Want to, need to, ought to: employee commitment to organizational change

The Authors

Janet Turner Parish, McCoy College of Business Administration, Texas State University, San Marcos, Texas, USA

Susan Cadwallader, College of Business and Economics, California State University, Fullerton, California, USA

Paul Busch, Mays Business School, Texas A&M University, College Station, Texas, USA

Abstract

Purpose – This study aims to focus on the role of employee commitment in the success of organizational change initiatives. The authors seek to propose and test a model that delineates antecedents and consequences of affective, normative, and continuance commitment to organizational change.

Design/methodology/approach – Data were collected via online survey from employees working in a large not-for-profit organization. A total of 191 responses (32 per cent) were obtained. The hypothesized model relationships were tested using structural equation modeling.

Findings – The results demonstrate that the antecedents: fit with vision, employee-manager relationship quality, job motivation, and role autonomy all influence commitment to change (C2C). Notably, affective commitment, which in turn influences employee perceptions about improved performance, implementation success, and individual learning regarding the change, had the greatest impact.

Research limitations/implications – Limitations that could lead to future research include three primary issues. First, the data were collected from one organization. Second, data were collected at one point in time. Third, the proposed model is far from exhaustive. Other antecedents to C2C could include risk acceptance, organizational culture, and/or leadership style.

Practical implications – In addition, given affective commitment to change (AC2C) has the greatest influence on outcomes such as implementation success and improved performance, the antecedents of AC2C warrant management's attention. If managers can influence AC2C through such factors as those proposed here, outcomes such as individual learning, perceived implementation success, and perceived improved performance will be influenced.

Originality/value – The paper integrates two models evaluating employee reactions to change, an under researched area, to help uncover ways to improve the success of change initiatives.

Article Type:

Research paper

Keyword(s):

Organizational change; Job satisfaction; Change management; Employee attitudes.

Journal:

Journal of Organizational Change Management

Volume:

Number:

Year:

2008

pp:

32-52

Copyright ©

Emerald Group Publishing Limited

ISSN:

0953-4814

Workplaces are faced with endless change (Herscovitch and Meyer, 2002), and effective management of that change is an important competency currently required by an organization (Paton and McCalman, 2000). The growing frequency and complexity of workplace change requires employees to adapt to change without disruption; however, resistance to change is the more common reaction (Caldwell et al., 2004). As managers make decisions for coping with change, they must consider not only how firm performance will be affected but also how employees will be affected. Can commitment to change (C2C) be obtained and, if so, will that help the implementation efforts? This issue warrants research attention.

As Herscovitch and Meyer (2002, p. 474) stated:

Given the accelerated rate and complexity of changes in the workplace, it is not surprising that there is a large and growing literature on the causes, consequences, and strategies of organizational change. What is surprising, however, is the paucity of research on employee reactions to change.

There is a growing interest in understanding how change is experienced by individual employees (Judge et al., 1999) and researchers are beginning to investigate the role of employee commitment in organizational change situations (Herscovitch and Meyer, 2002; Noble and Mokwa, 1999). To attain commitment, we believe that top management must strive to understand the drivers of C2C and its critical role in the successful implementation of strategic initiatives. We conducted our study to begin to understand this issue.

A primary contribution of our research is the integration of two models evaluating employee reactions to change. Herscovitch and Meyer (2002) introduced a three-component model of commitment to organizational change (based on Meyer and Allen, 1991) and demonstrated that employee commitment to a change is a better predictor of behavioral support for a change than is organizational commitment. The three components are: affective, normative, and continuance. Noble and Mokwa (1999) also studied multiple dimensions of commitment: organizational, role, and strategy. Though a broader concept, commitment to a strategy is conceptually similar to commitment to organizational change. Strategy commitment is defined as management's understanding and support of goals and objectives (Noble and Mokwa, 1999). Implementing change is a strategic decision. Noble and Mokwa's (1999) model considered strategy factors and role factors as antecedents of commitment. Their key outcome was the success of managerial implementation of initiatives. We seek to draw from these two models to increase our understanding of the antecedents and consequences of employee commitment to organizational change. To this end, we:

use Herscovitch and Meyer's (2002) three components of commitment to organizational change;
test Noble and Mokwa's proposed antecedents (along with additional antecedent and outcome variables); and
survey both employees and managers.

