i t i a l examination of the l i t e r a t u r e , a number of a r t i c l e s were reviewed which had implications f o r future research. 120 In r e l a t i o n to the development of a theory of organizational communication, many of the findings of Georgopoulos and Mann (1962) could be tested i n future convergent and discriminant validation of IPA. In addition, the r e l a t i o n s h i p between Bales* concepts and those of Transactional Analysis (Berne, 1964; James & Jongeward, 1971) should be explored. Methodologically, future studies should explore the use of longitudinal research designs such as that employed by Talland (1955), who applied IPA to the problem of analyzing therapeutic discussion groups. The simulation approach of Hess (1969) also holds considerable promise, p a r t i c u l a r l y i f used i n combination with one or more of Campbell and Stanley's (1963) guasi-experimental designs to test f o r causal r e l a t i o n s h i p s . In future research models, interaction as a horizontal coordinator i n organizations should be studied i n concert with other phenomena such as leadership. Because of the type of categorization, IPA has a s i g n i f i c a n t contribution to make to such investigations. The new Category 4, for example, i s hypothesized to be almost purely a "leadership" dimension. Therefore studies of convergent and discriminant v a l i d i t y , using t h i s and other measures of organizational leadership, could play an important role i n the understanding of organizational integration. Manheim's (1960) findings on the e f f e c t s of status and leadership should also be helpful i n future investigations.„ 121 The effectiveness of organizational communication should be a major concern i n future examinations of the u t i l i t y , of the method. Here the work of N. C. Scott (1973) and other nursing researchers should be p a r t i c u l a r l y h e l p f u l . For example, comparisons could be made of the r e l a t i v e v a l i d i t y and u t i l i t y of IPA and NOSY i n several d i f f e r e n t settings and across a variety of positions. This would a s s i s t researchers to determine tine appropriate degree of category s p e c i f i c i t y for interaction research i n various health care contexts. The foregoing discussion i n no way exhausts the p o s s i b i l i t i e s f o r future research i n the f i e l d of observation of organizational communication. Bather, i t i s intended to suggest a number of ways i n which the present research may stimulate others to explore the concerns raised i n the rationale of t h i s study. C. CONCLUSIONS! The methodological objectives of the present study, as outlined above (page 8), have been met. The study has demonstrated that communication behavior can be consistently measured i n the organizational context., The perceptual set of the observer, as determined Jby exposure to organizational s o c i a l i z a t i o n , i s believed to function as a major intervening variable. 122 Bales' Interaction Process Analysis can be refined f o r use i n the organizational environment. As long as subjects are f u l l y aware of the nature and objectives of the research, and have achieved some degree of commitment, the presence of a recording observer does not overly disrupt the organizational milieu. The method used i n the study f o r data c o l l e c t i o n might be improved by the use of e l e c t r o n i c tabulation devices. One such instrument i s now av a i l a b l e , with a ca l c u l a t o r keyboard connected to a casette tape recorder. However, i t has several drawbacks—size, weight, battery l i f e , cost, and d i f f i c u l t y of servicing. I f such devices are used i n future research, t h e i r potential impact on subject apprehension should be c a r e f u l l y assessed., The construct v a l i d i t y of IPA i s p a r t i a l l y upheld. Eleven of the twelve categories are c l e a r l y demonstrated to be independent dimensions of communication behavior. The hypothesis that Category 6 i s an adjunctive communication behavior i s strongly supported. Freguency of uncodable acts i s low, giving some support to Bales' contention that the categorization i s exhaustive. Sequential interdependence of the category pairs could not be assessed. Further val i d a t i o n of these findings i s recommended i f empirical research using the instrument i s to make a contribution to the development of a theory of organizational communication. 123 Implications for the f i e l d of Organizational Behavior are e x c i t i n g . As a r e s u l t of the present study, the researcher i s s i g n i f i c a n t l y closer to r e a l i z i n g the long-term goal expressed on page 2 of t h i s Dissertation. There i s promise that methods can be developed for t r a i n i n g organizational researchers and/or members to observe and analyze communication behavior. Such awareness should r e s u l t i n more e f f i c i e n t use of our health manpower resources, and eventually lead to more e f f e c t i v e i n t e g r a t i o n of patient care. 124 BIBLIOGRAPHY Amidon, E. J . , & Hough, J. B. (Ed.), Interaction analysis; Th~°gy« .ggsg^rch,_^?d application. Don H i l l s , Addison-Hesley, 1967 Argyris, C., "Some unintended consequences of rigorous research", Psychological ..Bulletin, 1968, v. 70(3), P. 185-197 Argyris, C., "Explorations i n con s u l t i n g - c l i e n t relat i o n s h i p s " , i n Bennis, W. G., Benne, Benne, K. D., & Chin, R. (Ed.), The planning of change, Toronto, Holt Rinehart, 1969, P. 434-457 Bales, R. F., Interaction .process analysis; A method for the study of small groups, Reading, Hass., Addison-Besley, 1950a Bales, R. F., "A set of categories for the analysis of small group i n t e r a c t i o n " , American Soci o l o g i c a l Review. 1950b, V. 15(2), P. 257-263 Bales, R. F., "Some s t a t i s t i c a l problems in small group research", Journal of the Amer^canStatistical Association, 1951, V. 45, P. 311-322 Bales, R. F., "The equilibrium problem i n small groups", i n Parsons, T. , Bales, R. F. , & S h i l s , E., 8pricing : papers i n the_theory_of_action. New York, Free Press, 1953a, P. 111-161 Bales, 1. F., "A t h e o r e t i c a l framework f o r i n t e r a c t i o n process analysis", in Cartwriqht, D., & Zander, A. (Ed.), Q rQMR^dynamics A_research andtheory, Evanston, Row Peterson, 1953b, P. 29-38 Bales, R. F., "Task roles and s o c i a l r o l e s i n problem-solving groups", i n Macoby, E. , E. , Newcomb, T. M., & Hartley, E. L. (Ed.), Readings i n s o c i a l psychology. New York, Holt, 1958, P. 437-447 Bales, R. F., "Conceptual frameworks for analysis of s o c i a l i n t e r a c t i o n " . Journal_of,„Experimental Education. 1962, V. 30(4), P. 323-324 Bales, R- F., Personality and interpersonal .behavior, Toronto, Holt RinehartT 1970 Bales, R. F., & Couch, A.,S., "The value p r o f i l e : A factor an a l y t i c study of value statements". Soc i o l o g i c a l .Inquiry. 1969, V. 39 (1) , P. 1 2 5 Bales, R. F. , fi Hare, A. P., "Diagnostic use of the inte r a c t i o n p r o f i l e " . Journal of Social_Psycholpgy, 1 9 6 5 , V. 6 7 , P. 2 3 9 - 2 5 8 Bales, R. F-, 6 S l a t e r , P. E. "Bole d i f f e r e n t i a t i o n i n small decision-making groups", i n Parsons, T., Family, s o c i a l i z a t i o n and intera c t ion ..process. New York, Free Press, 1 9 5 5 , P7 2 5 9 - 3 0 6 Bales, R. F., S Strodtbeck, F. L . "Phases i n group problem-solving". Journal of Abnormal and,Social.Psych 1 9 5 1 , V. 46 (4) , P . ~ 4 8 5 - 4 9 5 ~ Bales, R. F., Strodtbeck, F. L . , M i l l s , T. M., & Roseborough, M. E., "Channels of communication i n small groups", American Sociological.Review, 1 9 5 1 , ¥ . 1 6 ( 4 ) , P. 4 6 1 - 4 6 8 Bates, F. L. , "Position, r o l e , and status: A reformulation of concepts", Social_Forces, 1 9 5 5 - 6 , V. 3 4 , P. 3 1 3 - 3 2 1 Bennis, W. G., Benne, K. D., & Chin, R. (Ed.), The planning of_chan_e, Toronto, Holt Rinehart, 1 9 6 9 Berne, E., Games people play; The psychology of human rela t i o n s h i p s , New York, Grove, 1 9 6 4 ~ Biddle, B. J. , & Thomas, E. J. , Role^ theor. y,_;„Conee^ts_and research. New York, Hiley, 1 9 6 6 Borgatta, E. F., & Bales, R. F. "Task and accuaulation of experience as factors i n the inte r a c t i o n of small groups", Sociometry., 1 9 5 3 , V. 1 6 ( 3 ) , P. 2 3 9 - 2 5 2 Borgatta, E. F., & Bales, fi. F., "The consistency of subject behavior and the r e l i a b i l i t y of scoring i n Interaction Process Analysis", American S o c i o l o g i c a l Re view, 1 9 6 3 , V. 18 (5) , P. 5 6 6 - 5 6 9 Campbell, D. T. , & Stanley, J . C. , Experimental and ___§l~___--iiental designs for research. Chicago, Rand-McNally, 1 9 6 6 C a u d i l l , W. A,, The psychiatric hospital_as i M a s m a l l , society. Cambridge, Harvard, 1 9 5 8 Chappie, E. D., "The i n t e r a c t i o n chronograph: I t s evolution and present a p p l i c a t i o n " . Personnel. 1 9 4 9 , V. 2 5 ( 4 ) , P. 2 9 5 - 3 0 7 Cohen, J., "A c o e f f i c i e n t of agreement f o r nominal scales". Educational and„Psychological Measurement. 1 9 6 0 , V. 2 0 ( 1 ) , P. 3 7 - 4 6 126 Conant, L. H., "Use of Bales' Interaction Process Analysis to study nurse-patient i n t e r a c t i o n " , Nursing Research, 1965, V. 14(4), P. 304-309 Conant, 1. H. , Exploratory study of nujrse-patient^give and take, terminal progress report, 0. S. Public Health Service Research Grant No. N000181, Hew Haven, Yale University School Of Nursing, 1967 Crichton, A., TeaaworJc _tQgasfl§_iEgbafeiA:itatiQa-03|-r,:fe^ a ..Istflg term patient. Vol. 2, Vancouver, U. B, C. Faculty of Medicine, Department of Health Care and Epidemiology, 1975 Dalton, G., W., Lawrence, P. R., & Greiner, L. E., Organizational_changg_and development, Homewood, 111. , Irwin-Dorsey, 1970 Daubenmire, M. J . , & King, I. M., "Nursing process models: A systems approach". Nursing.Outlook, 1973, V. 21(8), P. 512-517 Davis, J. A., Elementary survey analysis, Toronto, Prentice-Hall, 1971 Davis, M. S,, "Variations i n patients' compliance with doctors' advice: An empirical analysis of patterns of communication", American Journal of .Public.