The Moderating Effect of (Non-)Threatening Change on the
Relationship between Calling and Well-being
Date: 23th of June, 2017 Student: Femke Aarts - 11417943 Education: Msc. Business Administration Track: Leadership & Management
University: Amsterdam Business School - University of Amsterdam Supervisors: H.A. Berkers & dr. S.T. Mol
ABSTRACT
A lot of researchers are looking into the concept of callings: looking for intrinsic meaning and an opportunity to make a difference in doing your work. Callings have a lot of positive outcomes (e.g. life satisfaction), but also negative outcomes are found (e.g. burnout). Yet, the field of research lacks knowledge about possible conditions under which these outcomes weighs heavier. Therefore, this study is investigating the moderation effect of change on the relationship between callings and happiness-, health- and relational-related well-being. Two ways of framing change are examined, non-threatening change and threatening change. Whereas non-threatening change, also seen as an opportunity, is expected to have a positive moderation effect on happiness and relations, threatening change is expected to have a negative moderation effect. The moderation effect of change on health is expected to be negative for both types of framing. For this study a Vignette Experiment Method is used to collect data among teachers and accountants. The results show that there are differences between teachers and accountants, as only for teachers callings have a positive effect on happiness and relations. No relationships are found for calling and health. For accountants, no significant results are found. The moderation effect of change of the relationship between callings and the three types of well-being is not supported for any of the professions.
Key words: calling, employee well-being, happiness, health, relations, threatening change, non-threatening change
STATEMENT OF ORIGINALITY
This document is written by Student Femke Aarts 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.
TABLE OF CONTENTS ABSTRACT ... 2 STATEMENT OF ORIGINALITY ... 3 1. INTRODUCTION ... 5 2. LITERATURE REVIEW ... 8 2.1 Calling ... 8 2.2 Employee well-being ... 10 2.2.1 Happiness ... 11 2.2.2 Health ... 13 2.2.3 Relations ... 16
2.3 The Moderating effect of Change ... 18
2.4 Research model ... 26
3. METHOD ... 27
3.1 Research approach ... 27
3.2 Sample and procedure ... 28
3.3 Design ... 31 3.4 Measurements ... 32 3.5 Analyses ... 35 4. RESULTS ... 37 4.1 Correlation analysis ... 37 4.3 Direct effects ... 38 4.4 Moderation effect ... 42 5. DISCUSSION ... 45
5.1 Theoretical implications and directions for future research ... 46
5.2 Practical implications ... 52 5.3 Limitations ... 53 6. CONCLUSION ... 54 7. REFERENCES ... 56 8. APPENDIX ... 65 Appendix A: Vignettes ... 65
Appendix B: Measurement scales ... 68 Appendix C: Normal distribution of Calling, Stress, Work Engagement & LMX . 69
1. INTRODUCTION
Do we work to live or do we live to work? Work is becoming more and more important in our lives and our motives to work have changed over time. We do not only work because we need to have money so we are able to live, but we also choose to work because it is meaningful to us (Chalofsky, 2003). Meaningful work is described as work that is “experienced particularly significant and holding more positive meaning for individuals” (Rosso, Dekas & Wrzesniewski, 2010; p. 95). Due to this change, there is an increase in people who are trying to pursue their “calling”. This means that individuals are thought to seek meaning and an opportunity to make a difference in doing their work, which are both subjectively by the individual and includes benefiting the individual itself, the family of the individual, and/or society (Cardador & Caza, 2012; Hall & Chandler, 2005). An increasingly amount of researchers is studying the concept of callings (Dik & Duffy, 2009; Hall & Chandler, 2005; Steger, Pickering, Shin & Dik, 2010; Wrzesniewski, McCauley, Rozin & Schwartz, 1997). Callings can have a big impact on the well-being of an employee, and thus also on the organization for which the employee is working (Torrey & Duffy, 2012). However, due to contradictory outcomes of having a calling on employee well-being that have been found in previous research, it is not clear for organizations whether it is beneficial to hire employees with callings or not.
On the one hand, research has found many benefits of having a calling. For example, Wrzesniewski et al. (1997) found that people with callings have a higher life and work satisfaction compared to people who are only focusing on money and necessity rather than pleasure or fulfillment. Callings are also linked to occupational identification and organizational commitment (Cardador, Dane & Pratt, 2011; Duffy, Dik & Steger, 2011). Therefore, organizations have good reasons to support employees in pursuing their calling. On the other hand, few, albeit a growing number of studies also found a dark side of callings. For example, Cardador and Caza (2012) suggest that an unhealthy type of calling exists which is
hypothesized to have a negative influence on the interpersonal relations of the employee. Furthermore, in a study among nurses it was found that the presence of a calling was positively related to burnout because meaningfulness and manageability were not in balance (Vinje & Mittelmark, 2007). Moreover, Bunderson and Thompson (2009) found that people with callings were willing to make whatever personal sacrifices that could help in performing on their work, and that this could lead to physically demanding and dangerous work. These negative outcomes do not only influence the well-being of the employee, but also the organization.
