Journal of Applied Psychology 1999. Vol. 84. No. 4, 514-528

Copyright 1999 by the American Psychological Association. Inc. 002I-9010/99/$3.00

What's a Good Reason to Change? Motivated Reasoning and Social Accounts in Promoting Organizational Change Denise M. Rousseau and Snehal A. Tijoriwala Carnegie Mellon University This study provides support for social accounts theory and motivated reasoning theory, both of which help explain how employees interpret the reasons that organizations undertake change. In a field study of a hospital implementing empowerment among nurses, staff nurses (n = 501) cited 3 types of reasons as motivating the change: economic, quality improvement, and self-serving or political reasons. (The formal, managerially stated reason for the change was quality improvement.) Results generally supported social accounts theory regarding the managerial explanation. In this setting, however, many nurses did not believe the explanation management offered. Their alternative interpretations of the change were investigated from the perspective of motivated reasoning. Findings indicated that nurses' trust in management, their psychological contracts with the hospital, and the beliefs of their coworkers affected the reasons nurses cited for the change.

"We want to care for patients not paper." "Nurses are able to share in their own future." "Added responsibility without monetary reimbursement." —Three contrasting views of the same restructuring

necessitate organizational changes and put pressure on employers to effectively manage these changes (Handy, 1989; Kanter, 1989). The causal frameworks employees use to understand change are not well understood, despite the general finding that the way employees interpret the reasons for change influences their reactions to it (Shapiro, Buttner, & Barry, 1994). How organizations frame change affects employee responses (Fairhurst & Sarr, 1996; Pondy, 1978). However, we know that not all reasons used to explain change are credible or acceptable to employees (Bies & Moag, 1986; Bies & Shapiro, 1993). The present study investigates the role of reasons in motivating complex organizational change from the perspective of social accounts and motivated reasoning. By joining these two perspectives to study complex organizational change, the present study is distinct from previous research on reasons and change. Typically, such studies are conducted in the laboratory (Bies & Shapiro, 1993; Shapiro, 1991) or involve field studies of a discrete event (e.g., introduction of a no-smoking policy; Greenberg, 1994). The present study, in contrast, investigates a major change intervention, where several competing reasons for change can exist simultaneously. (Indeed, the opening quotes are anecdotal evidence that recipients of the same change can have different interpretations.) Extending previous studies that limited their focus largely to managerial communication (Bies & Shapiro, 1993; Bies, Shapiro, & Cummings, 1988; Greenberg, 1994), this study also examines the roles played by individual, relational, and social information-processing factors in shaping interpretations of change. By considering factors at both the individual and

Downsizing and restructuring have brought about radical changes in employment relationships worldwide (Kanter, 1989). These events are blamed for increased employee stress and dissatisfaction (Mirvis & Hall, 1994) and, when poorly implemented, for reduced performance and commitment and increased perceptions of injustice (Novelli, Kirkman, & Shapiro, 1995; Schneider & Bowen, 1995). Global competition, cost pressures, innovations in information technology, and rising customer expectations are seen to

Denise M. Rousseau, H. John Heinz III School of Public Policy and Management and Graduate School of Industrial Administration, Carnegie Mellon University; Snehal A. Tijoriwala, H. John Heinz III School of Public Policy and Management, Carnegie Mellon University. An earlier version of this article was presented at the meeting of the Society for Industrial and Organizational Psychology, San Diego, California, April 1996. We appreciate the insightful comments provided by Mark Fichman, Paul Goodman, Susan Straus, Tom Tyler, and Laurie Weingart. Thanks are also due to Carole McCoy for word processing, Catherine Senderling for copyediting a previous version of the article, and Judy DePalma and Marge Hardt for help with data collection. Correspondence concerning this article should be addressed to Denise M. Rousseau, H. John Heinz III School of Public Policy and Management, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213. Electronic mail may be sent to rousseau+@ andrew.cmu.edu. 514

WHAT'S A GOOD REASON TO CHANGE?

work-group levels, we investigated the possible origins of the competing and diverse views employees can have regarding change.

How Reasons for Change Are Understood We review two major theories regarding causal explanations to understand how employees interpret the reasons for change (social accounts and motivated reasoning): Although social accounts theory focuses on reasons as a managerial intervention, motivated reasoning theory focuses on how individuals generate their own causal explanations. Social Accounts Social accounts have been defined as managerial justifications and excuses (Sitkin & Bies, 1993, p. 350) that are used to explain actions undertaken by persons or firms, such as citing economic conditions as a motivation for change. Sitkin and Bies describe three categories of social accounts studied by researchers: mitigating responsibility, which are used to alter perceptions of causality for an incident or action; exonerating motives, which attempt to legitimate the action by appealing to higher-level values; and reframing outcomes, which alter perceptions regarding the consequences of the action. Mitigating accounts have been found to lessen apparent responsibility for unfavorable outcomes and reduce feelings of unfairness and disapproval ("No one could have anticipated the decline in demand"). Exonerating accounts attempt to clarify basic premises underlying the actions, such as appealing to a shared goal (e.g., "competitive superiority," Sitkin & Bies, 1993, p. 356). Reframing accounts alters perceptions regarding consequences ("Downsizing today will create a more competitive organization and stable employment in the future"). Reasons that refocus attention externally (e.g., claiming economic justification for the change) can be a mix of all three forms of account; they are thought to unite manager and subordinate (Coser, 1956) and reduce the conflict between them in times of change (Sitkin & Bies, 1993). The social accounts managers offer influence employee reactions as a function of the adequacy or credibility of the reason and the sincerity of the account giver (Bies, 1987). Outcome negativity—the degree of adversity the change recipient associates with the change—has been also examined as a factor in social accounts research, but the results have been mixed: Shapiro et al., (1994) found that perceived adequacy of accounts is greater under conditions of low outcome severity, whereas Shapiro (1991) found the opposite. We note that in both studies, the expected outcomes of change were negative, without any potential benefits to employees, and the effectiveness measure of the account was its adequacy in lessening the outcome's perceived severity.

515

In-general, social accounts research focuses on changes that are adverse or negative in character (e.g., job losses, budget cuts, wage freezes, or failure to receive promised benefits or rewards). Accounts play the role of excuses or justifications used to influence a person's perceptions of responsibility for an action, the motives behind it, and its unfavorability. Typical accounts examined include external factors (e.g., economic downturns) or lack of managerial control (e.g., changes instituted by new owners many levels removed from the supervisor who offers the account; Bies & Moag, 1986). The most typical outcome focused on in this research is the perceived legitimacy of the explanation given. The purpose of social accounts is to create beliefs on the part of recipients that change is justified. In the context of organizational change, we are interested in two beliefs: the reasons employees perceive to motivate the change (why change occurs) and the legitimacy they attribute to those reasons (acceptability of a reason to the recipient). The present study distinguishes between perceived reasons and the perceived legitimacy of those reasons in investigating how change recipients understand the processes associated with managerially driven change. Consistent with social accounts theory, we made the following hypothesis. Hypothesis 1: Belief in the social account that management has expressed will be positively related to the perceived legitimacy of that explanation.