^Health, 1968, V. 58(2), P. 274-288 Diers, D., & Leonard, R. C., "Interaction analysis i n nursing research". Nursing Research, 1966, V. 15(3), P. 225-228 Diers, D., & Schmidt, R. L., "Transcriptions and tape recordings i n interaction analysis", Mursing^Research, 1968, V. 17(3), P. 236-241 Diers, D., Schmidt, R. L., McBride, M. A. B., & Davis, B. L., "The e f f e c t of nursing i n t e r a c t i o n on patients i n pain", Nursing_Research, 1972, V. 21(5), P. 419-428 Dubin, R., The world of work; I n d u s t r i a l society and human re l a t i o n s , Toronto, Prentice-Hall,~1958"" ~ Dubin, R., "Business behavior behaviorally viewed", i n Strother, G. B. (Ed.), Social_science_a£proaches_to business ..behavior, London, Tavistock, 1962, P. 11-55 Fayol, H., General,and i n d u s t r i a l management. London, Pitman, 1949 Fi t z g e r a l d , J . , "In-house s t a f f versus outside consultants". Proceedings of the 35th annual Academy .of Management, Auburn University Printing Service, 1975, P. 113-115 127 Flanders, N. A-, "Using Interaction Analysis i n the inservice t r a i n i n g of teachers", Journal_of Experimental Education, 1962, V. 30(4), P. 313-316 Flanders, N . A., Analyzing teaching behavior. Don M i l l s , Addison-wesley, 1970 Francis, V., Korsch, B. M., 5 Morris, M. J., "Gaps i n doctor-patient communication: Patients* response to medical advice". Hew_En_l_nd__ournal_of__ , 1969, V". 280(10), P. 535-540 Freemon, B., Negrete, V. F., Davis, a., 6 Korsch, B. M. , "Gaps i n doctor-patient communication: Doctors-patient int e r a c t i o n analysis". Pediatric Research. 1971, V. 5, P. 298-311 Georgopoulos, B, S., S Mann, F. c , The coimunit-y: general h o s _ i t a l . New York, Macmillan, 1962 ~~ Gibb, J . R,, "The r o l e of the consultant". Journal of Social I s s u e s , 1959, V.15 (2), P. 1-4 Gibb, J . R., "TORI theory and practise", i n P f e i f f e r , J . 8., S Jones, J . E., (Ed.), The 1972 annual handbook f o r group f a c i l i t a t o r s , Iowa C i t y , University Associates, 1972a, P. 157-162 Gibb, J . R., "TORI theory: Consultantless team-building". Journal of .Contemporary .Business. 1972b, V. .1(3), P. 33-41 Gordon, G., A., Role theory and illness„A s o c i o l o g i c a l perspective. New Haven, College and University Press, 1966 Gouldner, A. W., "The norm of r e c i p r o c i t y : A preliminary statement", American.Sociological Review. 1960, V. 25(2), P. 161-178 Gulick, L., "Notes on the theory of organization", i n Gulick, L., & Orwick, L. (Ed.), Papers on the .science of administration. New York, I n s t i t u t e of Public Administration, 1937, P. 1-45 H a l l , B. H., organizations:. Structure and process. Toronto, Prentice-Hall, 1972 Hare, A. P., & Bales, R. F., "Seating position and small group i n t e r a c t i o n " , Sociometry., 1963, V. ..26(4), P. 480-486 Hare, A. P., Borgatta, E. F., & Bales, R. F. (Ed.), Small groups; Studies i n s o c i a l i n t e r a c t i o n . New York, Knopf, 1965 128 flare, A. P., Waxier, N., Saslow, G. , 6 Matarazzo, J . D., "Simultaneous recording of Bales and. Chappie i n t e r a c t i o n measures during i n i t i a l p s y c h i a t r i c interviews". Journal of _ _ _ s _ l t i j i _ _ _ s y c ^ , 1960, V. 24(2), P. 193 Heinicke, C , & Bales, H. F., "Developmental trends i n the structure of small groups", Socigmetry, 1953, V. 16(1), P. 7-38 Heifer, fi. E., "An objective comparison of the pe d i a t r i c interviewing s k i l l s of freshman and senior medical students". P e d i a t r i c s , 1970, V. 45(4), P. 623-627 Heifer, B., S Hess, J., "An experimental model for making objective measurements of interviewing s k i l l s " . Journal of _li _ i _ _ l _ P s _ c h o l o q _ _ , 1970, V. 26(3), P. 327-331 Hess, J . W., "A comparison of methods fo r evaluating medical student s k i l l i n r e l a t i n g to patients". Journal of- Medical Education, 1969, V. 44 (2), P. 934-938 Hicks, L. ,E., "Some properties of i p s a t i v e , normative and forced-choice measures". Psychological B u l l e t i n . 1970, V. 74(3), P. 167-184 Hickson, D. J . , "A convergence i n organization theory". Administrative Science Quarterly, 1966, V. 11(2), P.,224-253 Hover, G. B., . _l_a-_Eg„ 1^ Analysis of~teacher verbal behavior a t . t h e _ u n i v e r s i t y _ l g v e l , unpublished Master*s thesis, Spokane, Gonzaga Oniv. School of Education, 1970 Howland, D., "Cybernetics and general systems theory". General systems; Yearbook.of the.society for,.general systems research, 1963a, ¥. 8, P. 227-232 Howland, D., "Approaches to the systems problem". Nursing Research, 1963b, ?. 12, reprinted i n General Systems, 1964, v7 9, P. 283-285 Howland, D. , "A hos p i t a l system model", Nursing-.-Research» 1963c, V. 12, reprinted i n General Systems. 1964, V. 9, P. 287-291 Howland, D., "Approach to nurse-monitor research", American _2M___I_2__Nursing, 1966, V. 66 (3), P. 556-558 Howland, D., & McDowell, W. E., "The measurement of patient care: A conceptual framework". Nursing Besearch. 1964, V. 13, reprinted i n General SvstemsT 1964,~?7 97 P. 293-296 129 Huse, E. F., Organizational development and change. Los Angeles, West, 1975 James, M., £ Jongeward, D., Born_.to .wjn; ^Transactional ^a4ll§i§,.,,»4th_gestalt experiments. Don M i l l s , Addison-Wesley, 1971 Johnson, B. S., "Relationships between verbal patterns of nursing students and therapeutic effectiveness". Nursing Research, 1964, V. 13(4), P. 339-342 Katz, D., S Kahn, R, L., The s o c i a l psychology of organizations. New York, Wiley, 1966 Kerlinger, F. N., Foundations of behavioral research, Toronto, Holt Rinehart, 1966 ~ Korsch, B. M«, Gozzi, E- K., S Francis, v . , "Gaps i n doctor-patient communication". Pediatrics, 1968, V. .42(5), P. 855-871 Korsch, B. M., 6 Negrete, V. F., "Doctor-patient communication", Scientific,American, August, 1972, P. 66-74 Landsberger, H. A., "Interaction Process Analysis of professional behavior: A study of labor mediators i n twelve labor-management disputes", American Sociological Review. 1955a, V. 20(5), P. 566-575 Landsberger, H. A., "Interaction Process Analysis of the mediation of labor-management disputes", Journal._of Abnormal afld_Socia.l_Psy.chology, 1955b, V. ,51, P. 552-558 Lawrence, P. R., S Lorsch, J, W., Organization and environment:Managing d i f f e r e n t i a t i o n and integration, ~ Cambridge, Harvard Univ. Graduate School of Business Administration, 1967 Leland, D. , Verbal interaction_in_closed and open classrooms,, unpublished Master* s research project, Waterloo, Oniv. of Waterloo Department of Psychology, 1974 Levinson, D. j . , "Role, personality and s o c i a l structure i n the organizational s e t t i n g " . Journal of Abnormal and S o c i a l _s_cholo__, 1959, V. 58, P. 170-180~ Levinson, H., Organizational Diagnosis, Cambridge, Harvard, 1972 Lewis, W. W., S Newell, J. M., "Analysis of classroom inte r a c t i o n through communication behavior". Journal of IXEeri_ental_Education, 1962, V. 30(4), P. 321-322 130 Lichtman, C . M., & Hunt, fi. G . , "Personality and organization theory: A review of some conceptual l i t e r a t u r e " , Psychological B u l l e t i n , 1971, V. 76(4), P. 271-291 Li k e r t , fi., „.___--_-___.gf Management* Toronto, HcGraw-Hill, 1961 L i p p i t t , G. L. , "A study of the consultation process", J o u r _ a l _ o f _ S o c i a l _ I ^ 1959, V. 15(2), P. ,43-50 L i p p i t t , fi., "Dimensions of the c o n s u l t a n t s job". Journal 2_____i_I_I_s_es, 1959, V. 15(2), P. 5-12 Lum, J . L. J . , "Interaction patterns of nursing personnel", I_£§ing__esearch, 1970, V. 19(4), P. 324-330 Manheim, H. L., "Intergroup i n t e r a c t i o n as related to status and leadership differences between groups", SocjQmetry. 1960, V. 23, P. 415-427 McBride, M. A. B., "Nursing approach, pain, and r e l i e f : An exploratory experiment". Nursing Research. 1967, V. 16(4), P. 337-341 McGill, M. ,£. , Staff „co mmunicati on z i n _ a r e habil i t a, t fon hospital^ and^its r e l a t i o n ,to integration of patient care, fiegina, Saskatchewan, National Health Grant application, 1968 McGill, M. E., The development_of_a model.o| organizational positions, unpublished manuscript, ,. Vancouver, U.B.C. Faculty of Commerce and Business Administration, 1975 McGrath, J . E., & Altman, I., Small,.group research: A synthesis and critigue.of„the_field, Toronto, Holt Rinehart, 1966 Merton, R. K., "The s o c i a l dynamics of status-sets and status-seguences", in Biddle, B. J . , & Thomas, E. J . , Role _heory. New York, Wiley, 1966, P. 74-76 Miles', R. E., Thforjes of management; Implications f o r organizational behavior and development. Toronto, McGraw-Hill, 1975 Mishler, E. G., & Waxier, N. E., "Family i n t e r a c t i o n and schizophrenia". Archives of General Psychiatry, 1966, V. 15, P. 64-74 Mishler, E. G., 6 Waxier, N. E., Interaction i n f a m i l i e s ; An ®SPerimental...study of family prpce.sses_and,„schxzophrenia . New York,"wiley, 1968 131 Mooney, J. D., "The p r i n c i p l e s of organization", i n Gulick, L., & Orwick, L. (Ed.), ___ers on the science of __ffljjl_§_i£_k__2_# N e w York, Institute of Public Administration, 1937, P. 89-98 Nagi, S. Z., "Some conceptual issues i n d i s a b i l i t y and r e h a b i l i t a t i o n " , i n Sussman, M. B. (Ed.), Sjociolo_y_and r e h a b i l i t a t i o n , American S o c i o l o g i c a l Association, 1965, P. 100-113 Nie, N- H., Hul l , C. H., Jenkins, J . G., Steinbrenner, K., & Bent, D. ,H., S t a t i s t i c a l , packages, for „ the social-..:sciences, Toronto, McGraw-Hill7""l975~ O'Reilly, C. A. I l l , & Roberts, K. H., "Information f i l t r a t i o n i n organizations: Three experiments", Organizational .Behavi 1974, V. 11, P. 253-265 O'Eourke, J . F., " F i e l d and laboratory: The decision-making behavior of family groups i n two experimental conditions". Sociometry, 1963, V. 26, P. 422-435 Pellegrino, 1. D., "The changing matrix of c l i n i c a l decision-making i n the hospital", 1970, i n Georgopoulos, B. S. (Ed.), Organization research on,health i n s t i t u t i o n s , Ann Arbor, I n s t i t u t e f o r S o c i a l Sesearch, 1972, P. 301-328 Pena, D., Assessing physician interpersonal s k i l l s : A bibliograp.hical_reyie_, unpublished manuscript, Detroit, Wayne State Oniv. School of Medicine, D i v i s i o n of Educational Services and Research, 1974 Pena, D., & Hogan, M. J . , Interaction analysis: A multidimensional.,a_proach. unpublished manuscript, Detroit, Wayne State Oniv. School of Medicine, D i v i s i o n of Educational Services and Research, 1975 P h i l l i p s , J . S., Matarazzo, J . D., Matarazzo, R. G., & Saslow, G., "Observer r e l i a b i l i t y of interaction patterns during interviews", Jqurnal_of^Consulting Psychology. 