UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Measuring staff outcomes following reorganization: participation in decision making, autonomy, and job… Hughes, Linda Rea 1997

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

Item Metadata


831-ubc_1997-0177.pdf [ 4.66MB ]
JSON: 831-1.0099135.json
JSON-LD: 831-1.0099135-ld.json
RDF/XML (Pretty): 831-1.0099135-rdf.xml
RDF/JSON: 831-1.0099135-rdf.json
Turtle: 831-1.0099135-turtle.txt
N-Triples: 831-1.0099135-rdf-ntriples.txt
Original Record: 831-1.0099135-source.json
Full Text

Full Text

Measuring Staff Outcomes Following Reorganization: Participation in Decision Making, Autonomy, and Job Satisfaction by LINDA REA HUGHES B.S.N., The University of British Columbia, 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NURSING in THE FACULTY OF GRADUATE STUDIES School of Nursing We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March, 1997 © Linda Rea Hughes, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract This longitudinal descriptive study was designed to evaluate staff outcomes at one hospital in Vancouver, British Columbia as a result of implementing an organizational model designed to enhance staff participation in decision making (PDM). The specific outcomes of interest were staff perceptions regarding PDM, autonomy, and job satisfaction. The purposes of the investigation were to explore the impact over time on the three variables, to examine the interrelationships among the variables, and to assess the effects of selected demographic characteristics on staff perceptions regarding the variables. A convenience sample of approximately 150 nursing staff completed a questionnaire at three distinct time periods: one year, two years, and three and one-half years after the initial implementation. Over the three-year period, the means for the three variables did not change significantly. Nurse managers experienced greater autonomy than did staff and were the group most vulnerable to organizational change. The interrelationships among the three variables were positive, but the strength of the relationships was inconsistent. The relationship between PDM and job satisfaction was moderately strong and comparable to that identified by other researchers. Autonomy and job satisfaction were positively, but weakly, correlated as were PDM and autonomy. The effects of demographic characteristics on the three variables are inconsistent and variable. Employment position, years experience and type of shift affected PDM, while employment position and type of shift affected autonomy. Employment position was the only nurse characteristic that significantly affected job satisfaction. The findings of this study do not support the popular notion that a participative organizational model will automatically increase PDM, autonomy, or job satisfaction. Such effects seem to require a change in more than structure. The study hospital did not implement any initiatives other than structural change and did not successfully increase staff perceptions regarding PDM, autonomy, and job satisfaction. The implication of the findings for nursing administration, and social and health policy and further research are discussed. iv Table of Contents Page Abstract ii Table of Contents iv List of Tables vi List of Figures vii Acknowledgments viii Chapter One: Introduction Purposes and Hypotheses 2 Definitions of Terms 3 Assumptions 4 Limitations 4 Significance of the Study 4 Organization of the Thesis 5 Chapter Two: Literature Review Environmental Context 8 Participation in Decision Making (PDM) 10 Autonomy 16 Job Satisfaction 20 Types of PDM Models 22 Summary 24 Chapter Three: Methods Instrument 28 Nurse Characteristics Questionnaire 28 Decentralization and Participation in Decision Making Scale 28 Sample and Setting 30 Data Collection 31 Data Analysis 32 V Chapter Four: Findings and Discussion Description of the Sample 33 Instrument Reliability 37 Means of the Variables 38 Participation in Decision Making 40 Autonomy 43 Job Satisfaction 45 Correlational Analyses 48 PDM and Job Satisfaction 48 Autonomy and Job Satisfaction 49 Autonomy and PDM 51 Demographic Differences 52 Degree of Participation in Decision Making 53 Autonomy 60 Job Satisfaction 61 Summary 62 Chapter Five: Summary, Conclusions, and Implications Summary 66 Conclusions 68 Implications 69 Nursing Adrninistration 69 Social and Health Policy 71 Future Research 72 References 74 Appendices: Appendix A - Questionnaire 81 Appendix B - Information Letter to Staff 92 vi List of Tables Page Table 1. Staff Response Rates 34 2. Education, Area Worked, Shift Worked, and Position 36 3. Reliability Data 38 4. Means of the Variables 39 5. Participation in Decision Making 40 6. Manager and Staff Participation in Decision Making 41 7. Autonomy - Manager and Staff Perceptions 44 8. Job Satisfaction - Manager and Staff Perceptions 46 9. Correlation of PDM and Job Satisfaction 49 10. Correlation of Autonomy and Job Satisfaction 49 11. Correlation of Autonomy and PDM 51 12. Autonomy and Involvement/Importance 52 13. PDM by Employment Position 54 14. Importance/Involvement by Position 55 15. PDM by Years of Experience 56 16. Importance/Involvement by Years of Experience 56 17. PDM by Area Worked 57 18. PDM by Shift Worked 58 19. Importance/Involvement by Shift 59 20. Autonomy by Position 60 vu List of Figures Page Figure 1. Working Profile of Staff. 35 2. Educational Preparation of Respondents 36 3. Employment Position of Respondents 37 4. Comparison of Means for PDM, Autonomy, and Job Satisfaction 39 5. Manager and Staff Participation in Decision Making 42 6. Correlations between PDM and Autonomy with Job Satisfaction 50 Acknowledgments No one person completes a thesis alone. I wish to acknowledge the people without whom the successful completion of this thesis would not have occurred. I would like to thank the members of my thesis committee, Dr. Sonia Acorn and Ms. Lynda Anderson, who offered the support, encouragement, and criticism necessary for me to complete such a project. I would also like to acknowledge Ms. Dianne Doyle who gave me the opportunity that started me on this journey, and to dedicate this thesis to the nursing staff who, through their valuable participation, made the research possible. I would like, finally, to thank my family for their support and encouragement over the past three years. To my sister who has acted as a sounding board, to my parents who have provided encouragement, laughter, and love throughout my life, and to my son, Sean, for his pride, forbearance, and sacrifices. I must finally acknowledge my husband, Bryan, who has acted as my editor and, more personally, taught me to fly. 1 Chapter One Introduction Increasing staff participation in decision making (PDM) is a central component of many organizational models currently being implemented by health care agencies. These models purport to increase nursing autonomy and job satisfaction by decentralizing decisions based on individual and role accountabilities; thereby increasing PDM that directly affects daily work . (Mcdonagh, Rhodes, Sharkey, & Goodroe, 1989; Porter O'Grady, 1994; Westrope, Vaughn, Bott, & Taunton, 1995). This type of organizational structure increases expectations for the professional nurse to participate in clinical, professional, and administrative decisions. One example of such an organizational model that provides both structure and process is shared governance. Although many opinion articles extol the virtues of these structures, few studies examine the impact upon nurses beyond one point in time (McCloskey et al., 1994). There is, more importantly, a paucity of longitudinal studies due to the complexity and cost of such endeavors (McCloskey et al.) Of the six studies in the literature which examined PDM models, (Bland Jones, Stasiowski, Simons, Boyd, & Lucas, 1993; Counte, Barhyte, & Christman, 1987; Ludemann & Brown, 1989; Weisman, Gordon, Cassard, Bergner, & Wong, 1993; Westrope et al., 1995; Zelauskas & Howes, 1992), only three (Bland Jones et al.; Westrope et al.; Zelauskas & Howes) extended beyond a 24-month period of implementing the new organizational structure. The results have been mixed due to differing structures, poor study designs, and often confounding variables, such as other initiatives being implemented concurrently (McCloskey et 2 al.). Effective evaluation must include a refocus on the intended outcomes rather than the structure itself. It can take as long as five years before an organization attains maturity (Peterson & Allen, 1986; Porter O'Grady, 1993) and organizational literature indicates the importance of longitudinal evaluation if one is to truly measure the impact of an organizational change. Therefore, it is imperative that staff outcomes are longitudinally studied (Blegen et al., 1993; Williams & Hazer, 1986). Purposes and Hypotheses Purposes The purposes of this study are: 1. To explore the impact over time of a participative decision making model on three variables: participation in decision making (PDM), autonomy, and job satisfaction 2. To examine the interrelationships among the three variables. 3. To assess the differences among the three variables for individual staff characteristics such as position in the hierarchy, education, tenure, area worked, and type of shifts worked. Hypotheses The following hypotheses will be tested: 1. There will be a significant increase in participation in decision making, autonomy, and job satisfaction for nursing staff over time as a PDM model is implemented. 2. There will be a significant difference between nurse managers and all other staff on all three variables over time. 3 3. There will be a positive relationship between participation in decision making and job satisfaction. 4. There will be a positive relationship between autonomy and job satisfaction. 5. There will be a positive relationship between autonomy and participation in decision making. 6. There will be significant differences among the study variables within the nursing sample for employment position within the hierarchy of the organization, education, years of nursing experience, area worked, and type of shift worked. Definitions of Terms Job satisfaction: an affective reaction or feeling reflecting how satisfied one is with facets of the job, supervisors, coworkers, pay, and current/future career progress and potential (Allen & Heidrich, 1986). Autonomy: the amount of discretion or influence one exercises in making job-related decisions (Allen & Heidrich, 1986). Participation in decision making: the discrepancy between how involved one wants to be in decisions that affect one's daily work and the actual amount of involvement in the decisions (Allen & Heidrich, 1986 adapted from Allutto and Vredenburgh, 1977). Participative decision making models: professional practice models that facilitate decision making at the lowest level possible in an organization, closest to where the work is performed. Structures and processes are in place to ensure staff have decision making authority about issues and resources affecting their work (Porter O'Grady, 1991a; Havens, 1994). Nurse manager: a person who is the first line manager and has 24-hour accountability for one or more nursing units. The manager is an RN who may have members of an interdisciplinary 4 team reporting to him/her. The manager reports directly to the chief operating officer or chief executive of the agency. LPN: a licensed practical nurse. Patient care aide (PCA): a staff member who has completed a 6-month course from a college and provides direct care under the supervision of an RN. Dedicated casuals: casual staff who work predominantly for the agency. More than 50% of their worked hours are at the agency in question. Assumptions It was assumed that scales using staff self-reports of perceptions provided an accurate measurement of staff outcomes, thereby providing valid information for the study. As well, it was assumed that measuring staff perceptions is a valid means of evaluating the effects of a PDM model. Limitations This study was carried out in one medium-sized community hospital in an urban setting. The findings cannot be generalized to the larger population of nurses working in other types of settings. Participation in the study was voluntary. There was no attempt to control for effects of downsizing and regionalization within the current health care climate nor was there a control sample. The study was also limited by the definitions and specific scales used. 5 Significance of the Study The current climate in health care requires effective, adaptive organizations . To be effective, organizations require staff who can make independent decisions that best serve the patient; this means staff must have control over their work and, therefore, must have clearly delineated lines of authority for decision making. Individuals who experience such authority are more likely to be satisfied with their jobs, to be more committed, and to feel they have more autonomy (Allen, Calkin, & Peterson, 1988; Mcdonagh et al., 1989; Weisman, Alexander, & Chase, 1980; 1981). The assumption is that workers whose needs are met in the workplace will exert greater effort on behalf of the organization yielding work group efficiencies in addition to promoting a climate in which nurse-patient interactions are more effective resulting in increased patient satisfaction (Weisman et al., 1993; Weisman & Nathanson, 1985). The purpose, therefore, of implementing PDM models is to increase worker autonomy and job satisfaction, ultimately serving the organizational goals of efficiency and effectiveness. It is crucial, therefore, to examine whether an organizational model such as shared governance, designed to increase staff participation, autonomy, and job satisfaction, yields the expected outcomes. According to McCloskey et al. (1994), organizational innovations must be evaluated with the same rigour as patient care innovations. Findings of this study will determine whether there is an increase in PDM, autonomy, and nurses' job satisfaction and can provide empirical data for the agency in question. The agency will be able to evaluate whether the model is working in the way that was intended when it was first implemented. The results can give direction for the executive within the agency to evaluate 6 the effectiveness of the model for decision making and to further improve and refine the processes to address any identified problems. Organization of the Thesis An overview of the problem; the purposes and hypotheses, the assumptions and limitations, and the significance of the study have been presented in this Chapter. Chapter Two includes the literature review for the three variables of concern for this study: participation in decision making, autonomy, and job satisfaction. In addition, literature pertinent to the organizational frameworks currently affecting participation in decision making are explored. Chapter Three includes the methods used for data collection and analysis. Results of the study are presented in Chapter Four and summary, conclusions, and implications are presented in Chapter Five. 7 Chapter Two Literature Review The complexity of the environment and the need for adaptable, responsive organizations are two reasons cited in the literature for decreasing tiers of management and increasing staff participation in decisions (Donaldson, 1995; Curtis, 1994). The current belief is that a work force that makes decisions closest to the work will survive the turbulent times in health care because they are flexible, efficient, and effective (Peterson & Allen, 1986; Porter O'Grady 1992a; Havens, 1994). Kanter (1977) hypothesizes that the bureaucratic structure is the major obstacle preventing employees from providing high quality, cost effective care; tightly defined reporting structures restrict an individual's capacity to mobilize the required resources (Laschinger & Shamian, 1994). Staff behaviours are, therefore, determined by the organizational structure (Hatcher & Laschinger, 1996) because it provides differential access to the required information, resources, and support necessary to accomplish work goals (Wilson & Laschinger, 1994; Sabiston & Laschinger, 1995). Organizations are adopting structures to overcome these barriers. The new structures increasingly demand staff participation in decisions beyond a clinical focus. Participation, however, must be perceived to be integral and meaningful to staffs work; this type of participation is associated with increased autonomy and job satisfaction (Acorn, Ratner, & Crawford 1997; Allen et al., 1988; Irvine & Evans, 1992; Jones & Ortiz, 1989; Skeleton-Green, 1996; Zelauskas & Howes, 1992). The environmental context fueling decentralization in hospital organizations, the types of decentralized organizational structures, and the purported staff outcomes are discussed in the 8 literature review. The review includes both research and opinion articles in order to draw conclusions as to the current state of knowledge regarding the impact of implementing a P D M model such as shared governance. The literature is reviewed in five sections. In the first section, the environmental context fueling health care restructuring is discussed, the second section examines the current state of knowledge regarding the effects of increasing participative decision making (PDM), and autonomy is discussed in the third section. The fourth section contains a review of the current state of knowledge regarding job satisfaction, predominantly in health care, and characteristics linked to it. The fifth section includes a review of the different types of P D M models and, more specifically, the types of shared governance models. E n v i r o n m e n t a l C o n t e x t A number of societal and economic changes have fueled the move to more streamlined, flattened organizations in both health care and the business world. As regionalization increases, hospitals annually face an average 4 to 6% decrease in their budgets, with continued increases in costs for employee salaries, and operating costs for goods and services. Increased patient acuity, decreasing budgets, increased worker expectations, the doubling of technology, and an increasing awareness of the quality of worklife have acted as stressors on an institution's ability to deliver services. Economic viability, while promoting quality health care, has forced hospitals to restructure and