As described, previous research found a lot of different outcomes of callings. Yet, it is still unclear under which circumstances these contradicted outcomes will occur. In this study, the moderation effect of threatening and non-threatening change will be examined. Nowadays, work is changing more often and faster, especially due to technological developments (Lau, Wong, Chan & Law, 2001). Yet, little research has been conducted about the possible different reactions to such changes between people with and without a calling. However, Doborow (2006) suggests that peoplewith a calling could develop a tunnel vision wherein they are less flexible at work and it is harder to adapt to new circumstances during change. In other words, people with callings could be a barrier in implementing changes. Therefore, it is important for organizations to find out how people with callings react in these situations. Moreover, change can also have a demonstrated impact on the well-being of people, which is affecting the organizational performance. For example, Rafferty and Griffin (2006) found that change, divided in three types of change (frequency, planning and transformational), can influence job satisfaction and turnover intentions by increasing uncertainty. Furthermore, Verhaeghe, Vlerick, Gemmel, Maele and Backer (2006) conducted a study on the impact of recurrent changes in the work environment on the well-being of nurses. They found that the occurrence of change was negatively related to job satisfaction, and that the nurses experienced more distress and had a higher sickness absence. However, positive outcomes were also found: when
the changes were experienced as challenging, change was positively related to job satisfaction (Verhaeghe et al., 2006). Therefore, it is important to make a distinction between two types of framing change, namely change framed as threatening and change framed as non-threatening (i.e. seeing change as an opportunity).
The current study sets out to provide new insights by investigating which of the contradictory outcomes of having a calling on well-being weighs heavier under circumstances of change. Because previous research found contradictory outcomes of callings on well-being, there are three types of well-being (happiness-, health- and relational-related well-being) investigated in line with the suggestion by Grant, Christianson and Price (2007). To elucidate the moderating effect of change on the relationship between having a calling and the three types of well-being, the following research question is formulated: “What is the moderating effect of (non-)threatening change on the relationship between callings and health-, happiness-, and relational-related well-being of employees?”. This study contributes to the existing calling literature because of two reasons. First, as there is ambiguity about the outcomes of having a calling, this study provides clearance about which of the contradictory outcomes of callings on employee well-being weighs heavier. These results give insights in the possible advantages or disadvantages for organization in hiring people with a calling compared to people without a calling. Second, as organizations constantly have to deal with changes, this study shows which outcomes of callings on employee well-being weigh heavier under circumstances of change. These results provide relevant information for managers, because they will know if people with callings need to receive extra support during circumstances of change. Moreover, if a difference is found between the two ways of framing the change, the management could frame the change in a way that could lead to desirable outcomes regarding to employee well-being. This is of practical relevance, because the well-being of an employee is positively related to
organizational performance and is therefore important to organizations (Wood, Van Veldhoven, Croon & de Menezes, 2012).
This study consists of five sections: the theoretical framework, method, results, discussion, and conclusion. In the first part, the theoretical framework, the calling literature is reviewed and the link between calling and well-being is examined. Furthermore, the moderating effect of change on this relationship is investigated. In the second part, the method, the sample and methodology of this study are described. In the third part, the results of the study are presented. Fourth, the results are discussed, the limitations of this study are described, and several recommendations for future research in this area are made. Finally, some conclusions of this study are drawn.
2. LITERATURE REVIEW 2.1 Calling
People can have different intentions to work; they can see their work as “a mean that allows individuals to acquire the resources needed to enjoy their time away from their job” (Wrzesniewski et al., 1997, p. 22). Other people might be more engaged to their work and see it as a career where they are focused on advancement. The third group of people takes the next step and finds their work inseparable from their life and see it as their calling (Wrzesniewski et al., 1997). A person who sees one’s work as a calling, does not work merely in order to earn money or to develop his or her career, but for the fulfillment that the work brings. The present study focuses on the third group: people who see their work as their calling.
Although the origin of calling is from a religious context where people were understood to be called by God (Duffy & Sedlacek, 2010), the religious connection is not made very often anymore. The modern sense of calling sees calling as a psychological construct, however, no consensus has been reached yet on how the term ought to be defined. However, Elangovan,
Pinder and McLean (2010) suggested that there are always three fundamental features that stayed constant across all interpretations of callings: (a) an orientation toward action, (b) a sense of clarity of purpose and personal mission, and (c) pro-social intentions. Dobrow and Tosti-Kharas, (2011; p. 1005) defined a calling as ‘‘a consuming, meaningful passion people experience toward a domain’’. Hall and Chandler (2005) consider calling as the highest level of subjective career success where work is seen as one’s purpose in life. These definitions differ in their views of what a calling is or where it exists (Dobrow & Tosti-Kharas, 2011). As callings can be directed to any domain, for example music or volunteer efforts, it is likely that the definitions also differ per domain. Because in the present study the focus is on the work domain, the more specific definition of calling towards work of Cardador and Caza (2012) is adopted. They describe people with a calling as: “individuals who are looking for their work to provide intrinsic meaning and an opportunity to make a difference” (Cardador & Caza, 2012; p.338). However, it is important to take into account that callings do not only influence intrinsic outcomes, but that they are also relate to extrinsic outcomes, illustrated by the research among musicians of Dobrow (2007). She found that musicians with a calling had a higher involvement in music activities, and enjoyed socializing with other musicians more. Therefore, in this study a broader definition of meaningful is used, such that the meaning of work is subjectively by the individual and can include benefiting the individual itself, the families of the individual, and/or society (Hall & Chandler, 2005).