1957, 7. 21 (3), P. 269-275 Porter, L. W., & Roberts, K., Communication i n organizations f U.S. Department of Health, Education and Welfare, Technical Report No. , 12, (ERIC ED microfiche No. 066 773), 1972 Psathas, G., "Phase movement and equilibrium tendencies i n in t e r a c t i o n process i n psychotherapy groups", Sociometry, 1960, V. 23, P. 177-194 132 Psathas, G., "Alternative methods for scoring Interaction Process Analysis", Journal of Soc i a l Psychology, 1961, V. 53, P. 97-103 Rackham, N. , & Honey, P., Developing_interacti-v$ sskills, Northampton, England, Hellens, 1971 Ring, K., "Experimental s o c i a l psychology: Some sober guestions about some f r i v o l o u s values". Journal of Experimental S o c i a l Psychology. 1967, 7. 3(2), P. ?13-123 Roberts, K. H., 6 O'Reilly, C A. I l l , "Failures i n upward communication i n organizations: Three possible c u l p r i t s " , Academy of Management Journal. 1974a, V. 17(2), P. 205-215 Roberts, K. H., & O'Reilly, C. A. I l l , "Measuring organizational communication", Jo.arnal,,Qf,- Applied Psycholo___ 1974b, V. 59(3), P. 321-326 Roberts, K. H., O'Reilly, C. A. I l l , Bretton, G. E., & Porter, L. H, "Organizational theory and organizational communications: A communication f a i l u r e ? " . Human Relations. 1974, V. 27(5), P. .501-524 Rosenthal, R., & Rosnow, R. L. (Ed.), A r t i f a c t i n behayiorai„research. New York, Academic, 1969 Scott F. G., "Family group structure and patterns of s o c i a l i n t e r a c t i o n " , American Journal of Sociology, 1962, V. 68(2), P. 214-228 Scott, N. C., Donnelly, M. B., Gallagher, R., & Hess, J . W., "Interaction analysis as a method f o r assessing s k i l i i n r e l a t i n g to patients: Studies on v a l i d i t y " , British,.Journal of_Iedical_Education, 1973, V. ,7(3), P. 174-178 Scott, N. C , Donnelly, M. B., 6 Hess, J. H., Longitudinal changes in interviewing .performance of .medical,.students, unpublished manuscript, Detroit, Sayne State Oniv. School of Medicine, D i v i s i o n of Educational Services and Research, 1974 Scott, W. A., " R e l i a b i l i t y of content analysis: The case of nominal scale coding". Public Opinion Quarterly, F a l l , 1955, P. 321-325 Seashore, C„, & Van Egmond, E., "The consultant-trainer role i n working d i r e c t l y with t o t a l s t a f f " , Journal, . o f S o c i a l Issues, 1959, V. 15(2), P. 36-42 S e l l t i z , C., Jahoda, M., Deutsch, M., & Cook, S., Research methods i n s o c i a l r e l a t i o n s . New York. Holt. 1959 133 Stevens, L. C , "Staff communications on the medical and su r g i c a l services; A research note", j_sr^a?-;r.gi_-ftlB|„4tfr,-a$d Social_Behavior, 1967, V. 8(2), P. 148-153 Suchman, E. A., Evaluative..research; .Principles and Practice i n . p u b l i c .service,and.social action programs. New York, Russell Sage, 1967 Sudnow, D. (Ed.), Studies i n s o c i a l i n t e r a c t i o n . New York, Free Press, 1972 ~ - -Swensen, C. H. J r . , Introduction^tp... interpersonal :/relotions, Glenview, 111., Scott Foresman, 1973 Sydiaha, D., "Bales' Interaction Process Analysis of personnel s e l e c t i o n interviews". Journal of Applied -Ps_cholo__, 1961, V. 45(6), P. 393-401 Talland, G. A., "Task and inte r a c t i o n process; Some ch a r a c t e r i s t i c s of therapeutic group discussion", Journal of _bn_______n__Social_Ps^ , 1955, V. 50, P. ,105-109 Thayer, L. O., Administrative cp^muBic§tign, Homewood, 111., Irwin, 1961 ~ Thayer, L, 0., Commuqication andcommunication :systgms_in organizational management and interperspnai_relat_ons, Homewood, 111., Irwin, 1968 Thompson, V- A., aodern_prqanizatiqn. New York, Knopf, 1961 Thorndike, R. L., Personnel selection; .Test and measurement techniques, Hew York, Wiley, 1949 Topf, M., "A behavioral c h e c k l i s t for estimating the development of communication s k i l l s " , Journal^pf.N ursinq Education, 196 9, V. 8(4), P. 29-34 Turner, R. (Ed.), Ethnpmethpdolpgy__Selected readings. Harmondsworth, England, Penguin Education, 1974 Warner, M. M. , An a.nnp.tat^d_biib_4agsa_fel-_i-feaa^fe_gft£^ tea_wor^_and_health_.cent , Vancouver, U. B. C. Faculty of Medicine, Department of Health Care and Epidemiology, Project TEAM, 1975 Waxier, N. E., & Mishler, E. G., "Scoring and r e l i a b i l i t y problems i n Interaction Process Analysis: A methodological note", Sociometry,, 1966, V. 29(1), P . 28-40 Webb, E., Campbell, D. T., Schwartz, R. D., & Sechrest, L., Onpb/trusive_measures;. Ngnreac^iye re^e^rch i n the s o c i a l sciences, Chicago, Rand-McNally, 1966 134 Werner, A., S Schneider, J- H . , "Teaching medical students i n t e r a c t i o n a l s k i l l s " . New England Journal of Medicine, 1974, V. 290(22), P. 1232-1237 Wessen, A. F., "Hospital ideology and communication between ward personnel", i n Jaco, E. G. (Ed.), Patients,.physicians afld_illnej;si_Sguxcebo science_and .medicine, Glencoe, I l l 7 , Free Press7~19587~P. 448-468 Wilkinson, G. s . , "Interaction patterns and s t a f f response to p s y c h i a t r i c innovations". Journal of Health and Soc i a l Behavior, 1973, ¥.,14(4), P. 323-329 135 APPENDIX A RESULTS OF PILOT.STUDY The following i s a summary of relevant findings from the 1967-69 p i l o t study conducted by the researcher at Wascana Hospital i n Regina, Saskatchewan. The r e l i a b i l i t y of the IPA instrument was not adeguately documented. This constituted a major flaw i n the study. When examining the data f o r construct v a l i d i t y , the multiple correlations were a l l found to be below .20, with the exception of Category 6 for sender, and 2, 6 and 11 f o r receiver. In a series of R-type factor analyses, a l l Categories except 2, 6 and 11 were r e l a t i v e l y orthogonal, i n d i c a t i n g that they represented independent types of i n t e r a c t i o n . A l l factors loading on the other nine categories had low E i g e n v a l u e s — i . e . none were predominant. The s p e c i f i c order of the factors varied with the data set (sender versus receiver) and the type of rotation. Categories 2 and 11 loaded moderately (.20 to .39) on f i v e of the factors for the receiver, and 6 loaded moderately on seven of the sender's factors. In both sets of data, the main loadings for these three variables were on the f i r s t three factors. 136 Categories 2, 6 and 11 were f e l t to be on a d i f f e r e n t conceptual l e v e l from the other nine. They were postulated to be adjunctiye eg^m^nic^tiye L behaviors working i n concert with the others to c l a r i f y (Cat. 6) and modify (Cats. . 2 and 11) the t o t a l i n t e r a c t i o n seguence. Because of the data arrangement, the interdependence of Bales' category pairs could not be thoroughly examined. , With respect to methodological a r t i f a c t , the e f f e c t of the observer was d e f i n i t e l y higher on the f i r s t day, but t h i s was not tested for l e v e l of si g n i f i c a n c e . In the p i l o t project observer communications were coded i n order to evaluate the maximum impact of the method. These values were i n f l a t e d since a l l observer interactions were recorded, whereas only a time sampling was taken f o r other communicators. After the f i r s t day, the observer had only a minimal e f f e c t on the ward p r o f i l e i n Categories 3 to 10. However, Categories 1 and 2 showed a d e f i n i t e observer e f f e c t as sender, and 1, 2 and 11 as receiver. The observer therefore increased the socio-emotional aspects of the ward p r o f i l e s . When the r e s u l t s were tested f o r concurrent v a l i d i t y using a univariate ANOVA, one of the major findings was that the who-to-whom f i r s t order interaction e f f e c t was highly s i g n i f i c a n t . This led to the suggestion that future studies use the role dyad, rather than an iso l a t e d sender or receiver, as the independent variable. 