MASTER THESIS
The relationship between participative and autocratic leader behaviors and commitment and support for organizational change and the impact of employee core
self-evaluation
Author: Eline Verbruggen Student number: 11111941
Amsterdam Business School (ABS) Faculty of Economics and Business
MSc Business Administration – Leadership and Management Track Thesis supervisor: dhr. dr. Merlijn Venus
Statement of Originality
This document is written by Student Eline Verbruggen who declared to take full responsibility for the content of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the content.
_________________________________________________________________ Abstract
Leader behavior is often assumed to play a significant role in an organizational change context. As a consequence, it is important to understand how leader behavior impacts employee support for change when taking employees’ core self-evaluations and their affective commitment to change into the picture. The data of the current study was gathered from 106 leader-employee dyads asked to fill in surveys in either German, Dutch, Hungarian, or English. This study investigates how participative and autocratic leader behavior impact change-supportive behavior and also the moderating effect of core self-evaluation. Further, the mediating effect of affective commitment to change on the relationship between the two leader behaviors and change-supportive behavior is explored. Results showed that no direct effect of either leader behavior on change-supportive behavior was found, and in addition, no moderation effect of core self-evaluation on this relationship was found. Partial mediation of affective commitment to change was found between the leader behaviors and change supportive behavior. These findings contribute to the understanding of how participative and autocratic leader behavior impacts affective commitment to change, and respectively increases and decreases support for change. This represents not only a basis for future research but to managerial decision-making during organizational change.
Keywords: participative leader behavior; autocratic leader behavior; core self-evaluation; affective commitment to change; change-supportive behavior
Table of Contents
1. Introduction 1
2. Literature review 3
2.1 Behavioral support for Change 5
2.2 Leader behaviors 7
2.2.1 Participative leader behavior 7
2.2.2 Autocratic leader behavior 10
2.3 Core self-evaluation 12
2.4 Affective commitment to Change 15
2.5 Derivation of Propositions 16
2.5.1 Leader behavior and change-supportive behavior (CSB) 16
2.5.2 Moderating role of CSE 17
2.5.3 Mediating role of affective commitment to change (ACC) and interaction
effects 18
3. Research Method 20
3.1 Procedure 20
3.2 Sample 21
3.3 Measures 22
4. Data Analysis and Results 23
4.1 Data analysis 23 4.2 Results 24 4.2.1 Descriptive statistics 24 4.2.2 Direct effects 26 4.2.3 Moderation 26 4.2.4 Mediation 27 4.2.5 Moderated-Mediation 29 5. Discussion 29 5.1 Theoretical implications 31
5.2 Limitations and directions for future research 33
5.3 Practical implications 34
5.4 Limitations and directions for future research 23
6. Conclusion 35 References 36 Appendix A – Surveys 41
1. Introduction
Managing organizational change is crucial to business survival. Not only an organization’s external environment is very dynamic and rapidly changing but also the way in how businesses operate and how they compete depends on an organization’s ability to change and its successful implementation. Work processes often need to be adapted or even completely reinvented so that the organization can stay ahead of competition. For instance, Makro Netherlands announced in February this year its reorganization plans, which also includes the transformation of 17 stores in The Netherlands. Makro Netherlands wants to redirect the organization in order to be successful in the long-term. Unfortunately, many employees will loose their job, which simultaneously, leads to changing job responsibilities within the organization.
The reasons for why organizational changes happen are either factors from outside the organization (external factors) or factors from within the organization (internal factors). Due to these factors businesses are constantly changing (Goodman & Loh, 2011). Nevertheless, the change can vary in terms of the extent and level from one business to another. However, the fact that people are involved in a change process cannot be denied and is crucial to recognize, since it all depends on them— what and how they do it. Hence, it is reasonable to assume that employee (leader and subordinate) attitudes and behaviors are one of the most significant factors in determining the success of an organizational change.
Scholars have investigated leadership and organizational change (e.g. Herold, Fedor, Caldwell, & Liu, 2008; Higgs & Rowland, 2005), however the literature is not clear in telling if leadership influences organizational change (Burke, 2008), which behaviors are exhibited by leaders who engage in change (Higgs & Rowland, 2005) or are critical to successful change. According to Herold et al. (2008) there are two approaches—namely the leadership style approach and organizational change approach—in how leadership shapes subordinates’ responses to change. The first assumes that specific leadership types are better able to deal with change situations as compared to others. On the contrary, the latter assumes that any leader can engage in decent and specific change-related behaviors. Since there has been done little integration of these two literature bodies, the question arises what kind of behaviors a leader should exhibit in order to trigger change-supportive behaviors from subordinates that are of ultimate significance for change success. Moreover, Herold et al. (2008) state that subordinates gauge the amount of effort and support towards an
initiative depending on how they view their leader. In their study they researched the relationship between transformational leadership and organizational change and found that those leaders appear to receive more subordinate input. In order to investigate how organizations can increase subordinates’ support for change, researchers have been interested in concepts including readiness for change (Armenakis, Harris, & Mossholder, 1993), commitment to change (Herscovitch & Meyer, 2002), openness to change (Miller, Johnson, & Grau, 1994), and cynicism about organizational change (Reichers, Wanous, & Austin, 1997).
The way in which leaders behave during organizational change might be appreciated by some subordinates but not valued by others depending on the subordinates’ personality. For instance, there may be subordinates who would like to be included and involved in decisions that need to be made during the change process. Others, however, might prefer to be directed by their leader throughout the change process instead of having to participate actively. Thus, subordinates might react differently to change depending on the behaviors exhibited by their leader, and additionally, on their own personality. This might lead to varying degrees of commitment to and support for the organizational change. Northouse (2012) argues that subordinates need different leadership styles, such as participative or directive leadership styles, based on their characteristics. However, like this, it has not been related to the literature on organizational change. The concept most often used in these studies to measure subordinate traits is core self-evaluation (CSE) (Chang, Ferris, Johnson, Rosen, & Tan, 2012). Chang et al. (2012) reviewed the literature on CSE and showed that the concept has been studied in relation to many other concepts such as satisfaction, commitment, motivation, and performance. Nevertheless, no study could be found on how CSE might influence the relationship between leader behaviors and subordinates’ reactions to change. Even though empirical evidence of a relationship between subordinates’ commitment to change and their support for the change initiative was found (Herscovitch & Meyer, 2002), this has not been studied in relation to leader behaviors and the influence of CSE.