Besides the direction of the calling towards a specific domain, research has found other characteristics that can influence the definition or outcomes of calling. First, a distinction is oftentimes made in research between the presence of a calling and searching for a calling (Duffy & Sedlacek, 2007). Whereas the presence of a calling leads to many positive outcomes such as decidedness and comfort, searching for a calling is just the opposite and is negatively related to decidedness and comfort (Duffy & Sedlacek, 2007). Second, Dik and Duffy (2009) suggest that
it is not the case that people “do” or “do not” have a calling, but that it is an ongoing process in which there is a continuum from weaker to stronger callings and in which callings can develop over time. The effects of strong callings could therefore be somewhat different than relatively weak callings. Third, even people who do not yet have access to a certain domain can already feel a calling, like students who can feel a calling towards a specific domain they aspire to access (Dobrow & Tosti-Kharas, 2011).
As the many definitions and characteristics of callings suggest, an increasingly amount of researchers started studying the concept of callings last decade. The primary focus of these studies was its relation to work and well-being outcomes (Duffy, Allan & Bott, 2012). For example, Vinje and Mittelmark (2007) found that among nurses the presence of a calling was positively related to burnout, and Bunderson and Thompson (2009) found a correlation between calling and work meaningfulness among zookeepers. However, the extant literature still lacks research examining the contradictory effects of callings on employee well-being simultaneously.
2.2 Employee well-being
In the last few years, a lot of research has been conducted on employee well-being. For many companies, employees are the most important resources they have, so it is important that they are feeling well, especially because the well-being of an employee is positively related to organizational performance (Wood et al., 2012). Many different variables have been used to measure well-being in previous research, for example life satisfaction (Wrzesniewski et al., 1997), and burn-out (Vinje & Mittelmark, 2007). To understand the different dimensions of well-being, there is a need to consider the meaning of well-being itself and therefore the definition that will be used is the overall quality of an employee’s experience and functioning at work (Warr, 1987). To take into account that calling could have a different influence on each
of the dimensions of well-being, namely happiness, health and relations (Grant et al., 2007), in this study all three dimensions of employee well-being are considered simultaneously. With this approach, it is possible to see how the three dimensions of well-being are affected differently. It is important to make this distinction because, according to Grant et al. (2007), it is possible that trade-offs between the different dimensions exists. Besides, this is relevant for organizations because when for example, only one of the three outcomes will be negative, the organization can focus more on this dimension and provide extra support to prevent a decrease in this type of well-being and thus also to prevent a decrease in the performance of the employee. The first dimension, happiness-related well-being, can be defined as “the degree to which an individual judges the overall quality of his life favorably” (Veenhoven, 1991; p. 2). The second dimension, health-related well-being, refers to “the physical, psychological, and social domains of health, seen as distinct areas that are influenced by a person’s experiences, beliefs, expectations, and perceptions” (Testa & Simonson, 1996; p. 835). Finally, relation-related well-being, refers to the degree to which relationships are experienced as mutually rewarding and create possibilities for personal development and growth (Dutton & Heaphy, 2003; Kahn, 2007).
2.2.1 Happiness
The most studied link of calling and well-being is the relation of calling and life satisfaction, in other words: the relation between calling and happiness-related well-being. This dimension of well-being is focused on subjective experiences and functioning at work, and is often assessed by employee satisfaction and employee commitment (Grant et al., 2007). Employees who experience high happiness-related well-being are important for organizations because they tend to have a higher job performance (Wright, Cropanzano & Bonett, 2007). The relation between job satisfaction and job performance is moderated by psychological well-being
(PWB), which can be described as the psychological functioning of an individual and is also often seen as happiness (Wright, 2005; Wright et al., 2007). Fredrickson’s model explains this by showing that positive feelings could broaden the awareness, thoughts and actions of someone (Fredrickson & Branigan, 2005; Fredrickson & Losada, 2005). Therefore, a high PWB can broaden a person’s action and thoughts and can help in building personal resources, such that this person will be more creative, socially connected, and physically and mentally healthy (Fredrickson, 2001; Wright, 2005). This means that people high on PWB are able to broaden and expand the positive evaluative sentiments associated with job satisfaction and perform better than people who score low on PWB and job satisfaction.
Linking callings to happiness, empirical research shows that having a calling is positively related to life satisfaction (Duffy & Sedlacek, 2010; Duffy et al., 2012). This relationship can be explained by the model of positive human health of Ryff and Singer (1998). They suggest that eudemonic behavior is the best way to ensure enduring happiness. Having a purpose and meaning in life is a really important part of eudemonia. The calling of employees could be seen as the purpose and meaning of the employees’ lives. This is in line with several definitions of calling, for example Dik & Duffy (2009) who describe calling as “‘ […] deriving a sense of purpose or meaningfulness and that holds other-oriented values and goals as primary sources of motivation”, and the definition of Cardador & Caza (2012) “[…] looking for their work to provide intrinsic meaning and an opportunity to make a difference” (Cardador & Caza, 2012; p.338). People with callings have a strong purpose and meaning in their life and thus experience more eudemonia which lead to an increase in happiness, compared to people without callings.