We first provide a brief review of the commitment to organizational change literature, followed by the development of our hypothesized model. Because the commitment to organizational change literature is early in its development, we rely heavily on the organizational commitment literature to develop our hypotheses. We then share the details of our analysis. After a discussion of the findings of the empirical study, we present implications for theory and practice. We conclude with study limitations and suggestions for further research.

Literature review

An important development in academic research is the recognition that employees can be committed to many different work-related foci. Recent work suggests the importance of considering objects of commitment in addition to the organization itself, such as supervisors or change initiatives (Herscovitch and Meyer, 2002). Noble and Mokwa (1999, p. 61) defined organizational commitment as “the extent to which a person identifies with and works toward organization-related goals and values”. Moving toward a more context-specific definition, Herscovitch and Meyer (2002, p. 475) defined C2C as “a force (mind-set) that binds an individual to a course of action deemed necessary for the successful implementation of a change initiative”.

Affective, continuance, and normative commitment

To better understand organizational commitment, Meyer and Allen (1991) proposed a three-component model comprising affective, continuance, and normative commitment. According to Meyer and Allen (1991, p. 67), affective commitment (AC) “refers to the employee's emotional attachment to, identification with, and involvement in the organization” (employees stay with a firm because they want to); continuance commitment (CC) “refers to an awareness of the costs associated with leaving the organization” (employees stay with a firm because they need to); and normative commitment “reflects a feeling of obligation to continue employment” (employees stay with a firm because they ought to). Drawing on this model in an effort to study a context-specific type of commitment, Herscovitch and Meyer (2002) proposed a three-component model of commitment to organizational change and suggested that it has advantages similar to those demonstrated in the organizational commitment literature (e.g. improved ability to predict employee behavior). Therefore, affective commitment to change (AC2C) refers to a desire to support a change, continuance commitment to change (CC2C) is based on a recognition that there are costs associated with resisting change, and normative commitment to change (NC2C) reflects a sense of obligation to be supportive (Herscovitch and Meyer, 2002).

Meyer and Allen (1991) noted that given their conceptual differences, it seems likely that the psychological states reflecting the three components of commitment will develop as the function of quite different antecedents and have different implications for work-relevant behavior. Empirical evidence suggests that this holds true. Meyer et al. (1989) found that AC correlated positively with supervisors' rating, whereas CC correlated negatively. Allen and Smith (1987) and Meyer and Allen (1986) found that self-report measures of citizenship behavior correlated positively with measures of affective and normative commitment but not with CC.

Antecedents of employee commitment to organizational change

As we show in the proposed model (Figure 1), the three components of employee commitment to organizational change are the focal constructs. Noble and Mokwa (1999) considered two categories of antecedents of commitment: strategy factors and role factors. These categories of antecedent factors were derived from their grounded theory work. They found support for these factors as antecedents of commitment. We include constructs representative of those categories (fit with vision and role autonomy) as well as two other types of antecedents: employee-manager relationship quality and job motivation. These antecedents emerged via a series of academic presentations, corporate workshops, and qualitative interviews. We solicited input on categories of antecedents of employee commitment and specific constructs that could serve as examples of these categories. We received input at a series of national and regional academic presentations. In addition, we conducted workshops and depth interviews with industry employees and managers experiencing change in their organizations. We sought individual employees' and managers' perceptions of relevant antecedents and reactions (outcomes) to organizational changes.

In general, we posit that fit with vision, employee-manager relationship quality, job motivation, and role autonomy are positively related to AC2C and NC2C. However, we propose a negative association with CC2C. Higher levels of CC2C are often viewed as undesirable (Sinclair et al., 2005). Indeed, studies frequently have shown that this form of commitment is negatively related or unrelated to such desirable outcomes such as job performance (Meyer et al., 2002). CC2C, as Herscovitch and Meyer (2002) conceptualized it, reflects an external pressure to support a change and a perceived cost or risk associated with not supporting the change. Further, they reported a negative correlation between AC2C and CC2C. This is different from an employee's internal desire (AC2C) or sense of obligation (NC2C) to support the change, thus the proposed effects. Hence, for each of the following hypotheses sets, we expect the effect of the proposed antecedent on CC2C to be opposite (negative) of the effect on AC2C and NC2C. We share a more detailed discussion of the antecedents of C2C subsequently.