For these reasons this study examines how participative and autocratic leader behaviors, subordinate characteristics, and commitment to change influence ultimately the behavioral support for the organizational change. More specifically, it investigates the relationship between the two opposing—participative and autocratic—leader behaviors and subordinates’ commitment to change looking at how
CSE might strengthen or weaken this relation. Thereby, it explores if certain combinations of leader behaviors with subordinate characteristics influence the support displayed by subordinates.
This research contributes theoretically to the existing literature in the fields of leadership and organizational change by integrating the two literature bodies to explore if subordinates exhibiting certain traits show higher commitment to and show more supportive behaviors for a change, when leaders engage with their subordinates and asks for their participation, or when leaders behave autocratically excluding their subordinates and not asking for their opinion. Many studies found evidence of a main effect of participation, however, not much is known about how different persons react on being involved, for instance, in decision-making by their leaders. Additionally, this research also contributes to the practical and managerial understanding of how commitment, and ultimately, support for change can be enhanced or degraded when considering subordinates’ CSE in combination with leader behaviors that either focus on participation or delegation.
First, this paper provides a literature review on organizational change and support for change, participative and autocratic leader behaviors, core self-evaluation, and commitment to change, whereupon hypotheses are derived to close the above mentioned research gap. Next, the methodology section describes the research procedure, the sample, and how the various concepts are measured. Subsequently, the data analysis and the results are presented, followed by a discussion of the results. The discussion includes a short summary of the results, theoretical as well as practical implication, limitations of the current study, suggestions for future research, and finally a brief conclusion.
2. Literature review
Today’s businesses face tremendous pressures from various factors. One might argue that globalization is one key driver of many changes as it has a significant impact on economies, and as a consequence, on each and every organization. Changes in the macro-economy drive changes at private and public levels (Goodman & Loh, 2011). Furthermore, change is a constant factor influencing how businesses operate. Therefore, some changes can affect an organization partially whereas others affect the whole entity depending on the nature of the change. Barnett & Carroll (1995) define organizational change as “a transformation of an organization between two points in
time” (p. 219). However, many changes are not successfully implemented. To make a change successful several factors need to be considered, namely content, process, context and individual differences (Walker, Armenakis, & Bernerth, 2007).
The change being implemented refers to the change content, which is distinctive for each organization. Walker et al. (2007) distinguish between fundamental and incremental change, wherein the former tries to respond to environmental demands (e.g. a new regulation imposed by the government), and the latter takes a gradual approach to reach an ideal set by the organization. According to Barnett and Carroll (1995) change content means that a change can be of incremental or radical nature. Further, Dunphy and Stace (1988) differentiate between incremental/evolutionary change and transformative/revolutionary change. Weick and Quinn (1999) again distinguish continuous change and episodic change. Continuous change is defined to comprise of “changes that tend to be ongoing, evolving, and cumulative” (p.375). This type of change often describes endless modification and a path of improvisation, since patterns emerge over time. Episodic change, in contrast, happens from time to time and often intentional.
Change content is very specific for each organization, but in addition, “the actions taken by change agents during the introduction and implementation of the proposed change” (Walker et al., 2007, p. 762) are highly specific and geared to an organization. Barnett and Carroll (1995) describe change process in how the change occurs, for example with what speed or in which order activities are executed. The fact that change agents are crucial for change success indicates that people play a key role in the process. Employees need to be prepared and ready for the proposed change. “Creating readiness involves proactive attempts by a change agent to influence beliefs, attitudes, intentions, and ultimately the behaviors of a change target“ (Armenakis et al., 1993, p. 683). These actions and decisions taken by change agents need to be viewed positively by other employees in order to counteract any potential resistance (Armenakis & Bedeian, 1999). One attempt to do this is by articulating a change message that includes a component on discrepancy, appropriateness, efficacy, principal support and personal valence (Armenakis, Harris, & Field, 1999). Nevertheless, the change determines the degree of resistance and the receptiveness of employees as it depends on how the change affects the employees’ self-interest (Clarke, Ellett, Bateman, & Rugutt, 1996). If it is threatened by the change the employee is more likely to resist the change even though they view the
change as necessary and beneficiary to the organization as a whole. Other employees may react resistant because the organizational identity may be threatened (Jacobs, Christe-Zeyse, Keegan, & Pólos, 2008). This means that some culturally unacceptable behaviors might result from the change processes whereupon employees generate intense reactions towards the change initiative in order to protect the identity.
In contrast to change content and change process, contextual issues arise due to factors that do already exist prior to the change, which can be categorized into internal and external contextual factors (Walker et al., 2007). The issues arising from the organization internally, for example, can be with regard to attitudes toward change or any type of resources such as a shortage of key people. This implies that internal contextual factors can be controlled from the organization itself. Contrary, contextual factors arising from the external environment are not in immediate control of the organization meaning that the organization needs to adapt and respond to those demands resulting into changes.
Lastly, differences between individuals in terms of their personality can impact the overall attitudes and behaviors of an organization, and hence, reactions to change which again might influence behaviors related to support change.
2.1 Behavioral support for Change
The existing literature on organizational change most often does not give an extensive explanation of outcome variables, and according to Weiner (2009) “outcomes are perhaps the least theorized and least studied aspect of organizational change” (p. 71). The same accounts for support-related behaviors. Nevertheless, some scholars (Herscovitch & Meyer, 2002; Kim, Hornung, & Rousseau, 2011) defined what they mean by supportive, change-related behavior.
Herscovitch and Meyer (2002) developed a continuum on which behavioral support for a change can be reflected. The continuum starts on the left with active resistance and proceeds to the right with passive resistance, compliance, cooperation to championing. If employees want that a change initiative fails and overtly showing resistance and counteraction, they are by definition engaging in active resistance. If employees respond to a change by demonstrating covertly opposing behavior in order to prevent the success of the change initiative, they are by definition engaging in passive resistance. If employees support the change only to the least extent and only reluctantly go along with it, they are complying with the change. Cooperation by
employees goes a step further in that they demonstrate “support for a change by exerting effort when it comes to the change, going along with the spirit of the change, and being prepared to make modest sacrifices” (Herscovitch & Meyer, 2002, p. 478). Lastly, championing means that the person is behaving very enthusiastic about the change and shows this by telling others about the change and making infinitely more effort to make sure that the change will be a success.
