D7 Culture

* Academy of Management Beview 2000, Vol. 25, No. 4, 783-794.

RETHINKING RESISTANCE AND RECOGNIZING AMBIVALENCE: A MULTIDIMENSIONAL

VIEW OF ATTITUDES TOWARD AN ORGANIZATIONAL CHANGE

SANDY KRISTIN PIDERIT Case Western Reserve University

In this article I review studies of resistance to change and advocate new research based on a reconceptualization of individual responses to change as multidimen- sional attitudes. A challenging question for research and practice arises: How can we balance the organizational need to ioster ambivalent attitudes toward change and the individual need to minimize the potentially debilitating effects of ambivalence? I conclude by highlighting the importance of examining the evolution of employee responses to change over time and the need to understand responses to change proposals that emerge from bottom-up, egalitarian change processes.

Adapting to changing goals and demands has been a timeless challenge for organizations, but the task seems to have become even more cru- cial in the past decade. In the for-profit sector, global population growth and political shifts have opened new markets for products and ser- vices at a dizzying pace. To respond to the pace of change, organizations are adopting flatter, more agile structures and more empowering, team-oriented cultures. As status differences erode, some employees are coming to expect involvement in decisions about organizational change. Successful organizational adaptation is increasingly reliant on generating employee support and enthusiasm for proposed changes, rather than merely overcoming resistance.

The concept of resistance to change has been widely studied, but it has limitations. Both Mer- ron (1993) and Dent and Goldberg (1999) have argued for retiring the phrase "resistance to change." The limitations of the concept can be framed in philosophical terms; for instance, crit- ical theorists and labor policy scholars argue that the interests of managers should not be privileged over the interests of workers when studying organizational change (Jermier, Knights, & Nord, 1994). Alternatively, the limita- tions of the concept can be framed in practical

I gratefully acknowledge the comments of Richard Bagozzi, David Deeds, Jane Dutton, Loren Dyck, Phoebe Ells- worth, Eric Neilsen, Mary Grace Neville, Janet Weiss, and the reviewers and special issue editor on earlier versions of this work.

terms; for instance, practical scholars and schol- arly practitioners argue that the concept might have outlived its usefulness (Dent & Goldberg, 1999; Krantz, 1999). My purpose here is to sum- marize a critique of existing views of resistance to change and to advocate a view that captures more of the complexity of individuals' responses to proposed organizational changes.

In the first part of the article, I suggest that in studies of resistance to change, researchers have largely overlooked the potentially positive intentions that may motivate negative re- sponses to change. I also show how studies of resistance have dichotomized responses to change and, thus, somewhat oversimplified them. Furthermore, I argue that varied empha- ses in the conceptualization of resistance have slipped into the literature, blurring our sense of the complexities of the phenomenon.

In the second part of the article, I propose a multidimensional view of responses to proposed organizational changes, capturing employee re- sponses along at least three dimensions (emo- tional, cognitive, and intentional). Within this view, "resistance to a change" is represented by the set of responses to change that are negative along all three dimensions, and "support for a change" is represented by the set of responses that are positive along all three dimensions. Responses to a change initiative that are neither consistently negative nor consistently positive, which were previously ignored but are poten- tially the most prevalent type of initial response.

783

784 Academy of Management Review October

can be analyzed as cross-dimension ambiva- lence in employees' responses to change.

In the third part of the article, I identify the implications of this new view for both research and practice. By highlighting the many other sets of responses that can occur, this new view shows the importance of ambivalent responses to change for research on exit, voice, loyalty, and neglect and for research on generating change within organizations.

A SYNTHESIS OF PAST CONCEPTUALIZATIONS OF RESISTANCE TO CHANGE

Unfavorable Responses to Change Might Be Motivated by the Best of Intentions

In the majority of work on resistance to change, researchers have borrowed a view from physics to metaphorically define resistance as a restraining force moving in the direction of maintaining the status quo (cf. Lewin, 1952). Fur- thermore, most scholars have focused on the various "forces" that lead employees away from supporting changes proposed by managers. As Watson (1982) points out, managers often per- ceive resistance negatively, since they see em- ployees who resist as disobedient. And as Jer- mier et al. put it, "The most prevalent way of analysing resistance is to see it as a reactive process where agents embedded in power rela- tions actively oppose initiatives by other agents" (1994: 9). Even if they only see employ- ees who oppose change as short sighted, man- agers are tempted by the language of resistance to treat their subordinates as obstacles.

Thus, the label of resistance can be used to dismiss potentially valid employee concerns about proposed changes. Of course, for a long time in the practical literature about managing change processes, researchers have been advis- ing practitioners to guard against this. For ex- ample, Mary Parker Follett pointed out in the 1920s that

we shouldn't put to . . . workers finished plans in order merely to get their consent one of two things is likely to happen, both bad: either we shall get a rubber-stamped consent and thus lose what they might contribute to the problem in question, or else we shall find ourselves with a fight on our hands—an open fight or discontent seething under- neath (reprinted in Graham, 1995: 220).

Likewise, Lawrence (1954) warns managers to avoid creating resistance in subordinates by as- suming that they will always be opposed to change. In the 1990s others have reissued simi- lar warnings (Dent & Goldberg, 1999; Merron, 1993). A prominent consultant noted that the con- cept of resistance to change "has been trans- formed over the years into a not-so-disguised way of blaming the less powerful for unsatisfac- tory results of change efforts" (Krantz, 1999: 42).

This tendency to dismiss employees' objec- tions to change simply may be another manifes- tation of the fundamental attribution error (Jones & Harris, 1967); that is, managers in charge of rolling out a change initiative blame others for the failure of the initiative, rather than accept- ing their role in its failure. Employees are likely to do the same thing—assigning blame for failed change attempts to their managers, rather than themselves. However, as Klein (1976) and Thomas (1989) argue, in most research on resis- tance to change, researchers have taken the per- spective of those in charge of implementing change, and so scholars have written less about the perspectives of those with less power. Per- haps scholars, as well as practitioners, need to be cautioned against playing the blame game unwittingly.