streamline to better meet patient needs (Porter-O'Grady, 1992a). Reform in health care was first sparked in the 1980s across Canada due to shrinking budgets, increasing health care costs, nursing shortages, increased consumer demands, and a 9 growing dissatisfaction in the nursing population. The Canadian Nurses Association (1990), reported a review of 23 studies across Canada and concluded that low involvement in organizational decision making, low autonomy, inflexible schedules, poor collaboration with colleagues, and lack of educational opportunities were major job dissatisfiers for the nursing population. Quality of worklife became a major focus of a number of studies (O'Brien-Pallas & Baumann, 1992) and both government and professional nursing associations urged changes in working conditions (RNABC, 1989a). As changes were studied and then implemented, the economic reality of health care altered; nursing shortages were no longer an issue. Satisfied, autonomous nurses were required for different reasons as health care moved into the 1990s. As the focus in hospitals became more sharply concentrated on the consumer, with increasingly fewer dollars, organizations implemented numerous concurrent initiatives. This has been referred to as "white water turbulence" in health care (Triolo, Allegeier, & Schwartz, 1995). These rapid and constant changes necessitate highly educated, technologically competent employees who play a larger role in the management of both their work and the governance of the organization (Havens, 1994). The strength of a bureaucratic organization lay in its ability to manage routinely in a static environment (Donaldson, 1995). The current rapidly changing environment introduces complexity beyond the capacity of the bureaucracy (Heckscher & Donnelon, 1994). Organizations recognized the need for change in the relationships with their staff. They needed partnerships with staff to effectively deliver quality patient care. Workers, then, had to become stake holders in the organization (Porter-O'Grady, 1992a). Organizational restructuring in the past decade has increasingly attempted to incorporate components that address the above 10 pressures. Organizations have reduced the tiers of management between front-line worker and senior administration; pushing the decision making closer to where the work is performed (Allen et al., 1988; Kramer & Schmalenberg, 1988a, 1988b; Naisbett & Aburdene, 1985; Porter-O'Grady, 1991a)~this is commonly referred to as decentralization. The implication of this type of restructuring is that nurses assume greater responsibility and accountability for their work environment and are required to assume more tasks (Curtis, 1994; Donaldson, 1995; Peterson & Allen, 1987; Havens, 1994; Weisman, et al., 1993). Although decentralization does not guarantee increased worker control, it is associated with effective organizations that adopt more innovations and have fairly complex environments (Weisman, Alexander & Morlock, 1981). Participation in Decision Making (PDM) A number of authors contend that streamlining the organizational structure decreases the number of decisions for any one person; thereby ensuring a more flexible organization (Marriner-Toomey, 1988; Hassen, 1988). Decentralization has been embraced not only by organizations facing economic restraint, but also by professional bodies such as the Registered Nurses Association of B.C. (RNABC, 1989a, 1989b) and the Canadian Nurses Association (CNA, 1990). Ringerman (1990) defines decentralization as the delegation of authority for decision making to the operational level. The result is often an increase in staff participation in decision making (PDM) which is linked to increased autonomy, increased professionalism, reduced absenteeism, reduced turnover, and increased job satisfaction (Alutto & Vredenburgh, 1977; Layton, 1988; Shoemaker & El-Ahraf, 1983; Ringerman, 1990). 11 Hrebeniak (1974) and Trump-Wells (1990) also link PDM with increased organizational effectiveness. Weisman and Nathanson (1985) reason that, although PDM is linked with job satisfaction, job satisfaction should not be viewed as an outcome in itself. Increased organizational effectiveness is an outcome linked with PDM that is further along the trajectory of providing quality patient care. Care givers who are more satisfied interact more effectively with clients. Effective interaction is a known determinant of patient satisfaction and compliance with a prescribed regimen. Wiesman and Nathanson further argue that there is a logical progression from this individual interaction to organizational effect. The climate that provides for effective nurse-patient interactions results in efficiencies produced by group synergy and a client-receptive atmosphere. Hage and Aiken (1967) believe that PDM has two components: (a) how much participation nurses have in decisions about the allocation of resources and (b) the organizational level within the hierarchy at which these decisions are made. They, therefore, empirically measured the amount of decentralization in the structure in order to ascertain the amount of PDM staff experience. Alutto and Belascoe (1972), however, felt that individual characteristics influenced the amount of PDM that one desired. They believed that organizational researchers were too optimistic in assuming that everyone desired the same amount of PDM. They demonstrated that individual desire for PDM was a crucial variable in the perceived value of participation. Alutto and Vredenburgh (1977) further refined the work by developing a scale that measured the discrepancy between individual preference for involvement versus actual involvement. They found that the discrepancy between the two was a better measure of nurses' perceptions about the amount of control over work decisions. 12 Skelton-Green (1996) measured the impact of committee participation on staff perceptions of PDM. Her findings supported those of Alutto and Vredenburgh (1977). Over-saturating nurses with too many committees leads to a perception of increased workload—a known dissatisfier for nurses (Blegen & Mueller, 1987; Hinshaw & Atwood, 1984; Weisman, Alexander & Chase, 1980). Nurses were most satisfied with increased decision making if committees were perceived to be efficient, had the authority to implement decisions, and were perceived as powerful within the institution. Agencies employ various organizational structures to ensure staff participation in decisions affecting daily work, work of the unit and, at times, organizational goals. A few studies have attempted to measure the impact of implementing models of PDM on the nursing staff (Bland Jones et al., 1993; Counte et al., 1987; Ludemann & Brown, 1989; Westrope et al., 1995; Zeulaskas & Howes, 1992; Weisman et al., 1993). Bland Jones et al. (1993) measured the impact of a PDM model by measuring job satisfaction, management style, and job stress at three different points in time; they did not attempt to measure staff perceptions of decision making. They found a general increase in job satisfaction by year two and a slight decrease between years two and three. They do not propose any rationale for the decline and suggest, therefore, that one need not measure beyond year two. Counte et al. (1987) measured staff perception of increased decision making and job satisfaction after units were invited to implement unit PDM models. They concluded there was no difference between the experimental and control units. Counte et al. do not discuss the limitation of the study design. They expected each unit to develop their own structure with little 13 education or organizational support beyond the unit level. In addition, two of the units did not really succeed at implementing a model. The post-implementation measure was taken a mere six months after the units attempted to implement change. It is not surprising that they found little difference in such a short time frame. Porter O'Grady (1993) cautions researchers and administrators that substantive outcomes cannot be expected from organizational and systems redesign for at least five years. Ludemann and Brown (1989) measured nurses' influence (defined as decision making for policies and resources), job satisfaction, and organizational commitment, before and after implementing a PDM unit structure. The authors' pretest measure was based on asking the respondents to recall how they felt before implementing a PDM model. This recall method was performed 18 months after the implementation of the new structure with a sample that represented a 28% response rate. Ludemann and Brown concluded that there were some improvements in nurse perceptions regarding decision making, but job satisfaction was negligible. The difficulty with this study is the recall method of pretesting. There is no rationale cited for this approach nor is there any indication that this is a valid means of measurement. Ludemann and Brown's findings would have been more useful if they had measured staff perceptions at two different points in time post-implementation to ascertain if the values of the three variables increased as the new model became integrated as part of the structure. In fact, they did carry out a second evaluation 6 months after the reported results, but did not report the findings. 14 Zelauskas and Howes (1992) measured staff perceptions following implementation of a unit-based PDM model. They attempted to match a similar unit that did not have a PDM structure as a control and collected information at 6, 12, and 30 months. They did not measure staff PDM, but rather job satisfaction and job characteristics, including autonomy. Again, they implemented primary nursing, peer review, and a revised salary structure. The researchers attributed the resultant increase in job satisfaction and autonomy to their unit model which included all of the initiatives outlined. They did not, however, discuss the demographic differences between their unit and the control unit. Seventy-one per cent of the experimental unit staff had seven more years nursing experience than did those on the control unit, and experience and/or age have been linked to job satisfaction (Blegen, 1993; Irvine, 1992). Weisman et al. (1993) evaluated the professional practice model (PPM) which had evolved over the previous 10 years at Johns-Hopkins. The study was a cross-sectional design that measured job satisfaction, specific work process characteristics, and retention. They carried out a follow-up measurement in 12 months, comparing staff in eight PPM units with those in / eight matched traditional units. The investigators developed a scale that measured decision making about unit management, control of the work (scheduling, shift assignments), coordination of care, and team performance. They concluded that job satisfaction was greater on the PPM unit than on the traditional units and that it was enhanced by team performance and coordination of care in the PPM units, not by the variables of control over work or management of the unit. 15 Westrope et al. (1995) measured staff perceptions for two years following the implementation of a PDM structure. These researchers used Allen et al.'s (1988) theoretical framework to longitudinally measure the impact on staff outcomes. They measured staff perceptions before implementing the new structure, and at two time intervals during implementation. Using Alutto and Vredenburgh's (1977) decisional scale, they indeed found that staffs perceptions regarding decisions affecting their work increased over time. Staff nurses found, however, that they had more control at the unit level than at the organizational level, but the difference was not statistically significant. Hatcher and Laschinger (1996) would, however, emphasize the need for nurses to have decisional control beyond the unit level. Hess (1994) observes that control is incomplete if decision making is limited to the unit level. Job satisfaction increased as well after three years in the study by Westrope et al. Other initiatives were started, such as case management and primary nursing. The authors conclude that some of the increase in PDM and job satisfaction could be attributed to the structure, but might well be related to the other initiatives. Of the six longitudinal studies that examined nurses' perceptions following implementation of a new structure, only three (Bland Jones et al. 1993; Westrope et al, 1995; Zelauskas & Howes, 1992) extended beyond 18 months after a major reorganization such as shared governance, and only the study by Westrope et al. measured the three variables of participation in decision making, autonomy, and job satisfaction. Five of the six found an increase in job satisfaction and two found an increase in PDM. Weisman et al. (1993) found that only certain types of decisions increased job satisfaction. The studies, however, measure different outcomes, use different measures for the same constructs 16 and, therefore, report inconsistent results (Ingersoll, Schultz, Hoffart, & Ryan, 1996). There are also serious limitations in the studies in relation to time frames used to evaluate change or impact. The majority of the studies evaluating staff outcomes were done within one year of implementing new structures. It takes much longer than one year to see the effects of such large scale organizational change (Ingersoll et al.). Causal models are just now emerging that link PDM with job satisfaction (Irvine & Evans, 1992). Hospital structures that enhance staff nurse PDM vary greatly. Some organizational structures stop at the unit level, others limit staff decision making to clinical issues, while others incorporate staff nurses in all decision making levels of the health care agency (Havens, 1994; Hess, 1995). Hess found that, although nurse administrators largely mean clinical practice when they discuss PDM, staff nurses mean control over resources and personnel that affect clinical practice. Hess argues that many PDM structures do include nurses in the decision making but do not give them the authority for implementing innovations. It is this authority that is commonly termed autonomy. Autonomy Autonomy is closely associated in the literature with PDM. Psychology and sociology literature discuss the beneficial aspects of workers having decisional control over their work and life domains (Bush, 1988; Dwyer Schwartz, & Fox, 1992). Decisional control is closely tied to job design and job characteristics research (Dwyer et al.). It is this latter body of research that has led to the conclusion that PDM leads to increased worker autonomy. Dwyer et al., in 17 reviewing the nursing education and organizational literature, found it steeped in the notion that professionals must be autonomous. Many definitions are used by various researchers for the concept of autonomy and, therefore, researchers measure the construct differently. A general definition includes the freedom to exercise professional judgment. Trump-Wells (1990) links this general statement to the issue of control over one's work, and Dwyer et al. (1990) link it to control over specific job tasks. Bush (1988) links the concept of autonomy with the concept of powerlessness by viewing them as either end of a continuum. He defines autonomy as the extent to which nurses believe they are permitted to exercise independent professional judgment. Hinshaw, Smeltzer, and Atwood (1987) define autonomy as a measure of centralization in an organization. Hess (1995) argues that much that is written in the literature confuses the terms centralization/decentralization with participation in decision making and/or autonomy. Blegen (1993), in her meta-analysis of job satisfaction research, concluded that autonomy was referred to by six labels: (1) autonomy, (2) control over work, (3) centralization, (4) participation, (5) powerlessness, and (6) personal control. Studies link nurse autonomy with job satisfaction from three perspectives: (1) as a structural characteristic of the organization, (2) as a higher order need satisfier and (3) as powerlessness, the lack of autonomy that negatively affects job satisfaction (Blegen et al., 1993). Many different scales are used to measure autonomy. Often it is subsumed as a facet of job satisfaction or as part of decision making (Blegen, 1993). Allen et al. (1988) measure autonomy using a distinct scale (labeled as locus of control/authority). Staff report how much influence they have and how much influence is 18 exerted by others in an agency in specific decisional areas: (1) professional, (2) clinical, and (3) resources. Autonomy is correlated with job satisfaction (Acorn et al., 1997; Irvine & Evans, 1992; Slavitt, Stamps, Piedmont, & Haase, 1979; Weisman, Alexander, & Chase, 1980, 1981). Irvine and Evans performed a meta-analysis of nursing turnover research and found that autonomy was moderately to strongly correlated with job satisfaction. They believe it is an important variable that must be incorporated in studies measuring job satisfaction. Theoretical testing has identified autonomy as a determinant of job satisfaction (Acorn et al.; Bush, 1988; Dwyer et al., 1992; Hinshaw, Smeltzer, & Atwood, 1987; Weisman, Alexander, & Chase, 1980, 1981). Autonomy is also correlated with other variables that fall within the categories of individual worker characteristics or work environment characteristics. Alexander, Weisman, and Chase (1982) found that autonomy was correlated with type of shift worked and amount of workload experienced by the staff. Ringerman (1990) reported correlations between autonomy and education and autonomy and type of work setting. Schutzenhoffer and Musser (1994) surveyed a random sample of nurses in four states in the United States to study the effects that nurse characteristics might have on autonomy. They found that nurses with higher education, those higher in the levels of the organization, and nurses in community health exhibited higher levels of autonomy. Kovner, Hendrikson, Knickman, and Finkler (1994) surveyed staff in 37 hospitals. Nurses were asked to rank order the importance of six dimensions of work. Nurses ranked autonomy as the second most important factor after pay. Bush (1988) performed multiple 19 regression equations and found that autonomy (defined as locus of control) or lack thereof accounted for 22% of the variance for job satisfaction in a random sampling of nurses in six hospitals using a forced-choice questionnaire. Alexander, Weisman, and Chase (1982) identified autonomy (defined as control over work) as an important determinant of job satisfaction by surveying nurses working in two large university teaching hospitals. This was measured by asking staff to report the amount of control they perceived they had. The questions were similar to that of Allen and Heidrich (1986). They linked autonomy with control over decision making and found autonomy was the strongest predictor of job satisfaction. Nurses working in areas with smaller patient-nurse ratios reported higher levels of autonomy. Attridge (1996), in a recent study of powerlessness among B.C. nurses, concluded that, despite studies and organizational changes, nothing has changed. Control over work was identified by the nurses as a critical component for their definition of power. They sought power for the altruistic reason of providing quality patient care, and failure to deliver that care caused feelings of guilt even though the failure was attributable to system dysfunction. Autonomy is clearly associated with job satisfaction in the research (Dwyer, et al., 1992; Hinshaw, et al., 1987; Kovner et al., 1994; Schutzenhoffer & Musser, 1994). McCloskey et al. (1994), in their review of nursing management innovations, reported that several studies found that nurse autonomy is an explanatory variable for job satisfaction. A general definition that encompasses most concepts discussed is the degree to which a job provides substantial freedom, independence, discretion, and accountability for the individual to carry out professional duties (Besel & Stock, 1988; Blegen et al., 1993; Weins, 1990 ). The difficulty is the overlap between the concepts of decision making, control over work, and power. 