Kim et al. (2011) however argue that the only change supporting and facilitating behavior of an active nature is championing. Change-supportive behavior (CSB) is defined as “actions employees engage in to actively participate in, facilitate, and contribute to a planned change initiated by the organization or, more precisely, the organization’s management” (pp. 1667-1668). This definition is distinguishable from existing constructs such as change-supportive psychological states (e.g. commitment to change), adaptive behavior, and proactive behavior by focusing on actual behavior, stressing active contributions to the change initiative, and supporting planned change. Scholars have argued that attitudes towards change are likely to influence an employees’ support or resistance towards change. Therefore, different but similar constructs of employees’ change-related attitudes have been investigated, that reveal either a positive or a negative judgment of the change (Choi, 2011). In his review, Choi (2011), compared frequently used key variables—readiness for change, commitment to change, openness to change, and cynicism about organizational change—that can be identified as cognitive antecedents of supportive or resistant behavior to change. Other scholars also argued for the importance of psychological ownership (Dirks, Cummings, & Pierce, 1996) in promoting change, and employees’ organizational attachment (Armenakis & Bedeian, 1999), which determines the effort they engage in to perform their work. Furthermore, as previously stated, people are crucial for organizational change. Change agents do not necessarily have to be leaders but it can be assumed that many are.
The relationship between leaders and organizational change is not clear (Herold et al., 2008), but nevertheless, researches argue that leaders occupy a key role in organizational change (Ford & Ford, 2012; Mangundjaya, Utoyo, & Wulandari, 2015). Herold et al. (2008) studied transformational leadership and change-specific leadership behaviors in the context of organizational change to determine how these influence employee reactions to change. Their findings indicate that leadership style does matter and that future work should investigate further how leadership influences
organizational change. Other scholars found evidence that leader behaviors as well as their characteristics do have an impact on change success or failure (e.g. Eisenbach, Watson, & Pillai, 1999; Higgs & Rowland, 2005). Yukl (2009) also included chapters on transformational and charismatic leadership and related leader behaviors to organizational change. Moreover, he includes a chapter on participative leadership, delegation, and empowerment but does not specifically put them into the context of organizational change.
Many organizations differ in how their leaders interact and work with their subordinates. Some leaders seem to be more authoritative compared to others, which might depend on their personality or on the task structure. But do these behaviors influence the behaviors of a subordinate in a change-related context? Do they ultimately lead to higher or lower commitment and hence, change-supportive behaviors? In order to answer these questions, this study focuses on participative and autocratic leader behaviors and how those might lead to change-supportive behaviors as until today researchers did not specifically look at this relationship. Therefore, the next sections elaborate on the two leader behaviors.
2.2 Leader behaviors
Based on early studies on leadership styles (e.g. Lewin & Lippitt, 1938) several dimensions of styles emerged, which are very similar to each other and have been built upon since then. These dimensions are participative versus directive and democratic versus autocratic leadership. This study assumes that participative and democratic, as well as directive and autocratic leaders show very similar behaviors, and thus are used interchangeably. The paper refers to participative and autocratic leader behaviors so make a clear distinction between the behaviors. The next section will first review the literature on participative leader behavior, and subsequently puts it in contrast to autocratic leader behavior.
2.2.1 Participative leader behavior
In an organizational setting, leaders can behave in diverse ways. Participative behavior means that leaders make decisions together with subordinates (Northouse, 2012). Subordinates are proactively invited and encouraged to participate, and consulted by the leader to share opinions and ideas. Further, their suggestions and recommendations to specific problems are incorporated into decision-making, and
hence, influence the path ahead of them. Whenever subordinates are involved in making decisions, they might approve and ultimately implement a decision a lot easier due to the political process (Yukl, 2009). Participative leaders do not only include their subordinates in decision-making but also show concern for them. After all this is said, participative behavior seems to be a very broad term. The question arises if one can exactly define participative behaviors or if the lines are blurry.
Yukl (2009) recognizes that decision procedures can range from autocratic decisions, to consultation, to joint decision-making, to delegation. In simplistic terms, making autocratic decisions means that others than the leader do not have any voice or influence. Thus, it is a contrasting concept to participation. If leaders consult their subordinates it means they are asking them for opinions and other perspectives. Nonetheless, the leaders make the decisions alone in the end. By comparison, joint decision-making means that leaders together with subordinates come to a decision. On the other side of the continuum is delegation. This managerial behavior of power sharing is characterized by delegating authority as well as responsibility to subordinates. The extent can vary greatly in terms of the magnitude of responsibilities, the task discretion, and the power for decision implementation (Webber, 1981). However, one needs to keep in mind that appearance and reality may be distinct from each other. Thus, overt procedures such as consulting subordinates in a pre-set meeting do not mean that others have actual influence on making a decision (Strauss, 1977). Also participative behavior can vary from time to time depending on the situation and the participants. Next potential benefits of participation are discussed.
Yukl (2009) elaborates on four potential benefits of participative leadership: 1) decision quality, 2) decision acceptance, 3) satisfaction with the decision, and 4) development of participant skills. Firstly, the quality of a decision can greatly be enhanced when considering different perspectives. Also there might be a participant who has expert knowledge or experiences that can be useful for the problem at hand. By incorporating this knowledge into the decision that is about to be made, might lead to a better decision (of higher quality). Secondly, a decision is more likely to be accepted if participants are involved in making the decision, because a form of ownership evolves as they identify with it. Also the decision problem as well as how it affects the participants is better understood which most probably has effects on how the participants react to it. As a result, participation has an influence on the affection a
subordinate feels towards the change. Thirdly, participants feel respected if their opinion is asked as it shows features of procedural justice, which likely influences their satisfaction with the decision process. Lastly, whenever participants take part in making decisions their skills and confidence are likely to advance. This has not only positive consequences for the participant but for the whole group or organization.
The previous paragraphs show that leader perspectives are at the core of research literature on participation and delegation. Perspectives of subordinates, on the other hand, are emphasized in the research on empowerment. Subordinates are more likely to feel empowered if they have the chance for job discretion and realize meaningful work. However, this paper focuses on leaders’ behavior and their perspectives. To shortly summarize, related concepts to participative leader behaviors are “consultation, joint decision making, power sharing, decentralization, empowerment, and democratic management” (Yukl, 2009, p. 133).
Participation has been studied in many different contexts. The majority of studies have investigated subordinate participation in relation to subordinate satisfaction and performance in order to evaluate leader effectiveness (Yukl, 2009). Bragg and Andrews (1973) studied how participation influences not only subordinates’ satisfaction but also their productivity. However, their findings were contradicting. In one group (medical section) participation had positive effects whereas for the other group (nursing) they showed much less success. Overall, Yukl (2009) argues that there is no agreement on reviewers’ conclusions but on one point. They do recognize that the inconsistent results might be due to the type of methodology used in the studies.
