THE RELATIONSHIPS OF CERTAIN STRUCTURAL" ‘ CHARACTERISTICS OF HOSPITAL WARDS AND THE ROLE CONFLICT AND JOB SATISFACTION 0F NURSESJLRL Thesis for the Degree 'of- Ph. D" MTCHIGAN STATE UNIVERSITY, RONALD WARREN RICHARDS 7? 1.968 . LIBRARY 1'" THEwb gan Stall! University This is to certify that the thesis entitled The Relationship of Certain Structural Characteristics of Hospital Wards and the Role Conflict and Job Satisfaction of Nurses presented by RONALD W. RICHARDS has been accepted towards fulfillment of the requirements for Educational _Bh.A_D_n_ degree in Wrat 1 on Q! A/ll ’2: /’ !_. Date M258.— 0-169 ABSTRACT THE RELATIONSHIPS OF CERTAIN STRUCTURAL CHARACTERISTICS OF HOSPITAL WARDS AND THE ROLE CONFLICT AND JOB SATISFACTION OF NURSES by Ronald Warren Richards In this study the relationships between certain structural characteristics of hospital wards and the role conflict and job satis- faction of nurses were examined. The problem was viewed as a specific case of the more general topic of professionals working in bureaucratic settings. Theoretically it was suggested that the nurse is undergoing a continual process of professionalization. She values a higher level of professional independence in her role behaviors and has a commitment to the profession's code of ethics rather than to bureaucratic rules and regulations. These values result primarily from the intensified and specialized nature of her training. In the typical hospital, stratification is high, freedom of role definition is low, the free flow of communications is low, and the professional nurse's predisposi- tions tend to be in direct conflict with the structural nature of her work setting. It was hypothesized that for the professional such structural conditions are related to low job satisfaction and to high role conflict. Ronald Warren Richards To test hypotheses regarding a relationship between the degree of structure of the nurse's work setting and her job satisfaction, an ex- perimental ward of a general hospital was compared with five other wards. The experimental ward was initiated for purposes of redefining the role of the nurse such that better patient care might be provided with fewer nurses, and for studying the feasibility of computer-based patient record keeping. To establish that a quasi-experimental design existed in a field situation hypotheses were tested regarding structural dif- ferences between the wards, differences in the need for order of the nurses of the wards, and no differences between the nurses on certain demographic factors. Hypotheses of differences on job satisfaction and role conflict between experimental and traditional ward nurses were then tested as were hypotheses of difference between the nurses with high perceptions of the structure of their wards and those with low per- ceptions of the structure of their wards, irrespective of specific wards. Data were obtained from a self-administered questionnaire. It was found that hospital wards differ in the degree to which nurses perceive that they are free to define their role. Nurses in the experimental ward were found to perceive their setting as low on the degree of structure, have a lower need for order and not significantly different levels of job satisfaction and intra-personal role conflict compared to nurses in more traditional settings. Significant differences between the perceptions of nurses, regarding the degree of stratification, free flow of Ronald Warren Richards task related communication, and freedom of role definition,cannot be accounted for by differences among nurses on certain demographic fac- tors.’ Generally, those with high perceptions of their settings were less satisfied with their job than those with low perceptions. Head nurses tended to have less role conflict than staff nurses and a lower need for order, while college degree nurses were found to have a higher need for order than nurses trained in hospital schools of nursing. THE RELATIONSHIPS OF CERTAIN STRUCTURAL CHARACTERISTICS OF HOSPITAL WARDS AND THE ROLE CONFLICT AND JOB SATISFACTION OF NURSES By Ronald Warren Richards A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Higher Education 1968 g ,3 3 J t. / x ///§”ER ,/ / ' (1“ .iICopyright by RONALD WARREN RICHARDS 1969 ACKNOWLEDGMENTS I wish to express my gratitude to the members of my doctoral com- mittee; Dr. Samuel A. Moore, II, Dr. Hilliard Jason, Dr. David C. Smith, and Dr. Richard L. Featherstone, for their help in this endeavor. To Dr. Moore, my committee chairman, I am indebted for his generous efforts and support throughout my doctoral program. I also wish to express special appreciation to Dr. Jason whose guidance and interest has been a special source of help. I am most appreciative of the extra time spent and interest shown by the hospital administrators, nursing supervisors and nursing staff of the several hospitals involved in this study. I wish also to gratefully acknowledge the financial support and sponsorship of the Michigan State University Division of the Michigan Association of Regional Medical Programs and the Michigan State University Office of Medical Education Research and Development. But most of all I am indebted to my wife, Ann, whose faith, encouragement and personal giving made the completion of this study and doctoral program a reality. To her, I dedicate this thesis. ii TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES CHAPTER I. II. III. IV. THE PROBLEM Purpose Hypotheses Theory Theoretical Definitions of Variables Overview REVIEW OF THE LITERATURE Structural Variables of a Social System Role, Personality and Social Structure in the Organizational Setting Implications Summary DESIGN OF THE STUDY The Sample Instrumentation Operational Definitions Statistical Hypotheses Analysis of the Data Summary ANALYSIS OF RESULTS Establishing the Treatment and Control Groups Testing for Differences Between Wards on Job Satisfaction and Intra-Personal Role Conflict Testing for Differences Between Nurses with High Perceptions and Low Perceptions of the Structure of Their Work Setting on Job Satisfaction and Intra-Personal Role Conflict iii PAGE ii r-F- baboon: 15 16 29 33 36 37 45 47 52 53 58 6O 62 62 72 74 Discussion Summary V. SUMMARY AND CONCLUSIONS Summary Conclusions Other Related Conclusions Theoretical Considerations Implications of the Study BIBLIOGRAPHY APPENDICES A. B. iv 84 90 92 92 94 97 97 104 109 114 133 TABLE 3.1 3.2 4.10 LIST OF TABLES Number of Nurses in Sample and Percent Returns Across Hospital Wards Demographic Description of Nurses in Sample Differences Between Hospital Wards or Stratification Differences Between Hospital Wards on Free Flow of Communications Differences Between Hospital Wards or Freedom of Role Definition Ranking of Six Hospital Wards According to Mean Differences Between Hospital Wards on Nurses' Need for Order Differences Between Need for Order of Nurses in Experimental Ward and Nurses in All Other Wards Combined Multiple Correlations, Beta Weights and Significance Levels for the Variable Free Flow of Communications Differences Between Experimental Ward and Traditional Wards on Job Satisfaction Differences Between Experimental Ward and Traditional Wards on Intra-Personal Role Conflict Differences Between High and Low Stratification Groups on Job Satisfaction PAGE 45 46 63 64 65 67 68 69 71 73 74 75 4.11 4.12 4.13 4.22 4.23 Differences Between High and Low Free Flow of Communications Groups on Job Satisfaction Differences Between High and Low Role Definition Groups on Job Satisfaction Differences Between High and Low Stratification Groups on Intra-Personal Role Conflict Differences Between High and Low Free Flow of Communications Groups on Intra-Personal Role Conflict Differences Between High and Low Role Definition Groups on Intra-Personal Role Conflict Differences Between High and Low Stratification Groups on Need for Order Differences Between High and Low Free Flow of Communications Groups on Need for Order Differences Between High and Low Role Definition Groups on Need for Order F Values and P Values for Job Satisfaction, Role Conflict and Order Scores Grouped According to High-Low Perceptions of the Structure of Their Work Setting The Pearson r Values for Stratification, Free Flow of Communications, and Role Definition and Job Satisfaction and Need for Order Number and Percentage of Nurses in Each Ward who Fall into High or Low Groups on the Structural Variables Wards Ranked According to the Percentage of Nurses who Fall into the Low Stratification and High Free Flow of Communications and Role Definition Groups Differences on Role Conflict Scores Between Head Nurses, Assistant Head Nurses, and Staff Nurses vi 76 77 78 79 80 81 82 83 84 86 87 88 89 CHAPTER I THE PROBLEM The study herein described represents an effort to systematically gather information on the nature of nursing in an exPerimental coronary care unit of a general hospital. First, the study was conducted to determine some of the changes in the social structure of hospitals that may result from computerization. Second, it was undertaken to assist those faced with the reSponsl- bility of training nurses. Finally, this study Was conducted to satisfy an academic curiosity regarding the structural nature of the nurse's hospital work setting and its effect on the professional staff. Not unlike other professions and organizations, the profession of nursing and the hospital are undergoing rapid changes. The shortage of registered nurses and advances in the technOIOQy of patient care are two of the forces causing these changes. in reSponse to these two forces a general hospital in a medium-sized mid-western city has estab- lished a coronary care unit as an eXperimental setting. TWO of its pur- poses are to redefine the role of the registered nurse, hereafter referred to as nurse, such that better patient care might be provided with fewer nurses, and to conduct a feasibility study of computer-based patient record keeping. This experimental hospital unit provided an Opportunity to examine the problens related to the changing role of health care professionals. -2- Little is known of the problems related to redefinition of the nurse‘s role, what these problems may be, their causes, or the means of resolving them. Changing the role of the nurse could alter her rela- tionships with physicians. it could cause dissatisfaction and personal conflict among those nurses accustomed to and preferring the "old way." it could require changes in the training of nurses, and it