137 S i g n i f i c a n t e f f e c t s were found for "ward" and minimal e f f e c t s for "day". I t was postulated that the weak "day" ef f e c t s were mainly attributable to f i v e a t y p i c a l days out of sixteen, but t h i s could not be v e r i f i e d because of inadeguate ethnographic material. For some categories, "locus" was a key variable, but t h i s was considered to be the r e s u l t of int e r a c t i o n of a strong "who-to-whom" e f f e c t , a moderate "task design" e f f e c t , and a weaker "pure locus" one. 138 APPEJ_____ ______M_____________I_ STAFF POSITION AND DEPARTMENT CODES-______ Grouped Positions Total N OOO - s t a f f nurse 41 001 - nursing attendant 40 002 - nursing student 20 003 - senior nurse 3 004 - nursing supervisor 8 005 — nursing administration 2 006 — nursing inservice education 3 010 — housekeeping aide 8 011 - houseman 14 012 - housekeeping supervisor 2 013 - dietary aide 13 015 — porter 5 020 - laboratory technician 3 021 - occupational therapy 18 022 - orthopaedic appliances 7 023 - pharmacy 6 024 - physiotherapist 25 025 — physiotherapy interne 2 026 - physiotherapy attendant 13 027 - medical s o c i a l worker 5 028 - speech and hearing 4 -029 — X-ray 2 030 — C.S.R. aide 3 031 - maintenance 7 032 - medical records 5 034 - purchasing and stores 3 036 - security 5 037 — admitting and switchboard 13 040 — general p r a c t i t i o n e r 8 041 - p h y s i a t r i s t 3 042 - orthopaedic surgeon 043 — other medical s p e c i a l i s t 050 — family 051 - volunteer 052 - clergy 053 - research observer 2 054 - admin i s t r a t i o n 2 139 Code # S in q u 1 a r P o s i t io n s 100 - assistant director of nursing 101 - ward clerk 102 - WIA&S nursing instructor 105 — d i r e c t o r of nursing 110 - assistant director of housekeeping 113 - food services supervisor 114 — d i e t i c i a n 121 — d i r e c t o r of occupational therapy 127 personal services worker 131 — director of maintenance 133 - outpatient c l i n i c attendant 134 - supervisor of purchasing 135 — driver 154 _ barber 155 - hairdresser 156 - health nurse 157 - psychologist 158 — V.O.N. 140 PATIENT POSITION.CODES LENGTH_OF_STAY: 3 = 2 to 7 days 4 = 8 to 30 days 5 = 31 to 90 days 6 = 91+ days MOBILITY: 0 = stays i n own room a l l the time, except f o r s p e c i a l appointments (e.g. X-Eay) 1 = up i n wheelchair out of room; cannot propel s e l f 2 = propels chair short distances; transfers not independent 3 = propels chair short distances; transfers independent 4 = moves f r e e l y (chair or walking); transfers not independent 5 = moves f r e e l y ; transfers independent COMMUNICATION_IMPAIRMENT: 0 = severe receptive impairment; severe expressive impairment 1 = moderate receptive impairment; severe expressive (or vice versa) 2 = moderate receptive impairment; moderate expressive 3 = s l i g h t receptive impairment; severe expressive (or vice versa) 4 = s l i g h t receptive impairment; moderate expressive (or vice versa) 5 = s l i g h t receptive impairment; s l i g h t expressive 6 = no receptive impairment; severe expressive 141 7 = no receptive impairment; moderate expressive 8 = no receptive impairment; s l i g h t expressive (or vice versa) 9 = no receptive impairment; no expressive Communication impairment i s defined as follows; RECEPTIVE: severe = understands less than 10% of what i s said to her/him moderate = understands 10% to 50% of what i s said s l i g h t = understands 51% to 90% of what i s said (or gets confused) none = understands more than 90% of what i s s a i d EXPRESSIVE: severe = unable to express desires i n any manner moderate = able to express some desires, but with d i f f i c u l t y s l i g h t = able to express most desires e a s i l y to a l l s t a f f none = normal expressive a b i l i t y Note: Communication code i s based on patient 1s s k i l l i n English. 142 APPENDIX C SUBMISSION ..TOSSHC MANAGEMENT COMMITTEE October 2, 1974 SUMMARY AND IMPLICATIONS of a £!0£OSEp_DOCTOR_OF_PH^ i n ORGA_IZA_IO__L_BE__yj[0___Np__ INTRODUCTION The following presentation outlines a study i n health team communication to be conducted j o i n t l y by a Canadian r e h a b i l i t a t i o n hospital and Miss Betty McGill, a.S-R- The bri e f i s prepared to e l i c i t i n t e r e s t i n and support for the project by hosp i t a l Boards and Administrators. The basic study has been designed by Miss McGill as a part of a Ph.D. diss e r t a t i o n i n Organizational Behavior and Health Care at the University of B r i t i s h Columbia. The f i n a l design of the project w i l l be worked out i n close consultation with the par t i c i p a t i n g unit. The hosp i t a l w i l l be asked to provide the s i t e f o r the f i e l d study and p a r t i c i p a t i o n of i t s s t a f f , plus a small o f f i c e for the use of the research team. The researcher w i l l assume r e s p o n s i b i l i t y f o r obtaining funding, carrying put the actual study, and interpreting the resu l t s to the hos p i t a l . The intent of the research i s to develop a method for observing and analyzing how hospital s t a f f from various departments communicate with each other and with patients. 143 This w i l l be achieved by a four-month f i e l d study during the Summer of 1975. A team of two observers w i l l record communication patterns within and between d i s c i p l i n e s on two wards of the p a r t i c i p a t i n g h o s p i t a l . Findings w i l l be given to the hospital Administration as soon as the data are analyzed (approximately four months afte r the end of the f i e l d study). The thesis of the researcher w i l l focus on the actual method f o r observing communication patterns, and i t s relevance f o r future studies i n health care and other organizations. The present outline focuses on the rationale f o r the research, the study design, and implications f o r the p a r t i c i p a t i n g h o s p i t a l . Relevant l i t e r a t u r e i s currently being reviewed, and technical d e t a i l s of research method and analysis of r e s u l t s have met with the approval of the researcher's Thesis Committee. This information i s available i n the second draft of the research proposal. The Canada Council has expressed support f o r the project through granting the researcher a Doctoral Fellowship. RATIONALE Health organizations of the 1970's are increasingly confronted by a dilemma i n the provision of high guality patient care. In order to cope with the explosion of knowledge, the health d i s c i p l i n e s have had to become highly s p e c i a l i z e d . Keeping pace with developments i n h i s own f i e l d reguires that the medical/paramedical s p e c i a l i s t spend l e s s time learning about findings i n other areas. This has resulted i n high c a l i b r e care for s p e c i f i c problems, but at the expense of integration of the needs of the t o t a l patient. Within the past ten years, the problem of how to coordinate the team of health s p e c i a l i s t s has been discussed with increasing freguency by health planners and administrators. Hospitals which have attempted to implement a "team approach to patient care" have met with varying degrees of success, but the findings from one unit have not generally been transferable to others. Most research i n the f i e l d has focused on the r e s u l t s of i n e f f e c t i v e teamwork. The reasons why i t i s d i f f i c u l t to coordinate a team of health s p e c i a l i s t s have been given l i t t l e attention, and ways have not been developed for objectively measuring h_ow a p a r t i c u l a r team i s functioning. Recognizing that communication of information re. Patient care provides a v i t a l l i n k for coordination of the various d i s c i p l i n e s , the proposed study focusss on t h i s aspect of teamwork. I t begins from the basic premise that people communicate in a hospital setting i s of greater concern for the development of an e f f e c t i v e team than what they say. It then develops a method for objectively assessing how persons from various d i s c i p l i n e s communicate with other s t a f f and with patients. Robert Bales* ___eraction_Process_An^ an observation scheme which has been widely used i n small group studies, w i l l be 145 modified f o r use on hospital wards, at the nursing s t a t i o n , i n team meetings, and at other locations. The intention i s to develop a simple method, so that h o s p i t a l s t a f f could be trained i n i t s use f o r ongoing diagnosis of communication problems. While data-gathering w i l l be r e s t r i c t e d to two wards of a single hosp i t a l , the method i s expected to be s u f f i c i e n t l y f l e x i b l e to allow transfer to other organizations within and outside the health f i e l d . , The study should therefore give the f i e l d of health care administration a diagnostic t o o l which can be used not only by the p a r t i c i p a t i n g hospital and researchers, but also by other administrators and researchers seeking to i d e n t i f y communication breakdowns i n t h e i r organizations. This, i n turn, could be the f i r s t step towards changes i n patterns of providing care, to the eventual benefit of both the i n d i v i d u a l patient and the general society. STUDY DESIGN The projected study w i l l span a period of two years (see _______7). In Phase A, the emphasis w i l l be on f i n a l i z a t i o n of the research proposal and selection of the s i t e for the f i e l d study. Phase B involves the actual gathering of data, while Phase C w i l l analyze the r e s u l t s , provide feedback to the h o s p i t a l , and prepare the f i n a l t h e s i s . 