Note that given the absence of a theoretical basis, we make no specific prediction regarding the relationship between perceived legitimacy and other reasons employees interpreted for the change, aside from the managerial account. However, we recognize that in complex organizational change, recipients may believe that a variety of reasons motivated it. Bies (1987) argues that employees who trust management are more likely to accept the managerial account as justifying the change. Trust can influence both the credibility of the actual reason (whether it is believed to be true) as well as belief in its legitimacy (whether it is justified). Where employees trust management, the managerial account will be more credible. In the context of a high trust manageremployee relationship, the account's credibility should promote its legitimacy by reducing suspicion and the search for disconforming information. Hypothesis 2: Trust moderates the relationship between belief in the managerial account and the account's perceived legitimacy. High trust levels will be associated with a stronger relationship between beliefs in the managerial account and its perceived legitimacy.

Limitations of Social Accounts Theory By focusing on reasons offered by managers, social accounts theory gives a framework for understanding the

516

ROUSSEAU AND TIJORIWALA

perceived legitimacy of reasons formally articulated during the change process. However, employees are not passive recipients of change; they can draw their own interpretations and act on their own understandings of it (Shapiro, Lewicki, & Devine, 1995). In complex organizational change, multiple competing (as well as complementary) reasons for change may exist simultaneously. In these situations, explanations managers offer for change may not be believed, heard, understood, or recalled. The same change intervention can give rise to very different causal explanations from the perspective of employees. Restructurings may be viewed as motivated by the desire for better market position (Porter, 1990), improved customer service (Schneider & Bowen, 1995), cost containment (Harrison, 1994), or a more motivating environment for the workforce (Reichheld, 1996). The latter, in effect, can be viewed as an end in itself or as a means of achieving sustainable competitive advantage through improved worker motivation (Pfeffer, 1994). Given the potential for several reasons to be attributed simultaneously, two employees may have very different views of why their organization has instituted change, as in the words of two nurses in the same hospital implementing a flatter organizational hierarchy: 1. "We are becoming more professional and better able to improve our own practice through the change." 2. "(The change was introduced) so some bigwig with an MBA gets to look good." To understand the origin of such divergent interpretations of the reasons for change and how they affect implementation, we examined the factors that affect causal reasoning generally. Motivated Reasoning People seek causal explanations for unusual or unexpected events (Weiner, 1985). Motivated reasoning addresses how a particular reason is adopted to explain such an event. From this perspective, employees are not passive recipients of change messages but active information processors who engage in causal selection (Einhorn & Hogarth, 1986; McGill, 1995). Moreover, the causal backgrounds people possess (e.g., facts, perspectives, emphases) are an amalgam of past experience, available information, and individual cognitive processes that can lead employees of the same organization to generate different explanations for the same event (McGill, 1995). Although managers may offer a social account for their actions, a variety of other factors can shape employees' information processing regarding change, including coworker opinions and past experiences with management. The informal network or grapevine frequently is an important source of information during change (Krackhardt & Hanson, 1993). Social information processing shapes the

causal, explanations people adopt, particularly where there is consensus among coworkers (Einhom & Hogarth, 1986; Salancik & Pfeffer, 1978). Thus, the beliefs of one's work group can be expected to impact the beliefs adopted by a given individual as to why change has occurred: Hypothesis 3: Work-group beliefs will be positively related to individual perceptions of the reasons for change.

Ample evidence exists that an employee's relationship with the organization shapes his or her interpretation of its actions, particularly because this relationship is a product of the employee's history with the organization (Rousseau, 1995). Employee reasoning may be shaped by perceptions of the organization's trustworthiness and other factors related to the employment relationship (e.g., high affective commitment or attachment to the firm). During planned change, the quality of the employment relationship plays an important role in promoting employee acceptance and involvement (Pfeffer, 1994; Rousseau, 1996). Mishra (1996) notes that a key aspect of trust is one's perception of frankness and honesty on the part of the other: Undistorted communication reinforces trust, whereas lies and distortions decrease it. Moreover, McAllister (1995) maintains that a perception of reliability and dependability may be a precursor to developing beliefs that the relationship entails reciprocated care and concern. Using social accounts theory as a basis, we argued previously that employees are more likely to believe the social accounts offered by credible managers. High trust increases the likelihood that reasons for change will be viewed as functional—that is to say, well-intentioned and constructive. However, social accounts theory does not indicate the kind of causal reasoning low trust might generate. Motivated reasoning theory suggests that employees seek explanations for change whether or not a managerial account is offered. Moreover, this theory predicts that under conditions of low trust, attributions consistent with dysfunctional experiences (i.e., nonconstructive or at odds with the broader interests of the firm and its members) are likely. To understand the effects of motivated reasoning, we differentiate the legitimacy attributed to a reason for change (i.e., whether or not an explanation is considered adequate or appropriate in a specific situation) from its functionality. Although legitimacy is situation-specific, functionality is a generic aspect of reasons, reflecting the extent to which a reason generally can be construed to support constructive goals of the larger firm such as progress or development (e.g., quality, continuous improvement, organizational survival) or nonconstructive goals, such as self-serving interests of particular persons or groups (e.g., self-aggrandizing senior managers). We postulate that low trust can itself generate explanations in the minds of employees that differ from those offered by managers and reflect negative evaluations of managers'

WHAT'S A GOOD REASON TO CHANGE?

intent (e.g., the real reasons may be seen as political or self-serving). Hypothesis 4: Trust will be negatively related to beliefs in dysfunctional reasons for change.

Conversely, we made the following hypothesis: Hypothesis 5: Trust will be positively related to beliefs in functional reasons for change.

According to social accounts research, high trust should generate greater acceptance of managerially offered explanations. Moreover, motivated reasoning suggests that hightrust individuals would be more likely to consider reasons for change legitimate (i.e., acceptable and appropriate), whether or not managers offer those reasons. Hypothesis 6: Trust will be positively related to perceived legitimacy of reasons for change.

Note, however, that the premise behind this main effect of trust on perceived legitimacy differs from that of the interaction effect in H2. Because motivated reasoning describes how individuals generate explanations, reasons need not be directly communicated by a trusted source to be viewed as credible. However, where managers are trusted, that trust should promote acceptance of change generally and therefore enhance legitimacy of any reasons associated with it. Another relational factor that can shape interpretations of the reasons for change is the psychological contract, a fundamental aspect of the employment relationship. Psychological contracts refer to beliefs that the employee and employer hold regarding their mutual obligations. Changes can be interpreted in the context of these beliefs (Levinson, 1962; Rousseau, 1995), and the nature of employees' psychological contracts can affect their reasoning regarding change. Relational contracts entail mutual obligations between employee and employer to support each other's interests, typically through commitments regarding job security and loyalty, and involve an open-ended, long-term exchange with a high degree of flexibility. These contracts are believed to have a broader zone of acceptance (Rousseau, 1995), suggesting that parties to such contractual arrangements are more likely to be accepting of change and the reasons offered for it. In contrast, transactional contracts are characterized by monetary obligations on the part of the employer and performance of specific, often narrow, job duties on the part of the employee (Macneil, 1985; Rousseau, 1989). With their limited, largely economic involvement and lower levels of emotional attachment, transactional contracts have narrower flexibility; changes would require a renegotiation of the contract and possibly changes in the parties involved (e.g., replacing one person with another). Thus, transactional contacts are likely to be associated with lower acceptance of change and the reasons offered by managers.

517

. Hypothesis 8: The extent to which an employee is party to a relational contract will be positively related to his or her perceptions of the legitimacy of the reasons for change. Hypothesis 9: The extent to which an employee is party to a transactional contract will be negatively related to his or her perceptions of the legitimacy of the reasons for change.