could change the type of individual attracted to and remaining in the profession. if the shortage of nurses is to be successfully Countered by using the nurse in new‘ways, systematic study is essential. Successful implementation of new technology also requires system- matlc study. Computers have been used extensively in the business area of hospitals but only recently have efforts been made to more directly use the machines to aid in providing patient care. Such use requires different skills and orientations of physicians and nurses to their respective roles. There is as much fallacy as fact in the conceptions of persons not familiar with what computers can do and how they are utilized. Some see the COMputer as dehumanizing their Jobs, transform- ing them into technicians rather than professionals. Others view com- puterization as an attack on the quality of the Job they have been performing. Thus, technological change, while it holds the potential for increasing the quality of patient care, creates many problems of personnel readjustment. Systematic study of this experimental unit -3- was undertakan with the intention of identifying these problems and gathering information that might serve as the basis for their resolution. Pur se It was the purpose of this study to examine two basic interrelated issues. They are: l) Do the work settings of nurses differ with regard to the structural characteristics of freedom of role definition, stratification, and free flow of task related communications? and 2) Do nurses‘ work settings that differ structurally, differ with regard to intra—personal role conflict and job satisfaction? flypgtheses in broad research form there are three hypotheses which guided the study; 0 2) 3) Freedom of role definition is related to intra—personal role conflict and Job satisfaction; Stratification is related to job satisfaction; and Free flow of task related communications is related to job satisfaction. Theory There are certain factors, inherent in both the predispositions of the nurse and the traditional orientations of the hospital, which tend to create role conflict1 and job dissatisfaction2 among nurses. The dilemma faced by the professional working in a bureaucratic organiza- tion is of central importance to the understanding of the nurse's role conflict and job dissatisfaction. Blau and Scott4 characterize the professional as one who adheres to norms of service and a code of ethics, and has a status achieved by his performance in accordance with norms laid down by his colleague group. Gouldner5 found professionals to have a high commitment to skills and an orientation to reference groups other than the organization. In sum, the professional is a highly trained specialist who believes that his actions are not subject to clearly enumerated organizational rules and regulations. 1Samuel Levey. "The Hospital Reconsidered." Hospital Administra- tion. CFall, 1961) pp. 6-21. 2Chris Argyris. DiagnosingFHuman Relations in Organizations: A Case Study of a Hospital. New Haven: Labor and Management Center, Yale University. 0956). 3H. L. Wilensky. "The Dynamics of Professionalism: The Case of Hospital Administration." Hospital Administration. Vol. 7. (1962). pp. 6-24. l'l’eter M. Blau and Richard W. Scott. Formal Orggnizations. San Francisco: Chandler Publishing Co. 1962. 5Alvin Gouldner. "CosmOpolitans and Localst'TOward an Analysis of Latent Social Roles, Part i." Administrative Science Quarterly, Vol. 2 (l957). pp. 28l-306. -5- The nurse has for years been a dedicated but low ranking subor- dinate in the hOSpital hierarchy. Changes in the means of providing patient care, however, demand an ever increasing leVel of skill. Con- tinual upgrading of the nurse's training has moved nursing closer to the normative conception of a profession. Today‘s nurses tend to be persons with a lower need for rigidity and order as a personality trait than nurses of years past. But the hOSpital organization has not been amenable to such change. The predi5position of the nurse tOWard independence of role behavior is rigidly circumscribed by the bureau- cratic rules and regulations of the organization. The result is lfl££§r personal conflict or what Levey6 calls "intra—individuai conflict." The nature of the nurse‘s work setting can be more fully under- stood if certain structural characteristics of the hoSpital are considered. As a bureaucratic organization, tasks that constitute given hospital roles are clearly defined and are not subject to any significant degree of choice by the person filling that role. in other words, the degree of role definition permitted of persons filling a role is low. Coser7 notes that nurses ". . . diSpiayed ritualism in their behavior" particularly in the medical ward. 6Samuel Levey. .92. £13, p. l5. 7Rose Laub Coser. "Authority and Decision-Making in a HosPitai." American Sociological Review, Vol. 23. (1958) pp. 56-63. -6- Another broad structural characteristic is the rigid status hierarchy of the h05pital8 in which norms of the hOSpitalls social system circumscribe acceptable behavior for each status leVel. This rigid status hierarchy, or stratification as Seeman9 calls it, serves to limit the tasks that a nurse is permitted to perform and also estab— lishes patterns for acceptable status interaction. in terms of this conceptualization, the typical hosPital ward can be described as having a high level of stratification, especially‘when considering the nurse vis a’vis the doctor. Such a rigid hierarchy ‘0 studied the strati- may decrease the quality of patient care. Seeman fication of hOSpitai wards as it relates to the quality of patient care. .Measuring quality on the basis of internes1 personal judgment, superiorsi ratings, and certain quantitatiVe measures such as length of patient stay, medication errors, and frequency of consultations, he found wards with high stratification to be low on these measures. 8Temple Burling. gt;- gi. The (live and Take in HOSpitals. Nev York: Putnam. i956. Basil S. Georgopoulos and Floyd C. Mann. "The HOSpltal as an Organization." HOSpitai Administration, Vol. 7. l962.. pp. 50-64. Hans 0. Mauksch. "The Organizational Context of Nursing Practice." in The NurslggiProfession: FiVe Sociological Essays. Fred Davis (ed) New York: John Wiley. 1966. 9Melvin Seeman. .E£-.El- "The Measurement of Stratification in Formal Organizations." Humgn Organiggtlon, Vol. 19. (Summer, 1960) PP- 90-95- loMeiVin Seeman, J. W. EVans. ”Stratification and HOSpital Care: Part i.” American Sociological Review. vol. 26. (l96l). - ..7.- Another structural characteristic of note is related to stratifi- cation. in the hOSpltal there is little sharing of ideas, Opinions, and knowledge between doctors and nurSes for the purpose of providing more adequate patient care. That is, the free flow of task related communications in the hOSpitai Is low 11 sIau‘Z discusses the nature of bureaucratic authority noting that the "pronounced inequalities that are often found in bureaucratic organizations are not essential for coordination and uniformity." He also contends that extensive and rigid definitions of role behaviors are dysfunctional since they tend to inhibit the correctiVe function of free interaction in group problem- soIvIng,13 The nurse is confined by theSe structural characteristics of the hOSpital. She is in the ambiguous position of being at the intersec~ tion of the hOSpital's dual authority system,14 She is reSponsible to both the lay administration and the medical staff. ‘The incongruency of role eXpectancies resulting from several superiors sharing the authority ”Peter M. Blau, Richard W. Scott. 92. gI_t_. p. 122, Burling,. 92- 913. pp. 101-105. 12Peter M. Blau. Bureaucracy_in Modern Society. New York: Random House. 1956. I3Peter M. Blau and Richard W. Scott. 92, 213, pp. 121-123. l1"1-lartrey L. Smith. "Two Lines of Authority: The Hospital‘s Dilemma.” in Gartley E. Jaco (ed) Patients, Physicians,_and illness. Glencoe, lliinois: The Free Press. 1958. pp.468—477 and Buriing, Op. Cit. pp; 101-110. -3- over a subordinate is similar to the role ambiguity that WispeIS found related to high anxiety. The nurse is also subject to what Smithl6 calls "blocked mobility," a situation which exists when promo— tion and advancement are blocked because the skills acquired in one area are not transferable to another. The organization has no way of rewarding high performance by advancement to a higher status level without promoting the nurse out of nursing into administration, a job viewed as demeaning paper shuffling.I7 One alternative method of rais- ing the nurse‘s status level is for nurses to Seek a role redefinition which includes tasks formerly associated with the doctor's role. in such a way a certain degree of UPWard mobility of the nurse is achieved. But it is not achieved without conflict behween doctor and nurse statuses regarding what constitutes their reSpective roles. That is, Inter-status role conflict exists over what tasks constitute the nurse's role. Some evidence is available indicating that such con- flict has an effect on patient care. Studies in mental hospitals found lsLauren G. WISpe and Paul Thayer. "Role Ambiguity and Anxiety in an Occupation Group." Journal of Social Psychology, Vol. 46. (August, 1957) pp. 41-48. l6Smith. 92, 21;, p. 473. i7Argyrls. Qp;_§l£. pp. 67-69. -9- that when there was conflict among the professional staff, the patients became pathologically agitated.l8 Finally, the structural characteristics of the hospital restrict the professional nurse such that her level of job Satisfaction may be low. This contention is supported by research in organizations employ- ing professionally trained personnel. Blaul9 studied professl‘gnals in a federal agency and found that the demand for worker conformity to task procedures tended to stifle initiative and decision—making by the workers. He also found that those working in the setting demanding high conformity were less satisfied with their jobs. Wilensky20 came to similar conclusions in his study of intellectuals working for labor unions. Katz and Kahn2] report that as the tasks performed become more varied and require greater skill more productive*workers become more satisfied with their job. For professional workers job satisfaction and performance are positively related.22 Blocked mobility and I8Alfred H. Stanton and Morris S Schwartz. The MentglfiHospltal. New York: Basic Books. (1954) pp. 3u2-365. William A. Caudill. The EgyChiatric H05pita1 as a Small Society. Cambridge, Massachusettsz Harvard University Press. (1958)» pp. 87-127. '9Peter M. Blau and Richard W. Scott. 