20 Further research is needed to contribute to the data that link autonomy with both job satisfaction and the lesser-studied variables such as worker and environmental characteristics. The concept must be measured as a separate entity distinct from such concepts as PDM, decisional control, and personal control. Job Satisfaction Job satisfaction has long been cited as a variable of interest to both workers and effective organizations (Hinshaw & Atwood 1984; Irvine & Evans, 1992). Irvine and Evans developed a simple model to explain turnover where it is "viewed as the outcome of behavioural intentions, which in turn are influenced by the level of an individual's job satisfaction" (p. 5-6). Glisson and Durick (1988) emphasize that it is an important focus for any human service organization. Job satisfaction is a complex concept that has many different definitions. Locke (cited in Glisson & Durick) describes it as "a pleasurable or positive emotional state resulting from the appraisal of one's job or job experiences" (p. 64). Weisman et al.(1980) define job satisfaction as an affective state or sense of well-being. Hinshaw and Atwood (1984) view it as an attitudinal state, while Mueller and McCloskey (1990) see it as a need fulfillment: "the degree of positive affective orientation toward employment" (p. 113). All of the definitions refer to a general state. Satisfaction, however, is measured in one of two ways— producing a general score or producing separate scores for specific facets of the job (Walker, 1990). Measurement of components frequently cited by nurses in relation to job satisfaction have to do with opportunities for professional growth such as challenging work, authority to carry out that work, recognition, feedback and support from managers, pay, and collegiality. Irvine and Evans (1992) concluded that neither approach is superior to the other; however, job satisfaction in relation to facets of work has received the greatest amount of empirical study (Glisson & Durick). Wanous and Lawler (1972), in a comprehensive review of job satisfaction measurement, found no fewer than nine different definitions. Irvine and Evans (1992) found that the job satisfaction research, although prolific, has yielded a "certain degree of conceptual overlap and models of varying complexity" (p. 26). This makes it difficult to compare and build on findings from the research to date. Blegen and Mueller (1987) found that job satisfaction was predicted by workload, distributive justice, increased autonomy, higher pay, and lower routinization. Hinshaw and Atwood (1984) categorized the variables under three broad concepts: personal, environmental, and job characteristics. Personal characteristics include age, sex, experience, tenure, education, and position in the hierarchy. Environmental characteristics include supervision, leadership, and care delivery model. Job characteristic variables include status, pay, and autonomy. In Irvine and Evans' meta-analysis of the job satisfaction studies, autonomy was consistently, moderately correlated with job satisfaction. In addition, there is consistent support for the role of education and inconsistent support for the effects of position within the hierarchy. Ringerman (1990) found relationships between job satisfaction, age, and tenure. Skelton-Green (1996) reviewed the literature and found an interconnected series of findings: (a) nurse retention is dependent upon job satisfaction, (b) participation in decision making affects nurses' work lives and enhances job satisfaction, (c) participation in decision making is most likely to increase job satisfaction when the decisions are deemed to be 22 meaningful, and (d) participation influences nurses' perception of their influence in the organization. Autonomy and PDM are two variables strongly linked with job satisfaction and of interest to organizations undergoing organizational change. Irvine and Evans (1992) believe further research is necessary to clarify concepts and relationships in the job satisfaction literature. Interestingly, the authors did not directly measure PDM nor discuss the relationship with job satisfaction, due, in part, to the differing definitions for PDM. The authors subsumed it as a concept of autonomy. Further study of the relationship between autonomy and PDM is needed in order to draw conclusions and further develop models that link these two variables with job satisfaction. Replication studies using similar measurement tools are an important step in further research. In order to provide background for the effects of restructuring addressed in this study, a brief review of PDM models follows, with particular emphasis on shared governance. Types of PDM Models According to Allen et al. (1988), many different types of organizational structures have been implemented to increase participation in decision making. These range from employee participation where the executive or manager retains the final decision through to joint decision making by consensus, to nurses controlling much of the decision making. Havens (1994) concurs that PDM models stretch out along a continuum. At one end, the autonomous mode, administrators delegate most of the professional control to the staff. At the other extreme, administrators still retain much of the control for professional decisions. Some of the labels 23 currently found in the literature for PDM include decentralized decision making, participative management, shared governance, product line or program management, and professional practice models (Acorn et al., 1996; Allen et al., 1988). Some models ensure decision making authority is in the hands of staff nurses, but only for clinical decisions or for decisions that have an impact on the unit. For these latter types of PDM models, critics such as Hess (1995) argue that full control for decision making has not been implemented. There are many decisions that are common concerns for staff on more than one unit or that have a professional practice component. There often is not a structure or process that ensures that staff can, as a group, make decisions that affect their larger scope of practice. The model of interest to this study is shared governance. There are many different models for this particular type of structure. Hess points out the difficulty and confusing use of terms found in the literature. Shared governance can be defined as a model that combines formal structures and processes to ensure staff have both the authority and accountability for making decisions about the content and performance of work (McDonagh et al., 1989; Minors, White, & Porter O'Grady, 1996; Porter O'Grady, 1987). Both Hess (1994,1995) and Porter-O'Grady (1987,1992b) argue that shared governance goes beyond a participatory model because the structure is such that a nurse not only can identify areas of concern and suggest changes, but also has the authority to implement the changes. The four areas defined as the purview of staff nurses, those practitioners providing the care, are: practice, education, quality assurance or improvement, and research (Porter-0'Grady, 1992b). Management of resources is a shared function between managers and staff and does not solely rest with the manager (Porter-O'Grady, 1992b). 2 4 Although there is no one "right" organizational model, there are three basic shared governance patterns: congressional, administrative, and councilor (Minors et al., 1996). The congressional model consists of a president and cabinet of officers, while the administrative method is a more traditional structure that uses elected representatives from the work force to provide advice to the decision makers already in management positions (Minors, et al.; Porter-O'Grady, 1992b). The most common model implemented is the councilor model which usually has five councils: (1) practice, (2) education (3) quality improvement (4) management, and (5) coordination (Minors et al.). The practice council handles all clinical issues such as setting practice standards, defining clinical competencies, defining policies and procedures, and evaluating the documentation system. The quality council evaluates the defined standards, develops a peer review or credentialling system, recommends corrective actions, and oversees research activities. The education council is responsible for coordinating orientation, continuing education, and initiating training for new skills. The management council facilitates systems operation and resources, and the coordinating council provides a structure to integrate the activities by developing the goals for the professional body within the institution and ensuring the mission of the organization is integrated with the professional body's activities. There are many articles identifying the benefits of shared governance. Most are anecdotal in nature and some have tried to show cost effectiveness. Few are empirical in nature. Bland Jones et al. (1993), Ludemann and Brown (1989), and Westrope et al. (1995) are the only empirically-based studies that specifically measure the impact on staff outcomes following implementation of shared governance. The results of these studies have been discussed earlier. Westrope et al. measure all three variables of interest to this study, using Allen et al.'s (1988) 25 theoretical framework. Very few studies measure the effect of hospital structures that are designed to facilitate and encourage staff PDM. Longitudinal studies are less common and most do not extend for the length of time necessary to fully evaluate the effects on staff outcomes. Summary A review of the literature and research findings related to PDM, organizational models that enhance PDM, autonomy, and job satisfaction has been presented in this Chapter. As regionalizaton of health care continues, hospitals will continue to decentralize organizational structures. The driving force is economic; however, articles predominantly focus on staff outcomes and patient care. There are few studies that address the impact of such organizational change for a satisfactory period of time. The terms used are confusing and often considered in opinion articles to be the same thing. Hess (1995) identifies the difficulty with the plethora of literature that extols the virtues of decentralized structures with few real outcomes or evaluations to rely on. Although job satisfaction has been well-studied, causal models are just emerging (Acorn et al., 1997; Cavanaugh, 1992; Weisman et al., 1980, 1981). The research related to job satisfaction is varied and complex. As Irvine and Evans (1992) state, one is "left with a long list of perceived predictors ... some of which have a certain degree of overlap" (p. 4). Autonomy is a construct that is moderately related to job satisfaction and has been identified as a predictor of job satisfaction, but has many definitions and is measured quite differently. It is closely identified with PDM as a critical variable, but findings from studies are not as clear as opinion articles. It is measured as part of job characteristics, a facet of job satisfaction, a facet of 26 professionalism, a facet of PDM or one that is measured as a distinct construct. Given the varied definitions and competing measurement scales, it is difficult to draw conclusions about the study findings to date, other than that the term is moderately related to job satisfaction. Participation in decision making has been less well-studied because of the confusing and overlapping definitions for the term. Decentralization, control over decisions, and decisional autonomy are all terms used in the literature to describe staff participation in decision making. The relationship between PDM and autonomy is less well established. There are a few studies that show positive correlations, but again the results are dependent upon the definition and the measurement scales used. Models of staff PDM are much discussed, but, to date, their impact has been poorly studied. Only six studies were found that specifically sought to evaluate the effects of a model on staff outcomes over time. Most of the study time frames were too short in duration to effectively measure impact and only one studied the three specific variables of staff PDM, autonomy, and job satisfaction. Clearly, it is important to study the impact of organizational restructuring upon the staff. Effective staff ensure quality patient care and assist the organization to survive turbulent times. To date, there are few studies and most do not extend for a sufficient time period to draw conclusions. In addition, the relationships between the three variables of interest need further study due to the overlap currently found in the literature. 2 7 Chapter Three Methods The research methods and procedures used in the study are described in this Chapter. The study design was a longitudinal correlational design with three data collection periods using a structured survey method. The first two data collections were at one year intervals and the third, 18 months following the second. Several instruments were examined by staff and managers of the study institution to determine which would best test the variables of interest. Fifteen staff nurses from the agency completed three different questionnaires: (1) The Minnesota Satisfaction Questionnaire (University of Minnesota, 1967), (2) Allen and Heidrich's (1986) questionnaire, and (3) Job Diagnostic Survey (JDS) by Hackman and Oldham (1975). The staff unanimously chose Allen and Heidrich's questionnaire as being the one which best captured areas of interest, specifically, control over decisions, autonomy, and job satisfaction specific to nursing practice. The questionnaire was suggested by Dr. Tim Porter O'Grady and had been chosen by the British Columbia Provincial Government to evaluate the effects of change in two hospitals within the province. Thus, results of this study could be compared to the results from those two hospitals. One of the strengths of the questionnaire is that it is a compilation of previously tested scales such as the Van de Ven and Ferry (1980) scale. Drawbacks of using this questionnaire are that the results of its previous use have not been published and it has not been widely used in studies by researchers other than Allen. 28 Instrument The study instrument consists of two parts: (1) a demographic questionnaire developed by the investigator, and (2) the Decentralization and Participation in Decision Making Scale developed by Allen and Heidrich (1986) that combines a number of organizational research scales measuring the variables of staff participation in decision making, autonomy, and job satisfaction (Appendix A). Two-medium sized community acute care hospitals in the province of British Columbia used this same instrument to evaluate the impact of shared governance on staff outcomes. Demographic Questionnaire The demographic questionnaire was developed by the investigator to identify selected demographic factors that have been identified in the literature as variables that affect job satisfaction. The information included: (1) job position, (2) education, (3) nursing experience, (4) area worked, and (5) usual shifts worked. Decentralization and Participation in Decision Making Scale The instrument is a fixed alternative questionnaire consisting of three scales and a total of 55 items. It uses a Likert scale of 1 to 5 where 1 represents "not very important/strongly disagree" and 5 represents "very important/strongly agree." Internal consistency for each of the scales is between 0.73 and 0.94. The questionnaire has been tested, revised, and used in at least 12 different hospital settings in both the US and Canada with 1,000 nurses (Allen, personal communication, February, 1997). Each of the scales is discussed separately. 29 Participation in Decision Making This two-dimensional 26-item scale measures the difference between the perception of how important it is to be involved versus the perception of the amount of actual involvement in 16 decisional areas. Scores for each item are summed and averaged resulting in a mean score for importance and a mean score for involvement. The perceived amount of participation in decision making results from subtracting each score for importance from each involvement score. These discrepancy scores are summed and averaged to obtain a total discrepancy score. The possible range is from 26 to 130 and the score can be either negative or positive. The scale is adapted from Allutto and Vredenburgh, (1977). Allen and Heidrich (1986) report Cronbach's alpha for the importance scale of 0.94 and for the involvement scale of 0.86 Job Satisfaction This scale consists of 11 items adapted from the Organizational Assessment Index used by Van de Ven and Ferry (1980). The scale measures satisfaction with job, pay, supervision, co-workers, past and present career advancement, and anticipated future career advancement. The degree of satisfaction is measured on a scale of 1 to 5. The total satisfaction score is the average of the scores on 11 items with a possible range of 11 to 55. Allen (personal communication, February, 1997) reports a Cronbach's alpha of 0.87. Autonomy (Adapted from Van de Ven and Ferry. 