The moderating influence of core self-evaluations and servant leadership on the relationship between qualitative job insecurity and commitment to change

(1)

Master Thesis

The moderating influence of core self-evaluations and servant

leadership on the relationship between qualitative job insecurity

and commitment to change

Executive Program in Management Studies Supervisor: dr. M. Venus

Student: Kees Verweij - 11132337 Date: 24-6-2017

(2)

Statement of originality

This document is written by student Kees Verweij, who declares to take full responsibility for the contents 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 contents.

(3)

Preface and acknowledgements

I work as a manager for a small municipal organization, which will merge with two other municipalities on January 1, 2019. Mergers of municipal organizations have in common that employees have the assurance of a job in the new organization, but the exact characteristics of the new job are not known. On a daily basis I notice that some employees in a certain way suffer from job insecurity, whereas others experience only an inconvenience due to the uncertainty. I am curious to which extent the personality disposition of the employee and the leadership style of the manager of the employee have influence on the commitment of the employee for the merger. This is the reason for me to conduct this study.

I thank dr. Merlijn Venus for being my supervisor during the master thesis process. I am grateful for his support, feedback on draft versions of this thesis and his suggestions for improvement of my thesis.

Last but not least, I am grateful to my beloved wife Greet and children Adriënne, Hans-Jan en Cédric for their love and unconditional support during the years in which I attended the Amsterdam Business School.

(4)

Abstract

In this study, the moderating influence of core self-evaluations and servant leadership on the relationship between qualitative job insecurity and commitment to change is investigated. The sample for this study comprised of employees of three municipal organizations who were preparing for a merger as partners. Results showed that qualitative job insecurity was negatively related to commitment to change. There was no support for the hypotheses which proposed a moderating influence of core self-evaluations and servant leadership, nor for the mediating role of satisfaction of basic psychological needs. Discrepancies with earlier research are discussed and practical implications and opportunities for future research are addressed.

Keywords: qualitative job insecurity, commitment to change, satisfaction of basic psychological needs, servant

(5)

1. Introduction

1.1 personal consequences of a turbulent business environment

In the last decades organizations have made use of various strategies such as downsizing and restructuring in response to increasing global competition. Likewise, in the public domain many organizations are engaging in mergers to be more equipped to adapt to a fast changing environment and to respond to varying demands from society. Although the goal of downsizing, restructuring and mergers is to improve effectiveness of organizations, these processes may lead to dysfunctional consequences for employees (Datta, Guthrie, Basuil, & Pandey, 2010), such as uncertainty and stress (Bordia, Hobman, Jones, Gallois & Callan, 2004). One type of uncertainty is job insecurity, which essentially is a stressor (De Witte, De Cuyper, Handaja, Sverke, Näswall & Hellgren, 2010; Loi, Lam & Chan, 2012). that could affect employees’ support for and eventually the effectiveness of the organizational change. For this reason it is of crucial importance to discover factors which can reduce the influence of job insecurity on employees.

1.2 Job insecurity and commitment to change

Greenhalgh and Rosenblatt (1984) defined job insecurity as employees’ concerns about their work-related future. Job insecurity was divided by Hellgren, Sverke, and Isaksson (1999) into a quantitative and a qualitative dimension. Quantitative job insecurity reflects the threat of losing one’s job, whereas qualitative job insecurity points to the threat of losing valued job features. Quantitative job insecurity has attracted considerable attention in both research and policy but little research has examined perceived threats to valued job features, i.e. qualitative job insecurity (De Witte et al., 2010). Recent research focused on the associations between qualitative job insecurity and well-being (De Witte et al., 2010), counterproductive work

(8)

behaviour (Van den Broeck, Sulea, Vander Elst, Fischmann, Iliescu & De Witte, 2014) and organizational citizenship behaviour (Stynen, Forrier, Sels & De Witte, 2015).

For organizational changes to be effective, active support of employees for the proposed changes is essential (Self, Armenakis & Schraeder, 2007). A leading construct with respect to individual support toward change is commitment to change, which reflects a proactive behavioural intention to support a change initiative and to work toward the success of this initiative (Choi, 2011). A diversity of antecedents of commitment to change has been researched, e.g. climate, communication and participation (Rogiest, Segers & van Witteloostuijn, 2015; Bordia et al. 2004), perceived supervisor support and competence (Neves, 2011) as well as leadership (Van der Voet, Kuipers & Groeneveld, 2016). However, little is known about the influence of qualitative job insecurity on commitment to change.

1.3 Possible factors affecting qualitative job insecurity and commitment to change All the above mentioned antecedents of commitment to change can be controlled by management in a certain way. However, qualitative job insecurity seems to be less controllable by managers. From a managerial point of view it is worthwhile to know which factors can be influenced to reduce the negative consequences of qualitative job insecurity on commitment to change. Possibly, the leadership style of an employee’s manager could be such a factor since leadership can have a great influence on the commitment of that employee (Hawkins & Dulewicz, 2009). However, no study has paid attention to the role of leadership in relation to qualitative job insecurity and commitment to change.

Another issue is the role of employee’s personality disposition in the relationship between qualitative job insecurity and commitment to change. In general, the influence of personality dispositions on job insecurity has received little attention in literature. Only few studies have considered general mood dispositions, especially negative affectivity, in relation

(9)

which personality dispositions are related to the relationship between qualitative job insecurity and commitment to change. It could be argued that employees with a certain personality disposition are less sensitive to uncertainty than others.

1.4 The research question

Several studies revealed that qualitative job insecurity has influence on the extent to which basic psychological needs are satisfied (Stynen et al., 2015; Van den Broeck et al., 2014). Remarkably, other studies indicated that basic psychological needs are fulfilled by servant leadership (Chiniara & Bentein, 2016; Van Dierendonck, Stam, Boersma, De Windt & Alkema, 2014). Combining these insights, the following research question is central for this thesis:

What is the influence of core self-evaluations and servant leadership on the relationship between qualitative job insecurity and commitment to change, mediated by basic psychological need satisfaction?

I propose that qualitative job insecurity is negatively related to satisfaction of basic psychological needs, which in turn will be positively related to commitment to change. Furthermore, it is expected that the negative relationship between qualitative job insecurity and basic need satisfaction will be amplified for employees who think more negatively about themselves. Finally, it is expected that servant leadership has an attenuating effect on the negative relationship between qualitative job insecurity and basic need satisfaction.