92. ELE- pp. 129-131. 20H. L. Wilensky. igtellectuals in Labor Uniogg, Glencoe, lliinois: Free Press. 11956) pp. 126-144. 2‘nanIeI Katz and Robert L. Kahn. "Some Recent Findings in Human Relations Research in industry." in Guy E. Swanson, 5; EL; (eds) Readlggs in Social Psychology, Rev. ed. New York: Holt. (1952) pp.650-665. 22Rensis Likert. New Patterns of Manggement. New York: McGraw- Hill. (1961). -10- sltuations in which the subordinate feels powerless to influence his superior were found by Pearlin23 to be related to anxiety. Such empirical studies along with the theoretical interpretations of jobs 24 and need satisfaction lead to the hypothesis that job satisfaction is related to role conflict and certain structural characteristics of the organizations. ‘The concept of role conflict requires clarification. Merton defines the term, role, as ''the behavioral enacting of the patterned expectations attributed to a position in a social system."25 Role as used here follows Merton but is more narrowly defined as the tasks attributed to a position in a social system. Role conflict exists either a) when there is lack of agreement as to the tasks that are attributed to a position in a social system, or b) when enactment of the tasks leads one to do what is incongruent with what the doer be- lieves ought to be done. 23Leonard Peariin. "Alienation from Work: A Study of Nursing Personnel.“ Americgg,Soc1010glcal Review, Vol. 27. (1962) pp. 314-326. 2“Likert. .Qp.‘§l£, p. 6& 16 and Chris Argyris. Personality;_ and Organization. New Ybrk: Harper. (1957). 25Robert K-.Merton. Social Theory and Social Structure. New York: 'The Free Press. (1957) p. 368. -11- A two dimensional classification of role conflict provides further clarification. On the one hand, role conflict can exist over which tasks are believed to constitute role, that is, the actual role: and role conflict can exist over which tasks are believed ought to consti- tute role, that is, the desired role. On the other hand, role conflict can be classified according to with whom the conflict exists. in this case it may be with oneself, intra—personal role conflict. Or it may be between persons of different social positions, inter-status role conflict. it follows from the above conception of role conflict that the degree of conflict depends in part on the number of tasks disagreed upon as constituting a role, or ought to constitute it. Each task, however, will not likely elicit the same degree of conflict when dISa- greed upon. Disagreement over whether a given task, X, constitutes part of the role, may produce greater conflict than disagreement over whether a QIVen task, Y, constitutes a part of that role. For one reason or another those holding the role expectations differentiate between Specific tasks. The degree of conflict then depends on both the number of disagreed upon tasks and relative weighting of the tasks in terms of their significance as perceived by those involved. it has been suggested that the nurse is undergoing a continual process of professionalization. Primarily resulting from the changes _12- in her intensive training she possesses a value orientation toward a higher level of professional independence in her role behaviors and a commitment to the profession‘s norms and values rather than to bureau- cratic rules and regulations. Because, however, the nurse practices her profession within the confines of a formal organization, the hOSpi- tal, her role is significantly influenced by certain structural charac- teristics of the work setting. in the typical hOSpitai, stratification is high; freedom of role definition is low; the free flow of communica- tions is low. Therefore, the professional nurse‘s prediSpositions tend to be in direct conflict with the structural nature of her work setting. This theoretical conceptualization leads to the hypothesis that for the professional such structural conditions are related to low job satis- faction and to high role conflict. if, as the case seems to be, one type of work setting is significantly lower in stratification and higher in freedom of role definition and free flow of task related communications, it follows that the professional nurse working in the less structured situatioanill have less role conflict and a higher level of .job satisfaction. For such a setting is more compatible with her pro- fessional pred15positions. -13- Theoretical Definitions of Variables The following terms are used throughout the discussion of the study. In order to provide for clarity of meaning they are stated here in glossary fashion. They are classified as being either inde- pendent or dependent variables, although in the strictest sense, field studies of this type do not have wholly independent and dependent variables. Independent Variables 1) 2) 3) Freedom of role definition--the extent to which persons performing a formal role in an organization are able to determine the tasks that will constitute that role. Stratification-~the degree of concentration of influence and decision-making, the degree of closeness, friendli- ness and intimacy of the relationships between personnel of a formal work setting, and the amount of deference expected and the emphasis upon matters of rank and status symbols. _Free flow of cogmynication--the sharing of ideas, opinions and knowledge between status levels for the purpose of more adequately performing one's job. -14- Dependent Variables: The dependent variables are so designated because their focus is on the nurse. 1) 2) 3) Overview The Intra-personal role conflicto-the conflict between what one believes are the actual tasks that constitute a role and what one believes are the desired tasks that should constitute a role. Job satisfaction--the positive feeling associated with the performance of one's formal role in an organizational setting. Need for order--to have written work neat and organized. to make plans before starting on a difficult_task, to have things organized, to keep things neat and orderly, to make advance plans when taking a trip, to organize details of work, to keep letters and files according to some system, to have meals organized and a definite time for eating, to have things arranged so that they run smoothly without change. nature of this study and the findings will be presented in the following manner. In Chapter 1, the study was introduced and a theore- tical framework explained. In Chapter 2, the pertinent literature is revised. An explanation of the design of the study constitutes Chapter 3. In Chapter 4, the results of the study are analyzed, and in Chapter 5 con- clusions and theoretical implications of the findings are presented. CHAPTER II REVIEW OF THE LITERATURE Empirical knowledge and theoretical speculation from a diversity of disciplines are relevant to the study of complex organizations, and probably every type of research methodology has been used by behavioral scientists to study this dominant form of modern institution. It is possible to construe this topic as part of several very different theoretical and empirical frames of reference. It can be seen, for example, as a study of communications, social stratification, nursing procedures, bureaucratization, or role theory. The viewpoint from which this study has been considered, however, is that area of pre- vious investigation known as the study of the professional in a bureau- cracy. The review of the literature is organized into two parts: 1) investigations related to organizational bureaucracy and structural social system variables; and 2) investigations related to the concept of "role" as it is applied to organizational settings. -15- -l6- Structuralfiyariables of a Social System Max Weber's ideal-type model of bureaucracy has served as the basis for much of the investigation of formal organizations.26 Accord- ing to the Weberian model a bureaucratized organization has six characteristics! 1) 2) 3) 4) Hierarchy of authority--the extent to which the locus of decision-making is pre-structured by the organization. Division of labor--the extent to which work tasks are sub—divided by functional specialization by the organization. Presence of rules--the degree to which the behavior of organizational members must follow organizationally defined procedures. Procedural specifications--the extent to which organiza- tional members must follow organizational defined procedures. 26Max Weber. The Theory of Economic and Social Organization. trans. sity. A. M. Henderson and Talcott Parsons. New York: Oxford Univer- (1947). -17- 5) Impersonality--the extent to which both organizational members and outsiders are treated without regard to individual qualities. 6) Technical competence-~the extent to which organizationally defined universalistic standards are utilized in the per- sonal selection and advancement process.27 Hall, noting that, "All too often organizations have been labeled bureaucratic for purposes of study when little evidence has been pre- sented that they are in fact bureaucratic,"28 studied several organiza- tions to ascertain their extent of bureaucratization.29 Using a questionnaire form containing items designed to measure Weber's six components of bureaucracy he found that, ". . . what is commonly approached as a totality (bureaucracy) is not such an integrated whole in reality."30 27Richard H. Hall. "Some Organizational Considerations in the Professional-Organizational Relationship." Administrative Science Quarterly, Vol. 12, No. 3. December (1967) p. 465. 28Richard H. Hall. "The Concept of Bureaucracy! An Empirical Assessment." American Journal of Sociology. Vol. 69. (July, 1963) p. 32. 291bid. 3°Ib1d. -13- His findings indicate that the commonly ascribed dimensions are present in organizations but the extent of their presence varies. The results of a second study by Hall, similar in many respects to the above mentioned study, indicate that the bureaucratic dimensions vary somewhat independently in diverse types of organizations.31 In both studies, however, the variability of the technical competence dimension is random. Litwak32 deve10ped two models of organizations that focus on con- flict. They are the human relations model and the professional bureau- cracy model. Whereas organizations which are typified by the Weberian Inadel are those which deal with uniform events, the human relations type organization deals with nonuniform events. The professional tuireaucracy model is, according to Litwak, the most common. It.applies tr) organizations which, a) must deal with events both uniforn and 'noruiniform, and b) need to have jobs requiring great social skills as 'welJl as jobs requiring traditional areas of knowledge. Litwak points to the hospital as a typical example of a professional bureaucracy. 