1980). This scale consists of 18 items. Respondents are asked how much control they have over their workload and who they perceive to influence key decisions i.e. how much influence do 30 individual respondents, other unit members, the manager, outside supervisors, and physicians exert on decisions affecting one's work. The possible range is 18-90. Allen and Heidrich (1986) report a Cronbach's alpha of 0. 60. Sample and Setting A questionnaire was sent to all nursing staff (approximately 250), including nursing managers, both permanent and casual registered nurses, patient care aides, and licensed practical nurses, working in one 250-bed community hospital in the Vancouver Lower Mainland. Each staff member received a questionnaire one year, two years, and three and one-half years after the decision to implement shared governance ( a further reorganization to program management delayed the third data collection). The population of staff remained relatively constant for the first two data collections. At the third collection, the population was smaller due to a reduction in numbers of positions. The nursing administration had identified a need for a change in the structure of the department to address staff nurse concerns identified in the nurses' labour strikes of 1988 and 1992. The members of the management committee searched the literature for available structures and after deliberation chose a shared governance model. Shared governance was formally initiated in September, 1992. A task force of managers and staff was struck to implement the structure and oversee the process. The process for implementation was decided upon, terms of reference were written for nursing councils, existing committees dissolved, and councils struck within 18 months of the initial task force. The first data collection took place one 31 year after the initial decision to implement the structure. The management councils and practice council were in place at that time and education council members were being recruited. Between the first and second data collections the hospital restructured to reflect the principles of shared governance throughout all departments. Interdisciplinary hospital councils were developed and all staff were educated about the principles of shared governance. Between the second and third data collections, the hospital again restructured to reflect an interdisciplinary patient focus with further decentralization for decision making to the units. Nurse manager positions were eliminated, service areas were defined, and a triumvirate management structure was operationalized to merge the vision of interdisciplinary service and shared governance. Data Collection The investigator met initially with all agency nurse managers to explain the purpose and nature of the study and secure their support and participation. The investigator then met annually with all nursing staff on each of the nursing units over a period of six weeks to answer questions, explain the study purposes, and ensure the staff of anonymity. A master list of all nursing staff was generated and a unique number for each staff member was assigned in addition to a number signifying the year that the survey was distributed. This master list was revised annually to ensure that the same employees were included in subsequent data collection times and to include new employees. A letter was sent each year to the potential participants to explain the purpose of the research and assure them that their responses were anonymous (Appendix B). Each individual staff member received a questionnaire, letter, and envelope in his/her mail box. Staff were asked to complete the questionnaire on their own time within an eight-week period and to 32 enclose it in the accompanying sealed envelope that contained a matching number. The sealed envelopes were deposited in a designated mail box on each unit and picked up by a research assistant. Data Analysis Data were coded and keypunched for computer analysis using the computer program Statistical Package for the Social Sciences (SPSS for UNIX, Version 4.0). Descriptive statistics applied to the nurse characteristics data included job category, educational level of the sample, mean years worked in the organization, mean years worked in nursing, area worked, and type of shift worked. Measures of central tendency (mean, median, mode, standard deviation, kurtosis) and dispersion assisted in establishing the homogeneity of the sample. Pearson correlation coefficients were used to examine the relationships between autonomy, job satisfaction, and participation in decision making for all three time periods. The correlation coefficients were compared to standard significance table values to determine if correlation coefficients were statistically significant. ANOVAs were used to test the differences in the measures in each of the three time periods in addition to testing for differences for the nurse characteristics of employment position, education, years experience in nursing, area worked, and shifts worked. Sheffe's test for significance was used to identify where significant differences existed (Lomax, 1992). Managers as a group were compared to all others using t-tests for each of the three data collection periods to ascertain differences and to distinguish differences between periods one and three. 33 Chapter Four Findings and Discussion The findings of this study are presented in six sections. The sample is described in the first section and instrument reliability is discussed in the second section. The third and fourth sections present descriptions of changes in the means and strength of relationships for the three variables of participation decision making (PDM), autonomy, and job satisfaction. In the fifth section, the analysis of selected nurse characteristics is discussed in relation to the three variables, and the results of the study are summarized in the sixth section. Description of the Sample As shown in Table 1, 255 questionnaires were distributed in 1993 and 155 were returned (response rate of 61%). In 1994, 257 questionnaires were distributed and 150 completed questionnaires returned (response rate of 58%); in 1996, 212 questionnaires were distributed and a total of 108 returned (51% response rate). The rate of return when broken down by nursing units was similar for all three years with the exception of the Intensive Care Unit. Staff from the Unit responded with a 90% return rate for the first two data collections, but only 50% of the staff returned questionnaires in 1996. The number of nurse managers responding to the questionnaire was consistently between 85% and 90% of the total population. Sixty respondents remained consistent for all three data collection periods and 76 respondents remained consistent for two of the three data collection periods. 34 Table 1 Staff Response Rates §11111 Number IS umber Respondent* . Retpiinsi- Rate tm 255 155 61% 1994 257 150 58% 1996 212 108 51% Analysis of demographic data for the three data collection periods revealed minimal differences. The distribution of the sample for the variables of education, experience, and tenure remained consistent for the three years as did the representation of professional and non-professional staff. The nurse manager group was the only group that experienced significant membership changes over the three data collection periods with a concomitant rise in educational preparation. In 1993, managers were predominantly diploma-prepared. By 1994, the predominant educational level rose to a baccalaureate, and by 1996 the predominant educational preparation was at the graduate level. Nurse Characteristics As of 1996, the respondents had worked an average of 21 years in nursing, 12 years at the current hospital, and 11 years on the current unit. The majority of the respondents had worked over half of their careers at the same hospital (See Figure 1). 35 Y e a r s 30 r— Y e a r s N u r s i n g Y e a r s at H o s p • 1993 • 19 94 M 19 9 6 Figure 1 Working Profile of Staff In 1993, 18.7% of the respondents were licensed practical nurses (LPNs) and patient care aides (PCAs), 74% were RNs, and 4.7% were nursing managers (Table 2, Figure 3). In 1994, 22.4% of the respondents were LPNs/PCAs, 73.4% were RNs, and 4.1% were nurse managers. In 1996, 25% of the respondents were LPNs/PCAs, 69.8% were RNs, and 4.6% were nurse managers. This sample is representative of the population working in the nursing department at the hospital. As shown in Figure 2, educational preparation of RNs remained relatively constant throughout all three years; with the majority being diploma-prepared. The percentage of RNs holding the baccalaureate or master's degree-13% in 1993, 12% in 1994, and 14% in 1996 is slightly lower than the provincial average of 18% (Statistics Canada, 1994). Table 2 Education. Area. Shifts Worked, and Position 1WJ 1994 Educational Preparation Diploma RN 65% 61% BSN/Ma.slers 13% 14% Aidc/LPN 22% 24% 21% Usual Shifts Days 25% 27% 26% Pcnn. shifts (7.5 hrs.) 14% 7</f< 16% Roiating shirts 34% 37% 36% J2hrs. 27% 29% 22% Area of Work Critical Care/OR/PAR 19% 23% 22% Acute Care 36% 32% 28% Extended Carc/SSATC 45% 44% 50% Position Manager 5% 4tt 5% 74% 73% 70% LPM/PCA 19% 22% 25% N.B. Some of the percentages are < 100 due to missing responses P er c e n t a g e 1 00 90 80 70 60 50 40 30 20 1 0 0 P C A / L P N D i p l o m a R N B S N / M S N • 19 9 3 • 19 9 4 ^ 19 9 6 Figure 2 Educational Preparation of Respondents 37 P er c e n t a g e 100 90 80 70 60 50 40 30 20 1 0 0 P C A / L P N R N M a n a g e r • 1993 • 1994 H 1996 Figure 3 Employment Position of Respondents Instrument Reliability Normative data for the questionnaire scales are the result of surveying of over 1000 nurses in at least 12 agencies in both Canada and the United States (personal communication, Allen, March, 1997). The results of reliability analyses using Cronbach's alpha scores for the sample and normative data for all scales are presented in Table 3. The PDM score is the difference between perceptions of importance and of involvement in decision making. The mean is compiled by subtracting each importance from each involvement response. The result is summed and averaged representing a discrepancy score between the two scales. 38 Table 3 Reliability Data (Cronbach's alpha) 1993 1994 im Normative Data Importance 0.96 0.96 0.96 8.94 Involvement 0.94 t>M 0.93 OJ6 PDM 0.94 tXB5 0.94 &SO Autonomy 0.54 OS2 0.55 0.60 Job Satisfaction 0.64 &7B 0.72 The original PDM scale asked nurses to respond to questions regarding specific questions in relation to 16 decisional areas (Allen & Heidrich, 1986 ). Contrary to Allen's description, factorial analysis for the current study revealed only five decisional areas: (a) professional practice issues, (b) patient care delivery, (c) administrative decisions affecting one's work, (d) control of admissions and other factors affecting one's work day, and (e) environmental issues affecting work. Means of the Variables Means were calculated for the sets of variables for the three data collection periods (Table 4). The means were stable for the three periods, showing no significant change for the sample as a whole. The means for periods one and three were compared using t-tests and showed no significant differences. There were, however, differences between specific groups in each year. The findings of this study are compared to the findings from the comparison hospitals to evaluate whether the effects are similar following implementation of a shared governance model. Of the two hospitals in British Columbia that used the same scales to evaluate the effects of a PDM model, the one located in the Lower Mainland was similar in size to the study hospital and provided similar services. Findings from this hospital are included in Table 4 and Figure 4 and the discussion of the findings. The findings are discussed specific to each of the variables. Table 4 Means of the Variables Variables m$ 1994 im Comparative Comparative Hospital Time 1 Hospital Time 2 *PDM -0.81 : -0.74 -0.81 4.0G -0.94 **Autonomy 3.24 . 3.21 3.20 3.00 3.40 **Job Satisfaction 3.61 • 3.70 3.69 3.60 3.73 *Possible range ± 4 ** Possible range 1-5 V a r i a b I e s P a r t i c i p a t i o n A u t o n o m y J o b S a t i s f a c t i o n ( 1 ) 0 1 2 3 4 5 S c a l e • 19 9 3 Hi 19 9 4 HI 1 19 9 6 Hi C o m p . Hosp. Yr. 1 EH Comp. Hosp. Yr.2 j l l l l i i i l l i M l p JllllllllllllllliilUHUilllHllUllllllllllltiiWll, Figure 4 Comparison of Means for PDM, Autonomy, and Job Satisfaction with Those of Comparative Hospital 40 Participation in Decision Making Participation in decision making is measured by subtracting scores for importance from scores for involvement and can result in a possible range of ± 4. The greater the negative number, the greater the discrepancy between how important nurses feel it is to be involved and how involved they actually are. Stated another way, the greater the negative number, the less participation in decision making. Comparing the results of 1993 and 1994 reveals a slight trend: The discrepancy lessened, that is, the nurses perceived that they were more involved than the previous year and that the amount of involvement had moved closer to their perception of how important it was to be involved (Table 5). This change was not, however, statistically significant. The PDM score for 1996 reverted to the original score of 1993, in part because staff again rated the importance higher than in 1994. There are no statistically significant differences between the means for any of the three years. The change between the first two data collection periods is identical to that for the comparison hospital (Figure 4). The findings do not support hypothesis one. Table 5 Participation in Decision Making L i I, Importance 3.36 3 32 3.36 Involvement 2.54 2.56 2.57 +PDM -0.81 -0.74 -0.81 + Possible range of± 4 There are differences in the two subscales of PDM when the sample is further divided into manager and staff groups (Table 6). The managers' perception of involvement is statistically 41 different from staff perception of involvement for 1993 and 1994, but not for 1996, thus offering partial support for hypothesis two. The lack of difference in 1996 is due to the decrease in managers' sense of involvement in decisions in the latter data collection. This can be clearly visualized by comparing the similar discrepancy between importance and involvement for the 1996 data collection in Figure 5. There is no difference between the two groups in overall PDM for any of the three years. Table 6 Manager and Staff Participation in Decision Making 1993 1994 1996 Manager Importance 3.84 3.60 Staff Importance 3 37 3.3- 3.35 Manager Involvement 3.38* JJfi* 2.87 Staff Involvement 2.50* 2,51* 2.56 Manager PDM -0.46 -0.3 -0.73 Staff PDM -0.86 -&76 -0.82 p<.05 The managers exhibit more changes in individual scores and in the overall PDM scores when comparing across years. Managers' perception of the importance of being involved in the types of decisions has consistently decreased since 1993. Their involvement in decisions has also decreased. The discrepancy between the two values has widened, however, resulting in a group whose PDM has decreased throughout the three years (the negative numeric value is greater). This decrease has occurred at a time when the group experienced a large change in their membership, in their numbers, and in their educational preparation. 42 Means Manage r s Manage r s M a n a g e r s Staff Staf f Staf f Involvement 94 1 Importance 94 • Involvement 93 Importance 93 H Involvement 96 H Importance 96 Figure 5 Manager and Staff Participation in Decision Making Because the sample of managers represents almost the entire population, it is important to note the yearly fluctuations. With a possible range of four, a small change in the numbers represents a noticeable percentage of change in manager perceptions. There was a 12% decrease in manager involvement in decisions from 1993 to 1996. The magnitude of change between 1994 and 1996 represents an 11% decrease in manager PDM. The results indicate that while the hospital structure was reorganized in a way that was supposed to support increased PDM for the managers, in fact the opposite occurred. Staff remained remarkably stable and seemingly impervious to the effects of numerous initiatives that were implemented during the study period while the managers exhibited more 43 variance in scores. Managers may be the level of staff most affected by the changes and may act as buffers for the staff. The scale may not be sensitive enough to capture staff change. However, the manager group represents almost the entire population, and a change in score more likely reflects a real change in perception. Westrope et al. (1995) found similar results using Allutto and Vredenburgh's (1977) discrepancy scale to evaluate staff perceptions. Two years after implementation, nurses' PDM had increased, but not significantly. In addition, the increase was directly related to decisions affecting unit practice such as clinical protocols rather than administrative decisions. Westrope et al. also discussed the confusion and concern that staff expressed regarding the amount of input they should have in decisions such as bed closures and restructuring. This mirrors some of the difficulties experienced in the study hospital during restructuring. Further analysis of the questions specifically pertaining to patient care delivery and control of factors affecting one's work would be of interest to confirm PDM results similar to those of Westrope et al. Autonomy The means for autonomy did not change over the three years, consistently remaining at 3.2 for the sample as a whole (See Table 4, Figure 4). When sorted by position separating managers from all others, there is marked difference (See Table 7). The managers report more autonomy one year after implementing shared governance, but drop below previous means in 1996 directly following a change to program management. This figure reflects some of the perceptions after a reorganization that had resulted in vacant support positions and a reduction in management positions. Despite the yearly differences in figures, the staff consistently perceive significantly less 44 autonomy in their jobs than do managers for all three years. These differences are significant for all three years offering partial support for hypothesis two. Staff autonomy has not changed over the course of this study. One could conclude that that either the staff are insulated from the effects of change or the change has not filtered to their level in any meaningful way. If, according to much of the literature regarding organizational change, it takes five years to achieve results, the time span included in this study may not be appropriate for evaluating the effects. The results, however, do show an increase and then a drop in the means for autonomy for nurse managers. The drop between 1994 and 1996 represents about an 8% change in autonomy for the managers; however, the change may reflect difficulty related to the latest reorganization rather than an effect from a PDM model. Table 7 Autonomy - Manager and Staff Perceptions mS 1994 IM Manager 3.66* 3.71* 3.40* Staff 3.24* 3.19* 3.19* *p<M There are few studies that measure autonomy longitudinally following implementation of a PDM model. Zelauskas and Howes (1992), in comparing an experimental unit to a control unit within one hospital, found an increase in staff autonomy 30 months after implementing a professional practice model. Ludemann and Brown (1989) also found a significant increase 18 months after implementing shared governance in a hospital, but the results are based on a small representative sample (28%), and are reliant upon asking staff to recall how they felt before implementation. The results from the comparison hospital are more marked (Table 4, Figure 4). 