Fortunately, in other types of literature—not yet well integrated into research on resistance to change—scholars also remind us of a wider range of reasons why employees may oppose a proposed organizational change. For instance, research on obedience to authority indicates that resistance might be motivated by individu- als' desires to act in accordance with their eth- ical principles (Milgram, 1965; Modigliani & Rochat, 1995). Similarly, the organizational dis- sent literature shows that some employee resis- tance to organizational actions is motivated by more than mere selfishness (Graham, 1984, 1986). Also, recent studies of issue selling (Ash- ford, Rothbard, Piderit, & Dutton, 1998; Dutton, Ashford, Wierba, O'Neill, & Hayes, 1997) indicate that employees might try to get top management to pay attention to issues that employees be- lieve must be addressed in order for the organi- zation to maintain high performance.

Rarely do individuals form resistant attitudes, or express such attitudes in acts of dissent or protest, without considering the potential nega- tive consequences for themselves. This point is documented in several studies. In the field of

2000 Pideiit 785

ethics, for instance, Clinard (1983) documents the "pressures on middle management," such a s threats to their opportunities for advancement or to their job security, that can discourage man- agers from speaking up about ethical concerns. Meyerson and Scully (1995) dramatize the dilem- mas faced by change agents when judging how far they can stretch those they wish to lead. Rodrigues and Collinson (1995) analyze the dif- ferent ways in which Brazilian employees use humor to "camouflage and express their dis- sent" (1995: 740), a s well a s the times when cam- ouflage was powerful (and the conditions under which more acerbic satire was used). Thus, friv- olous expression of resistance seems unlikely, since individuals who engage in it could face severe penalties and are aware that they should tread lightly.

Hence, what some may perceive a s disre- spectful or unfounded opposition might also be motivated by individuals' ethical principles or by their desire to protect the organization's best interests. It is worth entertaining efforts to take those good intentions more seriously by down- playing the invalidating aspect of labeling re- sponses to change "resistant."

Varying Emphases in the Conceptualization of Resistance >

Studies of resistance would also benefit from careful attention to the concept's meaning. As Davidson argues, resistance h a s come to in- clude

anything and everything that workers do which managers do not want them to do, and that work- ers do not do that managers wish them to d o . . . . resort to such an essentially residual category of analysis can easily obscure a multiplicity of dif- ferent actions and meanings that merit more pre- cise analysis in their own right (1994: 94).

A review of past empirical research reveals three different emphases in conceptualizations of resistance: a s a cognitive state, a s an emo- tional state, and a s a behavior. Although these conceptualizations overlap somewhat, they di- verge in important ways. Finding a way to bring t o g e t h e r t h e s e v a r y i n g e m p h a s e s should deepen our understanding of how employees respond to proposed organizational changes.

Portraying resistance in terms of behavior h a s been common since the earliest work on the topic. In his early theorizing, Lewin (1952) de-

fined resistance by using a metaphor from the physical sciences. In their classic study Coch and French (1948) focused on the undesirable behaviors of workers in response to manage- ment-imposed changes in jobs and work meth- ods. With their quasi-experiment they examined whether encouraging employee participation in planning a change would reduce resistance. Al- though their conceptual discussion indicated that resistance could involve undesirable be- haviors and/or aggression, their measures fo- cused on neither. Instead, the criterioh they used to compare the treatment and control groups was desirable behavior, in the form of compli- ance with the production rate standards set by management. (While strict compliance with the rate standards may or may not have been ac- companied by undesirable behaviors or aggres- sion, this possibility could not have been cap- tured in the measures reported.) This study generated a large body of work on the effects of participative decision making (see McCaffrey, Faerman, & Hart, 1995, for a recent review).

More recent studies of resistance also have focused on behavior. For instance, Brower and Abolafia (1995) define resistance as a particular kind of action or inaction, and Ashforth and Mael (1998) define resistance a s intentional acts of commission (defiance) or omission. Similarly, Shapiro, Lewicki, and Devine (1995) suggest that willingness to deceive authorities constitutes resistance to change, and Sagie, Elizur, and Greenbaum (1985) use compliant behavior a s evidence of reduced resistance.

In contrast, other scholars have described re- sistance in emotional terms. For example, Coch and French (1948) acknowledged a more emo- tional component of resistance (aggression), and in their preliminary theory of resistance de- scribed the forces that they believed produced frustration in employees and caused the unde- sirable behaviors. Similarly, Vince and Brouss- ine (1996) surfaced the responses of managers in public service organizations to a period of change in structure and financial constraints. They found that managers' responses were often paradoxically emotional. And, finally, the ideas of frustration and anxiety underpin Argyris and Schon's (1974, 1978) perspective that resistance arises from defensive routines. The approach that they advocate emphasizes the role of an external consultant in surfacing the defensive- ness inherent in those routines, finding ways to

786 Academy of Management Review October

minimize or dissipate the anxiety that reinforces those routines, and making time for calmer con- sideration of how to repair them (Argyris, 1993). As Diamond (1986) points out, although the rem- edy for resistance that they recommend involves a cognitive realignment of resistors' espoused theories and their theories-in-use, the underly- ing nature of resistance is portrayed as highly emotional.

The idea that resistance can be overcome cog- nitively suggests that it may include a compo- nent of negative thoughts about the change. Watson (1982) suggests that what is often la- beled a s resistance is, in fact, only reluctance. Armenakis, Harris, and Mossholder (1993) define resistance in behavioral terms but suggest that another state precedes it: a cognitive state they call "(un)readiness." A reinterpretation of the Coch and French quasi-experiment (Bartlem & Locke, 1981) suggests that participation might have motivational and cognitive effects on re- sistance to change, also implying that cognition is part of the phenomenon of resistance.

Each of these three emphases in conceptual- izations of resistance—as a behavior, an emo- tion, or a belief—has merit and represents an important part of our experience of responses to change. Thus, any definition focusing on one view at the expense of the others seems incom- plete. Therefore, rather than privilege one con- ceptualization over the others, I seek to integrate the three alternative views of resistance to change.