Managerially driven changes are not necessarily composed of a single event or unified message. Rather, change recipients must interpret a complex array of facts regarding the change, their relations to the firm, and their coworkers' perceptions in understanding the underlying motivations for changes management has introduced. The reliance of previous social accounts research on laboratory settings might have limited its ability to assess the effects of these contextual features. However, we note that proponents of social accounts have addressed aspects of motivated reasoning by addressing why some people are more likely to believe an account in the first place (e.g., trust; Bies, 1987). Further, we suggest that these two perspectives are not necessarily incompatible. Because motivated reasoning theory articulates more specifically the role that organizational context can play in shaping how employees interpret significant events, joining motivated reasoning with social accounts theory can be especially fruitful in understanding complex organizational change. Implementation of Change The reasons motivating organizational change, both real and perceived, have received relatively little attention in research on the implementation of complex organizational interventions. We postulate that reasons, particularly their adequacy or legitimacy, play an important role in shaping the success of the implementation process. When organizational actions are interpreted in a broader normative framework that legitimates their motives, individual acceptance and commitment can be enhanced (Fairhurst & Sarr, 1996). Research on transformational leadership stresses the importance of appealing to normative values (Bass, 1985). Moreover, research on social accounts suggests that accounts are more likely to be perceived as adequate where managers use normative appeals and consequently do not appear to have self-serving motives (Bies, 1987; Shapiro, 1991). In consequence, we postulate that legitimacy of reasons is directly related to employee implementation of change: Hypothesis 9: Perceived legitimacy of reasons will be positively related to change implementation.

Another major factor that motivates change is the perception that benefits (as opposed to losses) can be accessed through implementation. We note that social accounts research has not addressed changes that introduce benefits (e.g., new skills, development opportunities, greater challenge). Indeed, to date, it has focused on negative expert-

518

ROUSSEAU AND TIJORIWALA

ences (e.g., job loss). Social accounts theory also typically focuses on discrete events, such as smoking policies and layoffs, rather than more complex events, such as culture change or restructuring, where some benefits may be expected in the long term. In complex organizational change, individuals may experience both losses (e.g., uncertainty, deskilling, job insecurity) and gains (e.g., greater autonomy, development opportunities). However, losses are likely to occur relatively early in the process (particularly psychological stresses associated with change and uncertainty; Rousseau, 1995, 1996). The more fully a change is implemented, the more likely it is that employees will access its benefits. Hypothesis 10: The degree of implementation will be positively related to employee experiences of beneficial consequences of change.

Method Participants Registered nurses (RNs) in 34 inpatient units at a large northeastern U.S. hospital were the participants of this study.(n = 501). The average age at the time of the survey was 33 years, 92% were women, and 85% worked full time. On average, the nurses had been employed with the hospital for 7 years and had been with their current unit for 5 years. The sample included nurses with a wide range of educational qualifications: 23% had RN diplomas, 25% had associate degrees, 46% had bachelor's degrees, 4% had master's degrees, and 2% had other qualifications. All nurses surveyed had been with the hospital throughout the entire period of the change. The hospital, part of an integrated academic health care system, had 746 licensed beds and more than 5,000 full-time employees. At this site, we investigated the implementation of self-managing work groups among nurses. At the time of the survey, the hospital had implemented a program to promote the empowerment of nurses by giving them responsibility and accountability for decision making about clinical practice (referred to here as the change). Although the change began 2 years before the survey, our interviews with nursing leadership suggested that nursing units, and the nurses within these units, experienced varying degrees of implementation. This variation in implementation provided an opportunity to investigate cross-sectionally the factors contributing to effective change management in a single organization. Nursing units were grouped into sets of two to five, to form a cluster headed by a nurse manager/clinical director. The average number of nurses in a single unit was 30. Each clinical director oversaw 60 to 150 nurses.

The Change The organizational intervention consisted of a complex restructuring of the nursing division of the hospital following a program referred to as Shared Governance (SG; Porter-O'Grady, 1984). SG replaces the traditional hierarchy—charge nurses on each shift, head nurses on each unit, nurse managers over clusters of units, and higher-level nursing administrators—with a streamlined, de-

centralized decision-making system. Under SG, charge nurses and head nurses were eliminated, and nurse managers coordinated the efforts of several units (typically four). An array of decisions formerly made by nursing leadership were pushed down to the nurses who provide direct patient care. Proponents of SG maintain that autonomy, authority, and control should be vested in the person who provides services so he or she can decide what to do, execute the action, and enforce those decisions (Porter-O'Grady, 1984).1 In SG, a council, or subset of nurses in each unit, makes decisions formerly made by nursing managers. The SG model includes four councils within each unit: quality (focusing on continuous improvement), practice (focusing on patient care procedures), education (identifying training and development needs), and research (supporting development of nursing research programs intended to ultimately yield publications by unit nurses). To integrate activities throughout the hospital, representatives from different units sit on hospital-wide councils in each of these areas. A management council composed of nurse managers, which reports directly to the vice president of nursing, is intended to be the primary link between the units and hospital administration. PorterO'Grady (1984, p. 84) acknowledges While investing power and accountability in the professional staff may arouse some concern from the traditional nursing administrator, this leader will be relieved to know the exercise of individual power in a shared framework is limited at best. We note that cost cutting by eliminating expensive layers of management also is reported as a purpose of SG (Porter-O'Grady, 1984). 1

To understand Shared Governance in the context of the nursing profession, it is helpful to consider current trends in that profession. Goodrick and Meindl (1995), in their investigation of ideological transformation in the nursing profession, characterize the trend toward expanding the managerial responsibilities of nurses and giving them greater autonomy over their own clinical practice in terms of a century-long evolution of nursing. The traditional ideology of nursing was that of a feminine vocation, consistent with maternal devotion, focusing on the womanly duty of service to others and submission to authority, including that of physicians. The renewal of the women's movement in the 20th century helped loosen the constraints of nursing as womanly duty by helping to legitimate a professional agenda (Bullough & Builough, 1984; Goodrick & Meindl, 1995). Fundamental shifts in health care have occurred in conjunction with a spillover of the managerial ideology from the corporate sector, with its goals of maximizing efficiency, into contemporary hospitals (e.g., "Exmanager describes the profit-driven life inside Columbia/HCA," 1997). However, in nursing the humanistic emphasis on patient care has often been experienced as inconsistent with increased standardization and fewer personal relationships implied by professionalization (Gadow, 1980) as well as greater emphasis on health care as a business. Shared Governance and the appeal to concern for patient care quality expressed in its implementation melds the humanistic and managerial ideologies, consistent with recent trends in nursing to find a middle ground between humanism and managerialism (Goodrick & Meindl, 1995).

WHAT'S A GOOD REASON TO CHANGE?