Perceived fit with vision

Noble and Mokwa (1999) defined fit with vision as the degree to which a strategy being implemented is seen as congruent with the overall direction of the organization. People commit to an organization because they want to maintain organizational membership to facilitate its goals (Blau and Boal, 1987). An employee's perception that a change initiative is consistent with an organization's vision enables them to commit to the change. Noble and Mokwa found a significant association between fit with vision and manager strategy commitment. Dvir et al. (2004) provided evidence that forming a vision and working with employees to share in the vision is positively related to AC to the organization and is unrelated to CC. In their meta-analysis of the antecedents, correlates, and consequences of AC, NC and CC to the organization, Meyer et al. (2002) note that correlations involving work experience variables were generally much stronger than those involving personal characteristics. As expected, these variables correlated most strongly with affective commitment. In all cases, the sign of the correlation involving CC was opposite to that for affective and normative commitment. We posit:

H1A. Fit of the change with strategic vision is positively related to AC2C.

H1B. Fit of the change with strategic vision is positively related to NC2C.

H1C. Fit of the change with strategic vision is negatively related to CC2C.

Relationship quality: satisfaction, commitment, and trust

An employee's relationship to the organization shapes his or her interpretation of its actions (Rousseau, 1995). During planned change, the quality of employment relationships plays an important role in promoting employee acceptance and involvement (Pfeffer, 1994). Organizations are beginning to understand the value of intraorganizational relationships (Menon et al., 1996). In fact, leader-member exchange theory suggests that leaders influence organization members by developing social exchanges (Erdogan et al., 2006). Indeed, the quality of these complex relationships affects the overall health and well-being of the organization, its stakeholders, and, especially, its customers. Building on prior studies, Henning-Thurau et al. (2002) found that there is general agreement about three components of relationship quality which are treated as interrelated rather than independent components: satisfaction with the firm, commitment to the relationship to the firm, and trust in the firm. Research has shown that employees who believe that their managers are supportive tend to be more committed to their organizations (Johnson et al., 1990). Thus, we posit that, while employees are engaged in multiple types of workplace relationships, the relationship between worker and manager is of particular importance in the context of C2C.

Commitment is defined as the belief that a relationship warrants efforts to maintain it (Morgan and Hunt, 1994). Trust is described as confidence in a person's reliability and integrity (Morgan and Hunt, 1994). Satisfaction is an employee's judgment that organizational relationships are rewarding and fulfilling (Churchill et al., 1985). Employees highly satisfied with their relationships with their managers are likely to develop AC2C and NC2C. Satisfaction has been shown to lead to desirable employee behaviors, such as work performance (van Dyne and Pierce, 2004). Morgan and Hunt discussed multiple outcomes of trust and commitment, such as acquiescence (adherence to another's requests) and cooperation (working together to achieve mutual goals). Further, commitment leads employees to exert effort toward achieving goals (Hartline et al., 2000). Trust in management builds credibility and acceptance for employees who are faced with organizational change (Rousseau and Tijoriwala, 1999). Employees engaged in quality relationships (i.e. those with high levels of satisfaction, commitment, and trust) are likely to be committed to the organization's strategies. We posit that the quality of the relationship between employees and their managers is a significant predictor of employee commitment to organizational change. Thus, we hypothesize:

H2A. Employee-manager relationship quality is positively related to AC2C.

H2B. Employee-manager relationship quality is positively related to NC2C.

H2C. Employee-manager relationship quality is negatively related to CC2C.

Job motivation

Motivation is an energizing force with implications for behavior (Meyer et al., 2004). We believe that job motivation also influences attitudes. Specifically, it has implications for employee commitment to organizational change. In fact, there is a growing body of literature that supports the positive relationship between job motivation and employees' attitudes toward change (Coopey and Hartley, 1991). Paton and McCalman (2000) contended that organizational success is generated by motivated people. Furthermore, the greater the job motivation the greater the likelihood of employee commitment to the organization (Mathieu and Zajac, 1990; Thorsurd, 1972) – and to organizational change. Thus, we posit that employees experiencing job motivation develop AC2C and NC2C and positive feelings prompted by job motivation negatively influence employees' CC2C:

H3A. Job motivation is positively related to AC2C.

H3B. Job motivation is positively related to NC2C.