31 Richard H. Hall and Charles Tittle. "Bureaucracy and Its Corrwslates." American Journal of Sociology. Vol. 72. (November, 1966) PP- 267-272. :32Eugene Litwak. "Models of Organization Which Permit Conflict." Amman Journal of Sociolgy, Vol. 67. (September, 1961) pp. 177-184. -19- Hall hypothesized that if Litwak's professional bureaucracy model is the case, the "organizational divisions or departments that specialize in tasks nonuniform or difficult to routinize will be less bureaucratic on the six Weberian dimensions."33 Using his "bureaucracy instrument" he found that on the dimensions of hierarchy of authority, division of labor and procedural specifications the results supported the hypothesis but on the other three dimensions there was no particular pattern. Litwak's professional bureaucracy model can be described, in the words of Thompson, as an organization which ". . . attempts to accommo- date specialization within an hierarchical framework."34 Due, however, to the hierarchically arranged chain of command the organization is 35 the unable to deal with specialization. That is, in Udy's terms person "legally" vested by the organization with the responsibility to control the actions of a specialist subordinate does not possess the knowledge to make that control seemurationalfv Thompson explains the conflict that may occur between the organization and an individual mem- ber as follows: 33Richard H. Hall. "Intraorganizational Structural Variation." Administrative Science Quarterly, Vol. 7. (Dec. 1962) pp. 295-308. 34Victor A. Thompson. "Hierarchy, Specialization, and Organiza- tional Conflict." Administrgtive Science Quarterl , Vol. 5. (March, 1961) pp. 485-521. 35Stanley H. Udy, Jr. "Bureaucracy and Rationality in Weber's Theory." American Sociological Review, Vol. 24. (Dec. 1959) pp. 791-795. -20- The ability of an organization to satisfy the personal needs and motives of all its participants is compromised by the definition of hierarchical roles. Job satisfac- tion depends upon the degree of skill involved, the variety of activities, the degree of autonomy, the con- sistenqy of the job with the individual's self image and the predictability of work relations. These elements of job satisfaction may come into conflict with hierarchical rights to assign activities and to supervise them. The right of arbitrary command may conflict with cultural norms of independence, and the right to unusual deference, 'with norms of equality and dignity. Thus, the self images of subordinates are enlarged. 6 In general, then, Thompson views conflict to be of three types: 1) that due to violation of role expectations; 2) that concerning the reality of interdependence; and 3) that arising from blocks of spontaneous communication.37 A related conceptualization of organizational conflict is that 38 of Pondy. In his review of the literature he found the term con- flict used to describe: 1) Antecedent conditions (for example, scarcity of resources, policy differences) of conflictful behavior. 36Thompson. ‘92. Cit. p. 506. ’- 37Ibid. p. 11. 38Louis R. Pondy. "Organizational Conflict: Concepts and Models)‘ Administrative Science Quartegly, Vol. 12 (2), (Sept. 1967) pp. 296-320. -21.. 2) Affective states: (e.g. stress, tension, hostility, anxiety, etc.) of the individuals involved. 3) Cognitive states of individuals. i.e., their perception or awareness of conflictful situations, and 4) Conflictful behavior: ranging from passive resistance to overt aggression.39 Rather than argue for the validity of one usage over another, Pondy suggests that conflict be considered as a dynamic process. Thus, in ‘40 may exist as an early some instances "certain conflict potentials stage of the conflict process without the parties involved being aware of any bases for conflict. One most relevant source of such "latent conflict" is drives for autonomy, or, as Pondy puts it, ". . . autonomy needs form the basis of conflict when one party either seeks to exercise control over some activity that another party regards as his own pro- vince or seeks to insulate itself from such control."41 Assuming that 39Ib1d. p. 298. 4°Ib1d. p. 299. 411b1d. p. 300. -22- one seeks to control the conditions of one‘s existance and possibly those of others‘, such striving for autonomy is likely to create con— flict situations between superior and subordinate.42 Interestingly enough, claims Pondy, because the conflict is interpreted as a "resent- ment of the exercise of personal power," the organization's leadership reacts by substituting impersonal rules and regulations. The result is an even greater organizational rigidity which further limits the autonomy of subordinates.43 Underlying both Thompson's and Pondy’s conceptualizations of con- flict is the notion that conflict between an individual organization and the organization itself is inherent, and further, that for those persons with a higher need for autonomy, the conflict is greater. It follows that for professionals, typically viewed as highly valuing autonomy, conflict exists (with their employing organization) since their professional norms and values are inherently incompatible with 44 the organization. However, Hall hypothesized, that while conflict may exist, conflict may not be inherent since ". . . the organizational 421bid. p. 315. 431bid. p. 314. 44Richard H. Hall. "Some Organizational Considerations in the Professional-Organizationa1 Relationship." Administrative Science Qggrterly, Vol. 12 (3) (Dec., 1967) pp. 461-478. -23.. setting for professional employees may not confront the professional 'with organizationally generated normative systems which restrict him in his self-regulatory activity."45 To test the hypothesis,Hall studied twenty-three organizations varying in their degree of professionalization. The organizations were divided into three groups on the basis of the type of setting within which the professionals carried out their work. Autonomous organizations, such as a law firm, are characterized by the professional being subject to his own jurisdiction. Heteronomous organizations, such as a social work agency, are those in which professionals are subordinated to an administrative framework. The third group is the professional department where the professional is a member of a depart- ment which is a part of a larger organization. A questionnaire, composed of items intended to measure respondents' perceptions of the degree of bureaucratization on Weber's six dimen- sions, was administered to professionals in these organizations. Except for the technical competence dimension, on which the findings were reversed, the autonomous organizations were found to be the least bureaucratized. Hall views the technical competence results as suggest- ing that more emphasis on this dimension, for purposes of promotion and advancement evaluations, is highly compatible with professional standards. 45Ibid. p. 462. -24- Since the level of bureaucratization, which is seen as the source of professional-organization conflict, was not the same across types of organizational settings, Hall concluded that conflict is not in- herent in that relationship. 0f perhaps more importance, here, howb ever, are the findings related to the professional department setting. This setting was perceived by professionals working in it to be the most highly bureaucratized, and therefore, if Hall's reasoning is followed, it has the highest degree of professional-organization conflict. Etzioni interprets the conflict between professionals and their employing organizations as being a problem of administrative vs. pro- fessional authority. According to Etzioni: Only if immune from ordinary social pressures and free to innovate, to experiment, to take risks without the usual social repercussions of failure, can a professional carry out his work effectively. It is this highly indi- vidualized principle which is diametrically opposed to the very essence of the organizational principle of con- trol and coordination by superiors--i.e., the principle of administrative authority.46 LaPorte 47 studied the strains between scientists and administra- tors in industrial research organizations from a theoretical base 46Amitai Etzioni. Modern Orggnizations. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. (1964) p. 76. 47‘Todd R. LaPorte. "Conditions of Strain and Accommodation in Industrial Research Organizations." Administrative Science Quarterly, Vol. 10. (June, 1965) PP. 24-38. -25- similar to Etzioni's. From the literature he derived seven sources of possible conflict. Using data from several previous studies, his analysis found that only two sources caused conflict. goal orientation and restrictiveness of measures of control and coordination. In inter- preting his findings, LaPorte argues that there existed three "elements of accommodation." They were separation of major functional roles within the organization between scientists, managers, and administrators, mechanisms neutralizing major external strains from corporations and government, and the similarity in background of the scientists, managers, and administrators. This similarity of backgrounds was not far from what Goss found to be Operative among physicians in an outpatient clinic.49 She examined the generalization, similar to that questioned by Hall, that professionals require a freedom to make decisions according to their own trained judgment. Her results indicated that no conflict arose primarily because all persons in the hierarchy were physicians, and 48Ibid. p. 31. 