Besides satisfaction, work engagement is also associated with positive work-related feelings such as happiness and enthusiasm while performing tasks (Kahn, 1990; Schaufeli, Salanova, Gonzalez-Roma, & Bakker, 2002). Engagement is defined as “a positive, fulfilling,
work-related state of mind that is characterized by vigor, dedication, and absorption” (Schaufeli et al., 2002; p. 74). Vigor refers to having lots of energy and mental resilience while working. Dedication is characterized by a strong feeling of involvement in one’s work combined with experiencing a sense of significance, enthusiasm and challenge. Absorption refers to being fully concentrated and happily involved in one’s work, where the time passes quickly and where the person finds it hard to separate his or herself from work (Schaufeli et al., 2002). Although there is no empirical research done at the direct relationship between calling and work engagement, it is likely to assume that there will be a positive relationship. This can be explained by the research of Dobrow (2004). She suggests that the first component of an integrated view of having a calling is a sense of passion one’s work, which is characterized by “deep enjoyment and satisfaction from engaging in one’s work” (Dobrow, 2004; p. 3). Moreover, she suggests that people with a calling have always known that they would engage in a particular type of work. Furthermore, the absorption aspect of work engagement corresponds to the description of people with callings of Wrzesniewski et al. (1997). They describe that people with a calling find their work inseparable from their life. To summarize, people with a calling are more likely to experience work engagement because they have chosen that specific kind of work because it is their passion, which comes with positive feelings like happiness and enthusiasm. Therefore, in this study, happiness is assessed by work engagement, and thus I expect the following:
Hypothesis 1: Calling is positively related to happiness-related well-being (i.e. increase in work engagement).
2.2.2 Health
The second dimension of employee well-being is health (Grant et al., 2007). This type of well-being is studied in terms of objective psychological measures and subjective experiences of bodily health (Testa & Simonson, 1996). Researchers study the link between
work and health in many different ways, for example job stress, fatigue, emotional exhaustion, burnout, or turnover intentions. (Batt and Valcour, 2003; Kroon, van de Voorde & van Veldhoven, 2009; Macky & Boxall, 2008; Peccei, 2004). It is important to study this link because work can be a source of stress and a potential source of injury or disease (Danna & Griffin, 1999; Karasek & Theorell, 1990), which could not only influence the employee, but it could also cost a lot of money for the organization. For example, the costs can increase due to absenteeism, when employees fail in going to work as scheduled regardless of reason, or presenteeism, when employees come to work sick when they should have stayed home and are therefore less productive (Cascio, 2006). Therefore, it is important for organizations that employees are healthy, such that they are able to deliver an excellent job performance.
Linking health-related well-being to calling, Bunderson and Thompson (2009) found in a study among zookeepers that the costs of having a calling are grounded in a strong sense of personal moral duty and that they, because of that sense of moral duty, are willing to make whatever personal sacrifices that could help in performing on their work. This could lead to physically demanding and dangerous work (Bunderson & Thompson, 2009). Besides, they found that the calling of the zookeepers was so strong that when they were asked to do unpleasant tasks, the zookeepers still wanted to do their job. Therefore, they were at risk of being exploited by their managers and the zookeepers accepted this because of their strong calling. Elangovan et al. (2010) explain this by showing that people with a calling always experience a sense of dissatisfaction prompted by what is not being achieved. They are constantly hungry and pushing for the next higher standard. When someone with a calling would reject that calling, it could be seen as a moral failure and as a waste of someone’s gifts, talents, and efforts (Bunderson & Thompson, 2009). These symptoms are comparable to that of workaholism, which was first defined by Oates (1971) as “an addiction to work, the compulsion or the uncontrollable need to work incessantly" (p. 1). For most people work is
integral to their identity because people are able to develop strengths and skills, feel a sense of accomplishment and take their responsibility in the society. However, workaholics are obsessed by their work and when they have reached one goal, they immediately set another more ambitious one (Killinger, 2006). Therefore, it is likely that they are addicted to their work. Therefore, you could say that people with callings constantly want to work and set higher goals because they are addicted to their calling. They would do everything to pursue their calling, even if this takes personal sacrifices.
When people with a calling ask that much of themselves, this can eventually lead to a burnout. Burnouts are studied a lot by researchers and are known by the two central dimensions emotional exhaustion and a state of fatigue, and occurs often when somebody experiences moral distress (Sundin-Huard & Fahy, 1999; Vinje & Mittelmark, 2007). Moral distress can be described as the experience of a lack of resources when trying to accomplish a task (Vinje & Mittelmark, 2007). The job demands-resources (JD-R) model suggests that people experience a burnout when job demands are high and job resources are limited. Under these working conditions, the employee experiences energy depletion and therefore the motivation of the employee will decrease (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001). People with a calling ask more of themselves because they are always pushing for the next standard and therefore their workload increases. Workload can be seen as a job demand, or as a stressor, which is described as the stressful aspect or external environment of a job (Demerouti et al., 2001). Stress is in one of the six cases the reason for absenteeism (Cascio, 2006). Stress can be defined as “an unpleasant emotional experience associated with elements of fear, dread, anxiety, irritation, annoyance, anger, sadness, grief, and depression” (Motowidlo, Packard, & Manning, 1986, p. 618). When employees experience these stressful working conditions, they finally exhibit strains, namely physical illness, psychological distress and behavior that is harmful for his or her health (e.g., increased smoking).