Other scholars specifically researched the role of participation in the context of organizational change. Bordia et al. (2004) studied uncertainty during organizational change and examined the role of participation in decision-making. The authors found that participation in decision making positively influences control, and subsequently enhances acceptance of change. Especially in job-related issues that directly affect employees, it is important for them to have a sense of control over the outcomes. They also found evidence that control negatively relates to psychological strain, and additionally, that employee participation in decision-making directly reduces strain. Moreover, employee input has been linked to perceptions of fairness influencing four employee reactions to firm reorganization, namely organizational commitment, job satisfaction, turnover intentions, and management trust (Kernan & Hanges, 2002).
These four employee reactions are mostly affected by procedural justice, which again is influenced by employee input. According to the authors, allowing employees to give input provides them with the a chance to express their concerns and with the ability to make sure that procedures are fair. Furthermore, they are offered the opportunity to give form to procedures, hence, thereby perceiving the change as more fair.
Hence, it is not safe to say that participation leads to higher satisfaction, performance, or effort. Nonetheless, it can be argued that subordinates build a form of psychological ownership due to participating in and being an integral part of the change process, which may lead to affection towards and supportive behavior for the change initiative.
2.2.2 Autocratic leader behavior
The previous section specifically elaborated on participative leader behaviors. Leader behaviors can range from participative to autocratic as for example, discussed in Yukl (2009) on the issue of decision-making. It is largely discussed in the literature on leadership studies in how far leaders behave democratically encouraging subordinates to participate, or autocratically meaning that leaders make decisions on their own without asking subordinates for their input (Eagly & Johannesen-Schmidt, 2001). The literature on participation suggests that autocratic leader behavior is not relevant for change; however, it might be effective for change depending on the personality of an employee. Some employees might prefer a leader who explicitly provides them with guidance and direction. Thus, this research paper specifically studies the contrasting leader behaviors and their influence on employee support for change.
According to Pearce and Sims Jr. (2002) directive leadership stems from the issue of position power and origins from several leadership and management theories. Firstly, Douglas McGregor’s (1960) Theory X builds upon assumptions that leaders generally make about their subordinates in terms of their work motivation. Leaders assume that they need to control and direct their subordinates, and take responsibility since their subordinates do not like to work and try to avoid it wherever possible. Secondly, Ohio State studies (e.g. Fleishman, 1953) discuss initiating structure as a leader behavior directed towards task and goal attainment (Piccolo, Bono, Heinitz, Rowold, Duehr, & Judge, 2012). Leaders who initiate structure define their roles and
the roles of their subordinates to clarify responsibilities, coordinate and organize work, provide structure concerning the work context, and further arrange work-related activities (Northouse, 2012). Moreover, they instruct subordinates specifically regarding their task expectations of what, how and when a task should be completed, and additionally, make rules and establish performance standards. Lastly, research at University of Michigan (e.g. Katz, Maccoby, & Morse, 1950) identified task-oriented or production oriented leader behaviors paralleling behaviors of initiating structure (Northouse, 2012; Pearce & Sims Jr., 2002). Hence, core to directive leader behavior is to assign specific tasks and objectives, and to give guidelines and orders to subordinates often with one-way communication. However, it is questionable whether such behavior is efficient to ultimately achieve behavioral support for organizational change.
A change often implies something new which often leads to uncertain situations. Some organizational changes arise due to crisis situations that demand an organization to change how it conducts business. Yukl (2009) argues that subordinates have different expectations of their leaders when they are confronted with a task they need to perform under tremendous pressures. They prefer a more directive leader who is more decisive and guiding in terms of what and how to perform. However, an organizational change does not always imply that tremendous pressures are involved in task performance. Also individual differences in terms of traits may either enhance or diminish the feeling of pressure. Thus, it might not be true that subordinates prefer an autocratic leader at all times.
Northouse (2012) argues that subordinates prefer directive leader behavior if the task structure is ambiguous, complex, and involves unclear rules, and in contrast, participative leader behavior if the task structure is ambiguous, unclear and unstructured. One can argue that an organizational change can produce tasks that do not fall into one specific category. Furthermore, the author argues that subordinate characteristics influence the preference for a certain leadership style. In congruence, Yukl (2009) argues that according to contingency theory subordinate characteristics influence the preference for a leadership style, and subsequently influence subordinate effort and satisfaction. Therefore, it is reasonable to argue that leader behaviors influence subordinate support for change whether by influencing the psychological mind-set or expressed behavior. In their review, Oreg, Vakola, and Armenakis (2011), elaborate on antecedents of change recipient reactions to change, which include
recipient characteristics. According to the authors, researchers focused especially on locus of control (e.g. Naswall, Sverke, & Hellgren, 2005) and self-efficacy (e.g. Herold et al., 2007; Judge, Thoresen, Pucik, & Welbourne, 1999), and with a somewhat lesser extent on neuroticism (e.g. Rafferty & Griffin, 2006). These three factors together with a fourth factor encompass the construct called core self-evaluation. Evaluations of self are important in this context because they may determine what kind of leader they would prefer. For instance, someone who has very high self-worth is likely to prefer a leader who asks for employee input because they would want to voice their opinions and ideas. Hence, in the following, subordinates’ core self-evaluations are discussed in detail, which determine individuals’ core traits about how they see and value themselves.
2.3 Core self-evaluation
The previous section introduced two opposing leader behaviors—participative and autocratic leader behavior. As argued before change-supportive behavior depends on what behaviors the leader engages in, but in addition, it is reasonable to assume that subordinates’ personality traits have either an enhancing or a weakening effect on this relationship. Leader behaviors most likely have different effects on subordinates based on their personality and their self-assessment. For instance, subordinates who are insecure about their abilities in performing certain tasks might prefer a leader providing them with guidance. However, if they have a leader who expects their participation and wants them to work autonomously, there might arise unwanted pressures that may result into opposing behaviors than the leader desires. Therefore, this paper elaborates on the concept of core self-evaluation (CSE), which embodies broad and evaluative traits showing how individuals fundamentally assess themselves in terms of their self-worth and capabilities (Chang et al., 2012). However, it is important to note that core self-evaluations are assumed to be personality traits (Judge & Kammeyer-Mueller, 2011). Thus, it is necessary to briefly elaborate on what personality traits are in order to completely apprehend core self-evaluations.