49Mary E. Goss. "Influence and Authority Among Physicians in an Outpatient Clinic." American Sociological Review, Vol. 26. (l)(February, 1961) pp. 39-50. ~26- conflict was avoided because the professional norms governing doctor- doctor relationships deemed such conflict unacceptable. What evolved in this situation instead of a formal authority hierarchy was a "formal advisory relationship."50 Other investigators have examined the nature of the professional's conflict with his employing organizations, although each orientation to the problem varies somewhat. Kornhauser, for example, defined the problem as being the basic organizational dilemma of autonomy vs. inte- gration.51 That is, scientists must be given enough autonomy to enable them.to fulfill their professional needs, yet their activity must con- tribute to the overall goals of the organization. Miller?2 however, examined the relationship betweer1alienation from work and the degree of organizational control and the number of professional incentives. He hypothesized that: 1) the degree of alienation fromrwork is positively associated with the degree of organizational control and negatively associated with the number of professional incentives, and 2) these relationships are stronger for persons with Ph.D.'s than for those with M.A. or M.S. degrees. 51WilliamKornhauser. Scientists in Industry: Conflict and Accommodgtion. Berkeley: 'University of California Press. (1962) 52George A. Miller. "Professionals in Bureaucracy: Alienation Among Industrial Scientists and Engineers." American Sociological Review., Vol. 32 (5) (October, 1967) pp. 755-768. -27- To test these hypotheses, Miller devised a measure of alienation which contained five questionnaire items of intrinsic pride in work. For organizational control he used an item by which the type of work relationship between superior and subordinate was established, and also an item by which the degree of freedom to choose types of research projects was measured. For the professional incentives variable he locked at the freedom to publish and attend meetings, the facilities available, the opportunities to improve skills and the extent to which promotion was based on technical competence. His results suggested that alienation from.work is related to organizational control and professional incentives. He also found, however, that the length and type of training is less important than the type of supervision and professional incentives. Perucci studied social distance, bargaining power, and compliance with rules on a psychiatric hospital‘ward.53 An organization, accord- ing to Perucci, is, ". . . a network of social distance patterns built about a set of fixed positions, whose occupants behave in such a way as to maintain or enhance their status in the hierarchy.54 He hypothe- sized that, "the degree to which attendants on a hospital ward were 53Robert Perucci. "Social Distance, Bargaining Power and Com- pliance with Rules on a Hospital Ward." Psychiatry, Vol. 29 (l) (1966) pp. 42-55. 5411316. p. 42. -23- rule-oriented was indirectly related to their access to public and private knowledge about the doctors who were their super-ordinates."55 Using the participant observer technique of investigation Perucci com- pared the day shift of the ward with the evening shift, these shifts differing in the extent of interaction between staff members, He found that the evening attendants, a) complied more closely to the rules and refused to act if the directives were only slightly ambiguous; b) had a more formal relationship with doctors, and c) required more clarification of rules. These findings are supported by those of Segal56 who compared night shift and day shift nurses in a general hos- pital. In this case, the night shift nurses were more custodial and rule-oriented in their attitudes toward patients. The literature reviewed thus far has been characterized by its theoretical and empirical examination of the professional person work- ing in an organizational setting. In most of these studies there is implicit, and at times, explicit recognition that the personality of the professional effects the social structure of the organization. Yet the primary focus has been on such structural factors as authority 55Ibid. p. 43. 56Bernard E. Segal. "Nurses and Patients: Time, Place and Dis- tance." Social Problems, Vol. 9. (3) (1962) pp. 257-264. -29- hierarchy, rule orientation, congruence of goal perceptions, degree of specialization, and evaluation of technical competence. The point of view that seem to prevail is aptly exemplified by Thompson. Recogni- zing that role is a function of the personality he also concludes that, "prolonged enactment of a role reacts upon the personality. People become what they do."57 In order to more adequately examine the rela- tionship of the professional in an organizational setting a theoretical conceptualization is required that accounts for both the personality and the social structure. The gross concept of role has been utilized by many students for this purpose. Rolg, Personality and Social Stggcture in the Ogganizational Setting58 Levinson sums up the literature on role theory in these words, "The concept of role remains one of the most overuworked and under- developed in the social sciences "59 The purpose here is to first sort out the various meanings that have been ascribed to the term, role, 57Thompson. 492..g35. p. 495. 58This title, and the subsequent treatment of the subject is based on Levinson's paper of the same title. 59Daniel J. Levinson. Role, Personality and Social Structure in the Orggnizational Setting. Mimeographed paper p. 1. -30- examine the concept of "role consensus," and present a sample of those studies related to the topic under investigation.. Levinson suggests that different writers, and at times the same writer, have used three definitions of the global term role. It has been used to connote the structurally given demands of a given social position such as norms and responsibilities. In other instances it has been defined as the actions of individual members as seen in relation to prevailing norms. Or it has been defined as the member's orienta- tion or conception of the part he is to play in the organization. And Linton put them all together and defined role to "include the attitudes, values and behavior ascribed by the society to any and all persons oc- cupying a status."60 In this latter sense society shapes the atti- tudes and beliefs of all the occupants of a given role. An adjunct to the concept of role is the concept, role consensus. That is, there is consensus among persons as to what is expected of an 61 incumbent to a particular role. Gross in his exhaustive analysis of the role of the school superintendent, however, found little or no 60Ralph Linton. The Cultural Backggound of Personality; New York: Appleton-Century. (1945). 61Neal Gross, W. S. Mason, and A. W. McEachern. Explorations in Role Analysis. New York: John Wiley. (1958). -31- consensus to exist among such persons as teachers, school board mem- bers and community businessmen regarding what constitutes the role of the superintendent. This lack of role consensus is also the case in the health field. To assume that a role consensus exists for the hospital nurse would imply that ". . . her role-requirements will be understood and agreed upon by the hospital administration, the nursing authorities, the physicians, etc."62 The falaciousness of this assumption is borne out by Burling, Lentz, and Wilson's study of hospitals.63 Using the tech- niques of participant observation and structured interviews over a period of two years, these researchers found, in part, confusion re- garding nursing administration, differences in nurses' needs and expectations about supervision, and a breakdown of the status system, hierarchy and role expectations. The results of other studies of hospitals were similar.64 62Levinson. .92. Cit. p. 6. 63Temple Burling, Edith Lentz and Wilson. The Give and Take in Hospitals. New York: Putnam. (1956). 64Chris Argyris. Diaggosingiguman Relations in Organizatigns: A Case Study of a Hoppitgl. New Haven: Labor and Management Center, Yale University. (1956). and Alfred H. Stanton and Morris S. Schwartz. The Mental Hospital. New York: Basic Books. (1954). -32- An approach to the study of organizations which utilizes the con- cept of role as an integrating of the thoughts and actions of indivi- duals with the influence of social structure is that of Kahn and his colleagues.65 They had as their objectives to: 1) Explore the extent of role conflict and ambiguity in in- dustrial positions; 2) Identify high conflict and ambiguity situations; and 3) Relate conflict and ambiguity to personality factors.66 Viewing organizations as an array of overlapping role sets--in Merton's sense of a role set being, "that complement of role relations which persons have by virtue of occupying a particular social status"67--they compared the responses of persons influential in defining the role of "focal persons" with certain personality traits of the focal persons themselves. They found that high levels of conflict and ambiguity had 65Robert L. Kahn. t. al. Studies in Ogggnizational Stress. New York: Wiley. (1964). 661b1d. 67Robert K. Merton. Social Theogy and Social Structure. New York: The Free Press. (1957) p. 369. -33- as their major elements uncertainty over evaluation, opportunities for advancement, scape of responsibilities and expectations of others re- garding one‘s performance. Also of relevance to this review is the use in this study of an operational definition of conflict which was the extent of disagreement between what was perceived by role senders to be the preferred behavior of focal persons and the perceived actual behavior. The literature reviewed in this section is indicative of an approach to the study of professionals in organizational settings that uses the concept of role as a closer approximation of the integration of personality and social structure that exists in reality. The pre- vious section reviewed literature which emphasized social structure somewhat at the neglect of personality. In the following section the implications of these previous studies for the topic under investiga- tion will be considered. Implications The evidence drawn from previous studies of professionals work- ing in organizational settings strongly suggests that the individual professional‘s desire for autonomy is restricted by the social struc- tural dimensions of organizations, and that this restriction may lead to conflict. There is also evidence that the degree of conflict is related to both the degree of structural restriction (bureaucratization) -34- and to the degree to which the individual adheres to professional norms and values of independence and autonomy. The previous research indicates, further, that the hospital is an organization of the highly bureaucratized type including that area of the hospital which constitutes the work setting of the nurse. Stu- dies of hospitals indicate that there is considerable conflict regard- ing what constitutes the