A NEW VIEW OF RESPONSES TO CHANGE: AMBIVALENT ATTITUDES

These three emphases in the conceptualiza- tion of resistance to change can be reframed in a more integrative way by borrowing the con- cept of attitude from social psychology. Mindful adaptation of the concept might be required, because the research on attitudes does not al- ways provide clear guidance about which di- mensions of attitudes are most salient.

Multiple Dimensions of Attitudes

Early attitude theorists (Katz, 1960; Rosenberg & Hovland, 1960) argued that attitudes are struc- tured along three dimensions that roughly cor- respond with the three definitions that have dominated research on resistance to change. I

label these three dimensions of attitudes the cognitive, emotional, and intentional. This con- ception is known a s the tripartite view of atti- tudes (Ajzen, 1984).

In this view the cognitive dimension of an attitude refers to an individual's beliefs about the attitude object. In their review of the litera- ture on the tripartite view, Eagly and Chaiken define this dimension a s follows: "beliefs ex- press positive or negative evaluation of greater or lesser extremity, and occasionally are exactly neutral in their evaluative content" (1998: 271). The emotional dimension of an attitude refers to an individual's feelings in response to the atti- tude object. Eagly and Chaiken define this di- mension a s the "feelings, moods, emotions, and sympathetic nervous-system activity that peo- ple have experienced in relation to an attitude object and subsequently associate with it" (1998: 272).

The third dimension of attitudes is the most complex and controversial, both because in some studies researchers find evidence of only two dimensions and because others who find a third dimension label it inconsistently. The find- ings of past empirical studies of the tripartite attitude structure are mixed (e.g., Bagozzi, 1978; Breckler, 1984; Kothandapani, 1971), and a s Ea- gly and Chaiken conclude, "Evidence supports the empirical separability of three classes of evaluative responses under some but certainly not all circumstances" (1993: 13). In the tradi- tional tripartite view, the conative dimension of an attitude reflects an individual's evaluations of an attitude object that are based in past be- haviors and future intentions to act. Some re- searchers place more emphasis on past behav- iors, whereas others focus on future intentions. In some cases a separate attitude dimension concerning intentions or behavior h a s been identified, but in other cases intentions are so loosely connected with other dimensions of atti- tudes that they have been treated a s entirely separate constructs.

In the context considered here, because an employee facing a newly proposed organization- al change is responding to a novel event, the conative dimension is more likely to reflect in- tentions than past behaviors. (The employee might not find the change process particularly novel, but the specific proposal is likely to have some novel aspects.) Also, it seems more desir- able in this applied context to treat behavior a s

2000 Pideiit 787

a separate construct so that the mutual influ- ences of attitudes and behavior on one another are not buried in an already complex set of issues. In other words, it is useful to distinguish between an intention to resist at the attitudinal level and dissent or protest at the level of actual behavior, which might or might not be planned. By "an intention" I mean a plan or resolution to take some action, rather than a plan to try to achieve some goal (Bagozzi, 1992).

Much of the work on resistance in labor pro- cess theory (e.g., Jermier et al., 1994), a s well a s some recent work on extrarole behaviors, such a s taking charge (e.g., Morrison & Phelps, 1999), focuses on dissent or protest, whether inten- tional, habitual, or spontaneous. Distinguishing between intention and behavior will allow more careful study of the connections between the two concepts. Whether the intentional dimen- sion is sufficiently associated with individuals' cognitive and emotional responses to be treated a s a dimension of an employee's attitude re- mains an empirical question in the context of an a t t i t u d e about a p r o p o s e d o r g a n i z a t i o n a l change.

One remaining contentious question in atti- tude research concerns the causal relationships among the dimensions. Fiske and Pavelchak (1986) label the two dominant positions in the debate the "piecemeal" and "category-based" views. In the piecemeal view, advanced by scholars such a s Zanna and Rempel (1988), it is posited that variations in evaluation along the particular dimensions of an attitudinal response will cause variations in global attitude. In the category-based view (Ajzen, 1984; Davis & Os- trom, 1984), the global attitude is viewed a s pri- mary; changes in the global evaluation are mod- eled a s causes of variation in the cognitive, emotional, and intentional dimensions, rather than as results of variation in those dimensions. Unfortunately, these views are still the subject of continuing debate in social psychology, and competing interpretations and new data are still being advanced.

In summary, questions of how the multiple dimensions of employee responses to change should be defined—and how they are related to one another—remain open to further clarifica- tion through empirical research. Social psycho- logical research, however, clearly supports a multidimensional view of attitudes that can be used to integrate the inconsistent definitions of

resistance that have been found in organization- al studies. Thus, an employee's response to an organizational change along the cognitive di- mension might range from strong positive be- liefs (i.e., "this change is essential for the or- ganization to succeed") to strong negative beliefs (i.e., "this change could ruin the compa- ny"). An employee's response along the emo- tional dimension might range from strong posi- tive emotions (such a s excitement or happiness) to strong negative emotions (such a s anger or fear). An employee's response along the inten- tional dimension might range from positive in- tentions to support the change to negative inten- tions to oppose it.

The Possibility of Ambivalence in Response to a Particular Change Proposal

One key benefit of using this multidimen- sional definition to describe employees' atti- tudes toward proposed changes is that concep- t u a l i z i n g e a c h d i m e n s i o n a s a s e p a r a t e continuum allows for the possibility of different reactions along the different dimensions. In some cases this might only mean that beliefs about a proposed change are more positive than emotional responses to the change. However, with this definition we also recognize the possi- bility, in other cases, of ambivalent attitudes, where two alternative perspectives are both strongly experienced (Foy, 1985; Merton, 1976; Thompson, Zanna, 8f Griffin, 1995).

The simplest case of ambivalence to imagine is the case in which an individual's cognitive response to a proposed change is in conflict with his or her emotional response to the pro- posal. Furthermore, ambivalence within a di- mension is also possible, and, in fact, ambiva- lence within the emotional dimension already has been reported in research. In particular, Russell (1980) and Watson, Clark, and Tellegen (1988) have presented data suggesting that pos- itive and negative affect can co-occur. Similarly, Vince and Broussine's (1996) study of public ser- vice managers' responses to change shows that incongruent emotions, such a s excitement and fear, are often experienced simultaneously.