To summarize, people with a calling ask more of themselves by putting higher standards because they see their calling as a moral duty. The workload increases and thus the job demands and stressors increase, while their job resources remain the same. Therefore, people with a calling are expected to experience more stress, which means that callings are negative related to health-related well-being:
Hypothesis 2: Calling is negatively related to health-related well-being (i.e. increase in stress).
2.2.3 Relations
The third and last dimension of employee well-being is relations, which is somewhat different than the other two because relations are not focused on the individual but on their interactions with others, so the quality of relationships among employees and the relationship of employees with their supervisors (Grant et al., 2007). This dimension is often operationalized in terms of trust, social support, and leader-member exchange (LMX). Although little research has been conducted on callings and the relational aspect of well-being (in exception of Cardador and Caza (2012) who proposed a theoretical framework but which is not yet tested empirically), it is still highly relevant to the current endeavor since better relationships (i.e. high LMX) leads to higher performance of the employee (Jensen, Olberding & Rodgers, 1997).
To examine the relation between calling and relations, it is important to look at the most important relationship of an employee in pursuing his or her calling, which is likely to be the relationship with the most influence on pursuing callings, namely the leader-member relationship. The leader-member exchange (LMX) theory suggests that employees and their managers shape their own unique relationship through interpersonal exchanges and therefore they could expect certain behaviors of each other (Dansereau, Graen & Haga, 1975). When someone with a calling heavily invests in the relationship because it is an important mean in
pursuing his or her calling, this could shape the relationship with the leader. Through the interpersonal exchanges, the leader will be willing to invest in their relationship as well, and therefore the quality of the relationship will increase.
Wrzesniewski, Dutton and Debebe (2003) suggest that relationships and interpersonal sense making are important for understanding the meaningfulness of one’s work, which is an important aspect in many of the definitions of calling (e.g. Cardador & Caza, 2012; Dik & Duffy, 2009). They found that employees are constantly looking for others’ appraisals of their worth and that evaluations of others had a direct and indirect impact on the meaning of their work. Therefore, people with a sense of a calling are associated with more faith in management and better work team functioning than people without a calling. This is in line with the conceptual framework of Cardador and Caza (2012), who describe that the impact of calling is not only in one’s work but also in the quality of relationships with others. Callings should be associated with relationships that are enriching for the individual. This is because first, people with a calling have high levels of investments in work and this will result in mutually rewarding relationships (Cardador & Caza, 2012). Second, people with a calling are also more likely to have a visionary approach and inspire and support others in developing themselves at work which improves the relationships with colleagues (Seltzer, Numerof & Bass, 1989). Finally, people with a calling are more likely to see their organization as an important means in realization of their calling (Cardador et al., 2011). They see the organization as provider of the key resources, like coworkers and good management that are needed in pursuing their calling (Kennedy, Loughry, Klammer, & Beyerlein, 2009). In other words, people with a calling are aware of the need of good relationships in pursuing their calling.
However, Cardador and Caza (2012) also suggest that an employee can take the beneficial aspects of calling to the extreme, where it can be detrimental for the relationships of the employee. In that case, a calling is described as unhealthy and can be seen as
overcommitment or overinvestment. They provide two reasons for this. First, it can strain personal relationships because investing heavily in one’s work can make it difficult to fulfill other roles and relationships outside work (Greenhaus & Buetell, 1985). Second, they argue that people with a calling set too high standards for their colleagues and therefore strain also work-related relationships (Cardador & Caza, 2012). However, because people with a calling are aware of the need of good relationships in pursuing their calling, it is unlikely that their calling will be a strain in their work-related relationships. As explained earlier, people with a calling can indeed take it to the extreme, illustrated by the zookeepers with a calling who were willing to make whatever personal sacrifices that could help in performing on their work (Bunderson & Thompson, 2009), but only when this contributes in pursuing their calling. As long as the relationship is important in pursuing their calling, they would never let it be a barrier because then it could be experienced as a moral failure and as a waste of someone’s gifts, talents, and efforts (Bunderson & Thompson, 2009).
To summarize, employees with a calling are well aware of the need of certain relationships in pursuing their calling. The most important relationship for an employee with a calling is likely to be the leader member relation and therefore in this study, the relationship between the employee and his or her supervisor is analyzed. I expect;
Hypothesis 3: Calling is positively related to relational-related well-being (i.e. increase in LMX).