role of the nurse, and while no studies were found that treated the contemporary nurse as a professional in the same sense as a lawyer, engineer or scientist it seems pertinent and legitimate to do so. Given the nature of the nurse's organizational setting and her growing adherence to professional norms the existence of high conflict is a logical hypothesis. The literature on hospitals, specifically, was found to be either anecdotal or of the participant observation type. As Seeman notes with regard to hospital studies, "The bulk of work on the status sys- tem.in medical organizations has been anecdotal in character or has avoided altogether the measurement of medically relevant consequences."68 Also of note is the absence of any methodology utilizing comparison groups across hospitals. These weaknesses of previous research point to the need for studies which seek to measure quantitatively organiza- tional and personal variables and which compare hospital work settings 68Melvin Seeman and J. W. Evans. "Stratification and Hospital Care: Part I." American Sociological Review, Vol. 26. (1961) pp. 193-203. -35- in terms of their bureaucratization or degree of structuring. And as Hell points out it may or may not be correct to assume an organization is bureaucratized. Any legitimate classification of organizations as bureaucratized or nonbureaucratized requires empirical measurement of the variables. Finally, on the basis of this review of the literature it is evident that concepts such as role, role conflict, and organizational conflict have been used in a variety of ways. Smith's definition of the term organizational conflict as being ". . . a situation in which the conditions, practices or goals for the different participants are inherently incompatible,"69 is an example of broad statements that add to the confusion. No studies were found which used a list of all tasks conceivably performed by the actor in a given role to serve as the basis for operationalizing the conflict variables. The review of previous literature, then, points to certain direc- tions to be taken in a study of this kind. They are: 1) view the contemporary nurse as a professional desirous of autonomy and inde- pendenceg 2) examine the work setting of the nurse to determine the extent of its bureaucratization; 3) measure the variables to be 69Clagett G. Smith. "A Comparative Analysis of Some Conditions and Consequences of Intra-Organizational Conflict." Administrativg_ Science Quarterly, Vol. 10 (4). (March, 1966). pp. 504-529. ~36- considered in some quantifiable manner rather than rely on anecdotal evidence; 4) establish empirically whether, and to what extent, a given organization is bureaucratized; and 5) use a comprehensive task list as a means of operationalizing the concept of role conflict. Summagy In Chapter 2 the literature pertinent to the study of the pro- fessionals working in a bureaucratic setting was reviewed. The research studies and theoretical works reviewed included sociological studies of bureaucracy, studies of intellectuals, scientists and other pro- fessionals working in formal organizations, studies in role theory, and studies conducted in both general and psychiatric hospitals. Review of these studies indicated that the professional may be in conflict with his employing organization, that the hospital is highly bureau- cratized, and that there is considerable conflict regarding what con- stitutes the role of the nurses. Chapter 2 was concluded with a dis- cussion of certain directions that might be taken in a study of this kind. The nature of the research design and measurement instruments are considered in Chapter 3. CHAPTER III DESIGN OF THE STUDY In order to examine some of the possible relationships between the structural characteristics of work settings and the degree of intra- personal role conflict and job satisfaction of nurses an experimental coronary care ward of a general hospital was compared with five other wards in four general hospitals. These five comparison wards were similar in that they all provided some form of coronary care. The wards differed in the type of coronary care provided and in the manner by which they were organized. One ward was of the medical-surgical type caring only for those coronary patients who are in the later, less critical stages of their hospitalization. Two wards were exclusi- vely intensive care units caring for coronary and other patients re- quiring constant monitoring and more specialized nursing. A fourth ward was an intensive care unit similar to the other intensive care units with the exception that the coronary patients were patients of physicians who did not wish to admit them to the experimental coronary care unit in the same hospital. The sixth ward cared for both criti- cally ill coronary patients and other medical-surgical patients. This ward was organized such that the intensive care unit was integrated into a medical-surgical ward with the head nurse being responsible for -37- both. -38- The six wards studied constituted the total facilities of the community for the provision of coronary care. The experimental ward had the following identifying character- istics: a) b) e) d) e) f) Presence of a small computer to study the feasibility of computer-based patient record keeping. Doctors and nurses jointly work at programming the com- puter, the nurses as a formally designated part of their job, the doctors voluntarily. Doctors and nurses jointly interact in study and consul- tation on various aspects of computer usage. Nurses who are assigned to the ward express their willing- ness to work in an experimental setting. Hospital administration recognizes the ward as experi- mental in nature. Part of the selection of nurses is based on their willing- ness to examine the tasks of the nurse and the nursing role. -39- It was hypothesized that the presence of these factors in the experi- mental setting and their absence in the traditional settings would differentiate the two types on the basis of role differentiation, stratification and free flow of communications. To establish that the two settings did, in fact, differ structurally the following null and alternate hypotheses were tested: H18 There is no difference between the experimental setting and the traditional settings with regard to the degree of stratification. The degree of stratification is significantly lower in the experimental setting than in the traditional settings. There is no difference between the experimental setting and the traditional settings with regard to the degree of freedom of role definition. The degree of freedom of role definition is significantly higher in the experimental setting than in the traditional settings. There is no difference between the experimental setting and the traditional settings with regard to the degree of free flow of task-related communication. -40- H1: The degree of free flow of communications is signifi- cantly higher in the experimental setting than in the traditional settings. A critical assumption of the theory from which the research hypo- theses were derived is that nurses today tend toward the normative conception of a professional. Specifically, it was assumed that pro- fessionals tend, as individuals, to prefer less structured situations. It was essential to the design, therefore, to establish that nurses working in the experimental setting tend, as a group, to prefer less structured situations. The following hypothesis was derived to test the relationship between work settings and the degree to which nurses possess the need for order as a personality trait: Ho: There is no difference between nurses working in the experimental ward and nurses working in the traditional wards with regard to the need for order. H1: Nurses working in the experimental ward have a lower need for order than nurses working in the traditional wards. The purpose of the above four hypotheses was to establish a "treatment" group and several "controT' or comparison groups. In field studies of this type it is extremely difficult to devise a re- search design which is truly experimental in nature. When studying -41- ongoing human organizations it is often necessary to accept a design which controls for some but not all of the extraneous sources of variance in the dependent variables. For the purpose of establishing that varia- bility between the experimental and traditional groups on stratifica- tion, freedom of role definition and free flow of communications could not be accounted for by differences on certain demographic factors the following hypotheses Were derived: Ho: There is a relationship between the age of nurses and their perception of the degree of stratification, free- dom of role definition, and free flow of task-related communications. H1: There is no relationship between the age of the nurses and their perception of the degree of stratification, freedom of role definition, and free flow of task-related communications. H : There is a relationship between the marital status of the nurses and their perception of the degree of strati- fication, freedom of role definition, and free flow of task-related communications. H1: H1: -42- There is no relationship between the marital status of the nurses and their perception of the degree of strati- fication, freedom of role definition, and free flow of task-related communications. There is a relationship between the spouse's occupation and the nurse's perception of the degree of stratifica- tion, freedom of role definition, and free flow of task- related communications. There is no relationship between the spouse's occupa- tion and the nurse's perception of the degree of strati- fication, freedom of role definition, and free flow of task-related communications. There is a relationship between the shift worked and the nurse's perception of the degree of stratification, freedom of role definition, and free flow of task- related communications. There is no relationship between the shift worked and the nurse's perception of the degree of stratification, freedom of role definition, and free flow of task- related communications. H1: H1: -43- There is a relationship between the number of years of experience of the nurses and their perception of the degree of stratification, freedom of role definition, and free flow of task-related communications. There is no relationship between the number of years of experience of the nurses and their perception of the degree of stratification, freedom of role