146 FIG_RE_7 £H_SES_QF_RESE4iC__4_D_T_RG_T PHASE__ 1. explore alternatives for f i e l d s i t e 2. approach sources of funding 3. confer with Adminstration of hospital to f i n a l i z e f i e l d study 4. meet with hospital Administrators and Department heads to schedule study October 31,1974 November 30 January 30,1975 March 21 PHASE_B 1. begin f i e l d study May 1 2. orient hospital s t a f f and June 30 patients to study; f a m i l i a r i z e researchers with wards; researcher t r a i n i n g 3. gather t r i a l data; July 31 test computer programs 4. gather f i n a l data August 31 _______ 1. analyze results 2. communicate findings to hospital 3. prepare i n i t i a l a r t i c l e for publication December 15 February 15,1976 August 31 Phase_A focuses on the selection of a hospital f o r the f i e l d study. This i s the most c r i t i c a l aspect of the entire project. Minimal c r i t e r i a for consideration are: 1. hospital Board of Directors' consent to the project 2. top l e v e l Administrators* and Medical Directors' willingness to be committed to the study 3. support and i n t e r e s t of a majority of Department Heads, Nursing being a f o c a l concern 147 4. an inpatient unit, physical r e h a b i l i t a t i o n or extended care 5. adult patient population 6. minimum of two wards; more would be desirable, to allow choice 7. r e l a t i v e l y stable s t a f f i n g patterns ( i . e . few part-time or temporary) 8. reasonable length of patient stay In order to achieve maximum potential of the study, organizational support and willingness to p a r t i c i p a t e i n the study are e s s e n t i a l . Based on a 1969 p i l o t study using the proposed method i n a southern Saskatchewan ho p s i t a l , i t i s believed that the members of the p a r t i c i p a t i n g h o s p i t a l w i l l develop considerable i n t e r e s t i n the project. The hospital Administration w i l l receive extensive data on communication patterns on the two wards, for use i n diagnosis of present problems and for future planning. I f the findings are s u f f i c i e n t l y h e l p f u l to the unit, one or more s t a f f members could be trained i n the use of the method for ongoing diagnosis of communication problems. In addition, the hospital can gain recognition i n the f i e l d of health care research through an acknowledgement of i t s p a r t i c i p a t i o n i n a l l publication of r e s u l t s , should t h i s be desired by the Board. Phase B, the actual f i e l d study, w i l l span a period of four months. I t w i l l involve a team of two f u l l - t i m e researchers (Hiss McGill and another graduate student) observing communication behavior on the selected wards. 148 Choice of the l a t t e r w i l l be done j o i n t l y by the researchers and the hospital Administrator. The two observers w i l l record how s t a f f and patients communicate at d i f f e r e n t s i t e s on the wards. Generally they w i l l work separately, but during r e l i a b i l i t y checks they w i l l observe j o i n t l y . Since the observer w i l l minimize his interactions with s t a f f and patients during the data-gathering, interference with ongoing ward routines i s expected to be minimal. A major key to the success of the method w i l l be s t a f f and patient openness and willingness to communicate i n th e i r accustomed manner i n the presence of an observer. A relationship of trust must be established between the research team and persons on the wards. Considerable time w i l l therefore be spent during the f i r s t two months explaining the study to s t a f f and patients. In a small h o s p i t a l , t h i s w i l l involve meetings with a l l s t a f f and patient groups (approximately one hour of each s t a f f member*s time w i l l be reguired f o r formal o r i e n t a t i o n ) . In a larger unit, i t may be possible to do a general orientation f o r a l l s t a f f (15-30 minutes at a regular s t a f f meeting), plus detailed explanations f o r persons on the two study wards. Patients w i l l be oriented either i n d i v i d u a l l y or i n groups. The observation scheme and the type of data gathered w i l l be thoroughly explained, plus means f o r ensuring anonymity of a l l participants. No attempt w i l l be made at any time i n the study to i d e n t i f y an i n d i v i d u a l communicator. Staff c l a s s i f i c a t i o n s with only one member 149 w i l l be grouped to ensure c o n f i d e n t i a l i t y . Other communicators w i l l be i d e n t i f i e d by t h e i r occupational role {e.g. nurse, housekeeper, patient). Because considerable time w i l l be spent on practise observations and t r a i n i n g of the second observer, the research team w i l l reguire some space (however small) which could allow freedom of discussion during the f i e l d study. In £hase_C, the data w i l l be analyzed. Before any further work i s done on the thesis, the results w i l l be fed back to the pa r t i c i p a t i n g h o s p i t a l . They,/will^t^erelore•be f u l l y informed of,the.research findings before any presentation or .publication of the r e s u l t s . The extent to which the hospital w i l l be i d e n t i f i e d in the researcher's thesis w i l l be determined by the hospital Board or Administration. Time has also been allowed f o r preparation of some of the r e s u l t s f o r publication. I f the study can develop a stable method for measuring communication behavior within a hosp i t a l s e t t i n g , then the researcher w i l l have a d e f i n i t e o b l i g a t i o n to share the re s u l t s with others i n the f i e l d of Hospital Administration and Health Care fiesearch. At a l a t e r date, r e s u l t s w i l l be communicated to administrative researchers, for possible implementation i n other organizational contexts. _______ Implications of the proposed study are ex c i t i n g . If a 150 method can be developed whereby an untrained observer can be taught to objectively measure and analyze communication behavior, the t o o l can be used i n investigations of other factors influencing patient care. I t might also be adapted for Organizational Development and Inservice Education programs, enabling a h o s p i t a l to monitor i t s own communication patterns, and to adjust accordingly when those found to be detrimental to e f f e c t i v e and/or e f f i c i e n t patient care are uncovered. While these are mere speculations at the present time, i f the proposed study i s implemented, and i t s findings are s i g n i f i c a n t , then such "speculations" can be transferred to the realm of " p o s s i b i l i t i e s " . The p o t e n t i a l uses of the proposed method are considerable. The r i s k s inherent i n i t s development have been considered by the researcher's Thesis Committee. I t i s t h e i r opinion, as well as that of the researcher, that the scales of judgment are heavily weighted i n favor of attempting the project. This decision i s supported by evidence from the above mentioned p i l o t study which indicates that communication patterns of s p e c i f i c d i s c i p l i n e s can be i d e n t i f i e d using a modified form of I.P.A. The ensuing two years should therefore be an e x c i t i n g and challenging period, not only f o r the researcher, but also for Administrators and s t a f f of the hospital involved i n t h i s preliminary attempt to diagnose the "communication disorder" of a complex organization. 151 APPENDIX D PRESENTATION TO DEPARTMENT HEADS January 21, 1975 M1SC__A_HOSPITAL E_SEA_C__ST__Y in HEALTH_T_A__COMHOSICATIpN £2S£QSE; The purpose of the study i s to develop and test a method for observing and analyzing how hospital s t a f f communicate o r a l l y with each other and with patients and the i r f a m i l i e s . The high degree of s p e c i a l i z a t i o n of today's hospital departments has led to increased problems i n coordinating patient care, accompanied by greater d i f f i c u l t y i n communication. He are often not aware of where and why these communication breakdowns are occurring. Therefore, i t i s important to observe actual communication behavior on the hospital ward. Our ultimate aim i s to achieve more e f f e c t i v e and e f f i c i e n t use of hospital personnel, r e s u l t i n g i n a higher standard of patient care. ST_DY_DESIG__ 1. When: - May to August, 1975 (inclusive) - The f i r s t two months w i l l be spent t r a i n i n g the two observers, f a m i l i a r i z i n g ward s t a f f and patients with the study, and noting communication in s p e c i f i c areas. 152 The second two w i l l involve gathering information re. how s p e c i f i c groups communicate on the wards. - Most observations w i l l be done between 0800 and 1700 hrs., but some time w i l l also be spent on the wards i n the early morning and evening. 2. What: - He w i l l focus on how people communicate o r a l l y , rather than on what they are saying. - The recording scheme consists of twelve categories (see attached l i s t ) . Each time someone communicates with someone else, the process w i l l be coded into one or more of the categories. Only the coded information w i l l be recorded. , - Telephone conversations w i l l be included from the ward end only (plus noting which department sent or received the c a l l ) . - No written communication w i l l be examined. 3. Where: - Two wards of Wascana Hospital w i l l be s t u d i e d — S t a t i o n Five (rehabilitation) and Second West (extended care) . 