The Implementation At the outset, all nurses in the hospital received training in the SG model and were asked to volunteer to serve on councils within their units. When more than the requisite number volunteered, coworkers voted for their council representatives. Most of the head and charge nurses who were removed from their positions returned to the role of staff nurse. Nurse managers nominated nurses from within their clusters to participate on the four hospital-wide councils. At the time SG was introduced into the hospital, nursing leadership gave speeches, sent internal memoranda, and walked the halls to send a message that SG would promote nursing professionalism, improve care quality by empowering caregivers to make appropriate decisions without oversight, and enhance the skills and competence of nurses. As emphasized by the hospital's vice president of nursing, Continuous improvement of patient care is the point of Shared Governance. For this to happen, nurses need to think of themselves as professionals and let go of traditional ways of thinking about nursing. To make Shared Governance work, increased nursing professionalism is required. Note that the formal reason stated by nursing leadership for SG implementation was to improve the quality of patient care by enhancing nursing professionalism. Thus, improvement in care quality was the social account managers offered to justify the change. This conforms to what Sitkin and Bies referred to as an exonerating account—legitimating the motives for the change by placing management's actions within a broader normative framework. Thus, even negative consequences from the change (e.g., increased stress during the transition to SG, more work for the same pay) might be seen as derived from legitimate motives. In this context, according to Sitkin and Bies (1993), "anger and disapproval engendered... may be suppressed, in part because such feelings are not deemed appropriate given the legitimate purposes that ostensibly motivated the acts" (p. 356). SG had been initiated in this hospital 2 years prior to the survey conducted in this research. At this point in time, implementation across units within the nursing department varied widely, with some units having fully developed SG council decision systems and others having virtually no SG-related practices in place. The time lag between the implementation of the change and survey administration raises the issue of whether there was a mortality bias due to the attrition of employees dissatisfied with the change. However, during this period, voluntary attrition was under 2%, the latter due substantially to retirement. Thus, a mortality bias is unlikely.

Measurement Questionnaires were sent to RNs in the inpatient units who were listed in the Human Resource Department's database. Respondents filled out the questionnaire voluntarily, typically on the job. The unit of analysis in this study is the individual nurse. The total number of nurses participating reflects a response rate of 48.5%. Using a structured questionnaire, respondents were asked to give their perceptions of the change and to estimate its impact on their work life and the work practices in their units. Scales developed

519

for this study were subject to principal-axis factor analysis with varimax rotation. All scales were based on factors with eigenvalues greater than 1 and included those items meeting the requirements of simple structure (Comrey, 1973). The specific items used can be found in the Appendix. Reasons. The reasons respondents attributed to the change were measured using a 1 (not at all} to 5 (very great) scale. Respondents were asked to rate the extent to which they believed the managerial account (quality), economic factors, and selfserving managerial motives were the reason for the change. Reasons were derived from pilot interviews with staff nurses and managers prior to the development of the survey. Eleven items were used that yielded three factors. Alpha reliabilities were .81 (Quality), .71 (Economic), and .76 (Self-Serving). On the basis of information obtained from our pilot interviews and a content analysis of the reasons, Quality and Economic are treated as functional reasons for the purposes of hypothesis testing, and Self-Serving is treated as a dysfunctional reason for change. Legitimacy of reasons. Legitimacy was measured using the same items but with a different frame of reference (see Appendix) as described for Reasons above. Respondents were asked whether the reason was "legitimate, acceptable, and appropriate to you." The eleven items yielded the same factor structure for Legitimacy as for Reasons. Alpha reliabilities were .82 (Quality Legitimacy), .82 (Economic Legitimacy), and .75 (Self-Serving Legitimacy). Work-group level reasons. For each nurse, we computed the mean score of his or her coworkers' ratings for each of the three measures of reasons for change by summing the unit's total score, subtracting the nurse's own score, and taking the average. This coworjcer average was used as a measure of the social construction of the reasons within each unit. To determine whether aggregation was appropriate, we ran a Within and Between Analysis (WABA; Dansereau, Alutto, & Yammarino, 1984). Results indicated that the effects each Reasons measure had on other variables included in the study were both between as well as within the nursing units. That finding was consistent with the dual-role perceived reasons plays in this study, both as an individual perception as well as a work-group level variable. Psychological contracts. We examined nurses' beliefs about their obligations to their employer by asking them the extent to which they had committed to engage in specific behaviors as part of their employment with the hospital, using a 1 (not at all) to 5 (to a very great extent) scale. The obligations surveyed were derived from the literature on nursing roles (e.g., Gadow, 1980) and field interviews with staff nurses and nurse managers in this hospital. Principal axes factor analyses revealed two factors with eigenvalues greater than 1. The patterns observed here were consistent with previous research in a non-nursing population (Robinson & Rousseau, 1994). Scales are based on items with factor loadings exceeding .4. Relational contract (a = .82). Consistent with previous research on relational contracts reporting open-ended conditions of employment without clear performance specifications (Rousseau, 1989), this factor reflects obligations focusing on attachment to the employer, including loyalty, continuity, and willingness to accept new assignments. Transactional contract (a = .83). Reflects obligations to provide a basic level of performance with limited involvement with

520

ROUSSEAU AND TIJORIWALA

the organization (Rousseau, 1995). Because nurses are professionals, educated in nursing skills and codes of conduct, a transactional contract consists of adherence to basic professional standards that would apply to any organization in which the nurse might work. Trust. We measured the extent to which respondents experienced nursing leadership as trustworthy and credible regarding this particular change by using two items assessing the credibility of nursing administration and the nurse managers within each unit, using a 1 (not at all) to 5 (very highly) scale. The alpha reliability was .80. Implementation. The goals that hospital leadership established for this change included improvements in professional and leadership development, nursing decision making, and clinical practice quality. Nurses were asked to compare the outcomes before and after implementation of the change. Six items measuring implementation yielded a single factor (a = .87). Individual assessments of implementation are appropriate in this context because the change was intended to alter the day-to-day experience of the staff nurses within their unit. To evaluate the convergent validity of the implementation measures, three raters who were blind to the nursing units' questionnaire results scored each unit on its degree of implementation, using a scale from 1 to 10. The raters discussed their scores to reconcile any differences and generated a global implementation score for each unit (ranging from 1 to 8, with a median of 4). This global score was compared with the mean perceived implementation score for the nurses in each unit, producing a zero-order correlation between the global and individual implementation measures of .46 (p < .001). This validity coefficient suggests good convergence between individual nurse perceptions and the level of implementation within each unit. As an additional validity check, individual participation on hospital-wide councils (obtained from archival records) was dummy-coded. Council participation correlated significantly (r — .16, p < .001) with individual nurses' assessments of change implementation. Consequences. To examine the perceived benefits or losses nurses reported as a result of the change, we used two indicators: quality of care (a = .71) and employability (a = .67).

Analysis We used interaction effects by using hierarchical moderated multiple regression with significance tests based on difference in R2 with and without the interaction term. We tested hypotheses regarding direct effects by using a path analytic approach based on maximum likelihood estimation (MLE). Three separate path models were estimated, one for each of the three reasons (quality, economics, self-serving) studied here. Because of the complex set of factors postulated to affect the judgment processes of change recipients, the path model tested the effects of reasons on legitimacy, controlling for contextual factors. Effects of contextual factors on legitimacy were examined controlling for actual reasons. Effects of legitimacy on implementation were examined controlling for contextual factors and perceived reasons. Lastly, effects of implementation on consequences were examined controlling for all other variables in the model. This approach, consistent with Baron and Kenny (1986), makes possible a robust test of the hypotheses. We used MLE in the path analysis because of

the likelihood that correlated errors existed in our questionnairebased independent variables (Breen, 1996).