1.5 Contributions of this research

This thesis aims to contribute to the academic literature in various ways. First, by explicitly focusing on the relationship between the qualitative component of job insecurity and commitment to change. Second, by extending research on servant leadership specifically in

(10)

attempts to reveal servant leadership as a factor by which the negative consequences of qualitative job insecurity on commitment to change can be reduced.

(11)

2. Literature review

2.1Commitment to change

Employee support for change can be seen as an important antecedent of successful change implementation (Herold, Fedor, Caldwell & Yi, 2008). In general, commitment is considered to be one of the most important indicators in explaining employee behaviour and desirable work-related outcomes in organizations (Choi, 2011).

Herscovitch and Meyer (2002) defined commitment to change as “a force (mind-set) that binds an individual to a course of action deemed necessary for the successful implementation of a change initiative”. To put it shortly, commitment to change is considered “the glue that binds together people and change goals” (Conner, 1992). The construct of commitment to change is derived from a three-dimensional model of organizational commitment (Allen & Meyer, 1990).

Three types of commitment to change have been defined: affective, continuance and normative (Herscovitch and Meyer, 2002). Affective commitment to change depends upon the information and clarity regarding the need and purpose of change and the ability of the employee to cope with these changes. This type of commitment reflects the beliefs in the inherent benefits of the change. Continuance commitment to change is a cost-based commitment and concerns “a recognition that there are costs associated with failure to provide support for the change” (Herscovitch and Meyer, 2002). The cost factor depends on the lack of viable alternatives for employees. Normative commitment to change points to “a sense of obligation to provide support for the change”. This commitment develops when employees recognize that the organization is fulfilling its duties in honouring the psychological contract that exists between the two parties (Herscovitch and Meyer, 2002). Affective commitment to change and normative commitment have been related to both compliance to and behavioural

(12)

support for the change initiative (Herscovitch and Meyer, 2002), making both types of commitment an important prerequisite for a successful change process.

Although the construct of commitment to change was based on a model of organizational commitment, researchers have demonstrated through empirical studies that commitment to change is conceptually distinct from organizational commitment (Fedor, Caldwell & Herold, 2006). This is in particular related to the behavioural consequences of commitment to change versus organizational commitment: from a point of view of organizational commitment, a valued behavioural outcome is retention (the opposite of turnover), however, the aimed behavioural outcome of commitment to change is compliance with the change (as opposed to resistance). In addition, commitment to change appeared to be a better predictor of support for change than the broader organizational commitment (Meyer, Srinivas, Lal & Topolnytsky, 2007).

Researchers have tried to identify the antecedents of commitment to change. Most studies focused either on change characteristics, like change favourableness, magnitude, or turbulence (Fedor et al., 2006), or on employees’ self-perceptions, such as individual change self-efficacy or locus of control (Cheng & Wang, 2007). Other studies paid attention to some aspects of leadership style (Neves, 2011), the role of leadership in implementing change (Van der Voet et al., 2016) or the change process itself, for instance participation in and fairness of the change process (Fedor et al., 2006).

Although the goal of change processes is to improve effectiveness of organizations, these processes might have dysfunctional consequences for employees (Datta et al., 2010), such as uncertainty and stress (Bordia et al., 2004). One type of uncertainty is job insecurity. However, little is known about the impact of job insecurity on commitment to change. When people are supposed to work on an organizational change which is likely to result in a loss of valued job features or would mean saying goodbye to appreciated colleagues, would they

(13)

support the proposed change? In the next section I will clarify the concept of qualitative job insecurity in more detail.

2.2 Qualitative job insecurity

Job insecurity can give rise to inflexibility which is detrimental for the organizational change process (Chawla and Kelloway, 2004). Job insecurity can be divided into a quantitative dimension, which is essentially the threat of losing one’s job, and a qualitative dimension, which points to the threat of losing valued job features (Hellgren et al. 1999). To be more precise, qualitative job insecurity is the “perception of potential loss of quality in the employment relationship, such as deterioration of working conditions, demotion, lack of career opportunities, decreasing salary development and concerns about person–organization fit in the future” (Sverke & Hellgren, 2002).

Job insecurity is a stressor that leads to strain which can cause, for example, poor well-being. More specifically, qualitative job insecurity is associated with poor general well-being, like psychological distress and psychosomatic complaints (De Witte et al., 2010). Furthermore, job insecurity is also related to poor job-related well-being, such as job dissatisfaction and burnout (Ashford, Lee & Bobko, 1989). Job insecurity may not be an inherent consequence of downsizing or merger but rather a subjective definition by employees (Sverke and Hellgren, 2002), which can be culturally determined. For example, Sender, Arnold and Staffelbach (2016) investigated the relationship between job insecurity and satisfaction and turnover intentions and found that employee reactions to qualitative and quantitative job insecurity may differ in culturally distinct regions.

In general the effect of job insecurity on organizational commitment or commitment to change has been given little attention in literature. Ashford et al. (1989) found that the greater the perceived job insecurity, the lower the organizational commitment. To my knowledge the

(14)

change is by Kalyal, Berntson, Baraldi & Näswall (2010).They found that individuals who experienced higher levels of job insecurity also reported weaker affective and normative commitment as well as stronger continuance commitment to change. The direct relationship between job insecurity and commitment to change was expected to be moderated by employability, which was defined as a person’s self-perceived ability of finding a new job based on his or her competencies. However, Kalyal et al. (2010) found only partially support for this. Contrary to their expectations the interaction term failed to reach significance for continuance and normative commitment. However, for the affective component the interaction between job insecurity and employability was significant. It should be noted that both Ashford et al. (1989) and Kalyal et al. (2010) did not differentiate between quantitative and qualitative job insecurity but used the more general construct job insecurity. So, the exact relationship between qualitative job insecurity and commitment to change is not known.

Based on Kalyal et al. (2010) and De Witte et al., (2010), I expect that individuals who experience higher levels of job insecurity will report lower levels of commitment to change. This leads to the first hypothesis:

Hypothesis 1: qualitative job insecurity will be negatively related to commitment to change.

Up to this point I have elaborated on the relationship between qualitative job insecurity and commitment to change in general which suggests that this will hold for every employee under every circumstances to the same extent. However, it could be argued that there are factors that attenuate or amplify this relationship. In theory, this could be personal characteristics of the employee, e.g. personality disposition, or contextual factors like the leadership style of the manager of the employee. In the next section I will elaborate on the moderating influence of personality traits of employees and servant leadership.