H3C. Job motivation is negatively related to CC2C.

Role autonomy

People act autonomously when they believe that they have the freedom to choose and initiate their actions (DeCharms, 1968; Deci, 1875). Role autonomy refers to the extent to which one has freedom to make job decisions and adjust behaviors accordingly (Noble and Mokwa, 1999). A manager who grants employee autonomy is perceived as sharing control. Under conditions of greater autonomy, employees tend to have greater C2C (Dodd and Ganster, 1996; Thorsurd, 1972). Employees who have a sense of autonomy about their role in implementing workplace change initiatives (e.g. feel they influenced training or roll-out schedules) will develop AC2C and NC2C and reduce feelings of CC2C. Thus:

H4A. Role autonomy is positively related to AC2C.

H4B. Role autonomy is positively related to NC2C.

H4C. Role autonomy is negatively related to CC2C.

Outcomes of employee commitment to organizational change

There are few empirical studies in which researchers attempt to link change and organizational outcomes. According to Meyer and Allen (1991), normative commitment is positively associated with organizational outcomes but to a much lower extent than affective commitment. Meyer et al. (2002) found regarding the consequences of commitment that AC has the strongest positive correlation with desirable work behaviors, followed by normative commitment; CC is unrelated or negatively related to these behaviors. Our choice of desirable outcome variables (as suggested by the literature and our qualitative work) suggests that it would be likely that CC2C would not have a negative impact but would have no impact. Thus, we posit that AC2C and NC2C are significantly associated with change-related outcomes. We also expect interrelationships to exist among the outcomes. Noble and Mokwa's (1999) model included role performance and implementation success as outcomes of commitment. Their model included both individual-level and organizational-level outcomes. We included perceptions of performance (at the organizational level) and implementation success as outcomes of employee commitment to organizational change. Further, organizations that learn from their experiences are more likely to be successful than those that continue to repeat prior mistakes (Teare and Rayner, 2002). We found additional support to include individual learning during our organizational workshops on change; therefore, we included individual learning as an individual-level outcome in our model.

Perceived improvements in organizational performance

Improvements in organizational performance involve employee perceptions about financial and non-financial effects of organizational performance (Homburg et al., 2002). Non-financial performance includes variables such as customer satisfaction, customer loyalty, and market share (Menon et al., 1996). Financial performance includes measures for return on sales, return on investment, and return on assets (Chakravarthy, 1986). Paton and McCalman (2000, p. 151) noted: “Effective performance comes from having a committed workforce”. Indeed, Meyer and Allen (1991) found a relationship between AC to the organization and higher levels of performance. We expect that AC2C and NC2C positively influence an employee's perceptions of performance. Thus:

H5A. AC2C is positively related to perceived improvements in performance.

H5B. NC2C is positively related to perceived improvements in performance.

Perceived implementation success

Noble and Mokwa (1999, p. 60) identified implementation success as a primary outcome of commitment to a strategy and defined it as “the extent to which an implementation effort is considered successful by the organization”. Although researchers often evaluate strategic success from the viewpoints of managers (Noble and Mokwa, 1999), it is valuable to consider nonmanager perspectives as well. Implementation can be effective only when employees are committed (Paton and McCalman, 2000). In fact, Conner and Patterson (1982, p. 18) labeled the lack of employee commitment as the “most prevalent factor contributing to failed change projects”. We posit that AC2C and NC2C are positively related to implementation success, which is positively related to improved performance:

H6A. AC2C is positively related to perceived implementation success.

H6B. NC2C is positively related to perceived implementation success.

H6C. Perceived implementation success is positively related to perceived improvements in performance.

Individual learning

Learning has been defined as a knowledge-creation process in which information interpretation leads to a change in behaviors (Lehesvirta, 2004). Learning is part of the change process. For example, Gibb and Scott (1985) found that firms change by solving problems as they arise and by learning from the problem-solution process. Employees committed to change efforts are more likely to learn from the process. Loyal employees want to contribute to and see the results of their efforts, and they can do so through learning (Teare and Rayner, 2002). When employees learn from being involved in an organizational change, they consider that learning to affect the success of the implementation. Thus, we posit:

H7A. AC2C is positively related to individual learning.

H7B. NC2C is positively related to individual learning.

H7C. Individual learning is positively related to perceived implementation success.