Results Table 1 presents means, standard deviations, and zeroorder correlations among measures used in this study. Note that matched t tests contrasting the perceived reasons for the change demonstrate significant differences (p < .001) among all three. Economic factors, on average, are seen as the strongest actual reasons for the change, whereas quality reasons are the lowest rated, despite the fact that management offered quality improvement as its social account for the change. However, quality and economic factors were rated similarly in terms of legitimacy with each receiving a moderate rating (2.75 and 2.76, respectively). As expected, self-serving reasons were rated significantly lower (p < .001, M = 2.09) on legitimacy than either quality or economic reasons. Hypothesis 1 predicted that quality reasons—the social account offered for the change—would be positively related to the perceived legitimacy of quality as a reason for the change. This hypothesis is supported with both a significant zero-order correlation (r = .50, p < .01) and a path coefficient of .49 (p < .01; Table 2). Note that the correlations between economic reasons and self-serving reasons and their respective legitimacy measures are positive but significantly lower. This suggests that managerial accounts accepted as true by change recipients are more likely to be seen as legitimate than other interpretations of the change. Hypothesis 2 tested the moderating role of trust in the relationship between reasons and legitimacy, as predicted by social accounts theory (Table 3). High trust levels are associated with a stronger relationship between reasons and legitimacy, but only for the managerial account. A moderating effect was found for quality reasons—]8 = .40, p < .05, where the difference between R2b and R2a was significant, F(3,431) = 4.62, p < .01—but not for economic (/3 = .15, ns) or self-serving reasons (fi — .25, ns). We now turn to the hypotheses derived from the motivated reasoning framework (Figures 1-3). Hypothesis 3 predicted that work-group beliefs regarding reasons would be positively related to individual beliefs. A significant relationship is found for quality reasons (/3 = . 11, p < .05), and for self-serving reasons (j3 = .07, p < .05), but there is no relationship for economic reasons (|3 = —.01, ns). Trust was postulated to play a role in explaining perceived reasons for change. Hypothesis 4 postulated that trust would be negatively related to beliefs in dysfunctional (i.e., self-serving) reasons for change, which is supported (j3 = — .22, p < .01). Hypothesis 5 predicted that trust would be positively related to both economic and quality reasons, which also is supported: (quality: j3 = .35, p < .01; economic: 0 = .1 l,p < .01). Hypothesis 6 postulated that trust

521

WHAT'S A GOOD REASON TO CHANGE? * *

would have a positive relationship with perceived legitimacy of reasons for change. This prediction was supported for all three reasons: economic (/3 = .29, p < .01), selfserving ()3 = .15, p < .01), and quality reasons (j3 = .08, p < .05). Hypothesis 7 predicted that relational contracts would be positively related to perceptions of legitimacy, a finding supported for both self-serving (j3 = .21, p < .01) and economic reasons (/3 = .18, p < .01) but not for quality, £ = —.08, ns). Thus, relational contracts appear to make employees more accepting of the legitimacy of change, at least for those reasons that do not directly originate from formal managerial communication. In contrast, transactional contracts were predicted in Hypothesis 8 to have a negative relationship with perceived legitimacy of change, a finding supported for self-serving reasons (0 = -.12, p < .05) only. This finding suggests that transactional contracts are linked to low levels of tolerance for poorly justified change, perhaps because of the lower trust and flexibility associated with these contracts (Rousseau, 1995). Surprisingly, transactional contracts are positively related to legitimacy of quality reasons (|3 = .07, p < .05). The basis for this positive relationship is unclear. However, given that the transactional contract characterizing nurses in this study reflects meeting basic professional standards, without specific obligations to the particular employer, it is likely that transactional contract's relationship with the legitimacy of quality reasons derives from adherence to standards of nursing professionalism (Goodrick & Meindl, 1995). Hypothesis 9 predicted that the perceived legitimacy of reasons would be positively related to implementation, which is supported for all three types of reasons (p < .01). Finally, consistent with Hypothesis 10, implementation was positively related to improvements in quality of care and employability (p < .01) in all three models.

CN CO

* * m

* * ^O

OJ

in pp B co

* *

* * CO

^ *

o

* *

* *

CO

*

\O ON —« p

* * * * O ON

en en —<

CO

1 * * * * *

* * * * # *

CN -H \' i"

CN CN ^ l " i ' \'

•* o i2

&

CO C^-

* * *

*

m co O


•X-

*

o

O —<

i

^

* * o m

*

*

O

p f l'

^~~

* OX

~&

* * en

v-j

en en oo l" * * 1/1 --

^ *

si 1

0\

^r

—< o

o p

m TJpp

* # * CN ^

1

1

* *

* o\ oo

* * CN

—< ,-i

o

o

—i

I

* *

r-<

0

CN CN

0

a^n

en -^

OX —' -^ OO en CN

en O O en CNen O — i —< O O o

* p

CO CN

r

O ^0 p p

* O

O

2o

* * * * * * CO OO 9 CN CN

#•

* *

*

CO

1

I> ON 0 0

Discussion * CO O

en o

o •—

S i

r- co -3ooo

CN ON

o

CN

-~

O

O —

—

t

O CN CN

*

*

r~~ o eN NOo —

O

*

o

1

wo —< p 8p

* * * * * * CN CN CO

* * * xo en O en —<

ss o

•^

—;

i

XD O 1

O

^* ** ** m CO CO

p O p ]' |'

^

* * en

fr *

NO CN

—i

p p i"

* * rco

1

=

-^ 00 - 0

1—1

r- ^tNO

CO

oo

— O

Q

u

<

CN CN

0

W3 C ">

S3 "o C/5

ill

o

/

CO Tf ^ ON NO TT CN CN CN

CN

Self-serving

WO NO

ge . Implementation , Quality of care . Employability

m

Quality Economic

:edents Transactional Relational Trust Group quality Group economic Group self-serving

Variable

riptive Statistics y,

o\ r- c- p

O

CN en CO

XO — m r*

imacy of reasons

xo o m
^

o •-< CN "S j^ _J u

•<*• uo

p V 0 *

i! -Si ^ ^*

This study adopted a multifaceted view of the process whereby individual, relational, and social factors influence how recipients interpret complex change and the social accounts that are offered regarding it. To a great extent, motivated reasoning and social accounts seek to explain different phenomena. Although motivated reasoning seeks to explain how an individual arrives at a particular causal interpretation, social accounts research focuses on how a particular source of information regarding reasons, typically the managers involved in implementing change, shapes perceptions of adequacy or legitimacy of reasons. However, in the quest to understand the contextual factors shaping beliefs regarding social accounts, researchers have moved in the direction of motivated reasoning by examining such factors as employee trust in management (e.g., Bies, 1987).

522

ROUSSEAU AND TIJORIWALA

Table 2 Maximum Likelihood Estimates for Different Independent variable Quality reasons Transactional Relational Trust Group quality reason Quality reason Legitimacy Implementation Function of log likelihood = 1,621.59 Economic reasons Transactional Relational Trust Group economic reason Economic reason Legitimacy Implementation Function of log likelihood = 749.15 Self-serving reasons Transactional Relational Trust Group self-serving reason Self-serving reason Legitimacy Implementation Function of log likelihood = 1,531.68

Types of Reasons Reason

Legitimacy

-.04 .11** .35** .11**

.07* -.08 .08* .49**

.10* .04 .11** .03

Quality of care

Employability

.35** .18**

.10** .05 .58**

.07 .04 .36**

-.09* .27**

-.15** .05 .63**

-.04 .01 .40**

29** .12**

-.13** .06* .61**

-.08* .03 .37**

.01 .18** .29** .23**

.05 .02 -.22** .07*

Implementation

-.12** .21** .15** .20**

* p < .05. ** p < .01 (one-tailed).