45 The mean was lower (3.0) initially than that for the study hospital (3.4) and changed more drastically at collection period two, but again the change was not significant. The lack of significant results from the two hospitals may point to a scale that is not sensitive enough to measure discrete differences or that, contrary to what much of the literature claims, the changes may not have a meaningful effect on staff autonomy. Informal anecdotes and discussions with staff could give clues to a third interpretation. Staff from the study hospital have experienced increased workload for economic reasons and do not feel the PDM model has survived at the unit level. This means that there has been no appreciable changes from the staff's point of view. The data from this study do not, therefore, support hypothesis one that there will be an increase over time in autonomy as a PDM model is implemented. Job Satisfaction Job satisfaction was remarkably stable throughout the three years of data collection. It remained constant at a mean of 3.6 and 3.7 for the entire sample (see Table 4). The findings, therefore, do not support hypothesis one. When the sample is sorted into manager and staff, staff perceptions have remained stable while those of the managers have shown fluctuation (Table 8). There is no significant difference between the perceptions of managers and staff, thus hypothesis two is not supported for the variable job satisfaction. Staff have not experienced any change in job satisfaction despite efforts at unit level councils and department councils for practice, education, and research. In the past 18 months, after further restructuring, the staff have become despondent and negative, voicing concerns about safety to practice, workload, and lack of autonomy for decisions. One would expect that the mean for job satisfaction might reflect this 46 climate, but the 1996 results are exactly the same as those for 1994. The results from the comparison hospital (Figure 4) are similar to those from the study hospital with means of 3.6 and 3.73 for data collection periods one and two representing 6 months and 18 months following implementation of a PDM model. Table 8 Job Satisfaction - Manager and Staff Perceptions ~" " m$ i9M tm Manager 3.88 4.00 3.63 Staff 3.59 $.m 3.69 p > .OS The manager group exhibits significant changes between data collections. Their job satisfaction rose slightly in 1994, one year after implementing shared governance. This level dropped by 10% between 1994 and 1996 after the hospital reorganized to a program management model that was coupled with decreases in managerial and support positions. There were also changes in leadership and reporting lines for some of the individuals. In informal discussion, the managers concur that they currently feel they are making more day-to-day decisions putting out "brush fires," but that meaningful decisions affecting the vision and future of the organization are lacking. Their reflections could be interpreted to support Skelton-Green's (1996) findings that the decisions one is involved in must be perceived to be meaningful to one's work and influential within the organization. The results of the current study are at odds with those in the literature. Bland Jones et al. (1993) reported an increase in staff job satisfaction after two years of restructuring with a PDM model, but a decrease in year three with no explanation. Westrope et al. (1995) found an increase 47 in job satisfaction, but admitted to a number of concurrent initiatives such as case management and primary nursing which could have affected staff job satisfaction as much as a PDM organizational model. Ludemann and Brown (1989) found no significant change in job satisfaction when comparing experimental to control units. As previously discussed, however, they relied on staff recall to compare job satisfaction means. Bland Jones et al. (1993) found an increase in job satisfaction two years after implementing a PDM model. Zelauskas and Howes (1992) found an increase in job satisfaction 30 months after implementing a PDM model on one experimental unit. The satisfaction was significant after a peer review system was implemented. They admit that job satisfaction could well be specific to a number of initiatives rather than the organizational structure per se, but the initiatives are discussed in the literature as part of a PDM model. The PDM model in the study hospital was heavily reliant upon organizational structure and committee work rather than such initiatives as self-scheduling, case management, and clinical ladders. Educational and clinical support positions were not filled and this may reflect the limitations of implementing change without the money to support the necessary education, support, and leadership for meaningful change from the staff perspective. Skelton-Green (1996) discusses the importance of participation in decisions that have meaning to nurses' work and the need for both the committees and the nurse leaders to be perceived as influential within the organization to have an effect on job satisfaction. The changes implemented in the study hospital may have been insufficient to effect change in staff job satisfaction. 48 Based on the data analysis for the variables of PDM, autonomy, and job satisfaction, hypothesis one is not supported in the current study and hypothesis two is only partially supported. There are significant differences between the managers and staff for autonomy, but not for PDM or job satisfaction. There are significant differences between staff and managers for two of the three years regarding actual involvement in decision making. Weisman et al. (1993) found that staff job satisfaction did increase after implementing a PDM model, but the satisfaction was with better team relations and coordination of care and was not related to control over work or management of the unit. Correlational Analyses Pearson correlation coefficients were used to test the relationships among the variables over the three year period. According to Kidder (1981), in social sciences research, r-values between 0.15 and 0.30 are considered weak; between 0.30 and 0.50 moderate; and between 0.50 and 0.70 strong. Values of 0.71 and higher are considered very strong. Although causality cannot be implied by the correlation (Hair, Anderson, Tatham & Black, 1992), the strength or degree of association shows the direction or effect variables exert upon one another. Participation in Decision Making and Job Satisfaction There is a positive but weak relationship between PDM and job satisfaction in the first two data collections and a positive and moderately strong relationship in the third data collection (Table 9). Thus, the relationship between the variables appears to strengthen progressively over the study period which would signal a general recognition of the importance of participation to one's work. 49 The results support hypothesis three. The relationship is stronger for the sample data in comparison to Allen's normative data, and the first two data collections are similar to the value for the comparative hospital (Figure 6). Ringerman (1990) found a moderately strong relationship (r=0.35) between participation in decision making and job satisfaction for a set of nurse managers. The results of this study are consistent with that finding and confirm the strength of the relationship between the two values. The causal model developed by Ringerman included decentralization as a determinant of job satisfaction. A component of the concept of decentralization used by Ringerman was participation. Skelton-Green (1996) found that PDM enhances job satisfaction when nurses perceive it as meaningful. Table 9 Correlation of PDM and Job Satisfaction Nursing 93 0.20* Nursing 94 0.29* Nursing 96 0.37* Comparative Hospital 0.29* Normative Data 0.05* *p<„05 Autonomy and Job Satisfaction The strength of the relationship between autonomy and job satisfaction remained positive over the three data collection periods supporting hypothesis four (Table 10). However, the strength of the relationship was not consistent. Table 10 Correlation of Autonomy and Job Satisfaction Nursing 93 0.30* Nursing 94 0.06* Nursing 96 0.20* Comparative Hospital 0.28* Normative Data 0.37* *p < .05 50 Autonomy is moderately correlated with satisfaction in 1993, and weakly correlated in 1994 and 1996. The strength of the relationship is weaker than Allen's strongly moderate correlation cited for the normative data, but it is similar to the correlation for the comparative hospital (Figure 6). Figure 6 shows the relationships between PDM and job satisfaction and autonomy and job satisfaction for all three data collection periods in comparison to Allen's normative data and that obtained for the comparative hospital in the Lower Mainland. C o r r e l a t i o n V a l u e s 0 .6 0 . 5 -0 . 4 -0 . 3 -0 .2 -0 . 1 0 P a r t i c i p a t i o n A u t o n o m y A n t e c e d e n t V a r i a b l e s • 93 • 94 | H 96 H Comp. Hosp. HI Normative Data Figure 6 Correlations between PDM and Autonomy with Job Satisfaction The correlations are not as strong as those found by both Irvine and Evans (1995) and Blegen (1993) in their respective meta-analyses. Both calculated moderately strong mean correlations of 0.46 between autonomy and job satisfaction. Review of published research confirms that autonomy is a strong determinant of job satisfaction (Acorn et al., 1997; 51 Schutzenhoffer & Musser, 1984; Weisman et al. 1980). The weaker relationships found in this study could be due to the influence of other variables, not included in the study, that affect job satisfaction. As well, the scale may not be as sensitive as scales used in other studies. Autonomy and PDM The relationship between autonomy and PDM is positive, although weak and not statistically significant, thus supporting hypothesis five (Table 11). There are no correlation values available for the normative or comparative data because Allen conceptualizes job satisfaction as an outcome variable dependent upon the other two and, therefore, does not report the relationship between them. The literature suggests a strong link between the two, but many studies do not differentiate the two as distinct variables. Although weak, the relationship may still be important. Bivariate analysis may obscure the importance of the relationship due to the complex nature of the concepts. In addition, homogeneous scores may cause a weaker correlation for the sample than is found in the population (Burns & Grove, 1993). The reliability for this scale is lower (0.55) than that for the other scales (Table 3) and may be contributing to the weaker correlations. Because the correlation coefficient is not statistically significant, it is impossible to assume that the results reflect the population. This is contrary to findings from other research to date and may be due, in part, to the poorer rehability of the autonomy scale. Table 11 Correlation of Autonomy and PDM Nursing 93 0.10 Nursing 94 0.15 Nursing 96 0.11 *p< M 52 The means for both autonomy and PDM remained relatively consistent for all three data collections points. Although the correlation with PDM is not significant, there is a relationship with the subscales (Table 12). The relationship between autonomy and the involvement subscale is moderately strong but there is little or no relationship with the importance subscale. This is contrary to Allen and Heidrich's (1986) findings that the value nurses place on involvement, rather than a finite amount of involvement, is important to their sense of autonomy. Table 12 Autonomy and Involvement/Importance Ringerman's (1990) causal model indicated that decentralization predicted an autonomous climate for managers. One component of the concept used by Ringerman included participation. Very few published studies examine the relationship and effects between the concepts of PDM and autonomy. Further research is required to specifically examine the relationships between these two variables as distinct concepts. One way analysis of variance was carried out to determine if individual characteristics such as employment position within the hierarchy, educational preparation, years of experience, area worked within the hospital, or type of shifts worked significantly affect perceptions of the three variables under study. As suggested in the literature, one would expect that such nurse characteristics would affect staff perceptions; however, research findings to date are inconclusive Nursing 93 0.39* Nursing 94 0.32* Nursing 9$ 0.53* *p<X)5 .24* .15 .20 Demographic Differences 53 and reveal variable support for individual differences. The results of this study reflect the findings in the literature. Some nurse characteristics significantly affect the variables but are not consistent for the three data collection periods. The results are discussed under separate headings. Degree of Participation in Decision Making In 1993, there were significant differences in nurses' PDM depending upon their position in the hierarchy, years in nursing, area worked, and type of shift worked. In 1994, there were significant differences in nurses' PDM dependent upon their employment position, years of experience in nursing, and area worked. In 1996, there were no statistically significant differences in PDM for any nurse characteristics. In addition to the PDM scale, there were significant differences for each of the two sub scales (importance and involvement) comprising the PDM. The specific results are discussed below under each nurse characteristic. Employment Position Nurse managers reported the most PDM, indicated by the lowest negative number, for 1994 and 1996 (see Table 13). The differences are only significant for 1994 where they experience more PDM than RNs and LPNs. Although the PDM increased for managers between the first two years, a paired t-test for those managers who remained constant for the first two data collections showed a general decrease in PDM. It was not possible to run a test for the third year as the manager group experienced a complete change between 1993 and 1996. RNs experienced the least PDM for two of the three years (1993 and 1994). The difference is significant only in 1993, where the RNs perceived less PDM than PCAs/LPNs. RNs' PDM increased over the three 54 years, but the change was not statistically significant. Surprisingly, PC As are the group of staff who experienced a great amount of PDM--the most in 1993, and second to the managers in 1994 and 1996. Their PDM diminished over the study period, signified by an enlarging negative number, which could reflect an increased understanding of the need for more involvement in decisions. The differences between groups were not significant in 1996. The managers were significandy different from the RNs for only one of the three years. Table 13 PDM by Employment Position | LZZZZ." am tm ~ Meager PPM -0.50 -031* -0.73 RNPD&I -1.00* -0.87* -0.85 iJWDM -0.38* *&63* -0.92 PCAPDM -0.24* ~&37 -0.81 *~p<M There were significant differences by position when examining the subscales of PDM. Both managers and RNs rated importance of participation more highly than LPN/PCAs for all three years. In 1993, only the managers experienced significantly more involvement than the LPNs/PCAs. In the latter two data collections, both managers and RNs experienced significantly more involvement than LPN/PCAs. There was no significant difference for either scale between RNs and managers. RN involvement increased consistently over the study period, and the managers involvement decreased over the same time period, but the changes were not statistically significant across time periods. 