(15)

2.3 The moderating influence of core self-evaluations and servant leadership

In the last decade, several researchers have taken into account personal differences of employees which can have an influence on the way they deal with uncertainty and insecurity. For example, Stynen et al. (2015) found that the experienced stressor due to qualitative job insecurity affected all age groups, but most strongly the mature age workers. People differ with respect to their cognitive style of dealing with uncertainty too. For instance, Sorrentino & Short (1986) differentiated between individuals who are uncertainty-oriented (UO) and individuals who can be characterized as certainty-oriented (CO). Individuals with a UO-preference are positively motivated by the resolution of uncertainty. However, CO individuals are interested in maintaining clarity of what is already known, given that uncertainty resolution could threaten the status quo of an individual’s mental set. It seems reasonable to assume that personality traits can have an influence on the relationship between job insecurity and commitment to change.

Besides their cognitive style of dealing with uncertainty, people differ also on personality disposition. Roughly, a distinction can be made between positive and negative personality disposition. People with a positive personality disposition perceive the world as comprehensible, meaningful and manageable, whereas individuals with a negative personality disposition tend to appraise themselves, other people and the world in general negatively (Roskies, Louis-Guerin and Fournier, 1993). In general, employees with a negative personality disposition are inclined to focus on negative information in their environment (Debus, König & Kleinmann (2004).

Roskies et al. (1993) conducted two studies of the role of personality dispositions related to job insecurity: a study into long-term job insecurity and a study into short term, acute job insecurity. In the long-term job insecurity study, personality disposition was found to be the best predictor of how much danger of job loss was perceived: the higher the score on

(16)

positive personality disposition, the less the perceived risk. However, in the acute job insecurity study, personality disposition was not a significant predictor. It is noteworthy that also Roskies et al. (1993) did not distinguish between qualitative and quantitative job insecurity.

In literature there are some fundamental issues made about relying on the relationship between single traits and work stressors since they do not take the conceptual interrelations of personality traits into account (Judge & Bono, 2001). To overcome this discussion, Judge, Locke, Durham, and Kluger (1998) initiated the concept of core self-evaluations (CSE). CSE is a higher order trait and consists of a certain set of personality dimensions, i.e., self-efficacy, locus of control, neuroticism (i.e. emotional stability in reversed form) and self-esteem. CSE has been defined as the “fundamental, subconscious conclusions individuals reach about themselves, other people, and the world” (Judge et al., 1998). From research it is known that CSE is positively related to job motivation, task performance, organizational citizen behaviour (Chang, Ferris, Johnson, Rosen & Tan, 2012) and leader effectiveness (Judge, Piccolo & Kosalka, 2009).

Recently, Låstad, Berntson, Näswall and Sverke (2014) showed that CSE is negatively related to job insecurity, since CSE significantly predicted both qualitative and quantitative job insecurity. This means that people with higher levels of CSE report lower levels of quantitative as well as qualitative job insecurity. Remarkably, the authors measured the trait indirectly instead of by using the core self-evaluations scale (CSES) as developed by Judge, Erez, Bono and Thoresen (2003). The results of Låstad et al. (2014) are comparable with the results of the long term job insecurity of Roskies et al. (1993). However, both Låstad et al. (2014) and Roskies et al. (1993) did not examine a situation in which only qualitative job

(17)

People who score low on CSE are generally less positive about their ability to perform across a variety of situations (self-efficacy) and see events as happening beyond their control (external locus of control). They have a tendency to explain events in a negative way and to focus on negative aspects of the self (neuroticism). Furthermore, they have a low overall evaluation of themselves as a person. Due to their relative negative evaluations, I argue that people with a low score on CSE will perceive the stressor of qualitative job insecurity as more intense than people with a high score on CSE. In other words, I contend that CSE and

qualitative job insecurity interact in their influence on commitment to change, so I hypothesize:

Hypothesis 2: The relationship between qualitative job insecurity and commitment to change will be moderated by core self-evaluations (CSE) such that this negative relationship will be stronger at lower levels of CSE.

Besides characteristics of an employee, contextual factors can also have an influence on the relationship between quantitative job insecurity and commitment to change. Leadership style could be such a factor and I will focus on the role of servant leadership because as I will discuss later, servant leadership can mitigate the negative consequences of qualitative job insecurity by focusing on satisfaction of follower basic psychological needs.

The concept of servant leadership was introduced by Greenleaf (1970), followed by decades of debate in literature about what exactly comprises servant leadership. Part of this debate was the criticism that there is a considerable overlap between servant leadership, transformational leadership, authentic leadership and ethical leadership (Chughtai, 2016). Only few empirical studies have tried to determine the fundamental differences between servant leadership and transformational leadership. However, these studies only demonstrated

(18)

that servant leadership explained unique variance in outcomes (such as follower commitment) beyond the effects of transformational leadership (Van Dierendonck et al., 2014). In other words, these studies indicated that servant leadership and transformational leadership are different, without exactly explaining how these leadership theories differ.

Recently, in a study into the differential mechanisms of transformational leadership and servant leadership Van Dierendonck et al.(2014) found that transformational leadership worked mainly through perceived leadership effectiveness, whereas servant leadership worked primarily through follower need satisfaction. I will return to the point of follower need satisfaction for further explanation later, but will first elaborate in more detail on what servant leadership is.

Van Dierendonck (2011) reviewed the literature and identified six characteristics of servant leadership, which are humility, standing-back, empowerment, accountability, authenticity and stewardship (Van Dierendonck, 2011). Humility reflects the modesty of the leader and consists of three subcomponents: a) the ability to put one’s accomplishments and talents in perspective, b) admitting one’s fallibility and mistakes and c) understanding of one’s strong and weak points. Standing-back is the extent to which a leader gives priority to the interest of others first and gives them the necessary support and credits. Empowerment is about encouraging autonomous decision making, sharing information, and the coaching and mentoring of individuals for increased innovative performance. Accountability points to making sure that people feel responsible for their results. This ability allows the servant leader to provide direction to followers while considering the specific capabilities of people, as well as their particular needs. Authenticity concerns being true to oneself in showing sincere intentions and behaviour, both in private and in public. Stewardship is the notion of vision or long term orientation (Van Dierendonck, 2011). In servant leadership, humility and action-driven behaviour exist simultaneously. The humble service comprises the components

(19)

humility and standing-back, whereas the action-driven behaviour stems from the components empowerment, accountability and stewardship (Sousa & Van Dierendonck, 2015).