definition, and free flow of task-related communications. There is a relationship between the type of training received by the nurses and their perception of the degree of stratification, freedom of role definition, and free flow of taskerelated communications. There is no relationship between the type of training received by the nurses and their perception of the degree of stratification, freedom of role definition, and free flow of task-related communications. There is a relationship between the number of professional organizations belonged to by the nurses and their per- ception of the degree of stratification, freedom of role definition, and free flow of task-related communications. -44- H1' There is no relationship between the number of professional organizations belonged to by the nurses and their percep- tion of the degree of stratification, freedom of role definition, and free flow of task-related communications. HO: There is a relationship between the amount or type of additional training received since initial training and the nurse's perception of the degree of stratifica- tion, freedom of role definition, and free flow of task- related communications. H1: There is no relationship between the amount or type of additional training received since initial training and the nurse's perception of the degree of stratification, freedom of role definition, and free flow of task-related communications. These twelve hypotheses served to test whether a quasi-experimental design existed by showing that a) there is a difference between the experimental ward and traditional wards with regard to stratification, free flow of task-related communications, and freedom of role defini- tion; b) nurses working in the experimental setting have a lower need for order than nurses working in the traditional settings; and c) any differences between wards with regard to stratification, free flow of -45- task-related communications, and freedom of role definition cannot be accounted for by any differences among nurses on certain demographic factors. The ngple The sample population consisted of all registered nurses employed either on a part or full time basis in the six hospital wards. It included head nurses and assistant head nurses but did not include supervisory personnel holding positions higher than that of head nurse. In Table 3.1 the summarized information regarding sample size and per- cent response across wards is presented. Table 3.2 summarizes the general characteristics of the sample population. The data for all forty-four responding nurses was used in analyses which did not cate- gorize in terms of wards, but for ward comparison, ward number five was dropped due to the small number of nurses responding. Table 3.1. Number of Nurses in Sample and Percent Returns Across Hospital Wards. Ward Sample 1 2 3 4 5 6 Exp. Total Number of RN's Employed 8 12 6 8 3 7 9 51 Number Returning Questionnaires 5 ll 6 5 2 6 9 44 Percent Who Returned Questionnaires 62% 91% 100% 62% 67% 85% 100% 86.3% m m m mm m CH w m m o q 0 0H mu aa mm wcacaoufi oonHoo oow>uom mooz ouoe are. m N mesa o m N H ocoz powwow Hmuwomom Hocofiuwvom oEOm no“ no 0H no H umoh.w use» m 0H powooaom oocoauomxu mawoamua Hmoowuaoo< moooauoaxm mo ammo» .wuo .woum mo § mo mama UHHmHMuHuao oncam moaned: o>ocmuoo mmumm omiom mmnmm :qu“ omuoz .2 upon omusz mmoum .umw< poem coaummoooo m-omooom mououm Honeys: ow< F oowuamom UHHmHMMHU .05cx , the null hypothesis was not rejected. The results therefore, indicate that nurses of the experimental ward do have a lower need for order when compared with the five tradi- tional wards, but when compared with the mean score of nurses working in the traditional wards, irrespective of which ward, the results are not statistically significant at the .05 level. To determine whether any significant degree of the variance in the nurses' scores on the stratification, free flow of communications, and freedom of role definition variables could be accounted for by -70- variances between the nurses on certain demographic factors the multiple regression technique was used. The null and alternate hypo- theses for each of these possible relationships can be generalized in- to the following format There is a relationship between the age, marital status, spouse‘s occupation, shift worked, number of years of experience, type of training received, number of pro- fessional organizations belonged to, and the amount or type of additional training and the structural vari- ables stratification, free flow of communications, and freedom of role definition. There is no relationship between the age, marital sta- tus, spouse‘s occupation, shift worked, number of years of experience, type of training received, number of professional organizations belonged to, and the amount or type of additional training and the structural vari- ables stratification, free flow of communications, and freedom of role definition. Using multiple regression analysis to determine what variables could account for a significant (.05 OK) proportion of the variance in the stratification variable, it was found that only the free flow of -71- task related communications variable accounted for a significant pro- portion of the variance. The correlation between stratification and free flow of communi- cations was .41. The variance in stratification accounted for by age, marital status, spouse's occupation, shift worked, number of years of experience, type of training received, number of professional organi- zations belonged to, and the amount and type of additional training was nonsignificant at the .05 level. For the free flow of communications variable none of the demogra- phic factors accounted for a significant proportion of the variance at the .05 level. Stratification and role definition did, however, and these results are summarized in Table 4.7. TABLE 4.7. Multiple Correlations, Beta Weights and Significance Levels for the Variable Free Flow of Communications Independent Variable Seta Weights Significance Multiple R W Stratification -.388 .010 .59 Role Definition .427 .005 j For the variable role definition none of the demographic factors accounted for a significant portion of the variance at the .05 level. Free flow of communications did, however, with a correlation of .45. -72- This analysis indicates, therefore, that differences between the scores of nurses on the stratification, free flow of task related communica- tions, and freedom of role definition are not due to differences be- tween the nurses on certain demographic factors. A complete table of all intercorrelations is reproduced as Appendix B . Testing for Differences Between Wards on Job Satisfaction and Intra- Personal Role Conflict Three hypotheses were derived to test for differences between the mean scores of nurses working in the experimental setting and nurses working in traditional settings on job satisfaction and intra-personal role conflict. For each dependent variable the simple analysis of variance model was used to calculate a F ratio. The data were dichoto- mized such that there were two groups, the experimental and the traditional. H I There is no difference between the scores of nurses 'working in the experimental ward and nurses working in the traditional wards on the job satisfaction scale. H1: Nurses working in the experimental ward have higher scores on the job satisfaction scale than nurses working in the traditional‘wards. -73- TABLE 4.8. Differences Between Experimental Ward and Traditional Wards on Job Satisfaction Sources of Variance Sum of Squares df MS F __J: Between 54.608 1 54.608 1.157 Within 1839.441 39 47.165 TOTAL 1894.049 40 An F ratio of 1.157 with l and 39 degrees of freedom has an approximate significance probability value of .29. Since .29.> .05 the null hypothesis was not rejected. HO: There is no difference between the scores of nurses working in the experimental ward and nurses working in the traditional wards on the degree of intra-personal role conflict scale. H1: Nurses working in the experimental ward have lower scores on the intra-personal role conflict scale than nurses working in the traditional wards. -74- TABLE 4.9. Differences Between Experimental Ward and Tradi- tional Wards on Intra-Personal Role Conflict Source of Variance Sum of Squares df MS F W Between 41.170 1 41.170 .034 Within 43875.172 36 1218.754 m _. mat-=3:— TOTAL ' 43916.342 37 An F ratio of 0.034 with 1 and 36 degrees of freedom has an approximate significance probability value of .85. Since .85 .> .05 the null hypothesis was not rejected. The mean scores on job satisfaction for the experimental and traditional groups were 35.56 and 38.34 respectively. On the con- flict variable the experimental group had a mean score of 117.00 and the traditional group had a mean score of 114.55, differences between both sets of scores being nonsignificant at the .05 level. Testing for Differences Between Nurses with High Perceptions and Low Perceptions of the Structure of Their Work Setting on Job Satisfaction and Intra-Personal Role Conflict The purpose of the data analyses presented in this section is to determine if nurses who perceive their work settings to be high on the structural variables differ with regard to job satisfaction and role conflict from nurses who perceive their settings to be low on the -75- structural variables. The analysis ignores the particular wards in which the nurses work as a category classification. H : There is no difference between the job satisfaction scores of nurses who scored high on the stratification scale and the job satisfaction scores of nurses who scored low on the stratification scale. H1! Nurses who scored high on the stratification scale have lower job satisfaction scores than nurses who perceive their work setting to be low in stratification. TABLE 4.10. Differences Between High and Low Stratification Groups on Job Satisfaction Source of Variance Sum of Squares df MS F : Between 74.911 1 74.911 1.593 Within 1458.059 31 47.034 TOTAL 1532.970 32 An F ratio of 1.59 with 1 and 31 degrees of freedom has an approximate significance probability value of .216. Since .216.) .05, the null hypothesis was not rejected. -75- 0: There is no difference between the job satisfaction scores of nurses who scored high on the free flow of task-related communications scale and nurses who scored low on the free flow of task-related communications scale. H1: Nurses who scored high on the free flow of task-related communications scale have higher scores on the