In principle, ambivalence could occur within the cognitive or intentional dimensions a s well. For instance, an employee exhibiting cognitive ambivalence might simultaneously believe that the change proposed in his or her organization

788 Academy of Management Review October

is necessary for its future survival but is not yet sufficiently well researched. An employee ex- hibiting intentional ambivalence might plan to oppose a proposed change through anonymous comments in the suggestion box but might sup- port the change in public because of uncertainty about how top management will respond to crit- icism of the change initiative. Although re- search does not shed any light on the likelihood of intentional ambivalence, anecdotal evidence of its occurrence can be found; Drummond's (1998) case study of a site manager's indirect opposition to the proposed closure of his facility has similar elements.

The Prevalence of Ambivalent Attitudes

The following examples of employees' re- sponses to organizational change, drawn from interviews,' also illustrate the merits of assess- ing their attitudes toward change along three dimensions. In the first example an employee had learned that his budget for offering incen- tives to his distributors was disappearing. His emotional response to the announcement was quite negative. Because the budget cut was an- nounced late in his planning cycle, the an- nouncement shocked and frustrated him. How- ever, he also reported a positive cognitive response to the change: he believed the change would have positive effects, since the budget for product improvements was being increased to allow his distributors to offer their customers a more attractive product. Thus, this employee's response represents dn example of an ambiva- lent a t t i t u d e toward the proposed budget change, because of the incongruity between his cognitive and emotional responses to the pro- posal.

A second example comes from an interview with a middle manager in a large, diversified company, who described his response to the re- structuring and centralization of his organiza-

' To illustrate how the tripartite definition of attitudes could be used to describe employees' responses to organi- zational changes, I collected stories about employees' reac- tions to recently proposed changes in their organization. I conducted seven interviews with professionals and five with managers. The interviewees had varied functional back- grounds, and they described their reactions to three types of organizational changes (updating work processes, develop- ing new initiatives, and restructuring). The interview proto- col appears in Piderit (1999).

tion around a new enterprise-wide software sys- tem. His initial reaction to the restructuring included positive beliefs, because he felt the change was sorely needed, a s well a s positive emotions, reflected in expressions of enthusi- asm. However, he reported increasingly nega- tive intentions over time, and he planned to challenge his leadership to cancel the project if they would not provide the support that was needed. He later spoke out against the dangers of the "behemoth project." Although he still be- lieved the change was needed, he was discour- aged by his coworkers' lack of commitment. Thus, this manager's initial attitude can be rep- resented as initially supportive, but it evolved to a more ambivalent state as his negative inten- tions solidified and his negative emotions to- ward his coworkers' laxity emerged.

The third example is drawn from an interview with a consultant who learned that his firm was merging with another consulting company. He initially responded with a combination of ex- citement and fear, demonstrating ambivalence within the emotional dimension of his response to the change. In his case that ambivalence mo- tivated his efforts to gather information about the rationale for the merger and to assess the likelihood of job cuts in conjunction with it. Al- though he was not comfortable discussing the change with his superiors, because he did not want to reveal his fears and appear insecure, he and his peers were able to reassure each other through their surreptitious information gather- ing that the rationale for the merger was to acquire consulting skills in markets that his original firm had not already entered. As a re- sult, he became a n active supporter of the merger later on.^

In addition to this anecdotal evidence, there is also a theoretical reason to expect that most employees' responses to a proposed change will involve some ambivalence. We know from atti- tude research that the process of attitude forma- tion often b e g i n s with a m b i v a l e n c e (e.g..

^ These three employees' descriptions of their reactions to change were typical of the reactions reported in interviews, since most of the interviewees described their reactions to the organizational changes that they faced in terms of a mix of positive and negative thoughts, emotions, and behavioral intentions. Four of the twelve interviewees reported endur- ing ambivalence in response to the change they faced, and another five interviewees reported initial ambivalence.

2000 Piderit 789

Thompson et al., 1995). Furthermore, within the typology of alpha, beta, and gamma change, an initial response that is uniformly negative s e e m s possible only in response to a l p h a changes, which involve a "variation in the level of [a] state, given a constantly calibrated mea- suring instrument" (Golembiewski, Billingsley, & Yeager, 1976: 134). Because some employees will already have formed an attitude toward the current point, they may be able to infer their attitude toward the proposed shift immediately.

However, as Beer and Walton (1987) point out, beta change involves developing a new under- standing of what constitutes a shift on the refer- ence dimension (or a "variation in the level of [a] state, complicated by the fact that some inter- vals of the measurement continuum . . . have been recalibrated," according to Golembiewski et al. [1976: 134]). For example, a team trained in dialectic decision making might come to rede- fine what is meant by "too much conflict" in its meetings. Given the more complex process in- volved in making sense of a change proposal that involves such a recalibration, it seems un- likely that employees' inferences about their at- titude toward a change proposal, such as the proposal to engage in a structured decision- making process, could be immediate.

Similarly, the gamma change process, which involves "a complete conceptual redefinition" (Beer & Walton, 1987: 342) and which may in- volve either the addition of new dimensions or the complete replacement of old reference di- mensions with new ones (Porras & Silvers, 1991: 57), is even more complex. Thus, when facing beta or gamma change, employees seem more likely to engage in the formation of a new atti- tude, rather than simply shift their old attitude along a stable dimension. It seems reasonable to a s s u m e that most employees' initial re- sponses to a beta or gamma change will be ambivalent.

For these reasons I conclude that conceptual- izing employees' responses to proposed organi- zational changes a s multidimensional attitudes permits a richer view of the ways in which em- ployees may respond to change. Because of the potential for a multidimensional view of re- sponses to change to inspire future research in such directions, I join Dent and Goldberg (1999) and Merron (1993) in arguing that we should retire the phrase "resistance to change," and I advocate a new wave of research on employee

responses to change, conceptualized as multidi- mensional attitudes.