Servant leadership has several positive outcomes for individuals and organizations, as in research positive relations have been found between servant leadership and job satisfaction and work engagement (Van Dierendonck & Nuijten, 2011), task performance and organizational citizenship behaviours (Chiniara & Bentein, 2016), employee voice and negative feedback seeking behaviour (Chughtai, 2016), decreased follower turnover intentions (Hunter, Neubert, Perry, Witt, Penney & Weinberger, 2013) and commitment to change (Kool & Van Dierendonck, 2012). Sousa and Van Dierendonck (2014) conducted a study into work engagement in a merge process under high uncertainty and found that servant leadership strongly affected work engagement in a positive manner whereas organizational identification and psychological empowerment acted as mediating variables.

As mentioned before, the differential mechanism of servant leadership is satisfaction of basic psychological needs. A question could be: what are basic psychological needs? A central theory in the domain of basic needs is self-determination theory (SDT), which states that for the optimal functioning of an individual three universal basic psychological needs must be satisfied (Ryan and Deci, 2000): (1) the need for autonomy; (2) the need for relatedness or feeling connected to others; and (3) the need for competence or having an overall sense of effectiveness. According to SDT, when individuals’ basic needs are fulfilled, employees feel more engaged and will deploy favourable, productive behaviours. In contrast, when basic needs are frustrated, a negative spiral is precipitated. Individuals may instead develop defences or need substitutes (Deci and Ryan, 2000), which may decrease their overall functioning.

As servant leaders focus on follower needs satisfaction I argue that servant leadership can mitigate the negative consequences of qualitative job insecurity. To be more precise,

(20)

qualitative job insecurity acts as a stressor which can frustrate the satisfaction of basic needs (Van den Broeck et al, 2014). It is known from research that a high level of social support (i.e., the satisfaction of social needs) diminishes the negative outcomes of job uncertainty (Sverke & Hellgren, 2002; Sverke, Hellgren, & Näswall, 2002). Servant leaders can provide this social support as they, more than other leaders, exclusively focus on addressing the needs of their followers (Chughtai, 2016). Servant leaders will invest time and energy to understand the needs of all individual followers and subsequently work to satisfy these needs (Van Dierendonck et al. 2014) by listening, empathizing and acknowledging that change can bring many uncertainties that need to be addressed. This will allow employees to feel listened to (Sousa & Van Dierendonck, 2014). Although this does not affect the stressor itself directly, it may have beneficial effects for the individual by preventing the most negative reactions from occurring (Sverke & Hellgren, 2002). In other words, servant leadership can have a moderating effect on the negative relationship between job insecurity and commitment to change. I expect that in situations with high servant leadership, job insecurity will have a less strong impact on employees. This leads to the following hypothesis:

Hypothesis 3: The relationship between qualitative job insecurity and commitment to change will be moderated by servant leadership such that this negative relationship will be weaker at high levels of servant leadership.

Thus far I have elaborated on the direct relationship between qualitative job insecurity and commitment to change, which will be influenced by the aforementioned moderators. However, as stated in the previous section, Van den Broeck et al. (2014) found that qualitative job insecurity acts as a stressor which can frustrate the satisfaction of basic needs. This

(21)

indicates that satisfaction of basic needs could mediate the aforementioned relationship. In the next section I will focus on the mediating role of basic psychological needs.

2.4 The mediating role of basic psychological need satisfaction

As described in the previous section, SDT states that when individuals’ basic needs are fulfilled, employees feel more engaged and will deploy favourable, productive behaviours. In contrast, when basic needs are frustrated, a negative spiral is precipitated. Individuals may instead develop defences or need substitutes (Deci and Ryan, 2000), which may decrease their overall functioning.

Van den Broeck et al. (2014) tested the relationship between qualitative job insecurity and organizational counterproductive work behaviour. It was found that feeling insecure about one’s valued job aspects was associated with high levels of need frustration and, therefore, also with organizational counterproductive work behaviour. In this model the relationship between job insecurity and counterproductive work behaviour was fully mediated by frustration of basic psychological needs. An explanation for this finding is possibly that qualitative job insecurity is involuntary and undesired and will negatively affect the need for autonomy. Furthermore, employees experiencing qualitative job insecurity do not exactly know whether job features will change and to which extent. This undermines their possibility to undertake actions to deal with these changes which results in the frustration of the need for competence (Van den Broeck et al. (2014). Finally, qualitative job insecurity might signal a change for employees in their relationship with colleagues and the organization. Specifically, it may imply a breach of the psychological contract in which employees expect job security in return for their loyalty (Rousseau, 1995), thereby frustrating employees’ need for belongingness. In addition, changes regarding one’s supervisor or colleagues may also impact the need for belongingness.

(22)

Importantly, servant leadership has been found to enhance satisfaction of the three basic needs (Stynen et al., 2015; Van den Broeck, 2014; Chiniara & Bentein, 2016; Van Dierendonck et al. ,2014). So, I argue that servant leadership influences the relationship between qualitative job insecurity and basic need satisfaction. In other words: servant leadership and qualitative job insecurity will interact on basic need satisfaction. In environments where employees experience a high level of servant leadership, in general, their basic needs will be more satisfied compared to employees who do not experience servant leadership. When confronted with qualitative job insecurity, the basic needs of the employees who experience a high level of servant leadership will be less frustrated compared to employees who do not experience servant leadership.

Up to this point in this section I have focussed on the relationship between qualitative job insecurity and servant leadership on the one hand and satisfaction of basic psychological needs on the other. However, another question is what the effect would be of satisfaction or frustration of basic psychological needs on commitment to change? Greguras and Diefendorff (2009) found that satisfaction of the three basic psychological needs from SDT was positively related with affective organizational commitment. When applied to the context of this thesis, frustration of basic needs would lead to less affective organizational commitment. Van den Broeck, Vansteenkiste, De Witte, Soenens and Lens (2010) developed and validated a work-related basic need satisfaction measure and, furthermore, examined the associations between need satisfaction and organizational commitment. Specifically, affective organizational commitment was examined as this was considered to be the main component of organizational commitment. This is most predictive of, for instance, job satisfaction and positive affect, and is strongly influenced by organizational aspects such as leadership and job characteristics (Allen & Meyer, 1990). Van den Broeck et al. (2010) found that work-related basic need satisfaction was positively related to affective organizational commitment.