Personality traits can be distinguished from more transient mental states based on two key characteristics (McCrae, Costa, Ostendorf, Angleitner, Hrebickova, & Avia, et al., 2000; Roberts, O'Donnell, & Robins, 2004). Firstly, personality traits should not vary over time but instead stay the same within the same person regardless of time passing. Secondly, traits impact how individuals react and behave in certain
situations, and thus, the behaviors exhibited should be consistent across various situations. In their study, Judge, Bono, and Locke (2000), found evidence of the longitudinal robustness of CSE by measuring core self-evaluations in infancy and early adulthood, and relating them to job attitudes in middle adulthood. Therefore, we can assume that core self-evaluations are personality traits, which do not change over a long period of time. The next paragraphs elaborate on the theorization of CSE as a concept.
In her writings, Edith Packer (1985; 1985/1986) reasoned that fundamental appraisals—also referred to as core evaluations—influence how we evaluate specific situations. In 1997, Judge, Locke, and Durham proposed that individuals make bottom-line assessments about themselves, others, and the environment surrounding them. These assessments regard their self-worth, competence as well as their capabilities and can be positive as well as negative. Thus, these basic conclusions individuals make about themselves can be a key characteristic in how their reactions differ compared to others. Judge et al. (1997) introduced the concept of CSE and identified key criteria for selecting core traits to incorporate.
They argued that the traits inherent in the concept should meet the following three criteria: 1) evaluation-focus, 2) fundamentality, and 3) scope (cf. Judge & Bono, 2001; Bono & Judge, 2003; Judge, Bono, Erez, & Locke, 2005; Johnson, Rosen, & Levy, 2008; Chang et al., 2012). Firstly, evaluation-focus is the degree of traits comprising assessments as compared to descriptions of the self. Johnson et al. (2008) gives the example of agreeableness as a personality trait that rather describes a behavior of cooperation, in comparison to esteem that is an assessment of self-worthiness. In addition, the authors highlight that attitudes and perceptions of an individual and those of the environment are directly influenced by evaluation-focused traits. Fundamentality is the second criteria and describes the degree of centrality of traits to the concept of oneself. In contrast to peripheral traits, fundamental traits are more connected with other traits, beliefs, and appraisals. The third criterion is the scope of the trait, which can either be broad or narrow. For instance, the scope is broad if we look at a person’s general self-worthiness, but narrow if we look at the person’s self worthiness as an organizational member (Johnson et al., 2008). According to Allport (1961) core traits are those that are wide in scope. Based on these three criteria Judge et al. (1997) recognized three traits as indicators of CSE that obviously matched these criteria, and additionally, a fourth one that also qualified but
with much less clarity. Nevertheless, it is important to be aware that other traits (e.g. dispositional optimism) may be possible qualifiers as well (Judge, Locke, Durham, & Kluger, 1998).
The four CSE traits included in the theoretical concept proposed by Judge and colleagues (1997) are self-esteem, generalized self-efficacy, neuroticism, and locus of control. An individual’s overall appraisal or value of one’s self-worth refers to an individual’s self-esteem (Harter, 1990; Rosenberg, 1965), which is connected to cognitive, affective, and behavioral processes (Blascovich & Tomaka, 1991). General self-efficacy is a general evaluation of an individual’s ability to perform and cope across a wide-ranging variety of situations (Locke, McClear, & Knight, 1996; Bandura, 1997). The third trait neuroticism describes individuals with a negative mindset or attitude, and hence, individuals who have a predisposition to focus on negative features of themselves (Watson, 2000). Conversely, individuals who have the tendency to feel calm and secure are emotionally stable (Eysenck, 1990). Lastly, locus of control refers to one’s beliefs about who or what is responsible for causes of life incidents. If individuals believe that they can influence and control the outcomes with their behavior themselves, it is referred to as internal locus of control (Rotter, 1966). Contrasting, it is referred to external locus of control if individuals believe that they themselves cannot control but believe that external factors such as fate or luck impact the outcome. Often this means that people feel helpless and futile (Johnson et al., 2008). Based on these four core traits one can conclude that individuals can either be regarded as high or low on CSE.
In summary, the concept of CSE is a higher-order construct, based on evaluative, fundamental and broad traits. Moreover, individuals high on CSE appraise themselves with high self-worthiness (high self-esteem), are able to perform and cope successfully in various situations (high self-efficacy), do not doubt themselves and stay optimistic (emotional stability), and believe that they are responsible and in control of one’s own destiny (internal locus of control) (Judge & Kammeyer-Mueller, 2011). Contrasting, individuals low on CSE do not value themselves worthy (low self-esteem), are not adapting well in diverse situations (low self-efficacy), have negative mindsets (neuroticism), and believe that external factors are responsible for their destiny (external locus of control). The question arising is whether the level of subordinates’ CSE (high or low) influences the preference for either an autocratic or participative leader, and how this may impacts their cognitive mind-set and attitude
towards a change initiative. The ensuing section amplifies subordinates’ commitment to change, which precedes actual behavioral reactions to change.
2.4 Commitment to change
The conceptualization of commitment links back to the three-component model of organizational commitment proposed by Allen and Meyer (1990). They distinguish three mind-sets that together determine an individual’s commitment profile. The first mind-set is affective commitment, which is the desire to remain in the organization. Normative commitment is a mind-set where individuals feel obliged to remain in the organization. They call it continuance commitment when individuals remain simply because costs would occur if they leave the firm. Building on this conceptualization, Meyer and Herscovitch (2001) developed a more general model of workplace commitment and defined commitment as “a force that binds an individual to a course of action of relevance to one or more targets” (p. 301). A target in this sense can be, for instance, an organization, a unit, or a business objective. A lot of empirical support was found for this general conceptualization of commitment (Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). Based on the research that has been done another model has been developed to the issue of organizational change.
Herscovitch and Meyer (2002) suggested a new multidimensional model that specifically looks at commitment to change. Accordingly, affective commitment to change is a mind-set characterized by an individuals’ desire or want to support a change initiative due to their belief of its inherent advantages. Normative commitment to change refers to individuals’ mind-set to support a change initiative based on the feeling of obligation. The mind-set of continuance commitment to change indicates that individuals acknowledge that costs can occur if they fail to support a change initiative. In this sense, commitment means to show support intentions proactively and to perform in a certain way to make a change initiative successful (Herold, Fedor, & Caldwell, 2007).