4. Who: - To the greatest extent possible, the study w i l l be anonymous, i n that persons p a r t i c i p a t i n g w i l l be i d e n t i f i e d by the department (e.g. Purchasing) or position (e.g. Nurse attendant) to which they belong. - Persons who are the sole occupant of a p o s i t i o n (e.g. Ward Clerk) or the sole representative of a department (e.g. Social Service) on a ward w i l l have the option of being i d e n t i f i e d or being grouped with others. They are strongly encouraged to allow t h e i r position to be i d e n t i f i a b l e i n the data-gathering stage, with the option of deciding l a t e r the extent of d i s t r i b u t i o n of information re. their communication patterns. - A l l ward communications of the following groups w i l l be recorded: - patients - a l l persons resident on the wards being studied - physicians - those treating patients on the study wards 153 - s t a f f - a l l emplyees of Hascana Hospital who provide d i r e c t or i n d i r e c t care to patients on the two wards - persons employed by other organizations who partici p a t e i n di r e c t patient care on the wards (e.g. barber, hairdresser, clergy) - family - communication with s t a f f only - volunteers - guestionable - v i s i t o r s w i l l be excluded 5. How: - There w i l l be a team of two f u l l - t i m e observers. Generally one w i l l work on each ward, but they w i l l occasionally work together during training periods. - Each observer w i l l use a stenographer 1s notebook and penci l , or a mechanical punch which records the information i n a form which can be d i r e c t l y fed into the computer without re-coding. NOTE: at no time w i l l a tape-recording device be used. - He w i l l record i n the notebook who speaks to whom and how, phrase by phrase. - e.g., the seguence: DOCTOE TO NORSE : Good morning. How i s Mrs. Smith? Do you think we can discontinue the IV? NORSE TO DOCTOR : She had a good night, but s h e ^ r e s t l e s s today. would be recorded as follows: 040 : 189-000 : 56-- Some general categories for the type of content w i l l be developed and w i l l also be noted (e.g. patient care, s t a f f concern, administration) . - I t can be seen from the above example that no in d i v i d u a l s t a f f member (except as noted above) can be i d e n t i f i e d from the records of the observers, nor can what they have said be traced. 154 ________ After the data have been gathered, they - i l l be analyzed at the University of B r i t i s h Columbia, -here computer programs are being prepared. The results s i l l be fed back to the hospital approximately four months a f t e r the conclusion of the study. The exact feedback process M i l l be determined by the hospital Administration and Department Heads. Two types of information w i l l be provided: 1. Comparative: a) by department - comparison of the absolute volume of ward communication for each department i n each category, and of the p r o f i l e s giving the percentage of communications in each category. b) bY ward - comparison of absolute and percentage communications, broken down by day and time 2- l _ i r s of communicators: a) absolute and percentage p r o f i l e s for within -department communication (e.g. nurse to nurse, nurse attendant to ward clerk) b) absolute and percentage p r o f i l e s for between -department communication (e.g. housekeeping to maintenanace) c) absolute and percentage p r o f i l e s f o r communication between s t a f f of each department and patients,;;. a,-nd;i:.thgir • f a m i l i e s . The comparative s t a t i s t i c s (#1, above) should help the department heads to i d e n t i f y times and locations of heavy versus l i g h t volume, and to pick up general communication patterns. The p r o f i l e s of s p e c i f i c pairs of communicators should help the members of the two groups to i d e n t i f y areas of p o t e n t i a l l y more or less e f f e c t i v e communication. They 155 should then be better able to determine possible steps toward improvement of t h e i r methods of communicating., I f the information provided i s found to be h e l p f u l to the various groups within the h o s p i t a l — s t a f f , department heads, and a d m i n i s t r a t o r s — i t would then be possible to t r a i n several s t a f f members to use the method. Wascana might then be able to examine ad d i t i o n a l areas where s t a f f or patients f e e l there may be a breakdown i n communication. In t h i s way the hospital could begin to aore e f f e c t i v e l y diagnose i t s problems, attempt to solve them, and measure the r e s u l t s of that e f f o r t . The foregoing i s a summary of presentations made by Betty McGill, senior researcher, to f i v e groups of Adminstrators and Department Heads of Wascana Hospital on January 20 and 21, 1975. These w i l l be followed up by meetings with a l l s t a f f d i r e c t l y involved i n the study during A p r i l , 1975. 156 __________ _TAFF_AH__P__I__T_0_I____TX___S_EE_• I____SA_HOSPIT_L C0_M_NI____O_S_R_SEA_CH_PR_J_C_ _M_i The purpose of the study i s to develop a way of helping hospital s t a f f understand how, they communicate with other s t a f f and patients. WHERE; Station Five and Second West WHEN: May 5 to August 31, 1975 WHO: Betty McGill and Peter Bowman w i l l be the researchers. _ 2 _ i H e w i l l be observing how s t a f f and patients communicate at d i f f e r e n t locations on the two wards--ewg. at Nursing desk, i n patient rooms., Generally one researcher w i l l work on each ward, but sometimes we w i l l observe together to check our r e l i a b i l i t y . Host of the study w i l l be done between 0800 and 1700 hours, but some time w i l l also be spent i n the early morning and evening. WHAT: Staff w i l l be i d e n t i f i e d by t h e i r job t i t l e only. No names w i l l be used. Patients w i l l be grouped according to the communication demands which they place on the s t a f f . A conversation w i l l be coded, phrase by phrase, using twelve categories of _ow people communicate--for example, give or ask for information, agree or disagree. We w i l l not note what i s being said. Only the coded information of who speaks to w_om-and how w i l l be recorded. 157 For example, the conversation: DOCTOR TO NURSE : Good morning., How i s Mrs. Smith? Do you think we can stop the IV? NURSE TO DOCTOR : She had a good night, but she's r e s t l e s s today. would be recorded as follows: 040 : 189-000 : 56-RESULTS: afte r being analyzed, the r e s u l t s w i l l be discussed with a l l persons who have participated. They w i l l be able to see " p r o f i l e s " of how the nurse and doctor, or nurse attendant and patient, communicate, and to l e a r n where th e i r strengths and weaknesses i n communication may l i e . For example, the Communication P r o f i l e of Housekeeping aide to Patient might look l i k e t h i s : 1 2 3 4 5 6 7 8 9 10 11 12 VRSCflHS HOSPITAL COMMUNICATION PROFILE PERCENTRGE OF RCTS ]M EACH CATEGORY FOR STATIOM FIVE REG-STEREO NURSE PER!0D5= 25. T f l flCTS= 150. CCCUPaTIOHffl. THRflPY 0 1 2 M 3 4 5 6 7 8 9 J K L CATEGORY WSCRNfl HOSPITAL COWUNIOTNON PROFILE PERCENTRGE CF ACTS IN ERCH CATEGORY FOR STATION FIVE REGISTERED NIRSE TO PHTStOTHERSPIST 3 1 2 H 3 4 S 6 7 B 9 J K L CATEGORY PERIOOS= 28. ACTS: 203. WRSCRNA HOSPITAL COMMUNICATION PROFILE PERCENTAGE PF ACTS IN EACH CATEGORY FOR " STATION FIVE REGISTERED NURSE TO a. SOCIAL WORKER ta _ IN CATEGORY PERIODS: 10. ACTS- 124. VRSCAMA HOSPITAL COMMUNICATION PROFILE PERCENTAGE OF ACTS IN EACH CATEGORY FOR STATION FIVE SOCIAL WORKER PER1005= 10. TO «CT5= 92. REGISTERED NURSE WSCANS HOSPITAL COHHWICRTIO* PROFILE PERCENTAGE OF ACTS IN EACH CATEGORY FOR WSCANA HOSPITAL CCfWUCATtON PROFILE PERCENTAGE Of RCTS IN EACH CATEGORY FOR SECOND VEST 0 1 2 M 3 4 5 6 7 B 8 J K L CATEGORY 32.00 28.00-24.00-20.00-% OF ACTS 16.00-12.00 • .00 4.00 WASCANA HOSPITAL COMMUNICATION PROFILE PERCENTAOS OF ACTS IN EACH CATEGORY FOR STATION FIVE FT STAY 2-T OAYS . TO REGISTERED NURSE PERIODS* ACTS" • • # • * • * 28. »S. « * « * * • • * • * *> * * • • • •* 0 1 2 M 3 4 5 6 7 8 9 J CATEGORY 32.00 28.00 24.00 20.00 X OF ACTS 16.00* 12.00 8.00 4.00 WASCANA HOSPITAL COMMUNICATION PROFTLB PERCENTAGE OF ACTS IN EACH CATEGORY FOR STATION FIVE REGISTERF.O NURSE TO PT STAY 2-7 OAYS * * • * * * * • • * • * * • * * * * m m * * * * PERIODS" ACTS* 28. 250. * « * • *•' • • • • • ** * *• ****************************** 4 5 6 CATEGORY cr. 3 2 . 0 0 -28.00-24.00* 20.00-t OF ACTS 16.00* 12.00* 8.00* 4.00*-MASCANA HOSPITAL COMMUNICATION PROFUt PERCENTAGE OF ACTS XN EACH CATEGORY FOX STATION FIVE REGISTERED NURSE TO PT STAY 2-7 CAYS PERIODS" ACTS" 78. 761. • * « •* * * * * • * # * * • * * • * * 100* 40.00 35.00 30. 00 25. CO t OF ACTS 20.00-15.00 10.00 " 3.00 VASCANA HOSPITAL COHMUNICATION PROFIlff PERCENTAGE OF ACTS IN EACH CATEGORY FOR STATION FIV8 FT STAY 2-7 OAYS TO REGISTERED NURSE * * * * * # • • • • < • « • • • * * * PERIOOS" ACTS* 78. 392. F 3 4 5 6 CATEGORY ,00»***»**«*****»*rt***»«***«****•*****•«***»#***•**••***•**»••«• 0 I 2 M 3 4 5 6 7 8 9 J . K L CATEGORY 32.00-28.00* 24.00-20.00-t OP *' 16.00 12.00-8.00-4.00-WASCANA HOSPITAL COMMUNICATION PROFILE PERCENTAGE OF ACTS IN EACH CATEGORY FOR SECOND WEST REGISTERED NURSE TO PT COMNUNICATION-2 PERIODS-ACTS- 21. 180. • * * • • * • * * • * * * # * • * • « .00* 4 5 6 CATEGORY 40.00-39.00 30.00 25.00 * * OF ACTS * 20.00 WASCANA HOSPITAL COMMUNICATION PROFILE PERCENTAGE OF ACTS IN EACH CATEGORY FOR SECONO WEST PT COMMUNICATION** TO REGISTERED NURSE 13.00- * • 10.00- • • 5.00- * * ' * * » * m mm * •*• • • ** • * * • • • * * ** .OO****************************** * * * CATEGORY ***•••*«••«•**»*•*• 6 7 8 9 PERIODS- -21. ACTS- **62. « •***•*•» CO 6 4 . 0 0 -9 6 . 0 0 -WASCANA HOSPITAL COMMUNICATION PROFILE PERCENTAGE OF ACTS IN EACH CATEGORY FOR SECONO WEST REGISTERED NURSE TO PT COMMUNICATIONS PERIODS" ACTS' 8 . 58* 4 8 . 0 0 -4 0 . 0 0 -* O F A 3 2 . 0 0 2 4 . 0 0 -1 6 . 0 0 -8 . 0 0 -« • 0 0 * CTS * * * * * *** * * «» • • * **« * • * *•** * 4 3 6 CATEGORY _ 170 APPENDIX G FEEDBACK_INFO_M_IO__S_EET _ASCMi_aO_£I2_____I__NICAlI £_PLANATIO__OF_CO___N_ 1. At the top left-hand corner of each graph, the ward and two communicators are given. The f i r s t person i s the speaker, the second i s the l i s t e n e r . 2. At the top right-hand corner, two numbers are: given. "Periods" i s the number of diff e r e n t 5-minute time periods when these two persons were t a l k i n g . "Acts" i s the t o t a l number of communications. Generally, one act i s egual to a simple sentence. 