55 Table 14 Importance/Involvement by Position tm im tm Managers Involvement KN Involvement LPH Involvement PCA Involvement * p < .05 ftN Importance LPN Importance PCA Importance Managers Importance 3.66* 3.60* 2.66* 2.48* 3.15* 2.60 2.26* 2.17* 2.60* 172* 2.75* 1.94* 2M* 3.6 3.65 2.88 2.24 2.87* 2.80* 1.88* 1.79* Education There were no significant differences by education except in 1993 data collection. Diploma RNs perceived significantly less PDM (-0.98) than LPN/PCAs (-0.30), but baccalaureate-prepared RNs did not. This difference was not consistent for the other two years. There were no significant differences by education for the subscales except in 1996 where LPNs experienced significantly less involvement than all other staff. There were no significant differences between diploma-prepared and baccalaureate-prepared RNs. These findings confirm those of Skelton-Green (1996), who found no relationship between PDM and educational preparation of RNs. Years in Nursing There were significant differences in PDM affected by years of experience in nursing for 1993 and 1994, but not for 1996 (Table 15). In 1993, nurses with between 11 and 20 years nursing experience perceived more PDM than the other two groups. In 1994, the same set of nurses was only significantly different than nurses who had worked for more than 20 years. In 1996, there were no significant differences for PDM. These findings are contrary to those of Skelton-Green (1996) that years of experience did not affect PDM. Table 15 PDM by Years of Experience IW3 195M X9M yrs, PDM -1.00* -0.84 -0.95 11-2 ftyi*.PDM -0.50* -0.52* -0.70 31+ year* PDM -0.96* *Q.9i* -0.85 There were differences in the PDM subscales as well. Nurses who had 10 years or less experience rated importance and involvement higher each year, although the change is not statistically significant (Table 16). Nurses with between 11 and 20 years reported decreasing levels of importance and involvement throughout the three-year study period; however, the change is not statistically significant. Table 16 Importance/Involvement by Years of Experience tm tm mm 0*1(1 yrs Importance 3.39 3.50 3.68* 11-20 yns. Importance 3.17 3.13 3.05* 2t+-.yrs. Importance 3.48 3.48 3.51* 0*l&£r& Involvement 2.36 2.65 2.89 11*20 yr& Involyeniteitt 2.66 2.58 ! 2.35 21+yrs. Involvement 2.53 2.57 2.67 * p<0.05 57 Area Worked In 1993, staff working in ECU perceived more PDM than critical care and medical-surgical nurses in their PDM score (Table 17). The results were similar for 1994 with ECU staff again experiencing significantly more PDM than both critical care staff and the medical-surgical nurses in ACU. In 1996, there are no statistically significant differences between groups. The area that nurses work in significantly affects both subscales in 1993, but only the importance scale for 1994 and 1996. ECU staff increasingly rated importance higher for each successive data collection period and their involvement score increased as well, but the changes across years are not statistically significant. Critical care staff decreased the rating of importance for each successive year and reported decreasing involvement as well. Medical-surgical nurses were the only group who reported any increases in actual involvement in decisions over the three-year period, but again the changes were not significant. Table 17 PDM by Area Worked 1093 1994 CaORfSDC Importance 3.71* 3.42* 3.46 ACU Importance 3.67* 3.67* 3.79* JECU Importance 2.91* 3.02* 3.11* €C/0R/SI>C Involvement 2.72* 2.37* 2.41 ACUMvsivemeiit 2.65* 2.72* 2.88 iCUJ&valwmeiit 2.35* 2.54* 2.49 cammmmm -0.99* -1.05* -1.08 ACUPPM -1.03* -0.96* -0.90 JiCUPDM -0.56* -0.43* -0.66 58 Shift Worked Staff working a day/night rotation consistently experienced the most PDM compared to staff working any other shift throughout the entire study period (Table 18). The differences were, however, only significant in 1993. Staff working permanent shift experienced the least PDM. Again, this is only statistically significant in 1993. Staff working permanent shift were significantly different from all others except staff working a day/evening rotation. In 1994 and 1996, there were no significant differences by shift for PDM. Table 18 PDM bv Shift Worked tm tm vm Permanent Pa\s PDM -0.81* -0,71 -0.83 Permanent Shift PDM -1.28* -0.99 -0.98 Extended Hrs. PDM -0.83* -0.97 -0.96 D7E PDM -0.75 -0.50 -0.80 D /NPDM -0.39* -0.60 -0.45 There were also significant differences by shift on both subscales (Table 19). For all three years, staff working extended hours rated the importance of participation significantly higher than did staff working rotating shifts. In addition, in 1994, they rated importance higher than did staff working permanent days. Staff working extended hours also perceived significantiy more involvement in decisions than staff working a day/evening rotation at the time of the 1993 data collection. This did not hold true for 1994 and 1996 data collections. In 1996, staff working permanent shift experienced significantly less involvement than all other staff except those working a day/night rotation. 59 Table 19 Importance/Involvement by Shift vm im i i i i l l Permanent Days Importance 3.38 3.11* 3.60 Permanent Shift Importance 3.63* 3.59 2.87* Extended Hrs. Importance 3.62* 3.70* 3.75* D/E Importance 2.94* 3.01* 3.34 3.00* 3.31 2.79* Permanent Bayslnyolyement 2.52 2.40 2.80* Permanent Shift Involvement 2.38 2.61 1.88* Extended Hrs Involvement 2.79* 2.73 2.77* D/E Involvement 2.17* 2.45 2.65* D/M Involvement 2.58 2.71 2.37 Summary of PDM Results Nursing staff with jobs having more hierarchical power perceived more participation in decisions. This finding is similar to those of other studies (Kanter & Brown, 1982; Wilson & Laschinger, 1994; Sabiston & Laschinger, 1995). The managers experienced significantly more PDM than some staff groups for 1993 and 1994. The RNs were the least satisfied with the amount of PDM they experienced for the first two years of the study. In the last year of the study, all groups seemed to be equally discontent with the amount of PDM. Education was a significant variable for only one of the three years and tenure significandy affected PDM for two of the three years. Nurses who had worked 11 to 20 years experienced the most PDM. Type of unit affected PDM for two of the three years. ECU staff experienced the most PDM, due in part, to lower expectations of involvement, while ACU staff identified increased involvement over the three years. Type of shifts worked was significant to PDM for all three years. Staff working rotating shifts and permanent shifts experienced low PDM and low involvement for two of the three years. This is similiar to the findings by Hastings and Waltz (1995), where permanent shift staff reported difficulty staying in contact with organizational 60 decisions and activities. Staff working extended hours rated the importance of participation higher than did other staff for all three years. With the exception of shift worked, the effects of demographic differences have been variable and inconsistent throughout the three-year time period. The results are consistent with those found in the literature—there are no definitive characteristics that affect the three variables in question. The findings of this study partially support hypothesis six. Autonomy Employment position affected autonomy for the first two data collections (Table 20). Managers reported significantly higher autonomy than any other staff. Although the managers reported higher autonomy than the staff in 1996, the difference was not statistically significant and it was the lowest level of the three years. The results confirm those of Schutzenhoff and Musser (1994) that position in the hierarchy positively affected autonomy. LPNs reported a significantly lower sense of autonomy than staff nurses and managers in 1993 and significantly lower than all other categories of staff in 1994. Surprisingly, the patient care aides (working in ECU) expressed significantly more autonomy than did the LPNs (predominantly working in ACU) in the 1994 data. Table 20 Autonomy by Position Employment Position \m 1994 tm Patient Care Aide 3.24* 349* 3.18 Licensed Practical Nurse 3.04* 293* 3.09 Staff Nurse 3.26* 3,22* 3.21 Manager 3.57 3,71* 3.40 In 1993, there were also differences by education that did not continue into the next two data collection periods. RNs with baccalaureate preparation reported a significantly higher sense of autonomy (3.49) than nonprofessional staff (3.0). There was no difference between diploma-educated RNs (3.26) and baccalaureate RNs or between diploma nurses and non-professional staff. This is contrary to the findings of Schutzenhoffer and Musser (1994) that baccalaureate-prepared nurses experienced more autonomy than did diploma-prepared nurses. Ringerman's (1990) study of managers found that diploma-prepared RNs experienced more autonomy than baccalaureate-prepared nurses. The type of shift worked affected autonomy only for 1993 data. Staff working day/evening rotations reported a lower sense of autonomy (3.14) than staff working any other rotation. It is worthy of note that this set of staff also reported the lowest importance and involvement scores for the 1993 data. Staff working extended hours reported the highest sense of autonomy (3.33). This same set of staff reported the highest importance and involvement scores for 1993. There were no significant differences by shift for the 1994 or 1996 data. The findings of this study are consistent with those of Weisman et al. (1980) who found that rotating shifts did affect autonomy. No nurse characteristics were significant for autonomy for the 1996 data. Job Satisfaction There are significant differences by position in 1993, no significant differences in 1994, and differences by education in 1996. In 1993, the managers reported the highest level of job satisfaction (3.84). This was significantly different from both RNs and LPNs/Aides (3.67 and 3.51). In 1996, the only significant difference in job satisfaction was related to level of education. 62 The staff with the lowest level of education reported the highest job satisfaction. The PCAs reported the highest job satisfaction (4.25) while LPNs reported the lowest (3.39). Baccalaureate-educated RNs reported slightly lower levels (3.55) of job satisfaction than did diploma-educated RNs (3.74), but the difference is not statistically significant. Irvine and Evans' (1992) meta-analysis revealed that higher education was related to increased job satisfaction and that position within the hierarchy affected job satisfaction. Years of experience was weakly correlated with job satisfaction. Blegen's meta-analysis (1993 ) found that years of experience increased job satisfaction and higher education was associated with decreased job satisfaction. Blegen and Mueller (1987) found that shift, age, and tenure were related to job satisfaction. Hinshaw and Atwood (1984), in their review of variables, found that age, education, position, type of unit, and tenure affected job satisfaction. Ringerman's (1990) study found type of unit significantly affected job satisfaction. Summary The means of the variables of PDM, autonomy, and job satisfaction remained relatively constant over the three-year period of this study. Despite numerous changes in the organizational structure, there was no observed impact on perceptions regarding these variables for the total sample. The data did not support hypothesis one that there would be change in the variables over time. When the sample is divided into manager and staff components, the staff means remained stable. The managers, however, exhibited yearly fluctuations of about 10% for PDM and a 10% decrease between the latter two data collection periods for autonomy and job satisfaction. It 63 appears that managers are the most affected by the changes in the hospital environment. They either provide a buffer for the staff or real change has not filtered down to the staff level despite the profound change that both the institution and health care has experienced in the last three to five years. There are significant differences only between managers and staff for the variable of autonomy. This is consistent over the entire study period. Thus, the second hypothesis that there would be significant differences between nurse managers and all other staff on the three study variables was only partially supported. The strength of relationships between the variables was inconsistent over the study period, but predominantly weak to moderate. The relationship between PDM and job satisfaction was positive and strengthened over the study period, supporting hypothesis three. The relationship in this study was stronger than that in the normative data. The strength of the relationship was comparable to that found by other researchers and confirms earlier work that strongly links the variables of PDM and job satisfaction. Autonomy and job satisfaction were positively correlated, supporting hypothesis four, although the relationship was weak and fluctuated over the study period. This finding was different from that of the normative data, but was similar to that for the comparison hospital. The findings also ran contrary to the published research that reveals a moderately strong correlation between these two variables. The data are supportive of the original hypothesis, but the scale may not be a good choice in comparison to other scales used. The relationship between PDM and autonomy was positive and supported hypothesis five, although it remained consistently very weak and not statistically significant. The relationship between these two variables has not been well-studied and there is confusion and overlap between 64 the two. Further research is required to examine the strength of relationships since the findings from this study are contrary to claims found in much of the administrative opinion literature. Individual nurse characteristics did alter perceptions and experiences of the three variables, thus supporting hypothesis six, but again, most of the significant differences did not hold for the entire study period and reflected the confusing array of findings found in the published research. Employment position and type of shift were the two most consistent characteristics affecting PDM. Managers, despite the fluctuations in numbers, still experience the highest PDM, involvement in decisions, and autonomy using the measures of this study. The managers are similar to all groups of staff in that their actual involvement did not meet their expectations of participating in decisions for any of the three years, and their PDM actually decreased in 1996 after an organizational change to program management. RNs, in comparison to all other categories of staff, were the least satisfied with their involvement in decisions that affected their work for two of the three years of data collection. By 1996, there were no significant differences, in part, because all PDM scores were equally low. Staff working permanent shift experienced the lowest PDM. Staff working a day/evening rotation did not rate participation as highly as staff working other shifts. Despite reports in other studies, educational level did not make a consistent significant difference for any of the three variables. Within the RN staff, a baccalaureate education did not make a significant difference in PDM, autonomy, or job satisfaction. Years of experience only affected the variable of participation for two of the three years. Nurses with 11 to 20 years experience were the group whose involvement most closely matched their expectations. 65 Position within the hierarchy and type of shift affected autonomy. Managers reported, in addition to the highest PDM, the highest autonomy of all staff. Staff working D/E experienced the least amount of autonomy in decisions affecting their work, while staff working extended hours reported the highest sense of autonomy. Employment position was the only nurse characteristic that affected job satisfaction. The findings run contrary to much of the research literature that identifies many more variables as significantly affecting job satisfaction. Six hypotheses were tested in this study. The first of these was not supported while the second, fourth, fifth, and sixth were supported, but was weak or inconsistent over time. The third hypothesis is supported and is consistent with the findings of others linking PDM and job satisfaction. The conclusions arising from the findings and their implications are presented in Chapter Five. 66 Chapter Five Summary, Conclusions, and Implications Summary The impetus for this study emerged from a simple question—How will we measure the effects of reorganizing? The nursing department of one hospital had undertaken a major reorganization to increase staff participation in decision making, and autonomy. The literature claimed many advantages of restructuring and implementing shared governance; however, published, quantifiable data were noticeably absent. The few research studies revealed conflicting results about the impact of PDM models on staff outcomes. The study was designed to measure the longitudinal impact of such a reorganization on staff perceptions. A standardized tool already developed and used for a similar study was adopted to ensure the ability to compare data. A demographic questionnaire was developed by the researcher to identify personal and professional characteristics of the respondents in order to determine the effects, if any, on their perceptions. The premise upon which the study was based came from the literature that suggested increased participation in decision making was a result of implementing shared governance, which, in turn, resulted in increased autonomy and job satisfaction. Specifically, three questions were put forward: (a) what is the impact over time of a participative decision making model on the variables: participative decision making (PDM), autonomy, and job satisfaction?, (b) what are the interrelationships among the three variables?, (c) what are the differences among the three 67 variables for individual staff characteristics such as position in the hierarchy, education, tenure, area worked, and shifts worked? The study was carried out over a three and one-half year period. Data were collected and analyzed using the SPSS statistical package. The sample represented the nursing population within the study hospital. Most of the staff had worked an average of 21 years in nursing and over half of their careers at the study hospital. The educational preparation of nurses was slightly below that of the provincial population with the majority being diploma-prepared. The findings indicated that there was no impact over time on staff PDM, autonomy, or job satisfaction. There were no significant differences between nurse managers and other staff except in the area of autonomy. The managers as a group exhibited the most fluctuation in responses. With respect to the interrelationships among the variables, the findings confirmed results in the research literature for PDM and job satisfaction; the correlation between autonomy and job satisfaction was weaker than that found by other researchers. PDM and autonomy were very weakly related. The latter two relationships may be limited, in part, by the scale that was used in this study to measure autonomy. The two concepts of autonomy and PDM are not well-differentiated in the research and need further testing and study. Individual nurse characteristics did affect perceptions of PDM, autonomy, and job satisfaction. The characteristics did not remain consistent throughout the study period as has been true in other studies. Position in the hierarchy consistently affected PDM, autonomy, and job satisfaction. Type of shift worked most affected both PDM and autonomy. Tenure or years of experience affected PDM for only two of three years. 