2.3 The Moderating effect of Change
So far, we have seen that callings could have contradictory outcomes on the well-being of employees and that this is not always beneficial for the employee itself and for the organization. However, we still do not know how these well-being outcomes of people with callings develop in situations that are not stable. Today’s organizations have to cope with
constant change and therefore organizational change is a popular topic in research. Although organizational change is studied a lot (Herscovitch & Meyer, 2002; Lazarus & Folkman, 1984; Rafferty & Griffin, 2006), there is still a gap in literature on changes in employees’ jobs and how people with a calling react in these situations. It is particularly important to look at this kind of change because changes in one’s job are likely to influence the job attitudes of employees, which could affect the performance of employees (Ang & Slaughter, 2000; Riketta, 2008). Examples of changes are an implementation of a technical system, changes in the way of working, or changes in the job tasks of an employee. These changes can be gradual, such as the increase in e-mail traffic over the past 10 years that went from answering 5 to 100 emails per day, or sudden, such as the implementation of federal regulations (Layman, 2007). These changes can significantly modify jobs in organizations, for example due to new technologies, due to new tasks that can add value, or due to new regulations (Layman, 2007).
Besides the different types of changes, change can also be framed differently. According to the framing theory, an issue can be viewed from different perspectives (Chong & Druckman, 2007). This theory describes the process of people developing a conceptualization of an issue or change their way of thinking about an issue. In other words, the way of framing change could impact the influence it has on the relationship between calling and well-being, because employees will look with a different perspective to the change. The first way of framing that is studied in this research is threatening change, which can be described as a change in job tasks which is perceived as threatening by the employee, “which involves negative appraisal and consequently negative response” (Verhaeghe et al. 2006; p.648). The second way of framing change is non-threatening change, which can be described as a change in job tasks that is perceived as an opportunity or challenge by the employee, “meaning positive appraisal and consequently positive response” (Verhaeghe et al. 2006; p.648). In this research, both types of
framing are studied because it is expected that both ways of framing have different influences on the three types of well-being.
As discussed before, a positive relation between callings and happiness is expected. People with callings are expected to be happy employees because they are able to live their calling and be meaningful in their work (Duffy & Sedlacek, 2010; Duffy et al., 2012). However, this expectation is made based on a stable work environment and thus we cannot assume that this will be the same during circumstances of change. People with a calling always known that they would engage in a particular type of work (Dobrow, 2004). For example, a nurse who always wanted to become a nurse, because then he/she would be able to help people. If due to new technologies the nurse needs to answer a hundred emails instead of five, the nurse will spend fewer hours actually helping people that are visiting the hospital. Therefore, the nurse will spend less time in pursuing his/her calling.
In reaction to such changes, people will try to defend themselves (Baumeister, Smart, & Boden, 1996; Felson & Steadman, 1983). This is because in general, people strive to maintain positive selfs (Bies, 1999). In cases of people with a calling, this positive self will only exist when they are able to pursue their calling. Therefore, they will defend themselves against acts, such as changes in their job that can threaten their identity (Baumeister et al., 1996; Felson & Steadman, 1983). In other words, when someone with a calling is less able to pursue his or her calling due to changes in the job, it can harm their personal identity. An identity threat can be defined as “experiences appraised as indicating potential harm to the value, meanings, or enactment of an identity” (Petriglieri, 2011; p.644). Moreover, an identity threat can be experienced as a decrease in someone’s sense of competence, dignity, or self-, worth which will diminish the happiness of that person (Bies, 1999; Steele, 1988). Besides, when people cannot pursue their calling anymore, it is unlikely that they will still think that their lives are meaningful, and therefore their satisfaction may decline (Duffy et al., 2012). Petriglieri (2011)
explains this by one of the three types of identity threats he described in his study, namely ‘meaning’. The nurse from the previous example may associate the meaning of ‘helping people’ with his or her professional identity and the way he or she views him/herself (Anteby, 2008). The identity of the nurse will be threatened when the identity is no longer associated with the meaning. In other words, due to changes in the job tasks, the meaning of the work can disappear, which can be a threat to the personal identity and to the calling that fit within this identity.
On the other hand, when change is not framed as threatening but presented as a new opportunity to the employee, it is likely that this will even increase the happiness of the employee. According to Elangovan et al. (2010), people with a calling are always hungry for more and set higher standards for themselves and the people around them, and therefore people with a calling could also see a change as a new challenge. A challenge can be perceived as a learning opportunity for the employee. Job resources such as learning opportunities are positively associated with work engagement (Bakker & Demerouti, 2007; Schaufeli & Salanova, 2007). This is because resourceful work environments encourage the willingness to devote one’s efforts and abilities to the work tasks (Meijnman & Mulder, 1998). In such environments the likelihood of accomplishing work goals successfully will increase and therefore the outcome is positive and engagement is likely to occur (Schaufeli & Bakker, 2004; Schaufeli & Salanova, 2007). For example, the nurse could perceive the increase in emails as a challenge to improve her efficiency, such that the nurse could still help the same amount of people and answer her emails (and even help more people) in the same amount of time as before. Therefore, the nurse will be more willingness to accomplish the goal and he or she will be more engaged and more satisfied with the completed work.