job satisfaction scale than nurses who scored low on the free flow of task-related communications scale. TABLE 4.11. Differences Between High and Low Free Flow of Communications Groups on Job Satisfaction Source of Variance Sum of Squares df MS F ---—_-: =3: Between 33.456 1 33.456 0.640 Within 1620.059 31 52.260 TOTAL 1653.515 32 An F ratio of 0.64 with l and 31 degrees of freedom has an approxi- mate significance probability value of .43. Since the null hypothesis was not rejected. .43 > .05, -77- 0: There is no difference between the job satisfaction scores of nurses who scored high on the freedom of role definition scale and nurses who scored low on the free- dom of role definition scale. H I Nurses who scored high on the freedom of role defini- tion scale have higher scores on the job satisfaction scale than nurses who scored low on the freedom of role definition scale. TABLE 4.12. Differences Between High and Low Role Definition Groups on Job Satisfaction Source of Variance Sum of Squares df MS F u— Between 170.110 1 170.111 3.290 Within 1499.438 29 51.705 TOTAL 1669.548 30 The high perception group had a mean score of 35.0 on the job satisfaction scale, and the low perception group had a mean score of 39.69. -73- An F ratio of 3.29 with 1 and 29 degrees of freedom has an approximate significance probability value of .08. Since .08 I) .05, the null hypothesis was not rejected. Ho: There is no difference between the role conflict scores of nurses who scored high on the stratification scale and nurses who scored low on the stratification scale. H1: Nurses who scored high on the stratification scale have lower scores on the role conflict scale than nurses who scored low on the stratification scale. TABLE 4.13. Differences Between High and Low Stratification Groups on Intra-Personal Role Conflict Source of Variance Sum of Squares df MS F A M .- Between 12.033 1 12.633 .009 Within 38204.267 28 1364.438 -__. I ~ 5 J. m TOTAL 38216.300 29 An F ratio of .009 with l and 28 degrees of freedom has an approximate significance probability value of .926. Since .926.) .05, the null hypothesis was not rejected. -79- Ho: There is no difference between the scores on the role conflict scale of nurses who scored high on the free flow of task-related communication scale and nurses who scored low on the free flow of task-related communica- tion scale. H1: Nurses who scored high on the free flow of task-related communication scale have higher scores on the role con- flict scale than nurses who scored low on the free flow of task-related communication scale. TABLE 4.14. Differences Between High and Low Free Flow of Communications Groups on Intra-Personal Role Conflict Source of Variance Sum of Squares df MS F W Between 178.752 1 178.752 0.199 Within 25136.714 28 897.740 =—— T —‘ E TOTAL 25315.467 29 An F ratio of 0.199 with 1 and 28 degrees of freedom has an approximate significance probability value of .659. Since .659 )> .05, the null hypothesis was not rejected. -80- H 1 There is no difference between the scores on the role conflict scale of nurses who scored high on the freedom of role definition scale and nurses who scored low on the freedom of role definition scale. H1: Nurses who scored high on the freedom of role definition scale have higher scores on the intra-personal role con- flict scale than nurses who scored low on the freedom of role definition scale. TABLE 4.15. Differences Between High and Low Role Definition Groups on Intra-Personal Role Conflict Source of Variance Sum of Squares df MS F = m Between 344.286 1 344.286 .359 Within 25839.162 27 957.006 TOTAL 26183.448 28 An F ratio of .359 with 1 and 27 degrees of freedom has an approximate significance probability ratio of .554. Since .554 ). .05, the null hypothesis was not rejected. -81.. Ho: There is no difference between the scores on the need for order scale of nurses who scored high on the strati- fication scale and nurses who scored low on the strati- fication scale. H1: Nurses who scored high on the stratification scale have lower scores on the need for order scale than nurses who scored low on the stratification scale. TABLE 4.16. Differences Between High and Low Stratification Groups on Need for Order Source of Variance Sum of Squares df MS F Between 14.671 1 14.671 0.683 Within 644.204 30 21.473 W T m TOTAL 658.375 31 An F ratio of .683 with 1 and 30 degrees of freedom has an approximate significance probability value of .415. Since .415 ) .05, the null hypothesis was not rejected. Ho: There is no difference between the scores on the need for order scale of nurses who scored high on the free -82- flow of task-related communication scale and nurses who scored low on the free flow of task-related communication scale. TABLE 4.17. Differences Between High and Low Free Flow of Communications Groups on Need for Order Source of Variance Sum of Squares df MS F Between 54.746 1 54.746 2.566 Within 661.496 31 21.339 TOTAL 716.242 32 An F ratio of 2.566 with 1 and 31 degrees of freedom has an approximate significance probability value of .119. Since .119) .05, the null hypothesis was not rejected. Ho: There is no difference between the scores on the need for order scale of nurses who scored high on the free- dom of role definition scale and nurses who scored low on the freedom of role definition scale. 1: Nurses who scored high on the freedom of role defini- tion scale have higher scores on the need for order -83. scale than nurses who scored low on the freedom of role definition scale. TABLE 4.18. Groups on Need for Order Differences Between High and Low Role Definition Source of Variance Sum of Squares df MS F Between 19.307 1 19.307 1.326 Within 422.370 29 14.565 TOTAL 441.677 30 An F ratio of 1.326 with 1 and 29 degrees of freedom has an approximate significance probability value of .259. the null hypothesis was not rejected. Since .259 ) .05, The results of the several simple analyses of variance reported in this section are summarized in Table 4.19. -84- TABLE 4.19. F Values and P Values for Job Satisfaction, Role Conflict and Order Scores Grouped According to High-Low Perceptions of the Structure of Their Work Setting High-Low Categorization Dependent Variables Variable Job Satisfaction Role Conflict Need for Order Stratification 1.593 .009 .683 (.216) (.926) (.415) Free Flow 0.640 .199 2.566 Communication (.43) .659 (.119) Role Definition 3.290 .359 1.326 (.08) (.554) (.259) Discussion Choice of the .05 level of significance was governed primarily by precedent. While several null hypotheses could be rejected using the .05 rejection rule, many others could not. It is believed that one would err on the side of caution if certain subtleties of the data were not considered. Statistically, as the size of the sample decreases, the size of the F ratio required to meet a given significance level increases. That is, while a relationship between group mean scores of a large sample may have an F ratio significant at .05, that same relationship between mean scores of a small sample will not have an F ratio significant at .05. Thus the small number of persons in the -35- total sample of this study and especially the small number in each group, placed restrictions on the data, and it is suggested those hypothesized relationships with significance figures such as .098, and .08 deserve consideration. A similar statistical restriction exists for correlational coef- ficients. The larger the sample size the larger the correlation co— efficient. Since the sample size of this study is small, correlations in the range of .20 to .40, are worthy of more than passing considerar tion, as are the directional relationships of a series of paired variables. The possible implications of certain directional relation- ships are illustrated in Table 4.20. The correlations between job satisfaction and each of the three variables related to the degree of structure a setting has are consistent for each pair of variables. This consistency is interpreted to mean that nurses who perceive their settings as highly structured are less satisfied with their jobs. (Note that a high stratification score corresponds to low free flow of communication and freedom of role definitions scores in describing a setting as being high or low in structure. All three correlations are, therefore, indicating relationships in the same direction.) -86- TABLE 4.20. The Pearson r Values for Stratification, Free Flow of Communications, and Role Definition and Job Satisfaction and Need for Order Dependent Variable Independent Variable Job Satisfaction Need for Order Stratification -.12 +.09 Communications +.15 -.20 Role Definition +.26 -.23 The correlations between the need for order and the three struc- tural variables is also of interest. Although they are not high cor- relations, the fact that two are in the same direction suggest that the higher is the nurse's perception of the degree of free flow of communications and freedom of role definition the lower is her need for order. By increasing the size of the sample it is quite likely the coefficients would also increase. The third major question of the analysis dealt with nurses cate- gorized according to their perceptions of the degree of structure in their ward. It is informative to break down the high and low groupings on the structural variables in terms of the number and percentage of nurses from each ward that fell into the high group and low group. Table 4.21 shows this breakdown. The percentages for the experimental ward are of interest. As shown in Table 4.22, the experimental ward . id H . 0 - 0 a 00 a 0.00 N m.mm . N 0.00 N .m.mm N 0 m 0.00 m 0.0qM N 0.00 m 0.0N H 0.00 m 0.0a N _ m a n 00 q n.0HW H m.mm N 0.00 m 0.00 m n.0H H 0 m - .... . m.- ooH H 0.3.. a 0.3 a 0.3 s 98. N 0.8 m . S N L m . 0.00“ m 0 m 0 0.00 N 0.0a N 0.0N H 0.00 m M m _H HmcoHquuuH N.NN. N m.mm- m m.mm m a.se a m.mm m N.NN N w a N stcwsaumaxm N wash N comm N .voum R _ -vuum .Ml .vmum N .Vmum. onus zoom oH_ puuz macaw mocha nsoun 30H ououu Hm usouw “scum mucovcounom. HuuHemom 30H ame 39H ame 00 $0.52 :oHuHuHmoo oHoM mGOHumoHGDEEou mo son «mum COHumoHMHuuuum HuuoH oHnuHuu> aoquNHhomouuo msouu moHanHu> Hausa uuouum may no masouo 30A no swam ouufi HHuh oss wuss some cw momuaz wo omuuuoouom van nonsaz .HN.¢ mHm uohwz aoozumm macaum0muuoo m0mEHm umwuo How vomz Guesuaoo «00¢ nowuumm Imaumm new defiuaawmma O0om mo aovoohm maoauuu :0GHHEOU mo 3O0m menu coauaufima -uauum 196 6306 s llllll. Rl ll L| H u RH| " “II H Em Til “ T 193 03 31 llllllllllllil