Table 3 Summary of Hierarchical Regression Analysis for Legitimacy of Different Types of Reasons

R2

Independent variable

A/?2

Legitimacy of quality reasons Quality reason Trust Step 1 Interaction Step 2

.25* -.15

.28** .40*

.29**

.01**

Legitimacy of economic reasons Economic reason Trust Step 1 Interaction Step 2

.18 .18

.182 .15

.183

.001

Legitimacy of self-serving reasons Dysfunctional reason Trust Step 1 Interaction Step 2

.01 -.03 .05 .25 .06

.01

Note. A^?2 indicates the difference between hierarchical regression /?2 values when the interaction term is included (Step 2) and when it is excluded (Step 1). *p < .05. **p < .01.

We suggest that these perspectives can be effectively combined to better understand reactions to planned organizational change. By extending social accounts theory to better understand how managerial accounts function in complex organizational change, this study demonstrates that the context in which managerial accounts are offered can have powerful effects not only on the account's legitimacy (the traditional dependent variable in social accounts research) but also on the successful implementation of change. Moreover, it demonstrates that managerial accounts do not provide the whole picture in terms of the employee's understanding of the reasons for change. Consistent with motivated reasoning theory (Ajzen, 1985; McGill, 1989), change recipients do in fact form alternative explanations for change as a function of their trust in management, the psychological contract characterizing their relationship to the firm, and the beliefs held by their coworkers. Our results suggest that the managerial account was more effective at justifying change than the two other causal interpretations employees deduced, a finding consistent with the theory of social accounts (Bies & Moag, 1987; Sitkin & Bies, 1993). However, this study also demonstrates that managerial accounts are not always received or remembered as managers intend. As one nurse relates her experience, "Maybe the administration could have had a meeting

523

WHAT'S A GOOD REASON TO CHANGE?

.43

Figure 1. Maximum likelihood estimation for quality reasons. Double-headed arrows are correlations. *p < .05. **p<.OL ***p < .001 (one-tailed).

with staff to discuss/explain the hospital's rationale for going for Shared Governance. (But) one and half years ago, it was sprung upon us suddenly (and) no one was prepared." Because complex change involves a variety of events, a social account of change is not necessarily a one-time thing. Analysis of data regarding hospital-wide council participation (not presented here) indicated that nurses involved in the councils were much more likely to believe that the

change was associated with the managerial account, the need to improve care quality, than were nonparticipating peers because that topic was the repeated focus of council discussions. As one nurse who served on the quality council reported: "After being on quality council, I felt differently about myself, knew more about what was happening in the hospital, and was better able to persuade ... why we should make a change in our unit. I felt people

Figure 2. Maximum likelihood estimation for economic reasons. Double-headed arrows are correlations. *p < .05. ** p < .01. *** p < .001 (one-tailed).

524

ROUSSEAU AND TIJORIWALA

Figure 3. Maximum likelihood estimation for dysfunctional reasons. Double-headed arrows are correlations. *p < .05. **p<.01. *** p < .001 (one-tailed).

listened to me better. It changed the way I thought about myself as a nurse." To a great extent, research on change has focused on management's perspective. Blau (1996) has cogently argued for the existence of distinct frames of reference for evaluating organizational actions and accountabilities—and a managerial perspective is only one of many. In a sense, the present study reframes the focus of research on social accounts to address the individual employee's role as an active participant in change. As the change studied here demonstrates, the causal accounts managers offer are not always accepted and may not even be received in the way they are intended. Moreover, reasons for complex organizational change can be difficult for managers to explain and for change recipients to understand. (One nurse said she believed that the change toward self-managing work groups was a form of "communism."2) In a sense, all organizational transformations risk a crisis of legitimacy. Change can alter core features of the organization and its relations with employees (Trice & Beyer, 1986), creating a sense of both confusion and loss (Bridges, 1991) even when management intends for employees to benefit from the change. It is noteworthy that, in the organization we studied, economic motives were the most frequently attributed reason for change. However, despite the fact that economic reasons were rated to be as legitimate as quality improvement, economic reasons had comparatively weaker effects on implementation. This finding raises the question of whether economic factors alone are sufficient motivators of change for non-managerial employees. It is also possible

that economic reasons had a weaker effect on implementation because they were not the social account offered by management. It is plausible that reasons for change outside the managerial account might lack sufficient supporting information to effectively promote change. However, economic reasons might also be less motivating than quality reasons because economic factors, though perceived to be relevant and legitimate, carry with them the connotation of adverse consequences such as loss of job security. (We are unable to test for the connotations of losses and gains associated with the reasons for change in the present study because we measure actual consequences rather than anticipated ones.) Unless organizations implementing change in response to 2

Goodrick and Meindl (1995) maintain that despite resistance there has been a melding of managerial and humanistic ideologies in nursing to bridge two otherwise contradictory belief systems (humanistic concern and economic values). This evolution reflects a larger trend in the United States, with the increased popularity of business-management concepts extending the assumptions of private business firms into other organizations, including governments and nonprofits such as churches. However, such evolutionary processes are marked by resistance against revising the existing ideological order. In nursing, there has been resistance to the shifting emphasis on economic factors in clinical decision making (Bullough & Bullough, 1984), making it more likely that processes of social construction and resulting group-level effects on individual nurses would focus on the ideologically consistent message of quality improvement.

WHAT'S A GOOD REASON TO CHANGE?

economic factors make an effort to address the benefits workers can expect to receive, employees might be more anxious about such changes, particularly in settings such as hospitals where economic factors were not traditionally evoked to motivate change.3 In the case of this study's nurse population, quality improvement was consistent with the humanitarian core values of the nursing profession (Goodrick & Meindl, 1995)—and perhaps was an easier, more culturally compatible account for managers to offer. This study raises the question of how best to provide information about the reasons for change. Given the central role played by trust, consistent information from credible sources over time appears to be key. Motivated reasoning is particularly pertinent to complex organizational change because complex interventions are more likely to engender active employee information seeking than are simpler interventions such as discrete policy changes (e.g., introducing nonsmoking policies; Babocel, McCline, & Folger, 1997). During complex change, the individuals' need for information typically increases, whereas both the quantity and quality of available information often decline (Bridges, 1991; Rousseau, 1996). This information gap evokes active employee sensemaking (Weick, 1995); that is, greater efforts to gather information and interpret events, typically through reliance on informal sources, observation, and recall. Thus, change recipients are likely to rely on a variety of information sources to make sense of what is happening, leading to more complex understandings of any managerial accounts provided as well as personally generated alternative explanations. The role of work-group beliefs in shaping individual attributions suggests that group-level interventions (e.g., meetings within each unit to discuss the reasons behind the change and its implementation) may be important to reinforce organization-wide change efforts. Interestingly, the nature of the employment relationship plays a role in shaping whether the managerially offered account is received and believed, as well as in legitimating that account. Trust can operate here in two ways: first, by promoting a willingness to pay attention to managerial communications so the message is received, and second, increasing the believability of explanations that otherwise might appear to be imprecise, unclear, or confusing. Management credibility, on the basis of a history of good-faith relations with employees, is likely to engender a positive response to change and the perception that reasons offered are legitimate (Kramer, 1996). Although distrust tends to promote monitoring, the informal seeking of information to discredit managerial accounts (Clark & Reis, 1988), high trust is likely to be associated with acceptance of information provided by management and a reduction of information-gathering efforts. Our findings support the argument that high trust creates a broad zone of acceptance of the exigencies of complex organizational change, whereas