Method

Data collection

We recruited a transportation services department of a large public US university to participate in the study. The department was established about 20 years ago and has grown significantly. Starting with 16 budgeted employees, the full-time staff now numbers 108 with a current budget of $26 million. Owing to enrollment growth and campus expansion, the department continues to go through changes to meet the demands of a dynamic camp us. Some of the departmental services include parking administration and enforcement, shuttle and charter bus service, and traffic planning and control. We chose this organization because the department is fairly large, with approximately 600 full and part time employees and because it was undergoing several changes at the time of data collection (which is typical of many organizations). An e-mail message from the director encouraged participation among its 593 employees. Data were collected through a web survey. As a reward, respondents received a $5 restaurant certificate. Two follow-up e-mails were sent to thank respondents and to encourage others to respond. Data collection continued for three weeks and generated 191 responses (32 per cent response rate). Seven responses were removed from the analysis due to missing data. Of the respondents, 53 per cent were under age 25, 47 per cent were full-time employees, 51 per cent were female, 75 per cent were contact staff, and 15 per cent were managers.

Rousseau and Tijoriwala (1999) noted that managerially driven changes do not necessarily constitute a single event or unified message. Rather, change recipients must interpret a complex array of facts regarding the change. Thus, it was important that we obtain information about the changes our respondents were experiencing at the time of our data collection. During the survey administration, we elected to follow the procedure that Herscovitch and Meyer (2002) described and to collect written verbatim descriptions in response to the following statement: “Please describe a recent or ongoing organizational change that has had an impact on the way that you perform your job.” Our participants provided 237 verbatim responses about change in their organization. Approximately, 31 per cent of the respondents mentioned two or more changes. Employees primarily focused on organizational changes regarding such broad issues as service process redesign and technology implementations. Examples of change descriptions include:

“The combination of several groups into one to better serve our customers and cut costs.”
“We are in the process of linking all the garages. There are a lot of cameras and monitors to get familiar with.”
“Online citation payment and appeals has made a positive impact on our customers.”

Most of the respondents (73 per cent) indicated that the changes they described occurred within the past nine months and considered the change significant (M=5.28 on a seven-point scale). Each respondent indicated a change situation and responded to questions with that situation in mind.

Measure development

Our model variables are established in the literature and we adapted existing measures for our context. We used only nonfinancial items from the improved organizational performance scale because we collected our data from a nonprofit entity. We measured commitment to the manager using items from Morgan and Hunt's (1994) relationship commitment scale, and we measured the three components of C2C using items from Herscovitch and Meyer's (2002) scales. The questionnaire was pretested with a large organization, in which 190 employees completed the survey online. We subjected the pretest data to exploratory and confirmatory factor analyses (CFA) as well as reliability analysis. We made several revisions based on this pretest and the resulting measures, their sources and reliabilities, and the measurement model loadings from the main study appear in the Appendix.

Nonresponse bias

Though our sample provided a good representation of the department staff, it was underrepresented in terms of young, part-time employees. The staff consists of 80 per cent part-time workers, and 73 per cent are 25 years of age or younger. Further analysis shows that our sample is underrepresented in one unit: transit (young, part-time student bus drivers). Employees in this unit make up 53 per cent of the staff but only 32 per cent of our sample. Other departments (e.g. customer service, maintenance, parking services) were all fairly accurately represented. Our sample consists of 59 transit respondents. We compared the first 29 responses with the last 29 (the middle response was excluded) and found no significant differences among any of our model variables. Therefore, although one group (transit) was underrepresented in our sample, we do not believe nonresponse bias is a concern.

Results

The means, standard deviations, intercorrelations, variances, and covariances for all the constructs in this study appear in Table I. Note that the standard deviations for the ten variables, which are measured by seven-point multi-item scales, range from 1.83 to 2.45 (M=2.09), indicating a substantial amount of variance in the responses. The covariance matrix was the input matrix for measurement and structural model analyses. Convergent validity was supported in that estimated standardized loadings were significant (p<0.001); t-values ranged from 9.40 to 18.85. Furthermore, the variance extracted estimates (Table I) for all the multi-item measures ranged from 0.575 to 0.942, exceeding the suggested minimum levels (Fornell and Larcker, 1981).