IMPLICATIONS FOR RESEARCH AND PRACTICE

There are five key implications of this alter- native view for research and practice. First, a multidimensional view of responses to proposed change may enhance our accuracy in predicting employee behaviors that have been difficult to predict in past research.

For example, understanding exit, voice, loy- alty, and neglect h a s continued to challenge theorists and empirical researchers (Hirschman, 1970; Janssen, de Vries, & Cozijnsen, 1998; Rus- bult, Farrell, Rogers, & Mainous, 1988; Saunders, Sheppard, Knight, & Warshaw, 1992; Withey & Cooper, 1989). One premise that could aid in developing such a predictive framework is the idea that employees find it more difficult to ex- press negative emotions than negative beliefs. (This premise is certainly implicit in Argyris and Schon's [1974, 1978] work, although some em- ployees may exhibit more facility than others in expressing their emotions.) From that premise it would follow that employees would be more likely to engage in voice than in loyalty or ne- glect when they experience ambivalence within their cognitive response to a proposed change. Because they can easily articulate their beliefs about the change, they would be more likely to share their reflections with the managers intro- ducing the organizational change. Conversely, employees would be more likely to exhibit ne- glect when ambivalence occurs within the emo- tional dimension of their response to change or when an incongruity arises between their cog- nitive and their emotional reactions. Because it is difficult for them to articulate their negative emotional responses to change, they would be more likely to wrestle with their ambivalence alone or to avoid the subject entirely.

Similarly, understanding the nature of ambiv- alence in employee responses to change also might be useful in predicting the mode in which employees will communicate their responses to change agents and in identifying the most ap- p r o p r i a t e process for a d d r e s s i n g their re- sponses. For instance, when employees are ex- periencing emotional ambivalence rather than uniformly negative responses to a proposed change, they may be more likely to express their

790 Academy of Management Review October

responses through humor (e.g., Rodrigues & Col- linson, 1995) or other indirect modes of commu- nication (e.g., Drummond, 1998). In such a case, more data about the change initiative might not be very useful, even if it can be provided effi- ciently in large-scale rollout meetings. Instead, more impromptu and casual conversations might be more effective in creating an atmo- sphere in which employees feel safe expressing their negative emotional responses openly.

Conversely, when employees are experienc- ing cognitive ambivalence about a proposed change but no negative emotional responses, they may be quite direct in expressing their con- cerns. In such a case, change agents might find that their listening ability is more important than their ability to communicate their own per- spectives on the change to employees. Oversell- ing the benefits of the change may not be effec- tive in securing employee support, if employees already accept that the change will have some positive outcomes but feel a different perspec- tive is required.

Of course, the merits of these premises are empirical questions, to be examined in future research on predicting employee voice, loyalty, and neglect and in research on the modes in which employee responses to change are ex- pressed and managed.

A second key implication of the new multidi- mensional view of employee responses to pro- posed organizational changes is that the degree of ambivalence in an employee's attitude may have both desirable and undesirable conse- quences. Paying attention to balancing those consequences will help us understand how to manage change processes successfully. A vari- ety of research indicates that divergent opinions about direction are necessary in order for groups to make wise decisions and for organi- zations to change effectively. For instance, re- cent research on institutionalized dissent (Co- hen & Staw, 1998) shows that, sometimes, organizations encourage and plan for dissent and ritualize disagreement. Although the fact that organizations encourage dissent does not necessarily imply that dissent is functional, it is one reasonable explanation for the prevalence of such an organizational practice.

Furthermore, research on organizational learning indicates that disagreement and dis- confirmation of expectations can be important triggers for developing knowledge. In fact, Bar-

nett argues that "an emphasis on failure, nega- tive feedback, stress, or 'crisis' as a learning stimulus has eclipsed the potential importance of other meaningful stimuli (e.g., opportunities, people, and success)" (1994: 8) as conditions that foster learning. Similarly, research on strategic change processes indicates that disagreement can play a key role in supporting organizational renewal. Studies by Barr, Stimpert, and Huff (1992), Burgelman (1991), and Floyd and Wool- dridge (1996) show that if the organization's managers do not experiment, it seems unlikely that they will be able to carry out a renewal process. The implication of all this research is that moving too quickly toward congruent posi- tive attitudes toward a proposed change might cut off the discussion and improvisation that may be necessary for revising the initial change proposal in an adaptive manner.

It is not clear, however, whether the expres- sion of resistance (i.e., uniformly negative re- sponses to change) is likely to encourage con- tinued discussion, debate, and improvisation. Indeed, the honest expression of ambivalence seems more likely to generate dialogue than the expression of either determined opposition or firm support.

Several research pieces also indicate that am- bivalence and its acknowledgment might have positive effects. Pratt and Barnett (1997) argue that ambivalence is needed to stimulate un- learning (the discarding of obsolete and mis- leading knowledge), which is a necessary pre- cursor to change. Similarly, Weigert and Franks argue that the expression of ambivalence in public "is likely to lead to public collective re- sponses" (1989: 223), suggesting that acknowl- edging ambivalence can provide a basis for mo- tivating new action, rather than the continuation of old routines. Furthermore, recent research on creativity indicates that "insight is primarily de- pendent on analogical retrieval.... moreover, this retrieval usually is cued by some external event" (Sternberg, 1988: 3). Work by Langley and Jones (1988) and by Weisberg (1988) shows that the ability to perceive a situation from a differ- ent angle or to apply a novel analogy is often the key to finding a previously unconsidered alternative that may lead to novel behavior. All this work suggests that by fostering ambiva- lence and reframing our understanding of the status quo, we are better able to generate new possibilities for understanding and action.

2000 Pideiit 791

For change agents and for theorists, the strat- egy of fostering ambivalence rather than sup- port in the early stages of a change initiative invites a different view of how the first stage of a change process should play out. The first stage in creating change should be generating widespread conversation, rather than beginning the change process by engaging a small group of managers in identifying the desired change and later aiming to gain broader employee sup- port for that proposal. This strategy is less con- sistent with a view of change as a planned pro- cess (Porras & Silvers, 1991) and more consistent with a microlevel perspective on change as a continuous process in which "ongoing adapta- tion and adjustment" occur (Weick & Quinn, 1999: 362). Some models of this type of change process are emerging, such as the trialectic view of change advocated by Ford and Ford (1994) and the five-stage process model of break- away organizations developed by Dyck and Starke (1999). How change agents begin to gen- erate conversation around ambivalence about new possibilities is an important question for future research on the first phase of the change process.