Furthermore, researchers recognized commitment to change as better predictor of support for change (Herscovitch & Meyer, 2002; Meyer, Srinivas, Lal, & Topolnytsky, 2007). Individuals who are committed to a change contribute above and beyond organizational commitment. In their work, Allen and Meyer (1990) suggest that the three mind-sets together compile an individual’s commitment profile, which may then give an indication of their behavioral support. Based on this profile,
organizations could make distinctions between high and low contributors in terms of their supportive behavior. Nevertheless, scholars found that the three mind-sets lead to different change-related behaviors (Herscovitch & Meyer, 2002; Meyer et al., 2007). More specifically, they found that all mind-sets lead to compliance to change. However, only affective and normative commitments to change related positively to cooperation and championing.
This study focuses only on affective commitment to change as it is most likely influenced by specific leader behaviors (Herold et al., 2008). They argue that this mind-set best indicates individuals’ attitudes and cognitive states towards change initiatives. Moreover, Meyer and Allen (1991) proposed that certain personal characteristics are antecedents of affective commitment to the organization. Therefore, one can argue that personal characteristics may moderate the relationship between leader behaviors and affective commitment to change.
2.5 Derivation of Propositions
Based on the literature review this paper derives several hypotheses and proposes a conceptual model that illustrates the suggested relationships between the theories and concepts.
2.5.1 Leader behaviorand change-supportive behavior (CSB)
The literature on participative and directive leadership proposes that there is a preference for participative leaders if there is no clear task structure, and that subordinates have a general aversion towards autocratic leadership styles (Schoel, Bluemke, Mueller, Stahlberg, 2011). Since we can assume that an organizational change involves new tasks, they are most likely anything else than clear to subordinates. Furthermore, subordinate participation provides them with a voice and opportunity to get actively involved in, and also to better understand the change process. The leaders’ willingness to integrate their subordinates shows them that they are valued which is likely to motivate them to actively support the change. Moreover, if a person feels to be a part of something that is beneficial to them and/or others, they tend to talk about it to other people and show greater effort to make the change successful. In contrast, autocratic leaders only delegate their subordinates and make decisions alone which they may not comprehend. Subordinates may feel excluded and not valued leading to non-supportive behavior. Hence, this paper proposes that in an
organizational change context participative leader behavior relates positively, whereas autocratic leader behavior relates negatively to change-supportive behaviors.
H1a: Participative leader behavior is positively related to CSB. H1b: Autocratic leader behavior is negatively related to CSB. 2.5.2 Moderating role of CSE
As mentioned in the literature overview the relationship between leader behavior and subordinates’ CSB is likely to be influence by subordinates’ basic evaluations that they make about themselves. According to path-goal theory a participative leader is best for subordinates high on CSE. Firstly, if subordinates are confident and perceive themselves to be able to perform and to cope with certain tasks, they are less likely in need for a directive leader (Northouse, 2012). These individuals may even expect that their leaders ask them for involvement. Also they get the feeling that the leaders value their competence and abilities if they ask for their input. Furthermore, it is expected that subordinates with an internal locus of control prefer a participative leader due to their participation in decision-making and other change-related matters, which provides them with the opportunity to feel responsible for and in control of their own work. Control in this sense means that they can intervene and voice their opinions, and hence, have an actual influence on the change process.
In contrast, researchers found that subordinates who are not confident and have low ability are in need for a directive leader (Yukl, 2009). Additionally, an organizational change might be stressful now and then, which can have especially a negative influence if a person is neurotic. This is because neurotic subordinates are not inclined to emphasize goal-directed behavior but rather focus on the distress coupled to the situation (Parkes, 1986). Further, subordinates with an external locus of control should have a directive leader since they feel that external factors are responsible for their circumstances (Northouse, 2012).
In summary, based on previous literature we can assume that subordinates’ preference for a participative or autocratic leader depends on their core self-evaluations. Subordinates scoring high on CSE prefer to participate and to be an integral part of the change process, whereas subordinates scoring low on CSE feel more secure and conformable with a leader who provides guiding structures during
the change. Hence, one can argue that not only leader behaviors have an impact on subordinates’ change-supportive behavior (CSB) but that the level of CSE either strengthens or weakens this relationship.
H2a: The relationship between participative leader behavior and CSB will be moderated by CSE such that participative leader behavior will be positively related to CSB when CSE is high whereas that participative leader behavior will be negatively related to CSB when CSE is low.
.
H2b: The relationship between autocratic leader behavior and CSB will be moderated by CSE such that autocratic leader behavior will be positively related to CSB when CSE is low whereas autocratic leader behavior will be negatively related to CSB when CSE is high.
2.5.3 Mediating role of affective commitment to change (ACC) and interaction effects The relationship between the two contrasting leader behaviors and CSB is likely to be mediated by affective commitment to change. Subordinates with a participative leader are likely to get psychologically involved in the change process because subordinates feel that they are meaningful to the process, and feel valued and respected by their leaders due to the active involvement. Herscovitch and Meyer (2002) found evidence that affective commitment to change positively influences change-supportive behavior. Furthermore, the authors argued that widely recommended strategies to implement a change initiative, inter alia participation, increase employee’s identification, involvement, and value relevance, which again are enhancing employee’s affective commitment to change. Hence, this cognitive affection towards the change process influences employee’s effort and support for the change.
H3a: The relationship between participative leader behavior and CSB is mediated by ACC.
H3b: The relationship between autocratic leader behavior and CSB is mediated by ACC.
As argued before, it depends on subordinates’ level of CSE if they rather prefer a participative or an autocratic leader. Individuals high on CSE are confident about themselves and their abilities so that they prefer to be engaged in the change process. These individuals appreciate the respect the leader shows them by asking for their input simply because they feel that their competence and abilities are valued. However, subordinates low on CSE may not appreciate the involvement in the change process and feel an aversion towards the change initiative. This likely leads to lower levels of actual support behavior.
H4a: The interaction between participative leader behavior and CSE on CSB will be mediated by ACC such that participative leader behavior will interact with high levels of CSE in influencing ACC, which in turn will have a positive impact on CSB.
In contrast, subordinates scoring low on CSE like to be directed by their leaders providing them with guidance. With low confidence in their abilities they are likely to feel affective commitment towards the change if they feel secure rather than helpless. Therefore, it can be argued that autocratic leadership in interaction with low levels of CSE results to affective commitment to change. However, subordinates who are high on CSE but are directed and controlled by their (autocratic) leaders could get demotivated and as a consequence, extract themselves from the process since they feel that their competence and abilities are not valued. Hence, it may make them reluctant towards the change leading to a negative influence on their support for the change.