(23)

As commitment to change is a more specific and proximal variant of organizational commitment, it is reasonable to expect that work-related basic need satisfaction will be positively related to affective commitment to change. Therefore, combined with the statement that the interaction of qualitative job insecurity and servant leadership is negatively related to the satisfaction of basic psychological needs, I hypothesize the following:

Hypothesis 4: The interaction of qualitative job insecurity and servant leadership on commitment to change will be mediated by basic psychological need satisfaction such that:

a. qualitative job insecurity will be negatively related to basic need satisfaction and this relationship will be less strong at high levels of servant leadership.

b. basic psychological need satisfaction will be positively related to commitment to change

A similar reasoning can be followed with respect to the previous identified moderator CSE. In addition to the above described influence of qualitative job insecurity on the satisfaction or frustration of basic psychological needs, the interaction of qualitative job insecurity and CSE of people with a low score on CSE will amplify this influence. As mentioned before, CSE consists of a certain set of personality dimensions, which comprises of self-efficacy, locus of control, neuroticism and self-esteem. It is expected that qualitative job insecurity has a stronger negative influence on satisfaction of the basic need competence for people with a low score on self-efficacy in contrast to people who have a more positive evaluation about their potential to perform in a variety of circumstances. Similarly, satisfaction of the basic need autonomy under circumstances of qualitative job insecurity is likely to be more problematic for people with a external locus of control than for people with an internal locus of control. Finally, I expect that people with a high score on neuroticism do not satisfy their basic need

(24)

belongingness to the same extent as people who are more emotionally stable when confronted with qualitative job insecurity.

In conclusion, the interaction of qualitative job insecurity and CSE is related to the satisfaction of basic psychological needs. As stated above, I expect qualitative job insecurity to be negatively related to basic need satisfaction and I expect this relationship to be stronger for people with a low SCE-score. As I have already clarified the relationship between satisfaction of basic psychological needs and commitment to change, I can introduce the final hypothesis:

Hypothesis 5: The interaction of qualitative job insecurity and CSE on commitment to change will be mediated by basic psychological need satisfaction such that:

a. qualitative job insecurity will be negatively related to basic need satisfaction and this relationship will be stronger at low levels of SCE.

b. basic psychological need satisfaction will be positively related to commitment to change

(25)

3. Method and measures

The research model for this study is depicted in figure 1.

Fig. 1: Hypothesized moderated mediation model

3.1 Research design

As all variables of the research model reflect internal states or perceptions of employees, the most suitable research design was a self-report survey. To prevent the common method bias from occurring, commitment to change has been measured at two different moments (Spector, 2006) with an interval of two weeks. Data were collected using self-administered questionnaires, which were developed using Qualtrics online survey software. The questionnaires were tested by 3 persons as a pilot before distributing them to the participants. Employees were invited by e-mail to fill in the questionnaires. This e-mail served as an introduction and explanation of the purpose of the surveys and contained a link to the online questionnaires. Confidentiality and anonymity were stressed. Participation in this study was completely voluntary. The invitation e-mail can be found in Appendix I.

3.2 Sample

The study was conducted in three municipal organizations who were preparing for a merger as partners on January 1, 2019. Mergers of municipal organizations have in common that

qualitative job insecurity Basic psychological need satisfaction Commitment to change Core self-evaluations Servant leadership

(26)

employees have the assurance of a job in the new organization, but the exact characteristics of the new job are not known. In other words: there is qualitative job insecurity.

A recent culture survey in these three organizations resulted in a response rate of 35%. To make reliable inferences about the hypotheses in this thesis, at least 50 respondents were needed. For reasons of safety I counted on 80 respondents. This implied that 230 employees needed to be invited. As the three organizations in total have 240 employees with a permanent contract, the questionnaires were sent to all 240 employees. Two weeks after the initial invitation a reminder was circulated. A few employees had a labour contract with two of the three organizations. They received the survey only once, namely in the organization in which they work the most hours.

For the survey at time 1, 107 employees completed the questionnaire on the questions with respect to the variables of the research model. On the control variables, 18 respondents left open one or more answers. In total 89 employees filled in the entire questionnaire, which is a response rate of 37%. The sample consisted of employees aged 22–64 with an average age of 46.6 years (SD 11.13). About 42% of the sample was male and about 66% of the sample worked fulltime. Table 1 provides sample characteristics in more detail.

Table 1: Sample characteristics

Age 18-30 (n=9) Age 31-49 (n=42) Age 50+ (n=38) % % % 1. Gender Male 3 17 22 female 6 32 20 2. Education

no higher education degree 0 14 20

higher education degree 9 35 22

M (in years) M (in years) M (in years)

3. Tenure in organization 2.33 11.52 17.54

(SD=2.33) (SD=7.49) (SD=11.25)

4. Working with current manager 2.33 4.86 5.82

(27)

For the survey at time 2, in which only commitment to change was measured, 69 employees completed the questionnaire, which is a response rate of 29%. In questionnaire 1 as well as in questionnaire 2, employees were asked to create a personal code (Schnell, Bachteler & Reiher, 2010), which made it possible to match employees across the two surveys. The composition of the personal code is included in the questionnaires in the appendices. In total, 52 employees completed both surveys.

3.3 Concepts and measures

To measure the variables of the conceptual model, only validated scales were used.

Qualitative job insecurity: In this study qualitative job insecurity was measured using the 17-item scale as developed by De Witte et al. (2010). Participants were asked to indicate on a five-point scale whether certain job features would likely improve or deteriorate in the near future (1 = strongly deteriorate; 5 = strongly improve). Example items of job features are “the status that comes with your position in the company” and “the opportunity to interact with the public”. Reliability (Cronbach’s α) for this scale was 0.804. Since a higher score on this five-point scale would essentially imply lower job insecurity, all items of this scale were recoded.

Servant leadership: servant leadership was measured with the 14-item scale as used by Ehrhart (2004). Examples of sample items are “my manager holds employees to high ethical standards” and “my manager’s decisions are influenced by employees’ input” Participants were asked to rate the items on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s α for this scale equalled 0.931.

CSE: CSE was measured using the 12-item Core Self-Evaluations Scale (CSES) developed by Judge et al. (2003). Examples of sample items are “sometimes I feel depressed” and “when I try, I generally succeed”. Participants were asked to rate the items on a five-point

(28)

scale ranging from 1 (strongly disagree) to 5 (strongly agree). The Cronbach’s Alpha coefficient for this scale was 0.742.