To determine discriminant validity, we followed the procedure recommended by Fornell and Larcker (1981). Discriminant validity is supported when the variance extracted estimates of each of the constructs involved in the highest estimated correlations exceed the square of the correlation between the constructs. The largest correlation, between implementation success and improved performance, was 0.786. The variance extracted estimate for implementation success was 0.792, and for improved performance, it was 0.863; therefore, both exceeded the square of the largest correlation (0.786²=0.618), in support of discriminant validity. We also compared all construct pairs with the average variance extracted (AVE) estimates. In each case, the AVE estimate exceeded the square of the correlation, thus providing additional support of discriminant validity. Cronbach's α values ranged from 0.838 to 0.975 (Appendix), and composite reliabilities ranged from 0.844 to 0.993 (Table I), indicating reliability of the measures.

Measurement model results

We assessed measurement properties through CFA and included the items for fit with vision, commitment to manager, trust in manager, satisfaction with manager, job motivation, role autonomy, implementation success, individual learning, AC2C, CC2C, NC2C, and improved performance. One item for CC2C demonstrated an unacceptable factor loading (0.50) and squared multiple correlation (SMC; 0.25), and was removed. Loadings greater than 0.6 are considered acceptable (Bagozzi and Yi, 1988) as are SMC item values greater than 0.4 (Taylor and Todd, 1995). Thus, our measurement model consisting of 39 items measuring 12 constructs resulted in acceptable model fit (χ ² ₍₆₃₆₎=1,182.35, p=0.000; root mean square error of approximation [RMSEA]=0.058; standardized root mean square residual [SRMR]=0.07; nonnormed fit index [NNFI]=0.98; and comparative fit index [CFI]=0.98). All item loadings in the completely standardized solution exceeded 0.6, and SMCs were greater than 0.4.

Structural model analysis

We analyzed the hypothesized paths through structural equation modeling with the measures that resulted from the CFA described previously. We created composite indicators for each of the commitment components by averaging the items. Consistent with Herscovitch and Meyer's (2002) conceptualization, we included correlational paths among the commitment components. We created a composite indicator for employee-manager relationship quality by computing average composites for commitment to manager, trust in manager, and satisfaction with relationship with manager and then averaging the three composites. Error variances were fixed on the basis of construct reliabilities. We created a covariance matrix using the composites for these constructs and individual items for the others.

The results of this structural analysis appear in Table II. The proposed structural model had the following goodness-of-fit results: χ ² ₍₂₀₅₎=476.94, p=0.000; RMSEA = 0.082; SRMR = 0.085; NNFI = 0.97; and CFI = 0.97. Each of our seven hypotheses was at least partially supported by our data. Assessing the results in terms of paths, 14 of the 20 hypothesized paths were significant and in the hypothesized direction. Three antecedent paths – job motivation→to NC2C, job motivation→CC2C, and role autonomy → NC2C – were not significant. In addition, one path was significant but not in the hypothesized negative direction: fit with vision→CC2C. Two outcome paths – NC2C→implementation success and NC2C → improved performance – were not significant. The SMCs for this model were as follows: AC2C: 0.64; CC2C: 0.16; NC2C: 0.48; individual learning: 0.52; implementation success: 0.72; and improved performance: 0.75.

We removed the nonsignificant paths from the model, including the nonsignificant correlation between AC2C and NC2C. Thus, AC2C was correlated with CC2C, and CC2C was correlated with NC2C. This model respecification resulted in a more parsimonious model with acceptable fit (Table II). The respecified structural model had the following goodness-of-fit results: χ ² ₍₂₁₁₎=482.35, p=0.000; RMSEA = 0.082; SRMR = 0.087; NNFI = 0.97; and CFI = 0.97. The SMCs for this model were as follows: AC2C: 0.64; CC2C: 0.17; NC2C: 0.47; individual learning: 0.52; implementation success: 0.72; and improved performance: 0.75.

Summary of hypotheses testing

As we hypothesized, fit with vision was positively related to AC2C and NC2C (H1a and H1b, respectively). However, contrary to our hypothesis, it was positively rather than negatively related to CC2C (H1c). As we expected, employee-manager relationship quality was positively related to AC2C (H2a) and NC2C (H2b) but was negatively related to CC2C (H2c). Job motivation was significantly related to AC2C (H3a) but not as significantly related to the other commitment dimensions (contrary to H3b and H3c). Role autonomy was positively related to AC2C (H4a) and negatively related to CC2C (H4c), but it was not related to NC2C, contrary to H4b. Thus, the results showed that eight of the 12 hypothesized paths related to the proposed antecedents of commitment were supported, while four were not, and that the antecedent factors had differential effects on the commitment dimensions.