H4b: The interaction between autocratic leader behavior and CSE on CSB will be mediated by ACC such that autocratic leader behavior will interact with low levels of CSE in influencing ACC, which in turn will have a positive impact on CSB.
Figure 1. Conceptual Model
3. Research Method
This section provides an overview of the research method used for the current study. A detailed description of the research design, the respondent sample, as well as the instruments used in order to operationalize the different constructs is provided.
3.1 Procedure
This thesis is part of a research project on organizational change at the University of Amsterdam. In total four students participated in the project, which was under the supervision of one professor. Employees’ behavioral support for change was the outcome variable studied in all the theses, however, each student developed individual research questions and conceptual models. The project team created two separate surveys—one leader and one employee survey—with Qualtrics Survey Software and were distributed online. By using the back-translation method, the surveys were translated from English into three other languages (German, Dutch, and Hungarian). This made it possible to reach leaders and employees from different countries who work in different functions, departments, and organizations.
The data collection period lasted six weeks, starting mid April and ending in the end of May 2016. The project team emailed as many contacts as possible from their
personal network to invite for participation in the study. Many of those forwarded the survey to colleagues or other leaders from whom they knew were suitable survey respondents or who might be interested in the study. Furthermore, some smaller companies were visited in person to ask for participation. Therefore, non-probability convenience and snowball sampling was used. Since the surveys were returned anonymously matching codes needed to be inserted in the survey by the participating leaders and their employees. Thereby, the two separate surveys could be linked to each other.
The questionnaires began with a detailed but concise introduction to the topic, its purpose, structure, and other procedural and ethical aspects. Appendix A shows the introduction from a respondent’s view (both leader and employee). The leader survey includes a question on the type of organizational change that is currently taking place and one set of questions, which is on the leader’s perception of employee support for change. The employee survey contains multiple sets of questions—questions about their personality, their work, their direct supervisor, their relationship with the leader, the organizational change, and their demographics.
3.2 Sample
In total 112 employee and 115 leader questionnaires were returned. However, only complete leader-employee dyads can be used for conducting data analysis. As mentioned before, the questionnaires were linked to each other by matching codes, which resulted in a total of 106 leader-employee dyads. Nevertheless, the employee is the focal person in this research. The final employee sample consists of 54 females and 52 males (51% and 49%, respectively) among which 30 filled in the Dutch, 41 the German, one the English, and 34 the Hungarian survey. The age ranges from 21 to 64 years with an average of 38.1 years (SD = 12.071). The number of years the employee works in the current company varies greatly, from 0 to 46 years and an average of 9.10 years (SD = 9.56). Furthermore, the number of working hours a week ranges from 9 to 64 hours a week, with an average of 37.64 hours a week (SD = 11.12). More specifically, approximately 70% of the sample works more than 35 hours a week. Moreover, the amount of years the employee gets guidance from the current supervisor ranges from less than one to 19 years with an average of 4.46 years (SD =
1 In this case, average age and standard deviation is based on 105 instead of 106 respondents because of a mistake in the data.
4.92). Less than 55% of the employees have the same supervisor for more than 2 years. The days on which an employee has direct contact with the supervisor during one week significantly varies across the sample. It ranges from one day to 7 days a week with an average of 3.74 days a week (SD = 1.57). In addition, the type of change experienced in the various organizations include insourcing, mergers, reorganizations, relocations, new software technologies, and new work processes.
3.3 Measures
Since this study is part of a research project on organizational change, the distributed employee questionnaire included more variables than are of relevance for this study. Thus, this section only focuses on variables included in the conceptual model displayed in figure 1. To assess the various variable constructs, validated scales were adopted from extant literature. All items were rated on a 5-point Likert scale, ranging from ‘strongly agree’ to ‘strongly disagree’. By averaging item means, the overall scale scores were computed.
Change-supportive behavior is measured with 14 items from the supervisor’s perspective on subordinate championing, their willingness to contribute to change, and change commitment. All items are indicative. First, subordinate championing is assessed by four items adopted from Hercsovitch and Meyer (2002) and has a Cronbach’s Alpha of 0.81. Sample items include “The employee encourages the participation of others in the change” and ”The employee speaks positively about the change to outsiders”. Willingness to contribute to change is measured with seven items from Venus (2013) and has a Cronbach’s Alpha of 0.89.Example items include “The employee is willing to put extra effort in implementing the changes” and “The employee fully collaborates with implementing the current or future plans”. Last, change commitment is measured using Fedor, Caldwell and Herold’s (2006) scale shortened to three items. Cronbach’s Alpha is 0.82 and sample items include “The employee does whatever he/she can to help this change be successful” and “The employee is fully supportive of this change”. The Cronbach’s Alpha of the three scales combined is 0.95.
Participative leader behavior is measured with three items adopted from Arnold, Arad, Rhoades, and Drasgow’s (2000) scale to assess supervisor’s participative decision-making. Cronbach’s Alpha is 0.84 and example items include “My direct supervisor gives all work group members a chance to voice their opinions”
and “My direct supervisor uses my work group’s suggestions to make decisions that affect us”.
Autocratic leader behavior is measured with three items adopted from De Hoogh, Den Hartog, and Koopman’s (2004) CLIO. Cronbach’s Alpha for this scale is 0.87 and sample items include “This leader is bossy and orders subordinates around” and “This leader often pushes his/her opinions”.
Core-self evaluation is measured with a 12-item scale adopted from Judge, Erez, Bono, and Thoresen’s (2003) Core Self-Evaluations Scale (CSES). The CSES includes six indicative and six counter-indicative items and Cronbach’s Alpha is 0.77. Examples of indicative items are “I determine what will happen in my life” and ”I am confident I get the success I deserve in life”. Counter-indicative items include “Sometimes when I fail I feel worthless” and “I do not feel in control of my success in my career”.
Affective commitment to change is measured with three items adopted from Herscovitch and Meyer (2002). Cronbach’s Alpha is 0.90 and examples include “This change serves an important purpose” and “I believe in the value of this change”.
Furthermore, two control variables are included in this study, which are measured on single items. First, employee tenure is measured with “How many years do you work in your current company”, and second, contact is measured with “How many days per week do you have direct contact with your supervisor (e.g. meeting, call, email)”.
4. Data analysis and results
This section provides a detailed description on which steps were taken to analyze the collected data. First, the analytical strategy and procedure is described, and subsequently, results of the data analysis are presented. By providing detailed descriptions future research will be eased.