Basic psychological need satisfaction: to measure basic need satisfaction the 9-item scale of Sheldon, Elliot, Kim, and Kasser (2001), which was modified by Van Dierendonck et al. (2014), was used. Examples of sample items are “at work, I am free to do the things my own way” and “at work, I have a sense of contact with people who care for me, and whom I care for.” Participants were asked to rate the items on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s α for this scale was 0.755.

Commitment to change: commitment to change was measured with the 6-item scale of Herscovitch and Meyer (2002), which reflects only the affective component of commitment to change. This is in line with Van den Broeck et al. (2010), who considered the affective component as the most important component of commitment to change. In contrast to the original scale, the word ‘change’ was replaced by the word ‘merger’ in this study. Examples of sample items are “I believe in the value of this merger” and “things would be better without this merger”. Participants were asked to rate the items on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s α for this scale for time 1 was 0.942. For the time 2 questionnaire, Cronbach’s α equalled 0.684. After removing the item “this merger is not necessary”, Cronbach’s α increased to 0.942 also.

Control variables: the following control variables were included: gender (0 = men, 1 = women), education (0 = no education beyond secondary vocational school, 1 = education beyond secondary vocational school), working hours (0 = full-time, 1 = part-time, in which part-time is less than 32 hours a week), age, tenure in the organization and tenure with the current manager. These questions might be of value in explaining unexpected results or in offering alternative insights. For example, it is possible that men and women experience insecurity in a different manner. This could also be the case for young versus older

(29)

employees. In the same way, it is possible that employees with a low education would suffer more from qualitative job insecurity since their opportunities for applying for another job within a knowledge intensive organization as a municipality are scarce. This reasoning could also be applied to a lesser extent to employees who work part-time. Control variable ‘tenure with the organization’ is used because it could be argued that employees who work long for one organization are less attractive for other organizations. In other words, the employability of these employees is potentially reduced. As a consequence, these employees migth experience a higher job insecurity because their opportunities for finding another job outside the organization are limited in case their new job after the merger is not satisfactory to them. Finally, for truly experiencing the leadership style of a manager, it is necessary for an employee to work a certain time together with the manager. For this reason the control variable ‘tenure with the current manager’ was incorporated.

As only Dutch employees would fill in the questionnaire, the questionnaire was only available in Dutch. However, the scales adopted for the questionnaire stem from English studies, therefore the questions were first translated to Dutch by two native Dutch speakers. The questionnaires of time 1 and time 2 can be found in Appendices II and III. The English versions can be found in Appendices IV and V.

(30)

4. Results

4.1 Descriptive statistics

Before testing the hypotheses, counter-indicative items were recoded. These items are marked with (R) in the questionnaires in appendices IV and V. Furthermore, skewness, kurtosis, normality tests and reliabilities were computed for all variables before creating summated scales. As mentioned before, 52 employees filled in both questionnaire 1 and 2. From a paired-sample t-test it appeared that there was no significant difference between commitment to change at time 1 versus time 2 (t = -1.663, p > 0.05). Statistically, there is no reason to assume that commitment to change at time 1 is different from time 2. As sample 2 was small relative to sample 1, further analyses for hypotheses testing were based on the data of sample 1 only. Descriptives, inter-correlations and reliabilities of the scales based on the data from questionnaire time 1 are presented in table 2.

Table 2: Descriptives and intercorrelations, and reliabilities of the scales are presented in Table 1.

M SD 1 2 3 4 5

1.

qualitative job

insecurity 2.96 .34 (.804)

2. commitment to change 3.69 .78 -.552** (.942)

3. basic need satisfaction 3.71 .38 -.031 .158 (.755)

4. servant leadership 3.43 .67 -.104 .091 .283** (.931)

5. CSE 3.52 .41 -.209* .238** .451** .146 (.742)

N = 107. * p < .05. **p < .01.

The correlations between qualitative job insecurity, satisfaction of basic needs and commitment to change are in the expected direction. That is, qualitative job insecurity is negatively related to commitment to change and to satisfaction of basic needs, which in turn is positively related to commitment to change. Unfortunately, the latter two relationships were not significant.

(31)

4.2 Regression analyses

The hypotheses were tested with a hierarchical regression analysis. In step 1, the control variables were entered. This model was statistically significant (R2=0.162, F(6,81) = 2.606, p < 0.05) and explained 16.2% of variance in Commitment to Change. Independent variable Qualitative Job Insecurity was entered in step 2 (R2=0.345, F(7,80) =6.027, p < 0.001). Qualitative Job Insecurity had a significant negative effect on Commitment to Change

(β = - 0.483, p < 0.001), which explained additional 18.3% variance in Commitment to Change (Fchange = 22.420, p < 0.001). Thus, hypothesis 1 is supported by the data.

Subsequently, in step 2A, the variables CSE and Servant Leadership were entered (R2=0.374, F(9,78) = 5.183, p < 0.001) to calculate their effect on Commitment to Change. This increased the explained variance in Commitment to Change with 2.9% (Fchange = 1.804,

p > 0.05). However, the main effects of CSE (β = 0.182, p > 0.05) and Servant Leadership (β = 0.018, p > 0.05) on Commitment to Change were not significant.

Then, the interactions of CSE and Servant Leadership were tested. After removing Servant leadership, in step 3A, the interaction term between Qualitative Job Insecurity and CSE (QJI*CSE) was entered (R2=0.386, F(9,78) = 5.457, p < 0.001), which increased the explained variance in Commitment to Change with respect to the model of step 2 with 4.1% (Fchange = 2.613, p > 0.05). The interaction between CSE and Qualitative Job Insecurity was

not significant ((β = - 0.120, p > 0.05), so hypothesis 2 was not supported. In step 3B, after removing CSE and QJI*CSE, Servant Leadership was entered as a moderator (R2=0.364, F(9,78) = 4.950, p < 0.001). Introducing Servant Leadership increased the explained variance with respect to the model of step 2 with 1.8% (Fchange = 1.120, p > 0.05). The interaction

between Servant Leadership and Qualitative Job Insecurity was not significant (β = - 0.133, p > 0.05), so hypothesis 3 was not supported. When in step 3C both moderators CSE and Servant Leadership were entered, 39.9% of variance in Commitment to Change was explained