Ambivalence, however, must be fostered with care; we also know from other streams of re- search that acknowledging ambivalence might not always be optimal. On the one hand, Weigert and Franks warri that "if ambivalence is not ritually enacted and meaningfully inter- preted, its power to fuel extreme responses grows" (1989: 223). On the other hand, Schwartz (1986) examined the effect of inner dialogue on personal and relational well-being and found that an inner dialogue characterized by a high ratio of positive to negative statements was as- sociated with greater well-being. This finding suggests that acknowledging both polarities of an ambivalent attitude toward a change pro- posal with equal time might be unhealthy. Thus, the question that emerges for research and prac- tice concerns the tensions generated by foster- ing ambivalence: How can we balance the need for ambivalence with the need to limit its debil- itating effects?

A third key implication of the new multidi- mensional view concerns the need to expand our research beyond our past focus on top-down organizational change. Increasingly, change processes are managed in emergent and demo- cratic ways. However, our theorizing may not be

keeping pace, except in some emerging re- search. For example, in the appreciative inquiry process (Cooperrider, 1998; Cooperrider & Sriv- astva, 1987), the proposal emerges from and is tempered and repeatedly revised by an inclu- sive dialogue among a large number of employ- ees across many levels of hierarchy. In this ap- proach the important question of what it means to respond to a "proposed change" is framed, when the nature of the change that is proposed remains ambiguous for much of the process. Here, finding answers to the questions of how multiple dimensions of an employee's re- sponses to a change evolve over time and how such shifts are related to the effectiveness with which change is implemented seems even more important.

A fourth implication of these ideas is that employee responses to change may evolve over time, and paying attention to this evolu- tion might yield insights about how to manage change initiatives successfully. For example, a formal change announcement from the CEO may shift employees' cognitive responses to a change quite quickly from negative to posi- tive, but their emotional responses may re- quire more time to shift from negative to pos- itive, through many informal conversations after the formal rollout speech. Observing pat- terns of attitudes and ambivalence over time might be more useful in predicting the success of a change initiative than examining the fa- vorability of employees' attitudes toward the change at any one point in time. The implica- tion is that both scholars and managers need to pay more attention to the dynamic pro- cesses that help to acknowledge and sustain ambivalence without letting it impede the mo- mentum of change.

A final implication of these ideas is that schol- ars who wish to understand the full range of individual responses to proposed organization- al changes should assess those responses along multiple dimensions. Applied research is needed to continue the process of mindfully adapting the concept of tripartite attitudes from social psychology. Relevant methods for opera- tionalizing the dimensions could include inter- views (Piderit, 1998), surveys (Piderit, 1999), and even more novel approaches, such as drawing (Vince & Broussine, 1996).

792 Academy of Management Review October

CONCLUSION

There is power in metaphor, but the physical metaphor of "resistance to change" may have taken us as far as we can go. In this article I critiqued research on resistance to change for failure to take the good intentions of resistors seriously and for the varying emphases in con- ceptualizations of resistance. I proposed a new conception of responses to proposed organiza- tional changes as multidimensional attitudes. This new conception is intended to encourage an appreciation for the prevalence of ambiva- lence in individuals' responses to change. Inves- tigations of what motivates those responses to change also will be needed, as well as studies of both the positive and the negative conse- quences of ambivalence of different types.

These ideas are not all new to the field, but earlier admonitions about the benefits of em- ployee input and the drawbacks of dismissing subordinates' responses to change were not con- sistently brought to center stage in organization studies. If we can do better, we will be able to offer guidance to all employees involved in change processes and not just to change agents with official authority. Our research will begin to give equal attention to top-down, planned change and to bottom-up or egalitarian change processes. Finally, we will take on the challenge of helping organization members reap the ben- efits of ambivalence toward change for organi- zations while minimizing its potentially stress- ful effects for individuals.

REFERENCES

Ajzen, I. 1984. Attitudes. In R. J. Corsini (Ed.), WiJey encycJo- pedia of psychology, vol. 1: 99-100. New York: Wiley.

Ajzen, I. 1987. Attitudes, traits, and actions: Dispositional prediction of behavior in personality and social psychol- ogy. In L. Berkowitz (Ed.), Advances in experimental so- cial psychology, vol. 20:1-63. New York: Academic Press.

Argyris, C. 1993. Knowledge for action: A guide to overcom- ing barriers to organizational change. San Francisco: Jossey-Bass.

Argyris, C , & Schon. D. 1974. Theory in pracHce. San Fran- cisco: Jossey-Bass.

Argyris, C, & Schon, D. 1978. Organizational learning. Read- ing, MA: Addison-Wesley.

Armenakis, A. A., Harris, S. G., & Mossholder, K. W. 1993. Creating readiness for organizational change. Human RelaHons, 46: 681-703.

Ashford, S. J., Rothbard, N. P., Piderit, S. K., & Dutton, J. E.

1998. Out on a limb: The role oi context and impression management in selling gender-equity issues. Adminis- trative Science Quarterly, 43: 23-57.

Ashforth, B. E., & Mael, F. A. 1998. The power of resistance: Sustaining valued identities. In R. M. Kramer & M. A. Neale (Eds.), Power and influence in organizations: 89- 120. Thousand Oaks, CA: Sage.

Bagozzi j R. P. 1978. The construct validity of the affective, behavioral, and cognitive components of attitude by analysis of covariance structures. Multivariate Behavior Research, 13: 9-31.

Bagozzi, R. P. 1992. The self-regulation of attitudes, inten- tions, and behavior. Social Psychology Quarterly, 55: 178-204.

Barnett, C. K. 1994. OrganizaJiona7 learning theories: A re- view and synthesis of the literature. Working paper. University of Michigan, Ann Arbor.