4.1 Data analysis
To analyze the data IBM’s Statistical Package for Social Sciences (SPSS) version 24 was used. Before starting the data analysis the data were prepared and checked for missing values. A frequency test was performed in order to check all the relevant variables. The results showed no missing values. Since the CSES includes six
counter-indicative items they needed to be recoded, such that, for instance, rCSE_2 is now CSE_2.
Next, scale reliabilities were reviewed and results showed that all Cronbach’s Alpha coefficients were above 0.75 indicating that the instruments are reliable. Scale reliabilities are 0.84 for Participative leader behavior, 0.87 for Autocratic leader behavior, 0.77 for Core self-evaluation, 0.90 for Affective commitment to change, and 0.95 for Change-supportive behavior. For every variable the scale items exhibit a corrected item-total above 0.30, which means that all the items have a very good correlation with the total score of the scale. Also, none of the items would have a substantial impact on scale reliability if they were deleted. Only one item for the CSES, however, scores just below 0.30. Nonetheless, deleting this item neither decreases nor increases scale reliability substantially.
Scale means were computed as new variables in order to test the hypotheses. Moreover, to test the research hypotheses correlation and regression analyses were performed. Bivariate correlation analyses were used to examine the strength of the relationships between the different concepts that are displayed in figure 1. To test direct relationships (H1a and H1b) and mediation (H3a and H3b) simple linear regression analyses were performed. Hierarchical regression analyses were executed to test the moderating effect (H2a and H2b). In the analyses it was controlled for tenure and contact.
4.2 Results
4.2.1 Descriptive statistics
Means, standard deviations, correlations, and Cronbach’s Alpha coefficients are reported in Table 1. Since all correlations are below .70, all the variables can be included in the regression analyses to test the hypotheses.
Table 1 shows that the employees in the sample indicate a very high level of participative leader behavior (M = 4.09, SD = .76) and in contrast a low level of autocratic leader behavior (M = 1.92, SD = .90). Further, they have a rather high level of core self-evaluation (M = 3.98, SD = .46). Additionally, the employees show a high level of commitment to change (M = 3.77, SD = .91) as well as a high level of change-supportive behavior (M = 3.77, SD = .74). Furthermore, the table reports correlations between the relevant concepts. There is a negative correlation between the two independent variables, participative and autocratic leader behavior (r = -.679,
Table 1. Means, standard deviations, correlations and reliability scores
Variables M SD 1 2 3 4 5 6 7
1. Tenure 9.10 9.56 -
2. Contact 3.74 1.57 -.058 -
3. Participative leader behavior 4.09 0.76 .026 .165 (.838)
4. Autocratic leader behavior 1.92 0.90 .010 -.084 -.679** (.873)
5. Core self-evaluation 3.98 0.46 -.002 .029 .051 -.040 (.769)
6. Affective commitment to change 3.77 0.91 -.134 .157 .250** -.208* .293** (.904)
7. Change-supportive behavior 3.77 0.74 -.095 .085 -.023 -.123 .044 .264** (.948) **. Correlation is significant at the 0.01 level (2-tailed).
p = .01). This result is very logical as a leader either asks for his/her employees’ participation or not. Therefore, this negative correlation was to be expected. Moreover, a positive correlation between participative leader behavior and affective commitment to change (r = .250, p = .01), a negative correlation between autocratic leader behavior and affective commitment to change (r = -.208, p = .05), a positive correlation between core self-evaluation and affective commitment to change (r = .293, p = .01), and a positive correlation between affective commitment to change and change-supportive behavior (r = .264, p = .01). These correlations show that affective commitment to change is significantly correlated to all the other variables in the model. However, no significant correlation between the two leader behaviors and change-supportive behavior could be found. The correlation coefficient for participative leader behavior and change-supportive behavior is r = -.023 (p > .05) which suggests absence of a relations. The same accounts for the relation between autocratic leader behavior and change-supportive behavior (r = -.123, p > .05).
4.2.2 Direct effects
To test H1a and H1b two simple linear regression analyses with participative leader behavior and autocratic leader behavior as predictors were performed. Tenure and contact were added as control variables. Results revealed an insignificant effect of participative leader behavior on change-supportive behavior (b = -.035, t = -.347, p = .729). In addition, results revealed an insignificant effect of autocratic leader behavior on change-supportive behavior (b = -.116, t = -1.185, p = .239). Hence, no support for either H1a or H1b was found.
4.2.3 Moderation
To test the moderation effect of core self-evaluation on the relationship between the two leader behaviors and employees’ change-supportive behavior (H2a and H2b), hierarchical regression analyses were performed. The results are presented in Table 2 and 3. Table 2 shows that the interaction of participative leader behavior and core self-evaluation does not have a statistically significant effect on change-supportive behavior (b = -.128, t = -.660, p = .511). Table 3 shows that the interaction effect of autocratic leader behavior and core self-evaluation on change-supportive behavior is insignificant. Hence, no evidence of a moderation effect of core self-evaluation can be found, and thus, hypothesis 2a and 2b cannot be confirmed.
Table 2. Results of the Moderation effect H2a Variable Coeff. SE t p Constant i1 1.419 3.317 .428 .670 Tenure b1 -.007 .008 -.878 .382 Contact b2 .042 .048 .888 .377
Participative leader behavior
(PLB) b3 .488 .800 .610 .543
Core self-evaluation (CSE) b4 .587 .802 .732 .466 Interaction term (PLB x CSE) b5 -.128 .194 -.660 .511
R² = .023
F(1,100) = .436; p = .511
Table 3. Results of the Moderation effect H2b
Variable Coeff. SE t p Constant i1 3.140 1.406 2.234 .028 Tenure b1 -.007 .800 -.922 .359 Contact b2 .030 .047 .628 .532
Autocratic leader behavior (ALB) b3 .161 .616 .261 .795 Core self-evaluation (CSE) b4 .192 .355 .540 .590 Interaction term (ALB x CSE) b5 -.065 .155 -.418 .677
R² = .032
F(1,100) = .175; p = .677
4.2.4 Mediation
To test the mediation effect of affective commitment to change (H3a and H3b) multiple regression analyses with PROCESS (Model 4) were performed. According to Baron and Kenny’s (1986) methodology, four conditions need to be met in order to establish mediation. The first condition is that a significant relationship between the independent and the dependent variable must be found. The second and third conditions prerequisite that the mediator variable needs to significantly relate to the independent and dependent variable, respectively. The final condition for mediation requires that the main effect between the independent and dependent variable should be weaker or not significant anymore, after the mediator variable is added.