As the contradictory outcomes of the two ways of framing show, it is important to look at whether change is framed as threatening or non-threatening. If change is framed as threatening to a person’s calling, then it is more likely that this person will be less happy and
engaged because the change could be a threat for the personal identity and a barrier in pursuing one’s calling. When change is non-threatening to a person’s calling, such that the person is still able to pursue their calling, this person will still think that his life is meaningful and is thus more satisfied than people who experience change as threatening. Therefore, I expect:
Hypothesis 4: Change moderates the relationship between having a calling and happiness-related well-being, such that during threatening change the relationship between callings and happiness-related well-being will be weaker, and during non-threatening change the relationship will be stronger.
Although a lot of research is conducted into the effect of changing jobs on health, little research has focused on the influence of changes within jobs on the relationship between callings and health-related well-being. Bernstrøm (2013) found that people who are changing their jobs have an increase in long-term sickness absence before and after they change their job. After two years their long-term sickness absence seems to be normal again. A change in one’s job tasks is not the same as changing jobs but in both situations the person is in a transition period from the old situation to the new situation. During this period people need to change their social role and they often also change their identity structure (Cigoli & Scabini, 2006). Although adjustments in job tasks seems to have less impact than switching jobs, it is still important to study these adjustments, especially when experienced as threatening, to find out if they also influence the health-related well-being of an employee.
In contrast to the relationship between calling and happiness, where we expect positive and negative outcomes of change, the effects of the two ways of framing change on the relation between calling and health is expected to garner only negative outcomes. When change is framed as threatening to the employee, it is likely that they will experience more stress. This can be explained by the job demand-resources model that is discussed earlier. Organizational
change can be seen as a stressor, and therefore the job demands of the employee will increase (Verhaeghe et al. 2006). When the job demands increase, but the job resources will remain the same, the employee will experience more stress, which lead to job strain and it can also lead to a burnout (Demerouti et al., 2001). Therefore, threatening change will strengthen the negative effect of callings on the health-related well-being of the employee.
In a situation where people with a calling experience change as non-threatening, people could see this change as a challenge or an opportunity. Although this could lead to more satisfied employees, they will be asking more of themselves than before, so also in this case the workload and therefore the job demands will increase while the resources remain the same. As discussed before, when this happens, a high amount of stress is experienced (Demerouti et al., 2001). Furthermore, according to Xanthopoulou, Bakker, Demerouti and Schaufeli (2007) a balance between job demands and job resources is needed, because otherwise exhaustion is likely to result. So when the job demands of a person increase and the resources stay the same, it is likely that this person will feel exhaustion (Kroon et al., 2009), which is negatively impacting the health-related well-being. Furthermore, during change people will have a lowered amount of job control. Job control is defined by Karasek (1979; p.289-290) as “the working individual’s potential control over his tasks and his conduct during the working day”. According to the demand-control model (Karasek, 1979), job strain (anxiety, health complaints and exhaustion) exists when there are high job demands in combination with low job control. Therefore, it is likely that change, whether threatening or non-threatening, will strengthen the negative impact of having a calling on health-related well-being:
Hypothesis 5: Change positively moderates the relationship between having a calling and health-related well-being, such that during (non-)threatening change, the relationship between having a calling and health-related well-being will be stronger.
Relations, and LMX specifically, play an important role in implementing organizational changes, because supervisors can influence organizational members’ readiness to change by framing the change (Armenakis, Harris, & Mossholder, 1993). As employees are the ones that need to carry out and implement the new changes, managers must be able to “unfreeze” employees’ beliefs, and motivate them to make the desired changes (Lewin, 1951). The way change is framed by the leader influences the perception of the change by the employees. As discussed before, there is expected that callings are positively related to relations, however, under circumstances of change, the attributions of the employee can be influenced by the way change is framed.
When change is framed as threatening to employees’ calling, it is likely that the employee will provoke antisocial behavior (Aquino & Douglas, 2003). This is because people try to maintain a positive self and try to protect their identities (Bies, 1999). When an employee, who deserves fairness, considerations and respect, is not treated like this by his or her colleagues, this can be experienced as threatening to the personal identity of the employee (Bies, 1999; Steele, 1988). So, when the supervisor tries to implement change and the employee will not be able to pursue his/her calling anymore due to this change, the employee will perceive this as a threat to their identity and to the calling that fits with this identity. The supervisor will be seen as responsible for implementing the change. Employees may assume that the coercion to change shows that the manager dislikes or mistrusts the employee (Yukl, 1994). They will perceive their manager as a “pretty tyrant” who abuses his or her authority to implement the change and to threaten the employee’s calling (Ashforth, 1994). The interpersonal trust and loyalty that characterize a high leader-member relationship will decrease (Brower, Schoorman & Tan, 2000). This can eventually lead to an eroding of employees‘ sense of liking for the manager and to demotivation and actively resisting the change (Tepper, 2000). The employee
does no longer see their supervisor as supporting in pursuing their calling, but more as an obstacle. Therefore, threatening change will weaken the relationship between calling and LMX.