Barr, P. S., Stimpert, J. L., & Huff, A. S. 1992. Cognitive change, strategic action, and organizational renewal. Strategic Management Journal, 13: 15-36.

Bartlem, C. S., & Locke, E. A. 1981. The Coch and French study: A critique and reinterpretation. Human Relations, 34: 555-566.

Beer, M., & Walton, A. E. 1987. Organization change and development. Annuai Review of Psychology, 38: 339-367.

Breckler, S. J. 1984. Empirical validation of affect, behavior, and cognition as distinct components of attitude. Journal of Personality and Social Psychology, 47: 1191-1205.

Brower, R. S., & Abolafia, M. Y. 1995. The structural embed- dedness of resistance among public managers. Group and Organization Management, 20: 149-166.

Burgelman, R. A. 1991. Intraorganizational ecology of strat- egy making and organizational adaptation: Theory and field research. Organization Science, 2: 239-262.

Clinard, M. B. 1983. Corporate ethics and crime: The role of middle management. Beverly Hills, CA: Sage.

Coch, L., & French, J. R. P., Jr. 1948. Overcoming resistance to change. Human Relations, 1: 512-532.

Cohen, L. E., & Staw, B. M. 1998. Fun's over, fact checkers are here: A case study of institutionalized dissent in the magazine publishing industry. Advances in Qualitative Organization Research, vol. 1: 105-136.

Cooperrider, D. L. 1998. Capturing what matters most in the practice of appreciative inquiry: A positive revolution in change. Paper presented at the annual meeting of the Academy of Management, San Diego.

Cooperrider, D. L., & Srivastva, S. 1987. Appreciative inquiry in organizational life. Research in Organizational Change and Development, 1: 129-169.

Davidson, J. O'C. 1994. The sources and limits of resistance in a privatized utility. In J. M. Jermier, D. Knights, & W. R. Nord (Eds.), Resistance and power in organizations: 69- 101. New York: Routledge.

Davis, D., & Ostrom, T. M. 1984. Attitude measurement. In R. J. Corsini (Ed.), Wiley encyclopedia of psychology: 97-99. New York: Wiley.

2000 Pideiit 793

Dent, E. B., & Goldberg, S. G. 1999. Challenging "resistance to change." Journal of Applied Behavioral Sciences, 35(1): 25-41.

Diamond, M. A. 1986. Resistance to change: A psychoanalytic critique of Argyris and Schon's contributions to organi- zation theory and intervention. Journal of Management Studies, 23: 543-562.

Drummond, H. 1998. Go and say, "we're shutting": Ju Jitsu a s a metaphor for analyzing resistance. Human Relations, 51: 741-759.

Dutton, J. E., Ashford, S. J., Wierba, E. E., O'Neill, R., & Hayes, E. 1997. Beading the wind: How middle managers assess the context for issue selling to top managers. Strategic Management Journal, 15: 407-425.

Dyck, B., & Starke, F. A. 1999. The formation of breakaway organizations: Observations and a process model. Ad- ministrative Science Quarterly, 44: 792-822.

Eagly, A. H., & Chaiken, S. 1993. The psychology of attitudes. Fort Worth, TX: Harcourt Brace Jovanovich.

Eagly, A. H., & Chaiken, S. 1998. Attitude structure and func- tion. In D. T. Gilbert, S. T. Fiske, & G. Lindsey (Eds.), Handbooi of social psychology, vol. 2: 269-322. Boston: McGraw-Hill.

Fiske, S. T., & Payelchak, M. A. 1986. Category-based versus piecemeal-based affective responses: Developments in schema-triggered affect. In R. M. Sorrentino & E. T. Hig- gins (Eds.), Handbooi of motivation and cognition: Foun- dations of social behavior, vol. 1: 167-203. New York: Guilford.

Floyd, S. W., & Wooldridge, B. 1996. The strategic middle manager. San Francisco: Jossey-Bass.

Ford, J. D., & Ford, L. W. 1994. Logics of identity, contradiction, and attraction in change. Academy of Management Re- view, 19: 756-785.

Foy, N. 1985. Ambivalence, hypocrisy, and cynicism: Aids to organization change. JVew Management, 2(4): 49-53.

Golembiewski, R. T., Billingsley, K., & Yeager, S. 1976. Mea- suring change and persistence in human affairs: Types of change generated by OD designs. Journal of Applied Behavioral Sciences, 12(2): 133-157.

Graham, J. 1984. Abstract of a dissertation on principled organizational dissent. Paper presented at the annual meeting of the Society for Industrial and Organizational Psychology, Toronto.

Graham, J. 1986. Principled organizational dissent: A theo- retical essay. Research in Organizational Behavior, 8: 1-52.

Graham, P. (Ed.). 1995. Mary Parker Follett^prophet of man- agement: A celebration of writings from the 1920s. Bos- ton: Harvard Business School Press.

Hirschman, A. O. 1970. Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Cambridge, MA: Harvard University Press.

Janssen, O., de Vries, T., & Cozijrisen, A. J. 1998. Voicing by adapting and innovating employees: An empirical study on how personality and environment interact to affect voice behavior. Human Relations, 51: 945-966.

Jermier, J. M., Knights, D., & Nord, W. R. (Eds.). 1994. Resis- tance and power in organizations. London: Routledge.

Jones, E. E., & Harris, V. A. 1967. The attribution of attitudes. Journal of Experimental Social Psychology, 3: 1-24.

Katz, D. 1960. The functional approach to the study of atti- tudes. Public Opinion Quarterly, 24: 163-204.

Klein, D. 1976. Some notes on the dynamics of resistance to change: The defender role. In W. G. Bennis, K. D. Benne, R. Chin, & K. E. Corey (Eds.), The planning of change (3rd ed.): 117-124. New York: Holt, Rinehart & Winston.

Kothandapani, V. 1971. Validation of feeling, belief, and inten- tion to act as three components of attitude and their con- tribution to prediction of contraceptive behavior. Journal of Personality and Social Psychology, 19: 321-333.

Krantz, J. 1999. Comment on "challenging 'resistance to change.'" Journal of Applied Behavioral Sciences, 35(1): 42-44.