Regarding the outcome variables, AC2C was positively related to improved performance (H5a), implementation success (H6a), and individual learning (H7a). NC2C was significantly related only to individual learning (in support of H7b, but contrary to H5b and H6b). These results highlight the importance of AC2C and are consistent with prior research. Regarding the relationships among our outcome variables, implementation success was positively related to improved performance (H6c), and individual learning was positively related to implementation success (H7c). Thus, of the hypothesized paths related to the proposed consequences of commitment, six were supported and two were not.

Discussion

Motivations underlying employee acceptance of organizational change are important. However, these motivations have received little research attention (Rousseau and Tijoriwala, 1999). In this study, we attempt to examine this important area more closely and contribute to the understanding of why employees C2C and what that commitment can mean for the organization. The results of our model testing provide useful information for both managers and academicians. The data generally support our premise that employee commitment to organizational change has important consequences. We found that of the three commitment components, AC2C has the most significant effects on the outcome variables included in our model. Such a finding is consistent with previous studies. For example, Sinclair et al. (2005) found that at least a moderate level of AC is necessary to achieve desired outcomes. In addition, we show that select antecedents play an important role in influencing employee commitment to organizational change.

Theoretical implications

Organizations are constantly changing, and it is important for researchers to provide insight into opportunities to improve the potential for success of those changes. We demonstrate that employee C2C is influenced by several antecedent variables and that it influences employee outcomes. We also show that AC2C is of considerable importance, perhaps more than the other components in Herscovitch and Meyer's (2002) model. More work is necessary to fully understand the role of employee C2C in improving the likelihood of successful change initiatives; however, we provide initial empirical evidence that employee C2C is a variable worthy of additional study. Our findings extend and support the work of Noble and Mokwa (1999) through the testing of part of their model with employees of a service organization, thus evaluating the importance of C2C with a wider range of employees and in a different context. In addition, by using Herscovitch and Meyer's (2002) three-component model of C2C, we found that the different components of commitment are affected by various antecedents. For example, job motivation influenced AC2C, but not the other commitment components. Thus, future models of strategic change should consider including the three components of C2C in order to more fully understand the role of commitment.

Managerial implications

A recent Harvard Business School study of Fortune 100 firms that had each invested $1 billion in organizational change initiatives over a 15-year period shows that only 30 per cent of those initiatives improved bottom-line results that exceeded the companies' cost of capital, and only half resulted in an improved market share price (Pascale et al., 1997).

Based on our findings, we offer several insights to managers that could help improve change implementation efforts. Improving employee C2C initiatives can have a profound impact on the success of company efforts. For example, high levels of employee loyalty have been linked to an estimated 11 per cent boost in productivity (Mayfield and Mayfield, 2002). We found that AC2C influences important perceptions, namely, individual learning, improved performance, and implementation success. NC2C had some influence as well, but to a lesser degree than AC2C; again, this is consistent with prior research. Managers need to be aware that employee C2C has effects beyond the specific implementation of a change initiative. In particular, AC2C influences important attitudes about the change, such as perceptions of implementation success and improved performance. This is important because all employees should present a positive view of the change to customers and other constituents. In addition, both AC2C and NC2C to change are positively associated with individual learning, which should be beneficial for future change initiatives.

In terms of building employee C2C, our findings showed that fit with vision is positively related to all three forms of C2C. Thus, organization leaders must strive to demonstrate to employees how any change implementation fits with the firm's larger strategic picture. Employees who understand that congruence are more likely to be committed to the change, which ultimately influences important employee perceptions. Open and timely communication about the reason for change initiatives should be part of the implementation plan.