3. Across the bottom of the graph, the numbers and l e t t e r s stand for the 14 di f f e r e n t types of categories in t o which a l l communications were coded; Cat 0 1 2 M 3 4 5 6 not codable seems f r i e n d l y dramatizes re. work dramatizes re. non-work agrees, l i s t e n s L - seems unfriendly K - shows tension J - disagrees - gives suggestions, leads 9 - seeks suggestions gives opinions gives information 8 - seeks opinions 7 - seeks information 4. The numbers down the left-hand side give: the; percentage of the speakers*s acts which were i n a particular category. Note that t h i s scale may d i f f e r from one graph to another. Suggestions f©reinterpreting a_Pair_Of_£_g£___s 1. Compare the number of acts for each speaker., For example, i f NORSE to PATIENT has 570 acts, and PATIENT to NORSE has 220, then the nurse communicated twice as much to the patient as he did to her. 171 2. Look at the o v e r a l l shape of the two p r o f i l e s . Where are they similar? Where do they d i f f e r ? 3. Examine the frequency of s p e c i f i c categories. For example, are the percentages of Cat. ,1's equal for DIETARY AIDE to PATIENT versus PATIENT to DIETARY AIDE? 4. Look f o r marked e g u a l i t i e s or i n e q u a l i t i e s between pairs of categories on each p r o f i l e . , ( i . e . . 1-L; 2+8-K; 3-J; 4-9; 5-8; 6-7) e.g. DOCTOR to NORSE - high 5 (giving opinion) and NURSE to DOCTOR - low 8 (seeking opinion) versus DOCTOR to NURSE - high 5 and NURSE to DOCTOR - high 8 5. Relate each pair of p r o f i l e s to what you know about the work you do on the ward. Which findings are easy to explain? Which aren't? 6. Discuss what you've learned with other s t a f f members. Compare your p r o f i l e s with t h e i r s . , 7. Write down any guestions which you have, or points which you'd l i k e to discuss further. (Note: This sheet was accompanied by a copy of the patient position code given in _2_endijc_B.) 172 A__E_DI__H IIMAL_SISQRl_o__Gg____i_ATig_s_^ February 28, 1976 To: Hr. ,R. Osiowy, Administrator, Wascana Hospital, Eegina, Saskatchewan He: Depart-ental_ffqllow-UB^^ During my with the f o l l o w i Assistant Direct - M. Lindberg, C - R. Osiowy, A. - G. Pernisie, P - M. Kaplunsky - R. Beebe - c. M i l l e r , M. - F. Richardson - F. Ford a l l s t a f f - L. Johnson - D. Leland - A. Sloan - D. Rackley ten-day stay i n Regina, meetings were held ng Wascana Hospital Department Heads and ors of Nursing: Staff Development Administration Dietary Housekeeping Haintenanace Nursing Occupational Therapy Porters Pharmacy Physiotherapy Psychology Social Work Speech and Hearing Spooner Tatarinoff Skaien Cook, G. De-ay C. Spooner attended a majority of the above meetings. M. Lindberg and the Staff Development Instructors also attended sessions with the two A.D.N.*s. At each meeting, persons present were asked: 1. What addit i o n a l work they or their s t a f f had done with the communication feedback since mid-December? 2. Did they see any need for action? I f so, what had they already done? What should be suggested for the future? 3. What type of help would they l i k e , either from myself during my stay, or from Staff Development i n the future, f o r : - general s t a f f members? - supervisors and/or department heads? 173 I__§_^§E___l__tal_2_ob_ems were discussed as follows: 1- Problem: low communication by s t a f f with aphasic patients on Station Five Discussion: Therapists f e l t that they do t h i s when the patient i s i n the department for treatment., Nursing believed that more communication occurs i n evenings when patients are on the ward. However, i n sp i t e of these factors, most s t a f f recognized t h i s as a problem. Mt_2___a_en: Most s t a f f believed that being aware of t h i s d e f i c i t was helping them to overcome: i t through i n d i v i d u a l e f f o r t . D. Rackley (Speech Therapist) i s now providing more support to the team by attending c l i n i c s and giving suggestions re. s p e c i f i c patients. This was appreciated by several departments. Suggestions; Several departments reguested Staff Development classes on p r a c t i c a l suggestions f o r communicating with the aphasic patient; This was discussed with Ms. Rackley and a f i r s t session for professional s t a f f was scheduled f o r Thursday, February 26, at 15:30 hrs. 2» Problem: team conferences on Station Five were not f e l t by some participants to be functioning as e f f e c t i v e l y as they might Discussion: Informal discussions were held with senior nurses, occupational therapists, physiotherapists and s o c i a l workers from both teams. I t seemed that i t was mainly the Senior Nurses who were d i s s a t i s f i e d with present functioning. Action^Taken: A. Balon, Staff Development Instructor for that ward, and I sat i n on one conference for each team. We then met with the four senior nurses and Mrs. Demay to discuss how. they, i n th e i r present r o l e as coordinator of the conferences, might improve th e i r effectiveness and e f f i c i e n c y . 3» Problem: desire of Department Heads and/or Supervisors for some t r a i n i n g i n how to help s t a f f communicate more e f f e c t i v e l y A______X_l£.§2.'• Four one-and-one-half-hour sessions on " E f f e c t i v e Communication" were conducted f o r Department Heads and Supervisors during my second week at Wascana. Topics covered were: - understanding the communication process - assessing communication effectiveness using the twelve Study categories - improving l i s t e n i n g s k i l l s 174 Attendance at these classes was as follows: 15 :00_;_o_16_30_hrs_ y Staff Development Second West Physiotherapy C l i n i c a l Supervisor Housekeeping Dietary Interne D i e t i c i a n Pharmacy S.S.H.C. Staff Development lli20_to_13__0__rs_ Personnel Station Five Staff Development Health Nurse Staff Development Administration 15__0_to_16_30_hrs. Outpatient C l i n i c s Dietary Speech & Hearing Speech 6 Hearing Pharmacy 15x00_to_16_30_hrs. Hedical Records Security Occupational Therapy Senior O.T.*s Social Work (substitute) A l l persons present participated a c t i v e l y i n discussions. In addition to the above, a separate one-hour class was held with a l l the Senior Physiotherapists and L. Johnson, focusing on th e i r s p e c i f i c concerns. 4. Problem: general s t a f f wanted to know more about e f f e c t i v e commnication 4__i2__2_ll§ n .s T o assess i n t e r e s t i n t h i s t o p i c , and provide a model for future Staff Development sessions, I conducted f i v e half-hour classes. P a r t i c i p a t i o n was as follows: Tues._ Feb_ 17 - 13:30 hrs. (5) 22:00 (12) 2M§§___Feb_l7 A. Balon M. Cook L. Golby M. Kaplunsky C. Liu P. Skaien (2) H. Lindberg ______________ E. B e i t e l G. Demay J. Kletke B. McCudden C. Spooner A. Tatarinoff F. Ford G. Pernisie E. Small M..Hunter (3) ____________ D. rJones P. Hatwiy F. Richardson (3) A. Sloan _____ Feb. v_18 - '15:30 (5) Thurs. _Feb. , 19-13:30 (4) 18:45 (16) 175 Suggestions; In addition to the above, opportunities for follow-up for general s t a f f exist through; a) i n d i v i d u a l or departmental education by Department Heads and A.D.N.'s b) orientation of new s t a f f by s t a f f Development c) follow-up classes on " E f f e c t i v e Communications" by Staff Development In addition to the above, the following ___£_--_§E___S__i_l_£_2_l§l§ were discussed during the meetings with i n d i v i d u a l Department Heads: Dietary: 1. Concern: low communication of Dietary Aides with a l l patients Discussion: Ideally, s t a f f f e l t that i f they had more Aides they could spend more time communicating with patients. However, given present l i m i t a t i o n s , t h i s problem cannot be a l l e v i a t e d . 2. Concern: Aides showed low communication with confused or aphasic patients Discussion: Dietary Aides suggested that a brief discussion with P. Skaien, either bi-weekly or monthly, giving them more information re. patients, would help them to make better use of the time they do have : with these patients. Action Taken: a) P. Skaien w i l l hold b r i e f informal meetings with Aides from each ward to give them suggestions re. communication with certain patients and an explanation of why the patient i s on a pa r t i c u l a r d i e t . Emphasis w i l l be placed on the patient's present communication a b i l i t y and diet rather than his/her d i s a b i l i t y or history. G. Pernisie w i l l inform s t a f f of t h i s at t h e i r next meeting, and discuss t h e i r r e s p o n s i b i l i t i e s re. c o n f i d e n t i a l i t y . b) Staff Development w i l l conduct a twenty^minute cl a s s for non-professional s t a f f (Dietary, Housekeeping) on Reality Orientation and general approaches to communication with such patients. 176 3) Concern: strong communication role of D i e t i c i a n with Station Five patients on therapaeufic diets Discussion: The communication p r o f i l e s of the D i e t i c i a n on Station Five during the p i l o t project on Therapaeutic Diets showed that a constructive r o l e was being f i l l e d in seeking patient input (opinions, suggestions) re. t h e i r d i e t s . Recent elimination of t h i s service cannot help but cause a deterioration i n the guality of t o t a l care offered to the patients on that ward. Suggestion: I would strongly recommend that t h i s service be made available to patients on a l l wards as soon as possible. P r o f i l e s supporting t h i s suggestion are available from P. Skaien. Housekeeping: 1. £2S£___* tension and unfriendliness between Housekeeping Aides and Housemen on Second Best Discussion: B. Kaplunski has held meetings with h i s staff~and~supervisors re., t h i s issue. They f e l t part of t h i s problem was a c o n f l i c t over job duties, which has been somewhat a l l e v i a t e d by compiling the job evaluations. Suggestion: I spent considerable time discussing t h i s c o n f l i c t with H. Kaplunski and his two Supervisors. Thay f e e l that the major cause of t h i s c o n f l i c t , which i s present throughout the Department, i s the large wage spread between the two c l a s s i f i c a t i o n s (H..Aide = $3.22/hr.; Houseman = $4.04), and that i f t h i s spread were reduced i t would be possible to get increased productivity from the female s t a f f . , 2. Concern: low motivation of Housekeeping employees; : lack of pride i n work - i s S - S - i S S - I n addition to the low wages of H. Aides, there are only two steps to the salary scale. Therefore, there i s no incentive to long-term employees to maintain a high l e v e l of productivity... Compounding t h i s problem i s the fact that on most wards. Housekeeping s t a f f receive l i t t l e or no recognition from other departments for a job well donei , Suggestion: a) The Supervisors recommended more steps on the salary scale to allow recognition for long-service employees. 