525

low trust necessitates greater care in the implementation process (Rousseau, 1996). Psychological contracts play a particularly important role in influencing legitimacy in the eyes of employees. During change, transactional contracts can lead to a greater focus on the monetary nature of the exchange and less willingness to accept new responsibilities without increased compensation. In the words of one nurse with a transactional contract, "The hospital takes advantage of staff—we take on more responsibility and workload with no compensation or reward for all the hard work." In contrast, a relational contract reflects other-oriented and needs-based thinking, where a broad array of inducements and contributions form the exchange. As one nurse with a highly relational contract indicated, "The individual employee (has) more opportunity to develop his or her professional goals, which leads to growth in organization and client satisfaction." Employee psychological contracts with a strong relational quality have been theorized to create a greater zone of acceptance, that is, a willingness to accommodate to needs and requests expressed by the employer (Rousseau, 1995; Shapiro et al., 1995). Promoting such a broad zone of acceptance is likely to be affected by organizational conditions that strengthen the employment relationship, such as perceived organizational support (Eisenberger, Huntington, Hutchinson, & Sowa, 1986) and a history of good faith and fair dealing (Rousseau, 1996). Lastly, reasons for change are likely to play a greater role in motivating change in its early phases. Once implementation is underway, the extent to which the change takes hold influences whether people experience its intended benefits. Although several of the work units studied here were still in the earliest stages of implementation after 2 years, our results demonstrate that the extent of implementation is positively related to perceived gains from change, specifically improvements in patient care quality and in nursing employ ability. Because implementation is a matter of degree, studies of planned organizational change might do well to consider the extent to which change recipients have accessed gains (or losses) from the change as one factor affecting responses to the change's later phases. 3 Using scenarios based on the facts of the organizational setting studied here, a follow-up study using masters-level students in business and health care management contrasted the impact of economic versus quality reasons on perceptions of losses and gains (Rousseau & Tijoriwala, 1997). Results indicate that economic reasons are associated with greater anticipated losses from change and quality reasons are associated with greater anticipated gains. Thus, we infer that at the time changes are introduced, economic reasons might tend to signal to change recipients that they stand to lose something (e.g., job security) as a result of the change, whereas quality reasons signal potential improvements and benefits (e.g., improved customer satisfaction or patient care).

526

ROUSSEAU AND TIJORIWALA

There are several methodological concerns with regard to this study. First, a one-time questionnaire was used to gather most of the data. Although path analytic methods were used to analyze the data, longitudinal data would have provided a more robust test of causation. Second, the alpha reliability of one measure of benefits, employability, was .67, below the conventional standard of .70. We included this measure in our analyses because research on this important variable is absent, despite the widespread concern that employees and the public have regarding the impact of organizational changes on job security or loss. We note that this variable enters into only one of the ten hypotheses, and its results are significant. Third, the reasons employees attributed to the change, although plausible in a variety of organizational settings, do not fully represent the domain of potential reasons that change recipients in other organizations might attribute. The three reasons used here were identified by extensive interviews with nurses prior to the construction and administration of the questionnaire. Thus, the reasons are applicable in this setting, and the scales developed to assess them may be idiosyncratic to this organization. However, grounded in the context of this organization, the reasons measured here do yield results largely consistent with motivated reasoning and social accounts theory. A final concern we raise is the use of questionnaire data for all major variables studied. Although our results do show fairly high discriminant validity across different measures, relational factors, in particular trust, play a central role in most observed relationships. For this reason, we have used statistical tests that reduce some of the biases associated with multicollinearity (e.g., use of differences in our R2s to test for moderator effects).

Conclusion The causal interpretations employees rely on to understand change reflect a system of beliefs regarding their relationship with the organization and its management. The social influence of coworkers also plays a role in shaping how change is interpreted. For highly committed employees who trust management and are party to strong relational contracts, managerial accounts may be enough to legitimate change and begin the process of motivating employee participation. However, in complex organizational change, managerial accounts are not necessarily always heard, let alone understood. Our findings suggest that employees hear different messages from management depending on the nature of their relationship with the organization. Although managerial accounts are often needed to motivate change, such accounts may not be sufficient to directly alter behavior for workers low on trust or in settings where their coworkers do not believe management. In such circumstances, it may be critical to provide sufficient justification for change in the form of demonstrable benefits and valued

consequences over time. Although a strong employment relationship is a predictor of implementation success, it can also be an outcome of effectively implemented change.

References Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhn & J. Bechmann (Eds.), Action control: From cognition to behavior (pp. 11-39). Berlin, Germany: Springer-Verlag. Babocel, D. R., McCline, R. L., & Folger, R. (1997). Letting them down gently: Conceptual advances in explaining controversial organizational policies. In C. L. Cooper & D. M. Rousseau (Eds.), Trends in organizational behavior (Vol. 4, pp. 73-87). London: Wiley. Baron, R., & Kenny, D. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual strategies and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press. Bies, R. J. (1987). The predicament of injustice: The management of moral outrage. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 9, pp. 289-319). Greenwich, CT: JAI Press. Bies, R. J., & Moag, J. S. (1986). Interactional justice: Communication criteria of fairness. In M. H. Bazerman, R. Lewicki, & B. Sheppard (Eds.), Research on negotiations in organizations (Vol. 1, pp. 43-55). Greenwich, CT: JAI Press. Bies, R; J., & Shapiro, D. L. (1993). Interactional fairness judgements: The influence of causal accounts. Social Justice Research, 1, 199-218. Bies, R. J., Shapiro, D. L., & Cummings, L. L. (1988). Causal accounts and managing organizational conflict: Is it enough to say it's not my fault. Communication Research, 15, 381-399. Blau, J. R. (1996). Organizations as overlapping jurisdictions: Restoring reason in organization accounts. Administrative Science Quarterly, 41, 172-179. Breen, R. (1996). Regression models: Censored, sample selected or truncated data. [Quantitative Applications in the Social Sciences Series No. 07-111]. Thousand Oaks, CA: Sage. Bridges, W. (1991). Managing transitions: Making the most of change. Reading, MA: Addison-Wesley. Bullough, V. L., & Bullough, B. (1984). The politics of nursing. East Norwalk, CT: Appleton-Century-Crofts. Clark, M. S., & Reis, H. T. (1988). Interpersonal processes in close relationships. Annual Review of Psychology, 39, 609-672. Comrey, A. L. (1973). An introduction to factor analysis. New York: Academic Press. Coser, L. A. (1956). The function of social conflict. New York: Free Press. Dansereau, F., Alutto, J. A., & Yammarino, F. J. (1984). Theory testing in organizational behavior: The varient approach. Englewood Cliffs, NJ: Prentice Hall. Einhorn, H. J., & Hogarth, R. M. (1986). Judging probable cause. Psychological Bulletin, 99, 3-19. Eisenberger, R., Huntington, R., Hutchinson, S., & Sowa, D. (1986). Perceived organizational support. Journal of Applied Psychology, 71, 500-507.