Furthermore, we found that employees who have quality relationships with their managers feel more desire and duty to support an organizational change. However, this relationship decreased CC2C. Managers should aim to create relationship commitment, trust, and satisfaction among their employees. In addition, if senior leaders manage employee attitudes, they will likely influence employee predispositions toward change (Albrecht, 2005). Successful managers will pay attention to factors such as investment in people, technology that supports frontline workers, revamped recruiting and training practices, and compensation linked to performance for employees at every level (Heskett et al., 1994). Managers need to understand that investment in internal relationship management has the potential for important payoffs. Organizations should assess, both formally and informally, employees' feelings about relationships and learn how to better nurture and develop them. Of course, it is imperative that management act on what they learn.

We also found that job motivation influences only AC2C and not the other components of commitment. Herscovitch and Meyer (2002) argued that AC2C is a stronger binding force than NC2C or CC2C. The latter are more likely to conflict with the perceived short-term self-interests of some employees (Meyer et al., 2004). In addition, given AC2C has the greatest influence on outcomes such as implementation success and improved performance, the antecedents of AC2C warrant management's attention. Motivating employees who are anticipating an organizational change should be part of the implementation plan for the change. When employees are motivated to support the change and to develop AC2C, outcomes such as individual learning, implementation success, and improved performance are enhanced.

In addition, managers must recognize that when employees feel a sense of role autonomy, they are more likely to commit positively to workplace change. When managers plan for the change implementation, they should consider when it might be practical to give employees some decision-making authority and control over their part of the implementation process.

Research limitations and directions for further research

Our context of a nonprofit organization employing a large number of part-time employees provides an important initial test of our model; however, other contexts should be studied. For example, are there differences in the significant antecedent factors of C2C in a for-profit or a manufacturing context? To determine the generalizability of our model and identify potential boundary conditions, the model should be tested in a variety of industry and organizational contexts and with different types of change initiatives. Because our model items were collected from the same data source, common-method bias is a potential concern. However, our measurement model findings provided evidence that our variables were distinct. This was also true of our pretest data. Additionally, our data are cross-sectional. Longitudinal studies could provide greater understanding of the causal relationships among commitment to organizational change and its antecedents and outcomes.

We included four antecedents in our empirical model. However, there are likely to be other antecedents that influence C2C. For example, Noble and Mokwa (1999) studied buy-in and strategy importance which are similar to the antecedent fit with vision and could be associated with C2C. Together, fit with vision, buy-in, and importance could be considered a set of “change factors” that influence commitment. Although we included the quality of employees' relationship with their manager as an antecedent to C2C, there are other relationships that could play a role. For example, relationships with coworkers or customers might influence employee C2C. Manager relationship quality (MRQ) could be considered one “relationship factor.” With regard to individual characteristics, such as motivation, there are others that could be considered, such as risk acceptance or innovativeness. Again, these variables, along with job motivation, might be grouped into a set of “personal factors” influencing C2C. Other items similar to role autonomy, such as involvement or significance (Noble and Mokwa, 1999), might also impact commitment to organizational change. These “role factors” may individually or holistically influence C2C. Future research could extend this work by testing all of the antecedents included in Noble and Mokwa's (1999) model and could build on the antecedent factors we have tested to develop a larger conceptual model of the potential antecedents of the three components of commitment. More information regarding how to build commitment, especially AC2C, could be useful for managers.

Further, we did not explore the role of firm characteristics, such as corporate culture (Schneider and Bowen, 1993). Of particular interest might be the evaluation of the role of service climate or a culture for change. Another area important for future exploration includes manager characteristics related to leadership and communication. These could be potentially rich areas for future study in the context of organizational change and commitment to that change.

Conclusion

The fundamental purpose of our study was to draw scholarly attention to and to demonstrate empirically the role of employee C2C in the perceived success of the implementation of strategic initiatives. To that end, we reviewed literature from a variety of disciplines and proposed a model of antecedents and outcomes of the three components of employee C2C. The results from a test of our model suggest that employee C2C, particularly AC2C, is important. Thus, the means to success in implementing change initiatives include choosing the right strategies and effectively implementing them (Zeithaml et al., 2005). A key to effective implementation appears to be in the development of AC2C. Our results offer insights for managers and identify opportunities for future empirical research on change initiatives in organizations.

Figure 1Hypothesized empirical model of antecedents and consequences of employee commitment to organizational change

Table ICorrelation/covariance matrix

Table IIStructural model results

Table AI

References

Albrecht, S. (2005), "Leadership climate in the public sector: feelings matter too!", International Journal of Public Administration, Vol. 28 pp.397-416.