(32)

by the model (R2=0.399, F(11,76) = 4.585, p < 0.001), which is an increase of only 5.4% with respect to the model of step 2 (Fchange = 1.695, p > 0.05). Finally, in step 4 the mediator Basic

Need Satisfaction was entered, combined with both moderators. This model was analyzed with the PROCESS tool developed by Hayes (2012). It appeared that the model explained almost 38% of variance in Commitment to Change (R2=0.3798, F(8,79) = 6.048, p < 0.001). For the conditional indirect effects of Qualitative Job Insecurity on Commitment to Change through Basic Need Satisfaction, PROCESS has estimated the indirect effect and its corresponding 95% confidence intervals at different values of the moderators CSE and Servant Leadership. These values are the mean and plus/minus one SD from mean. As all calculated confidence intervals included zero, it must be concluded that the indirect effect of the interaction between independent variable Qualitative Job Insecurity and the moderators CSE and Servant Leadership on Commitment to Change was not significant. Thus, the effect of the interaction between independent variable Qualitative Job Insecurity and the moderators CSE and Servant Leadership on Commitment to Change was not mediated by Basic Need Satisfaction at none of the levels of the moderators, nor partially mediated. Hypotheses 4 and 5 must be rejected. Appendices VI and VII contain details about the results of the regression analyses and the PROCESS analysis.

4.3 Additional analyses

To gain more insight in the research model, some additional analyses have been performed. First, the influence of the independent variable and moderators on the mediator and the effect of the mediator on the dependent variable were investigated in more detail. From a regression analysis it appeared that the moderators CSE (β = 0.432, p < 0.001) and Servant Leadership (β = 0.214 p < 0.05) had a significant direct effect on Basic Need

(33)

Satisfaction. There was no significant direct effect of Qualitative Job Insecurity on Basic Need Satisfaction (β = 0.057, p > 0.05).

The interaction between Servant Leadership and Qualitative Job Insecurity on Basic Need Satisfaction was significant (β = -0.248, p < 0.05), which is in line with the assumptions on which hypotheses 3 and 4 were based. The nature of this interaction is depicted in Fig. 2, which shows that, regardless of the level of qualitative job insecurity, the levels of basic need satisfaction are higher in circumstances of high servant leadership opposed to circumstances of low servant leadership.

Fig.2 Plot of interaction effect between Servant Leadership and Qualitative Job Insecurity on Basic Need Satisfaction

The interaction between CSE and Qualitative Job Insecurity on Basic Need Satisfaction was not significant (β = 0.064, p > 0.05). Furthermore, from the regression analysis it appeared that mediator Basic Need Satisfaction had a significant direct effect on dependent variable Commitment to Change (β = 0.209, p < 0.05), which is in accordance with hypotheses 4B and 5B. Further details about these analyses can be found in appendix VIII.

As second additional analysis, as only the interaction between Servant Leadership and Qualitative Job Insecurity was significant, the moderated mediation research model was tested without moderator CSE. The model explained almost 38% of variance in Commitment to Change (R2=0.3798, F(8,79) = 6.048, p < 0.001). All calculated confidence intervals for the conditional indirect effects of the interaction between independent variable Qualitative Job

2,9 2,95 3 3,05 3,1

Low QJI High QJI

B asic Ne ed s S at isf ac tion Low SL High SL

(34)

Insecurity and moderator Servant Leadership on Commitment to Change and also the confidence interval for the index of moderated mediation included zero, which means that the indirect effect of the interaction between independent variable Qualitative Job Insecurity and moderator Servant Leadership on Commitment to Change was not significant. Thus, the effect of the interaction between independent variable Qualitative Job Insecurity on Commitment to Change is not mediated by Basic Need Satisfaction, nor partially mediated. Further details about these analyses can be found in appendix IX.

(35)

5. Discussion

5.1 Summary

The current study examined the potential moderating influence of CSE and servant leadership on the relationship between qualitative job insecurity and commitment to change. I proposed a research model in which the relationship between qualitative job insecurity and commitment to change was mediated by satisfaction of basic psychological needs. In other words, it was expected that employees who experienced higher levels of qualitative job insecurity would be less satisfied on the basic psychological needs autonomy, relatedness and competence and in turn would be less committed to the change. Furthermore, it was hypothesized that CSE and servant leadership would act as moderators. More specifically, it was supposed that the negative relationship between qualitative job insecurity and commitment to change would be stronger at low levels of SCE and less strong at high levels of servant leadership. Results revealed that indeed employees who experienced higher levels of qualitative job insecurity showed less commitment to change. There was no support found for the expected mediating role of satisfaction of the basic psychological needs: employees who suffered from qualitative job insecurity did not report lower levels of satisfaction of autonomy, relatedness and competence. Finally, it was found that CSE and servant leadership did not intensify or attenuate the relationship between qualitative job insecurity and commitment to change.

The proposed research model explained 37.98% of variance in commitment to change. Regrettably, there was no significant difference in explained variance between the proposed research model and the model of step 2 of the regression analysis, in which besides the control variables, only qualitative job insecurity was entered. The hypotheses with respect to the moderating influence of CSE and servant leadership and the mediating role of satisfaction of basic psychological needs had to be rejected.

(36)

5.2 Theoretical implications

Qualitative job insecurity and commitment to change

In volatile business environments, employees in many organizations are faced with ongoing changes. Commitment of employees to the proposed changes is an crucial prerequisite for organizational success (Self et al., 2007). At the same time, these changes come with uncertainty for employees. Little is known about the influence of job insecurity on commitment to change. Ashford et al. (1989) reported a negative relationship between job insecurity and organizational commitment and Kalyal et al. (2010) found a negative relationship between job insecurity and commitment to change. However, job insecurity can be distinguished in a quantitative and a qualitative dimension. While job insecurity in general has been studies extensively, attention for qualitative job insecurity has increased only the last decade. In this study, in line with hypothesis 1, a significant negative relationship was found between qualitative job insecurity and commitment to change. In this way this study contributes to current knowledge as it goes beyond existing studies in which job insecurity in general is related to organizational commitment (Ashford et al., 1989) or commitment to change (Kalyal et al., 2010) and it demonstrates the relationship explicitly for the qualitative version of job insecurity.

Based on this finding, a critical remark towards the ‘best-practice school’ in strategic human resource management can be made. Pfeffer (1998) summarized seven best practices, among which was employment security. However, it is clear that even if employment is guaranteed, qualitative job insecurity can occur. Consequently, as qualitative job insecurity is negatively related to commitment to change, employees’ support for the change is not guaranteed.