6 8 Conclusions Analysis of the data suggests the following major conclusions: 1. Despite great organizational change and uncertainty in both the study hospital and the health care environment, there has been no appreciable change in the variables under study. 2. Nurse managers as a group are the most vulnerable to the effects of change. 3. Other staff have remained protected or impervious to the profound changes in the study hospital. 4. Much of the change has not altered the basic way and type of work that nursing staff perform. 5. A P D M model as it was implemented in this hospital is not successful by itself in altering staff perceptions of P D M , autonomy, and job satisfaction. 6. The most profound changes in all variables appear in the last year of data collection—1996. 7. The means for most of the variables for individual groups have fallen, resulting in very few significant differences or a more homogeneous group. 8. Type of shift worked affects P D M and autonomy. 9. A higher position within the organizational hierarchy increases autonomy, job satisfaction, and P D M . 69 Implications The findings of this study have implications for nursing administration, social and health policy, and nursing research. Nurse administrators have the responsibility to provide leadership in developing and implementing structures and initiatives that directly affect nursing practice. They must ensure that both managers and other staff are able to develop the requisite skills for today's professional practice. Nursing researchers can guide theory development regarding the work of nursing, but can also evaluate the effects of organizational change on both the quality of life and the quality of patient care. Social and health policy must reflect the requirements for administrator and practitioners to function in the current climate of health care. Nursing Administration One of the findings of the study was that PDM, autonomy, and job satisfaction did not increase after restructuring. Drawing also from the findings of other published research, it is clear that more than changing structure is required to make a measurable shift in staff perceptions. The findings indicate that managers are the group most directly affected by organizational change. They are expected in a PDM model to provide leadership and direction to the staff. It is important, therefore, for nurse administrators to ensure that managers receive the education necessary to understand the changes in function required when a PDM model is implemented. Continuing education, leadership development, and ongoing education regarding communication skills, conflict resolution, and team building are crucial if staff are to successfully participate in decision making (Triolo et al., 1995; Hastings & Waltz, 1995). Nurse managers were vulnerable not only in the reorganization, but also in the economic climate of restraint. Support positions, 70 restraint. Support positions, such as clinicians and administrative assistants, in addition to effective functioning hospital systems are required if the role of the manager is to successfully change from one of daily operations and supervising to mentoring and providing leadership. Comparing the findings of this study to findings from other research indicates that the lack of change in the variables may be related to an absence of initiatives such as self-scheduling, primary nursing, and peer review that are often a part of a PDM model. PDM is moderately correlated with job satisfaction. This is encouraging despite the lack of change found in this study. It should direct nurse administrators to continue to implement PDM models. Initiatives at the unit level that are seen to be meaningful and have an effect on the way care is delivered and/or facilitate control over the work environment are crucial to changing PDM. Ensuring that mechanisms are in place for staff to identify issues of concern is a cornerstone of a PDM model and requires commitment from the nurse administrator and nurse managers. The PDM model itself needs to be evaluated to ensure it is functioning as intended. Organizational change requires a commitment from nurse administrators and administrators to provide a vision that will see the change implemented. They should understand the length of time required before the effects of organizational change become evident. This study covers only a three-year period and is really midpoint in evaluating impact on staff perceptions. Nurse administrators must include an evaluative phase to all reorganization initiatives, including a multi-year commitment of funds for research and clerical support for the length of time required to evaluate the impact of the organizational change. Too often claims are made in opinion literature and implemented without formal mechanisms for evaluation. 71 Specific variables such as type of shift worked directly affected staff perceptions of the variables. Staff working rotating shifts and permanent shifts had the most difficulty remaining involved in P D M and feeling autonomous. Nurse administrators and nurse managers need to work with the staff to identify ways to ensure ongoing communication and to facilitate staff control over work issues such as bed control and workload patterns on shift. These directly affect P D M and autonomy of workers. Social and Health Policy Social and health policy are important at both the macro and the micro level. At the micro level, hospital and regional boards must develop vision statements that reflect the linkages between quality patient care and nursing autonomy and job satisfaction. The boards' vision and priorities must value the supports and education necessary for the workers and, in particular, the nurses to function autonomously and, as a result, become more effective in their jobs. At the macro level, health policy is concentrated on spending cuts and is too often primarily concerned with physician costs and resource planning. Both provincial and regional boards must address resource and educational planning for nursing and other allied health workers. Rather than focusing solely on the financial ledger, boards must develop a vision that not only encourages decentralization but also recognizes the need for supports and resources for the resulting structure to work. 72 Future Research Implications for research arise from a number of findings in the study. The relationship between autonomy and PDM is weak or non-existent. The research literature has not differentiated between the two concepts and further study is required to examine the areas of overlap and further differentiate the concepts. Testing Allen's conceptual framework, upon which this study was, in part, based, is an important piece of research to contribute to the understanding of the effects of the concepts and further theoretical understanding of PDM, autonomy, and job satisfaction. A further study could analyze the data for the 60 respondents in this study who remained constant for all three data collection periods. There are few longitudinal studies in the published research and more such studies are required to examine the effects of implementing a PDM model for more than three years. The components of a PDM model are not clearly identified or understood. Organizations have implemented different models and different components such as self-scheduling, clinical ladders, and case management. Too often, studies measure one or two of the components, but do not examine the effects of all of the initiatives and their interconnectedness. Researchers must start to identify the relationships between the PDM model and the specific initiatives that are crucial for successful change in staff perceptions. A qualitative component as part of the current study would have been beneficial in identifying specific staff concerns and perceptions regarding the organizational structure and areas in which they desired further decision making control. Nurse 73 characteristics and their relationship to PDM, autonomy, and job satisfaction also require further study. Job satisfaction has been greatly studied, but is a very complex concept. Causal models are really just emerging. One of the limitations of the findings of this study is the limited number of variables. Longitudinal studies that explore causal modeling of a large number of variables is crucial if the required changes in the work place and their effects are to be identified. This study was conducted to provide evaluative empirical data for the study hospital, so that the senior executive could evaluate the effectiveness of the model that was implemented. The agency can conclude that the model is not working in the way that it was intended. The conclusions may assist the CEO in refining the model and in implementing other required initiatives necessary to achieve a more self directed, autonomous staff. The executive of the hospital must, in particular, evaluate the effects on its most important and vulnerable level of staff—the managers. The findings of the study could be helpful to any organization contemplating a reorganization with a view to empowering staff and ultimately improving patient care. 74 References Acorn, S., Ratner, P., & Crawford, M. (1997). Decentralization as a determinant of autonomy, job satisfaction and organizational commitment. Nursing Research. 46 (1), 52-58. Allen, D., & Heidrich, S. (1986). Staff nurses' participation in decision making: Correlation with job satisfaction, organizational commitment, motivation, role conflict and performance-reward expectations. Unpublished manuscript, University of Washington, Seattle. Allen, D., Calkin, J., & Peterson, M. (1988). Making shared governance work: A conceptual model. Journal of Nursing Administration. 18 (1), 37-43. Alexander, C , Weisman, C.S., & Chase, G.A. (1982). Determinants of staff nurses' perceptions of autonomy within different clinical contexts. Nursing Research. 31. (1). 48-51. Alutto, J. A., & Bellascoe, J.A. (1972). A typology for participation in organizational decisionmaking. Administrative Science Quarterly. 17. 117-125. Alutto, J. A., & Vredenburgh, D. J. (1977). Characteristics of decisional participation by nurses. Academy of Management Journal. 20 (2), 341-347. Attridge, C. B. (1996). Analysis of powerlessness in nursing workforce. Canadian Journal of Nursing Administration. 9 (2), 36-59. Besel, L., & Stock, R. (1988). Benchmarks 1: A sourcebook for Canadian nursing management. Toronto: Carswell. Bland Jones, C , Stasiowski, S., Simons, B. J., Boyd, N. J., & Lucas, M. D. (1993). Shared governance and the nursing practice environment. Nursing Economics. 11 (4), 208-213. Blegen, M.A. (1993). Nurses' job satisfaction: A meta-analysis of related variables. Nursing Research. 45 (1), 36-41. Blegen, M. A., Goode, C , Johnson, M., Maas, M., Chen, L., & Moorehead, S. (1993). Preferences for decision making autonomy. Image: Journal of Nursing Scholarship. 25 (4), 339-344. Blegen, M. A., & Mueller, C. W. (1987). Nurses' job satisfaction: A longitudinal analysis. Research in Nursing & Health. 10.227-237. Burns, N., & Grove, S.K. (1993). The practice of nursing research: Conduct, critique, and utilization (3rd ed.). Philadelphia: W.B. Saunders. 75 Bush, P. (1988). Job satisfaction, powerlessness, and locus of control. Western Journal of Nursing Research. 10 f61. 718-731. Canadian Nurses Association, Canadian Hospital Association. (1990). Nurse retention and quality of worklife: A national perspective. Ottawa: Author. Cavanaugh, S. J. (1992). Job satisfaction of nursing staff working in hospitals. Journal of Advanced Nursing 17. 704-711. Counte, M.D., Barhyte, D.Y., & Christman, L.P. (1987). Participative management among staff nurses. Hospital and Health Administration Services. 32 (1), 97-108. Curtis, L. (1994). Restructuring: What works - and what does not! Nursing Management. 10 (3), 7-8. Donaldson, S. (1995). Organizational restructuring. Nursing BC. March/April. 12-15. Dwyer, D.J., Schwartz, R.H., & Fox, M. L. (1992). Decision-making autonomy in nursing. Journal of Nursing Administration. 22 (2), 17-23. Glisson, C . , & Durick, M. (1988). Predictors of job satisfaction and organizational commitment in human service organizations. Administrative Quarterly. 33. 61-81. Hackman, R. & Oldham, G. (1975). Development of the job diagnostic survey. Journal of Applied Psychology. 60 (2), 159-170. Hage, J., & Aiken, M. (1967). Relationship of centralization to other structural properties. Administrative Science Quarterly. 12 (6), 72-91. Hair, J. F., Jr., Anderson, R. A., Tatham, R. L., & Black, W. C. (1992). Multivariate data analysis with readings (3rd ed.). New York: Macmillan. Hassen, P. (1988). Decentralization as a means to achieve a dynamic organization. In L. Besel & R. Stock (Eds.) Benchmarks 2: A sourcebook for Canadian nursing management, (pp. 177-181). Toronto: Care well. Hastings, C , & Waltz, C. (1995). Assessing the outcomes of a professional practice design: Impact on staff nurses' perceptions. Journal of Nursing Administration 25 (3), 34-42. Hatcher, S., & Laschinger, H. K. (1996). Staff nurses' perception of empowerment and level of burnout: A test of Ranter's theory of structural power in organizations. Canadian Journal of Nursing Administration. 9 (2), 74-94. Havens, D. S. (1994). Is governance being shared? Journal of Nursing Administration. 24 (6), 59-64. 7 6 Heckscher, C. & Donnelon, E. (Eds.). (1994). New perspectives on organizational change. London: Sage Publications. Hess, R. G. (1994). Shared governance: Innovation or imitation? Nursing Economics. 12 (1), 28-33. Hess, R. G. (1995). Shared governance: Nursing's 20th century tower of babel. Journal of Nursing Administration. 25 (5), 14-17. Hinshaw, A. S., & Atwood, J. R. (1984). Nursing staff turnover, stress, and satisfaction: Models, measures and management. In H. H. Werley and J. J. Fitzpatrick (Eds.), Annual review of nursing research. Volume 1 (pp. 133-153). New York: Springer Hinshaw, A.S., Smeltzer, C.H., & Atwood, J.R. (1987). Innovative retention strategies for nursing staff. Journal of Nursing Administration. 17 (6), 8-16. Hrebeniak, L. G. (1974). Effects of job level and participation on employee attitudes and perceptions of influence. Academy of Management Journal. 17. 649-662. Ingersoll, G. L., Schultz, A. W., Hoffart, N., & Ryan, S. A. (1996). The effect of a professional practice model on staff nurse perception of work groups and nurse leaders. Journal of Nursing Administration. 26 (5). 52-60. Irvine, D., & Evans, M. (1992). Job satisfaction and turnover among nurses: A review and meta-analysis. Toronto: University of Toronto, Quality of Nursing Worklife Research Unit. Irvine, D., & Evans, M. (1995). Job satisfaction and turnover among nurses: Integrating research findings across studies. Nursing Research. 44 (4), 246-253. Jones, L., & Ortiz, M. (1989). Increasing nursing autonomy and recognition through shared governance. Nursing Administration Quarterly. 13 (4). 11-16. Kanter, R. M. (1977). Men and women of the corporation. New York: Basic Books. Kanter, R. M., & Brown, J. C. (1982). Empowerment: Key to effectiveness. Hospital Forum. 25 (3), 6-13. Kidder, L. H. (1981). Research methods in social relations (4th ed.). New York: Holt, Rinehart and Winston. Kovner, C. T., Hendrickson, G., Knickman, J. R., & Finkler, S. A. (1994). Nursing care delivery models and nurse satisfaction. Nursing Administration Quarterly. 19 (1), 74-85. Kramer, M., & Schmalenberg, C. (1988a). Magnet hospitals: Part 1: Institutions of Excellence. Journal of Nursing Administration. 18(1). 13-24. 7 7 Kramer, M., & Schmalenberg, C. (1988b). Magnet hospitals: Part 2: Institutions of Excellence. Journal of Nursing Administration. 18(21 11-19. Laschinger, H. K., & Shamian, J. (1994). Staff nurses' and nurse managers' perceptions of job-related empowerment and managerial self-efficacy. Journal of Nursing Administration. 24 (10), 38-47. Layton, D.G. (1988). Influence of the workplace on nurse manpower supply in British Columbia: An exploratory study. In The nurse manpower study. (Volume III). Vancouver: University of British Columbia, Health Manpower Research Unit. Lomax, R. G. (1992). Statistical concepts: A second course for education and the behavioral sciences. New York: Longman. Ludemann, R.S., & Brown, C. (1989). Staff perceptions of shared governance. Nursing Administration Quarterly. 13 (4), 49-56. Marriner-Toomey, (1988). Guide to nursing management (3rd ed.). St. Louis: Mosby. McCloskey, J. C , Mass, M., Huber, D. G., Kasperek, A., Specht, J., Ramler, C , Watson, C , Blegen, M., Delaney, C , Ellerbe, S., Etscheidt, C , Gongaware, C , Johnson, M., Kelly, K., Mehmert, P., & Clougherty, J. (1994). Nursing management innovations: A need for systematic evaluation. Nursing Economics. 12 (1), (35-43). McDonagh, K., Rhodes, B., Sharkey, K., & Goodroe, J. (1989). Shared governance at St. Joseph's Hospital of Atlanta: A mature professional practice model. Nursing Administration Quarterly. 13 (4), 17-28. Minors, S. P., White, J. B., & Porter-O'Grady, T. (1996). Assessing shared governance: An example of instrument development in a hospital setting. In M. A. Rahim, R.T. Gblembiewski, & C.C. Lundberg, Current topics in management. Volume 1 (pp. 187-198). Greenwich: Jai Press. Mueller, C.W., & McCloskey, J.C. (1990). Nurses'job satisfaction: A proposed measure. Nursing Research. 39 (2). 113-117. Naisbett, J., & Aburdene, P. (1985). Reinventing the corporation. New York. Warner Publications. O'Brien-Pallas, L. L., & Baumann, A. (1992). Quality of nursing worklife issues—A unifying framework. Canadian Journal of Nursing Administration. 5 (2). 12-16. Peterson, M.E., & Allen, D.G. (1986). Shared governance: A strategy for transforming organizations. Part 2. Journal of Nursing Administration. 