WHAT'S A GOOD REASON TO CHANGE? Ex-manager describes the profit-driven life inside Columbia/HCA (1997, May 30). Wall Street Journal, pp. Al, A6. Fairhurst, G. T., & Sarr, R. A. (1996). The an of framing: Managing the language of leadership. San Francisco: Jossey-Bass. Gadow, S. (1980). Existential advocacy: Philosophical foundation of nursing. In S. F. Spicker & S. Gadow (Eds.), Nursing: Images and ideals (pp. 79-101). New York: Springer. Goodrick, E., & Meindl, J. R. (1995, August). Evolutionary bridging: The ideological transformation of the nursing profession. Paper presented at the meeting of the Academy of Management, Vancouver, British Columbia, Canada. Greenberg, J. (1994). Using social fair treatment to promote acceptance of a work site smoking ban. Journal of Applied Psychology, 79, 288-301. Handy, C. (1989). The age of unreason. Cambridge, MA: Harvard Business School Press. Harrison, B. (1994). Lean and mean: The changing landscape of corporate power in the age of flexibility. New York: Basic Books. Kanter, R. M. (1989). When giants learn to dance. New York: Simon & Schuster. Krackhardt, D., & Hanson, J. R. (1993, July-August). Informal networks: The company behind the chart. Harvard Business Review, 71, 104-111. Kramer, R. M. (1996). Divergent realities and convergent disappointments in the hierarchic relation: Trust and the intuitive auditor at work. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 216-245). Thousand Oaks, CA: Sage. Levinson, H. (1962). Organizational diagnosis. Cambridge, MA: Harvard University Press. Macneil, I. R. (1985). Relational contract: What we do and do not know. Wisconsin Law Review, 483-525. McAllister, D. J. (1995). Affect- and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38, 24-59. McGill, A. L. (1989). Context effects in judgments of causation. Journal of Personality and Social Psychology, 57, 189-200. McGill, A. L. (1995). American and Thai managers' explanations for poor company performance: Role of perspective and culture in causal selection. Organizational Behavior and Human Decision Processes, 61, 16-27. Mirvis, P. H., & Hall, D. T. (1994). Psychological success and the boundaryless career. Journal of Organizational Behavior, 15, 365-380. Mishra, A. K. (1996). Organizational responses to crisis: The centrality of trust. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 261287). Thousand Oaks, CA: Sage. Novelli, L., Kirkman, B. L., & Shapiro, D. L. (1995). Effective implementation of organizational change: An organizational justice perspective. In C. L. Cooper & D. M. Rousseau (Eds.), Trends in organizational behavior (Vol. 2, pp. 15-36). Chichester, England: Wiley. Pfeffer, J. (1994). Competitive advantage through people: Unleashing the power of the workforce. Boston: Harvard Business School Press.

527

Pondy, L. R. (1978). Leadership is a language game. In M. W. McCall & M. M. Lombardo (Eds.), Leadership: Where else can we go? Durham, NC: Duke University Press. Porter, M. E. (1990). The competitive advantage of nations. New York: Free Press. Porter-O'Grady, T. (1984). Shared governance for nursing: A creative approach to professional accountability. Rockville, MD: Aspen Systems. Porter-O'Grady, T. (1992). Implementing shared governance: Creating professional organizations. St. Louis, MO: Mosby. Reichheld, F. F. (1996). The loyalty effect: The hidden force behind growth, profits, and lasting value. Boston: Harvard Business School Press. Robinson, S. L., & Rousseau, D. M. (1994). Violating the psychological contract: Not the exception but the norm. Journal of Organizational Behavior, 15, 245-259. Rousseau, D. M. (1989). Psychological and implied contracts in organizations. Employee Rights and Responsibilities Journal, 2, 121-139. Rousseau, D. M. (1995). Psychological contracts in organizations: Understanding written and unwritten agreements. Newbury Park, CA: Sage. Rousseau, D. M. (1996). Changing the deal while keeping the people. Academy of Management Executive, 10, 50-61. Rousseau, D. M., & Tijoriwala, S. A. (1997). Reasons for change and the attribution of loss. Pittsburgh, PA: Carnegie Mellon University, Heinz School of Public Policy. Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23, 224-253. Schneider, B., & Bowen, D. E. (1995). Winning the service game. Cambridge, MA: Harvard Business School Press. Shapiro, D. L. (1991). The effects of explanations on negative reactions to deceit. Administrative Science Quarterly, 36, 614-630. Shapiro, D. L., Buttner, E. H., & Barry, B. (1994). Explanations: What factors enhance their perceived adequacy? Organizational and Human Decision Processes, 58, 346-368. Shapiro, D. L., Lewicki, R., & Devine, P. (1995). When do employees choose deceptive tactics to stop unwanted organizational change? A relational perspective. In R. J. Bies, R. J. Lewicki, & B. Sheppard (Eds.), Research on negotiations in organizations (Vol. 5, pp. 155-183). Greenwich, CT: JAI Press. Shortell, S. M., Morrison, E. M., & Friedman, B. (1990). Strategic choices for America's hospitals: Managing change in turbulent times. San Francisco: Jossey-Bass. Sitkin, S. B., & Bies, R. J. (1993). Social accounts in conflict situations: Using explanations to manage conflict. Human Relations, 46, 349-370. Trice, H. M., & Beyer, J. M. (1986). Charisma and its routinization in two social movement organizations. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior (Vol. 8, pp. 113-164). Greenwich, CT: JAI Press. Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage. Weiner, B. (1985). "Spontaneous" causal thinking. Psychological Bulletin, 97, 74-84.

528

ROUSSEAU AND TIJORIWALA

Appendix Questionnaire Scales Reasons and Legitimacy Measures "We would like to ask you two questions regarding the reasons for change at . First, in your opinion, what are the reasons why is implementing Shared Governance? Second to what extent do you think these reasons are legitimate, that is appropriate and acceptable to you? Please go down the list twice. First answer the question in terms of whether each item is a reason for the change to Shared Governance at . Then, go down the list again and answer the question in terms of whether the reason is legitimate, acceptable, and appropriate to you."

Travel if necessary Accept a transfer to another geographic location

Transactional Contract Meet minimal acceptable standards for performance Perform your job in a reliable manner Behave in a professional manner Protect confidential information Dress in a professional manner Follow hospital's norms and policies

Quality Improve quality of work life Improve quality of care Increase nursing professionalism

Economic Respond to changes in government regulation Increase organizational profitability Cut costs Adjust to greater competition Keep the hospital from losing money

Self-Serving

Trust "How credible or believable is the information each of the following provides to you regarding the change toward Shared Governance?" Nursing administration Clinical directors

Actual Changes "What is the impact, that is, the gains and losses of Shared Governance, to you personally? Compare your present situation to what it was before Shared Governance was introduced."

Quality of Care

Get more work out of nurses for the same pay For political reasons Following current management fads

Performance quality Ability..to provide quality care

Employability Psychological Contract Measures "Let's consider the promises you have made to . In exchange for its commitments to you, you have made certain promises. To what extent have you promised, implicitly or explicitly, to provide each of the following? (Indicate your answer using the 1—5 scale). Keep in mind we are not asking what you should provide or what you believe prefers. Instead, we are interested in what you actually promised to provide."

Relational Contract Stay with hospital a minimum of two years after accepting job Not look for a job elsewhere Not support hospital's competitors Accept a job change within your unit Accept an internal transfer if necessary

Job security Employability throughout one's career

Implementation Skills used in the job Professional status Personal development Leadership development Control over important aspects of my work Understanding my role as a nurse

Received September 16, 1997 Revision received September 24, 1998 Accepted September 28, 1998 •