(37)

CSE and commitment to change

No support was found for hypothesis 2 which holds a moderating influence of CSE on the relationship between qualitative job insecurity and commitment to change. From table 2 it can be derived that CSE and qualitative job insecurity were significantly negatively correlated, which corresponds with Låstad et al. (2014). However, table 2 shows that CSE was also significantly positively correlated with commitment to change and basic needs satisfaction. In other words, CSE was correlated with all variables of the research model, except the other moderator servant leadership. Apparently, CSE is directly related to qualitative job insecurity and commitment to change instead of influencing the relationship between qualitative job insecurity and commitment to change.

Servant leadership and basic need satisfaction

Hypothesis 3, which stated that servant leadership has a moderating influence on the relationship between qualitative job insecurity and commitment to change, was also needed to be rejected. From literature (Van Dierendonck et al., 2014) it is known that servant leadership works through basic need satisfaction. Apparently, servant leadership does not have an influence without this differential mechanism basic need satisfaction. A possible explanation for not finding the expected results from hypothesis 3 appears when a closer look is taken on the analyses of hypothesis 4. This hypothesis holds that the interaction of qualitative job insecurity and servant leadership on commitment to change will be mediated by basic need satisfaction. The regression analysis (appendix VIII) revealed that qualitative job insecurity had no significant main effect on satisfaction of basic psychological needs (β = 0.057, p > 0.05). This is contrary to research findings of Van den Broeck et al. (2014), who found that the relationship between job insecurity and counterproductive work behaviour was fully

(38)

mediated by frustration of basic psychological needs. Surprisingly, Stynen et al. (2015) found a partial mediation of basic need satisfaction in the relationship between qualitative job insecurity and OCB, in which the direct effect was weaker in magnitude than the mediated effect. Possibly, the way of measurement of the variable Basic Need Satisfaction could be an part of the explanation for these contradictory results. I followed Van Dierendonck et al. (2014) in measuring satisfaction of basic psychological needs as a unifying construct, whereas Van den Broeck et al. (2014) and Stynen et al. (2015) measured the three components of SDT individually. In the applied measure in this thesis the three components of SDT are reflected in three items for each component. When testing the reliability for each separate SDT-component, only the items measuring relatedness exceeded the threshold of 0.70 (Cronbach’s α= 0.759). The reliabilities for competence and autonomy were respectively 0.623 and 0.600, although the reliability for the unified construct was 0.755. To rule out potential measurement errors, it is recommendable to replicate this study in future research with the measures used by Van den Broeck et al. (2014) and Stynen et al. (2015).

Interaction between qualitative job insecurity and servant leadership

Although the interaction of qualitative job insecurity and servant leadership had no significant influence on commitment to change, there was a significant influence on satisfaction of basic psychological needs (β = -0.248, p < 0.05). The nature of this interaction is depicted in Fig. 2. In circumstances of low servant leadership it appeared that basic need satisfaction is higher under conditions of high qualitative job insecurity than for conditions of low qualitative job insecurity, which is contrary to prior expectations. This phenomenon is also present under conditions of high servant leadership, but then the difference between the levels of basic need satisfaction is smaller than in circumstances of low servant leadership. However, irrespective of the level of qualitative job insecurity, the levels of basic need

(39)

satisfaction are higher in circumstances of high servant leadership opposed to circumstances of low servant leadership. This means that employees experience higher levels of satisfaction of basic psychological needs when their managers show more servant leadership behaviour.

Given the null-findings with respect to CSE, the moderated mediation research model was, additional to the hypotheses testing, tested without moderator CSE. Although the interaction term did not reach significance on the 0.05-level, it came close with a p-value of 0.081. Since in this model the effect of basic need satisfaction on commitment to change was also significant (β = 0.385, p < 0.05), this model has potential for further investigation when taken into account the potential measurement errors of basic need satisfaction. In this way this thesis may have a small contribution to literature by pointing to servant leadership as a style of leadership which is conducive to commitment to change and thereby to the success of the proposed change eventually. After all, the levels of basic need satisfaction appeared to be higher in conditions of high servant leadership and, subsequently, basic need satisfaction is positively related to commitment to change. So, servant leadership could be a new link in the chain of variables for creating more commitment to change and, thereby, increasing the possibility of a successful change.

5.3 Strengths, limitations and future research

There are two strengths in this study that are worth mentioning. First, studies in which job insecurity is related to commitment to change are relatively scarce. This study extends the study of Kalyal et al. (2010) as it focuses on qualitative job insecurity instead of job insecurity in general. Second, the survey took place in the period around the final decision making about the merger. At this time, there was very little known about the new organization that will arise out of the three merger partners. This reflects a pure situation of qualitative job insecurity.

(40)

This study also has some limitations. First, all variables were measured through self-report surveys, suggesting a potential common-method bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). To overcome this bias, the variable Commitment to Change was measured at two different moments with an interval of two weeks. However, in future research all variables could be measured on more different moments than just once. Second, due to the cross-sectional design of this study, it is not possible to make inferences about causality. In order to determine causality, in future research a longitudinal research design is recommended. Third, there was a relative small response to the questionnaire (N=89). Although enough for statistical analyses, a larger sample size would result in more statistical power. Fourth, the questionnaire was distributed in three different organizations, which each contribute to a certain amount of variance due to differences between these organizations. Respondents were instructed to create a personal code, which started with the first letter of their municipality, for two reasons. Primarily, to match the answers of the questionnaires on time 1 and time 2. Secondarily, to distinguish between the three organizations. Unfortunately, quite a lot of the employees did not conform to the instruction but created a personal code in a different way. In future research this study could be replicated in one organization to rule out variance due to differences between these organizations. This goal could also be attained by introducing a control variable which distinguishes between the organizations. Finally, the respondents of this study all worked in small municipal organizations. This might limit the generalizability of the findings to other industries. Future research could incorporate other industries and other cultural contexts.

From a theoretical perspective, some directions for extension of this research can be mentioned. First, in the research model of this study servant leadership has been incorporated as a moderator. Respondents were asked to rate the extent of servant leadership behaviour of their direct manager. However, it is questionable whether this direct manager makes the

The moderating influence of core self-evaluations and servant leadership on the relationship between qualitative job insecurity and commitment to change

Master Thesis