16 (2). 11-16. 78 Porter-O'Grady, T. (1987). Shared governance and new organizational models. Nursing Economics. 5 (6), 281-285). Porter-O'Grady, T. (1991a). Shared governance for nursing, Part II: Putting the organization into action. American Operating Room Nursing Journal. 53 (3), 694-703. Porter-O'Grady, T. (1991b) Changing realities for nursing: New models, new roles for nursing care delivery. Nursing Administration Quarterly. 16 (1), 1-6. Porter-O'Grady, T. (1992a). Of rabbits and turtles: A time of change for unions. Nursing Economics. 10 (3), 177-182. Porter-O'Grady, T. (1992b). Implementing shared governance: Creating a professional organization. New York: Mosby. Porter-O'Grady, T. (1993). Patient focused care service models and nursing: Perils and possibilities. Journal of Nursing Administration. 23 (3), 7, 8, 15. Porter-O'Grady, T. (1994). Worker partnerships in hospitals: Constructing new paradigms for a transformational age. Manuscript submitted for publication. Registered Nurses Association of British Columbia, (1989a). Registered nurse manpower planning guide. Vancouver: Author. Registered Nurses Association of British Columbia, (1989b) Increasing nurse satisfaction: A guide to improving retention. Vancouver: Author. Ringerman, E.S. (1990). Characteristics associated with decentralization experienced by nurse managers. Western Journal of Nursing Research. 12 (3), 336-346. Sabiston, J.A., & Laschinger, H.K. (1995). Staff nurse work empowerment and perceived autonomy: Testing Ranter's theory of structural power in organizations. Journal of Nursing Administration. 25 (9), 42-49. Schutzenhoffer, K. K., & Musser, D. B. (1994). Nurse characteristics and professional autonomy. Image: Journal of Nursing Scholarship. 26 (3), 201-205. Shoemaker, S., & El-Ahraf, A. (1983). Decentralization of nursing service management and its impact on job satisfaction. Nursing Administration Quarterly. 7. 69-75. Skelton-Green, J. (1996). The perceived impact of committee participation on job satisfaction and retention of staff nurses. Canadian Journal of Nursing Administration. 9 (2), 7-35. 79 Slavitt, D. B., Stamps, P. L., Piedmont, E. B., & Haase, M. M. (1979). Measuring nurses' job satisfaction. Hospital and Health Services Administration. 24(3). 621-76. Statistics Canada (1994). Registered Nurses Management Data 1994. Ottawa: Author. Triolo, P. K., Allgeier, P. A., & Schwartz, C. E. (1995). Layoff survivor sickness: Minimizing the sequelae of organizational transformation. Journal of Nursing Administration. 25 (3), 56-63. Trump-Wells, G. (1990). Influence of organizational structure on nurse manager job satisfaction. Nursing Administration Quarterly. 14 (4), 1-8. University of Minnesota, (1967). Minnesota satisfaction questionnaire. MN: Author. Van de Ven, A. H., & Ferry, D. (1980). Measuring and assessing organizations. New York: Wiley. Walker, J. H. (1990). Job satisfaction among hospital-employed nurses. Unpublished master's thesis, University of British Columbia, Vancouver, British Columbia. Wanous, J. P., & Lawler, E. E. III. (1972). Measurement and meaning of job satisfaction. Journal of Applied Psychology. 56 (2), 95-105. Weins, A. G. (1990). Expanded nurse autonomy: Models for small rural hospitals. Journal of Nursing Administration. 20 (12), 15-22. Weisman, C. S., Alexander, C. S., & Chase, G. A. (1980). Job satisfaction among hospital nurses: A longitudinal study. Health Services Research. 15. 341-364. Weisman, C. S., Alexander, C. S., & Chase, G. A. (1981). Determinants of hospital staff nurse turnover. Medical Care. 19 (4). 431-43. Weisman, C. S., Alexander, C. S., & Morlock, L. L. (1981). Hospital decision making: What is nursing's role? Journal of Nursing Administration. 11 (9), 31-35. Weisman, C. S., Gordon, D. L., Cassard, S. D., Bergner, M., & Wong, R. (1993). The effects of unit self-management on hospital nurses' work process, work satisfaction, and retention. Medical Care. 31 (5). 381-393. Weisman, C. S., & Nathanson, C A . (1985). Professional satisfaction and client outcomes: A comparative organizational analysis. Medical Care. 23.1179-1192. Westrope, R. A., Vaughn, L., Bott, M., & Taunton, R.L. (1995). Shared governance: From vision to reality. Journal of Nursing Administration. 25 (12). 45-54. 80 Williams, L. J., & Hazer, J. T. (1986). Antecedents and consequences of satisfaction and commitment in turnover models: A reanalysis using latent variable structural equation methods. Journal of Applied Psychology. 71 (2), 219-231. Wilson, B., & Laschinger, H. K. (1994). Staff nurse perception of job empowerment and organizational commitment: A test of Kanter's theory of structural power in organizations. Journal of Nursing Administration. 24 (4s), 39-47. Zelauskas, B., & Howes, D.G. (1992). The effects of implementing a professional practice model. Journal of Nursing Administration. 22 (7/8). 18-23. 8 1 A P P E N D I X A DECENTRALIZATION AND PARTICIPATION IN DECISION MAKING SCALE -1996 DEMOGRAPHIC PAGE 1. Position: Permanent Care Aide 1 Permanent LPN 2 (circle one) Permanent Staff Nurse 3 Department Director 4 Clinical Support 5 Supervisor 6 Other 7 2. What percent time do you work in a typical week? (Circle one) 1. 100% 4. 25-49% 2. 75-99% 5. Less than 25% 3. 50-74% 3. What shift(s) do you usually work? (Circle one) Permanent Days 1 Permanent Evenings 2 Permanent Nights 3 12 Hour Shifts 4 Extended Work 9 Day Fortnight 5 Days/Evenings 6 Days/Nights 7 4. What area do you work in? (Circle one) 1. Critical Care/O.R./S.D.C. 2. Acute Care 3. Extended Care/SSATC 5. Years worked at this organization. 6. Years worked on current Unit. 7. Years worked in Nursing. 8. Nursing Education: (Circle one) Care Aide Certificate LPN Diploma BSN MSN Other 1 2 3 4 . 5 6 82 DIRECTIONS 1. Please circle the number which m o s t reflects your feelings or beliefs. Do not circle more than one number and don't choose a position midway between two numbers. 2. Use pen or pencil, but if you change your mind, please draw a distinct "X" through the incorrect response. 3. Please return the completed Questionnaire within seven days to the boxes which will be placed on each Unit. We appreciate your cooperation a great deal and sincerely hope your efforts will make this organization a better place for you to work. We will make the results readily available to you as soon as the data is analyzed. The next questions are about decisional areas that B.C. nurses have identified as being important to them. For each decisional area, please indicate both I M P O R T A N C E / I N V O L V E M E N T a) just how important it is for you to be involved in the decision and b) your present level of involvement in that decision. Please use the following scales: a) I m p o r t a n c e 1 2 3 4 5 Not important to be involved in this decision Slightly important to be involved Somewhat important to be involved Very important to be involved Extremely important to be involved b) P r e s e n t I n v o l v e m e n t 1 2 3 4 5 Not at all involved in this decision Slighdy involved Somewhat involved Very involved As completely involved in this decision as I want to be 83 1 2 3 4 5 Not Important Slightly Somewhat Very Extremely Important 9. Providing information to patients and families. a) Importance [ ] b) Involvement [ ] 10. Assigning Unit personnel to their daily work. a) Importance [ ] b) Involvement [ ] 11. Coordinating overall patient services (i.e., Physical Therapy, X-ray, etc.) with nursing care. a) Importance [ ] b) Involvement [ ] 12. Hiring new Nursing personnel. a) Importance [ ] b) Involvement [ ] 13. Creating Nursing Standards and Procedures for the Unit. a) Importance [ ] b) Involvement [ ] 14. Detennining Break Schedules. a) Importance [ ] b) Involvement [ ] 15. Determining Annual Unit Objectives. a) Importance [ ] b) Involvement [ ] 16. Determining annual Nursing Division Objectives. a) Importance [ ] b) Involvement [ ] 17. Creating Hospital Policies and Procedures. a) Importance [ ] b) Involvement [ ] 18. Determining the over-all Nursing Service Budget. a) Importance [ ] b) Involvement [ ] 19. Goal Setting with Patients. a) Importance [ ] b) Involvement [ ] 20. Assigning Patients to Nurses. a) Importance [ ] b) Involvement [ ] 21. Coordinating Discharge Plan Information with Patient. a) Importance [ ] b) Involvement [ ] 22. Determining Assignment of Patients to Rooms. a) Importance [ ] b) Involvement [ ] Questions 9-34= PDM (Importance - Involvement) 84 1 2 3 4 5 Not Important Slightly Somewhat Very Extremely Important 23. Determining Number of Admissions for Elective O.R. a) Importance [ ] b) Involvement [ ] 24. Determining Bed Moves on same Unit. a) Importance [ ] b) Involvement [ ] 25. Determining Bed Moves to Different Units. a) Importance [ ] b) Involvement [ ] 26. Evaluating the Implications of Decisions made by Other Disciplines/Areas (e.g. product choices) on Nursing Practice. a) Importance [ ] b) Involvement [ ] 27. Determining Inservice and other Educational Programs for Staff, a) Importance [ ] b) Involvement [ ] 28. Impacting Multidisciplinary Patient Care Decisions. a) Importance [ ] b) Involvement [ ] 29. Determining Hospital Environment Issues (such as parking, unit moves, space design), a) Importance [ ] b) Involvement [ ] 30. Determining Admission to a Unit for a particular Shift. a) Importance [ ] b) Involvement [ ] 31. Impacting collaboration between Nursing and Medicine. a) Importance [ ] b) Involvement [ ] 32. Determining Introduction and Process for Implementing new Patient Care Technology, a) Importance [ ] b) Involvement [ ] 33. Determining Standards and Procedures for Nursing Practice. a) Importance [ ] b) Involvement [ ] 34. Determining Unit Work Schedules (Self Scheduling), a) Importance [ ] b) Involvement [ ] Please identify other areas of decision-making where you want to be but are not currently participating to your satisfaction: Questions 9-34= PDM (Importance - Involvement) 85 PLEASE ANSWER THE FOLLOWING QUESTIONS BY THINKING ONLY ABOUT THE TYPES OF TASKS THAT OCCUPY MOST OF YOUR WORKING TIME 35. To what extent does your work alone give you many clues to figure out how well you rate doing your job (without relying on feedback from your Department Director or co-workers)? My work gives me no clues on how welll do my job My work gives a few clues on my job performance My work gives some clues on my job performance My work gives many clues on my job performance My work gives all clues needed to know how well I do my job *36. How heavy was your workload during the past 3 months? Often not enough to keep me busy Sometimes not enough to keep.me busy Just about the right.amount Hard to keep up.with Entirely too much for me to handle 1 *37. How far in advance do you generally know how much work will be required of you? About an hour or About a day in About a week in About a month in About 6 months or less in advance advance advance advance more in advance 1 2 3 4 5 *38. During the past 3 months, how much control did you have in setting the pace of work? None Very Little Some Quite a Bit Very Much 1 * Autonomy 86 39. Listed below are four common decisions about your work. How much authority do you have in making each of the following decisions about your work? By "authority", we mean how much weight does your own opinion/decision have in determining what actually happens? AMOUNT OF AUTHORITY I HAVE IN EACH DECISION QUITE V E R Y I A M NONE L I T T L E S O M E A B I T M U C H N O T SURE *a. Determining what tasks I will 1 2 3 4 5 6 perform day to day. *b. Setting limits on how much 1 2 3 4 5 6 work/how many patients I have. *c. Establishing Policies and 1 2 3 4 5 6 Procedures about how my work is to be done. *d. Determining how unusual 1 2 3 4 5 6 situations are to be handled. 40. How much does your Supervisor or Department Director hold you personally accountable for the work decisions you make in your job? Not at All Very Little Some Quite a Bit Very Much I 2 3 4 5 41. How clearly do you know what level of work performance is expected from you (in terms of amount, quality and timelines?) Very Unclear Quite Unclear Somewhat Clear Quite Clear Very Clear 1 2 3 4 5 * Autonomy 87 Y O U R FEEL INGS ABOUT Y O U R JOB WE WOULD LIKE TO ASK YOU SOME QUESTIONS ABOUT HOW YOU PERSONALLY FEEL ABOUT YOUR JOB. Very Somewhat Somewhat Very Unsatisfied Unsatisfied Neutral Satisfied Satisfied 50. **b. How satisfied are you with each of the following: Your job? Your Dept. Director immediate Supervisor? Your pay? The friendliness and cooperativeness of your co-workers? The career progress you have made in this organization up to now? 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 **f. Your chances for career advancement in this organization in the future? None A Little Some Quite a Very Much Bit 51. How much effort do you put 1 into your work? 52. How much did you try to 1 improve your job performance in the past 3 months? ** Job Satisfaction 88 YOUR FEELINGS ABOUT YOUR JOB (continued) The following items ask about your feelings towards your job. Please read them carefully and be frank and honest in the feelings you report. **53. How do you feel about the amount of work you're expected to do? 1. Very Dissatisfied. 2. Somewhat Dissatisfied. 3. Neither Satisfied nor Dissatisfied. 4. Somewhat Satisfied. 5. Very Satisfied. 54. How does the amount of work you're expected to do influence the way you do your job? 1. It never allows me to do a good job. 2. It seldom allows me to do a good job. 3. It has no effect on how I do my job. 4. It usually allows me to do a good job. 5. It always allows me to do a good job. *55. I feel my workload is: 1. never too heavy. 2. seldom too heavy. 3. sometimes too heavy. 4. often too heavy. 5. almost always too heavy. Please use the following scale to indicate your agreement with the following statements. (Use whichever scale makes the most sense to you). 1 2 3 4 5 Strongly Disagree Disagree Undecided Agree Strongly Agree 1 2 3 4 5 Never Rarely Sometimes Frequently Always 56. I receive an assignment without 1 2 3 4 5 adequate resources and time to execute it. 57. I have to go against Hospital 1 2 3 4 5 Policies or Procedures to carry out an assignment. * Autonomy; ** Job Satisfaction 89 YOUR FEELINGS ABOUT YOUR JOB (continued) 1 = Strongly Disagree 1 = Never 2 = Disagree 2 = Rarely 3 = Undecided 3 = Sometimes 4 = Agree 4 = Frequently 5 = Strongly Agree 5 = Always 58. I feel certain about how much authority I have. 59. I am under incompatible Policies and Directives concerning how to do my work. 60. I am able to carry out my nursing responsibilities in the same way whether I'm around other nurses, supervisors or physicians. 61. I work with two or more groups of professionals who operate quite differently and each expects me to conform to the way they do things. 62. I receive incompatible requests from two or more people. 63. I do things that are apt to be accepted by one person and not accepted by others. **64. My job requires that I keep learning new things. **65. On my job I get a chance to use my skills and abilities. ** Job Satisfaction 90 YOUR FEELINGS ABOUT YOUR JOB (continued) 1 2 3 4 5 Strongly Disagree Disagree Undecided Agree Strongly Agree **66. On my job I don't perform a 1 2 3 4 5 complete service. My contri-bution to a patient's care consists of little bits and pieces which are added to by many other people. **67. My work makes a visible impact 1 2 3 4 5 on the patient's ultimate outcome. 68. Most things in life are more 1 2 3 4 5 important than work. 69. Each of the statements below is something that a person might say about his or her job. Please indicate your own, personal feelings about your job by cncling a number on the scale to the right of each statement to indicate how much you agree or disagree with each statement. Disagree Disagree Neutral Agree Agree Strongly Somewhat Somewhat Strongly a. It's hard on this job for me to care very much about whether or not the work gets done right. b. My opinion of myself up when I do this job well. c. I feel a great sense of personal accomplishment when I do this job well. d. I feel a very high degree of personal responsibility for the work I do on this j o b . 1 ** Job Satisfaction 91 THE ORGANIZATION AND WORK OF YOUR UNIT So far you have been asked questions about your work and your job. This next part asks how your Unit is organized to do its work and achieve its performance goals. Please keep in mind that your Unit consists of your Department Director or immediate Supervisor and all individuals (your co-workers) who report directly to your Department Director or Supervisors. 78. How much say or influence do each of the following have in deciding what kinds of work or patient care will be performed in your Unit? AMOUNT OF SAY IN DECIDING UNIT'S WORK/PATIENT CARE None *a. People in line management 1 or staff positions outside of your immediate work Unit? *b. Your Dept. Director or 1 Supervisor? *c. Unit members, 1 individually? *d. The Unit Dept. Director 1 and staff members as a group in Unit meetings? *e. Physicians? 1 Little Some Quite a Bit Very Much 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 *Autonomy 92 A P P E N D I X B Department of Nursing September 3,1993 To: All Nursing Staff, Department Directors, Supervisors, From: Linda Hughes, Nursing Q.A. Coordinator Re: Shared Governance Questionnaire It is time to evaluate some of the differences in the Nursing Division since the introduction of Shared Governance. Attached is a questionnaire which we are using to collect information about staff perceptions regarding Shared Governance. We are trying to evaluate what are the effects by reorganizing in in this way. We would appreciate your most candid responses and are asking for everyone to assist by completing this questionnaire and returning them by September 25,1993. Your responses will be confidential and anonymous. The questionnaires are numbered to ensure the same people are included in a follow up questionnaire in 1 year's time. These numbers will only be seen by someone outside the agency who will analyze the questionnaires. You will need approximately 30 minutes to complete this. We appreciate your assistance. Other factors such as work productivity and staff turnover will be collected and used in conjunction with the results of this questionnaire when we evaluate the effectiveness of implementing Shared Governance structure in the Division. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items