Langley, P., & Jones, R. 1988. A computational model of sci- entific insight. In R. J. Sternberg (Ed.), The nature of creativity: Contemporary psychological perspectives: 177-201. New York: Cambridge University Press.

Lawrence, P. R. 1954. How to deal with resistance to change. Harvard Business Review, 32(3): 49-57.

Lewin, K. 1952. Group decision and social change. In G. E. Swanson, T. M. Newcombe, & E. L. Hartley (Eds.), Read- ings in social psychology (2nd ed.): 459-473. New York: Holt.

McCaffrey, D. P., Faerman, S. R., & Hart, D. W. 1995. The appeal and difficulties of participative systems. Or- ganization Science, 6: 603-627.

Merron, K. 1993. Let's bury the term "resistance." Organiza- tional Development Journal, 11(4): 77-86.

Merton, R. K. 1976. Sociological ambivalence and other es- says. New York: Free Press.

Meyerson, D. E., & Scully, M. A. 1995. Tempered radicalism and the politics of ambivalence and change. Organiza- tion Science, 6: 585-600.

Milgram, S. 1965. Some conditions of obedience and disobe- dience to authority. Human Relations, 18: 57-76.

Modigliani, A., & Rochat, F. 1995. The role of interaction sequences and the timing of resistance in shaping obe- dience and defiance to authority. Journal of Social Is- sues, 51(3): 107-123.

Morrison, E. W., & Phelps, C. C. 1999. Taking charge at work: Extrarole efforts to initiate workplace change. Academy of Management Journal, 42: 403-419.

Piderit, S. K. 1998. United we stand: Embedding attitudes toward change in social relationships. Paper presented at the annual meeting of the Academy of Management, San Diego.

Piderit, S. K. 1999. Wavigating relationships with coworiers: Understanding employees' attitudes toward an organi- zational change. Unpublished doctoral dissertation. University of Michigan, Ann Arbor.

Porras, J. L, & Silvers, R. C. 1991. Organization development

794 Academy of Management Review October

and transformation. Annual Review of Psychology, 42' 51-78.

Pratt, M. G., & Barnett, C. K. 1997. Emotions and unlearning in Am way recruiting techniques: Promoting c h a n g e through "safe" ambivalence. Management Learning, 28(1): 65-88.

Rodrigues, S. B., & Collinson, D. L. 1995. "Having fun"?: Hu- riior as resistance in Brazil. Organization Studies, 16: 739-768.

Rosenberg, M. J., & Hovland, C. I. 1960. Cognitive, affective, and behavioral components of attitudes. In M. J. Rosen- berg (Ed.), Attitude organization and change: 1-14. New Haven, CT: Yale University Press.

Rusbult, C. E., Farrell, D., Rogers, G., & Mainous, A. G. 1988. Impact of exchange variables on exit, voice, loyalty, and neglect: An integrative model of responses to declining job satisfaction. Academy of Management Journal 31: 599-627.

Russell, J. A. 1980. A circumplex model of affect. Journal of Personality and Social Psychology, 39: 1161-1178.

Sagie, A., Elizur, D., & Greenbaum, C. W. 1985. Job experi- ence, persuasion strategy, and resistance to change. Journal of Occupational Behavior, 6: 157-162.

Saunders, D. M., Sheppard, B. H., Knight, V., & Warshaw, P. R. 1992. Employee voice to supervisors. Employee Respon- sibility and Rights Journal, 3: 241-259.

Schwartz, R. M. 1986. The inner dialogue: On the asymmetry between positive and negative coping thoughts. Cogni- tive Therapy and Research, 10: 591-605.

Shapiro, D. L., Lewicki, R. J., & Devine, P. 1995. When do employees choose deceptive tactics to stop unwanted organizational change? Research on Negotiation in Or- ganizations, 5: 155-184.

Sternberg, R. J. (Ed.). 1988. The nature of creativity: Contem- porary psychological perspectives. New York: Cam- bridge University Press.

Thomas, R. J. 1989. Participation and control: A shopfloor perspective on employee participation. Research in the Sociology of Organizations, 7: 117-144.

Thompson, M. M., Zanna, M. P., & Griffin, D. W. 1995. Let's not be indifferent about (attitudinal) ambivalence. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Anteced- ents and consequences: 361-386. Mahwah, NJ: Lawrence Eribaum Associates.

Vince, R., & Broussine, M. 1996. Paradox, defense, and attach- ment: Accessing and working with emotions and rela- tions underlying organizational change. Organization Studies, 17: 1-21.

Watson, D., Clark, L. A., & Tellegen, A. 1988. Development and validation of brief measures of positive and nega- tive affect: The PANAS scales./ournal of Personality and Social Psychology, 54: 1063-1070.

Watson, T. J. 1982. Group ideologies and organizational change. Journal of Management Studies, 19: 259-275.

Weick, K. E., & Quinn, R. E. 1999. Organizational change and development. Annual Review of Psychology, 50: 361-386.

Weigert, A., & Franks, D. D. 1989. Ambivalence: Touchstone of the modern temper. In D. D. Franks & E. D. McCarthy (Eds.), The sociology of emotions: 205-227. Greenwich, CT: JAI Press.

Weisberg, R. W. 1988. Problem solving and creativity. In R. J. Sternberg (Ed.), The nature of creativity: Contemporary psychological perspectives: 148-176. New York: Cam- bridge University Press.

Withey, M. J., & Cooper, W. H. 1989. Predicting exit, voice, loyalty and neglect. Administrative Science Quarterly, 34: 521-539.

Zanna, M, P., & Rempel, J. K. 1988. Attitudes: A new look at an old concept. In D. Bar-Tal & A. W. Kruglanski (Eds.), The social psychology of knowledge: 315-334. New York: Cambridge University Press.

Sandy Kristin Piderit is an assistant professor of organizational behavior in the Weatherhead School of Management at Case Western Reserve University. She re- ceived her Ph.D. at the University of Michigan at Ann Arbor. In her research she examines the roles of individuals in organizational change and the dynamics of relationships among coworkers.