177 b) Staff Development w i l l encourage ward s t a f f to thank and praise Housekeeping s t a f f for jobs well done. , This i s already being done by the Stations 5-6 Nursing s t a f f , and was f e l t to contribute to t h e i r being considered the most popular s i t e s for s t a f f to work. c) Staff Development w i l l also emphasize t h i s communication role to new employees during orientation, and by example on the wards. d) Staff Development w i l l conduct bi-weekly sessions with supervisors to improve t h e i r effectivenes. I_rsing_-_Second__est: 1« Concern^ between R.N. and G.P, on Category 1, Seems Friendly (R.N. = 3%; G.P. = 20%) there ;is an imbalance Suggestion: Balanced communication should be stressed during Orientation. This could also be pointed out to present s t a f f as part of a general session on e f f e c t i v e communication. However, i t was not f e l t that s t a f f would be w i l l i n g to attend such a class at present. . 2. Concern: low p a r t i c i p a t i o n by 28 and 2E Nursing s t a f f i n Inservice classes Discussion: This concern arose from the discussion of #1, above, rather than from the study feedback. , I t i s considered a major problem. Suggestion: a) Involvement of H. Cook, as a resource person f o r classes (staff are presently not making maximum use of her knowledge and experience). b) I n s t i t u t e a reguirement that a l l hos p i t a l s t a f f p a r t i c i p a t e i n a certain number of s t a f f development classes per year (e.g. 6) . ,. A previous Mascana system of each s t a f f member having his or her own educational card, on which such p a r t i c i p a t i o n was checked o f f , i s recommended. Staff would thus assume r e s p o n s i b i l i t y for deciding which classes were of most relevance to them, and of attending the yearly reguirement^ This recommendation i s now being considered by Staff Development. 178 _u_sin____Station_Fi_e: 1. Concern: Senior Nurse over-dramatizes A_ii2Bi Awareness of t h i s tendency provided through the December feedback seems to have a l l e v i a t e d the problem. Her communication i s now - perceived to be more functional. 2. Concern: a l l health team members are low on seeking the professional opinion of t h e i r own and other d i s c i p l i n e s Action: Nurses have discussed t h i s as a group with G. Demay, and have agreed to make an e f f o r t to correct t h i s problem. 3. Concern: low communication with aphasics Suggestion: In addition to the general points discussed on page 1, the nurses requested that the classes by D. Rackley be conducted at 15:30 hrs. i f possible, as evening s t a f f have the greatest communication with patients. Action: The cl a s s has been scheduled accordingly. Occupational Therapy: no intra-departmental issues were discussed; low communication with dysphasics i s aggravated by lack of s t a f f and lack . of Senior Occupational Therapist on Station Five. Pharmacy: 1. Concern: low communication with a l l patients Discussion: Pharmacists f e e l that t h e i r education does not prepare them s u f f i c i e n t l y for communication with patients, whether with or without the addi t i o n a l problem of aphasia. Action: They reguested that a l l „ f i v e members of the department part i c i p a t e i n the longer sessions on " E f f e c t i v e Communications" for Department Heads. 2. Concern: low communication with aphasics Suggestion: Classes by D. Rackley be held at the end of the morning or afternoon so that s t a f f can p a r t i c i p a t e . 179 Physiotherapy-1- Concern: frequency of negative emotional reactions (tension, unfriendliness) from patient to Behab. Attendant on 28 Discussion: The Senior Therapists, when discussing t h i s , f e l t that at l e a s t part of the cause of t h i s might be that attendants were working too f a s t . Action: The department has already implemented a series of weekly classes for Rehab. Attendants to review basic techniques and discuss the use of Reality Orientation. 2. Concern: lack of communication between Physiotherapist and Registered Nurse on Second Hest (no s i g n i f i c a n t p r o f i l e ) Action: L. Johnson w i l l work with the Senior Physio, i n that area to t r y to increase t h i s communication. Staff Development w i l l r einforce the need f o r communication with other departments during Orientation and through follow-up on the wards. 3. , Concern: lack of communication between Physiotherapist and patient on 29 Action: L. Johnson w i l l make s t a f f aware of t h i s d e f i c i t . 4. Concern: Physiotherapists are somewhat lower on Category 1 (seems friendly) than are the other departments with whom they are communicating Discussion: Because s t a f f are working under very heavy pressures at the moment, i t was not f e l t to be appropriate to discuss t h i s issue d i r e c t l y with thenu Action: This was handled by discussing " E f f e c t i v e Communication" with the senior therapists. Psychology: Although she received no direct feedback from the study, I have kept Ms. Leland informed on the feedback process for two reasons. The f i r s t i s that she has a d e f i n i t e interest i n t h i s area, since her Master's thesis was on Teacher-Pupil Communication. The second i s that she i s a potential resource f o r a l l Hospital departments on t h i s topic (including Staff Development for c l a s s e s ) . 180 S o c i a l .Work: 1. Concern: how to provide communication stimulation for patients on 2W Suggestion: Increase the use of volunteers through: a) A. Sloan providing greater support to the Personal Services worker, (G. McClelland has been carrying an exceptionally heavy load for many years, and i s needing increased professional support to maintain her effectiveness). b) More direc t contact with volunteers by A. Sloan, Close l i a i s o n between A. ,Sloan and G, McClelland i s necessary to the above two suggestions. . I t i s therefore c r i t i c a l for the Personal Services Worker to remain within the wascana Department of So c i a l Work. Given t h i s administrative arrangement, the above suggestions are feasi b l e with present s t a f f i n g . 2- Concern: low communication with new patients on S5 Action: S o c i a l Workers are now asking for n o t i f i c a t i o n from Admitting when a new patient a r r i v e s , and for notice from the Ward when there i s a discharge. They w i l l also be encouraged to see the patients more i n th e i r f i r s t week on the ward. . S_eech_and ^Hearing: 1. Concern: low communication with other departments on Station Five Action; D. Backley i s attempting to p a r t i c i p a t e i n Station Five c l i n i c s on a bi-weekly basis. The team discuss her patients f i r s t , so that reduction i n treatment time i s minimized. , She has also been phoning the ward with suggestions re. s p e c i f i c patients. Suggestion: D. Backley w i l l attempt to attend ward planning conferences every two to three weeks, to provide a l l s t a f f with suggestions on patient communication. While t h i s w i l l s l i g h t l y decrease her direct treatment s t a t i s t i c s , Ms. Backley f e e l s i t w i l l provide better o v e r - a l l care f o r patients because s t a f f w i l l be able to coordinate t h e i r approaches to the patient's communication d i s a b i l i t y . 2. Concern; need to a s s i s t patients to improve t h e i r communication s k i l l s Suggestion: Ms. Backley w i l l discuss with Ms. Demay the p o s s i b i l i t y of making language materials available on the ward for evenings and weekends. 181 _______: The above are the major areas of concern which were uncovered during my v i s i t . Most Department Heads expressed a d e f i n i t e i n t e r e s t i n following through on the i n i t i a l feedback from the Communication Study. As you w i l l note from the above, some action has been taken to resolve most of the problems and concerns i d e n t i f i e d by the Study. While most of the problems can be resolved at the departmental l e v e l , I f e e l that several have emerged which require Administrative support and/ or decision making. These are as follows: Dietary (#3); Housekeeping ( a l l ) ; Nursing - Second Best (#2); Soc i a l Work (#1); Speech and Hearing (#1). Throughout a l l of the discussions, i t was evident that Wascana s t a f f at a l l l e v e l s are aware of the importance of communication, and of maximizing i t s effectiveness i n times of s t a f f i n g pressures. I believe that the impetus of the study has helped to focus attention on t h i s problem, and on constructive solutions, and hope that t h i s concern w i l l continue i n the future. Thank you f o r your i n t e r e s t i n the the study. Respectfully Subaitted, B. HcGILL (SIGNED) Betty McGill, Research Project Director 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0093937/manifest

Comment

Related Items