In contradiction, when change is perceived as non-threatening by the employee, thus as an opportunity, the supervisor will be seen as the responsible in providing this extra opportunity. Eisenbach, Watson, and Pillai (1999) linked the organizational change literature with the leadership literature. They found that in order to attract followers to change possibilities, leaders must have a clear vision that takes into considerations the underlying needs and values of the key stakeholders. By setting challenging goals for employees and motivating them to change the old way of working, the leader frames the change by appealing to the follower’s needs for achievement and growth. Therefore, the follower will find the change attractive. Moreover, the leader provides guidance, support and coaching to the followers, which indicates a good relationship between the supervisor and the follower (Eisenbach et al., 1999). Furthermore, employees are more likely to interpret influence tactics of their manager in a way that reinforces their existing perceptions of their leader-member relationship (Furst & Cable, 2008). This means that when the leader-member relationship is already high (which is expected by people with a calling because these employees are aware of the importance of this relation in pursuing their calling), they will perceive this influence tactics of their supervisors also as positive. Therefore, the positive effect of callings on relational-related well-being will be strengthened during non-threatening change. Linking this to the nurse example, when the supervisor of the nurse has framed the change in a way that the nurse thinks that he or she will be able to make the work even more meaningful by helping more people through answering emails, their relationship will be strengthened. The nurse will be thankful to the supervisor who provided the opportunity to make it possible for the nurse to pursue his/her calling even better. Therefore the interpersonal trust and loyalty will increase and the LMX relationship will improve.
To summarize, the two ways of framing change have contradictory outcomes on the relationship between callings and relational-related well-being. Therefore, it is really important for a leader to make sure how change is framed when introducing to the employee. When change is framed as threatening this will weaken the relationship between calling and relations, whereas changes is framed as non-threatening, this could strengthen the relationship. Therefore, this study expects:
Hypothesis 6: Change moderates the relationship between having a calling and relational-related well-being, such that during threatening change the relationship will be weaker, and during non-threatening change the relationship will be
stronger.
2.4 Research model
To summarize the hypotheses of this study, the following research model is drawn:
3. METHOD 3.1 Research approach
For this empirical quantitative study, an experimental vignette study among accountants and teachers in the Netherlands is undertaken. Four reasons are found to explain why especially these two professions were chosen for this study. First, teaching often comes together with high initial commitment and can be said to be a calling for many teachers entering the profession (Hakanen, Bakker & Schaufeli, 2006). Many teachers know from a very young age that they want to become a teacher and this goes along with strong positive feelings about the job. However, the Central Bureau of Statistics in the Netherlands reported in 2014 that the burn-out rate in the Netherlands is the highest among teachers (CBS, 2014). This makes it interesting to study teachers because we can expect that they will have a strong calling, but they also have to deal with negative employee well-being outcomes (i.e. burnouts). Second, accountants often have a strong identity related to their work because they have chosen a very specific education. Furthermore, accountants find themselves in functionally structured organizations where it takes junior accountants a lot of effort to develop themselves and get higher on the career ladder (Ahrens & Chapman, 2000). Because of the strong identity and the effort they made, we can expect that accountants have a strong calling. Third, both professions have to deal with the technological developments, for example for accountants the implementation of enterprise information systems, which can lead to changes in their daily work tasks (Ahrens & Chapman, 2000). Therefore, it is interesting how accountants and teachers will react to these technological changes. Finally, focusing on two professions instead of one, enables the examination of calling within its context and to make comparisons.
The vignettes used in this study are short hypothetical stories that describe a situation (Aguinis & Bradley, 2014). The respondents were asked to answer the questions and statements as if they had experienced this situation in real life (Finch, 1987). The two professions were
presented different vignettes that described a situation that is likely to occur in their real life and fit their profession. When the vignette is seen as a real and plausible situation, it is more likely that respondents will reflect their personal experience and that, next to the internal validity, the external validity will also be high (Finch, 1987). The alternative, finding organizations that are willing to participate in research where you want to manipulate situations with threatening and non-threatening change and want to exclude other possible influences was deemed not to be viable. The experimental vignette experiment allowed us to manipulate the hypothetical situations and to causally trace the effects on well-being.
3.2 Sample and procedure
The population of this study are Dutch speaking people working as teachers or accountants because this makes it more likely that they can identify themselves with the situation in the vignette. An advantage of these lifelike scenarios to respondents is that it can help to increase the generalization of results (Aguinis & Bradley, 2014). Furthermore, it can help us selecting a more representative sample that brings more accurate results than using participants with all kind of professions.
There were 253 respondents who participated in the experiment, of which 89 accountants and 164 teachers. Because the sample of this study was limited to a small subgroup of the Dutch population, a combination of two non-probability sampling techniques was used, convenience sampling and snowball sampling. The surveys were administered online via a weblink to make it easier for the respondents to participate. Furthermore, participants could win an incentive of a hundred euro gift card of a webshop. Several actions were undertaken to reach as many respondents as possible. First, the research is conducted together with a group of six other master students and a PhD candidate, such that we were able to reach more participants than I would be able on my own. Second, the contacts of our own network were asked to
participate with the question if they could forward the survey to teachers and accountants in their network. Third, the research was disseminated on our personal social media (e.g. Facebook and LinkedIn) and in different community groups for teachers and accountants on social media (e.g. Accountant & Audit – LinkedIn, Dutch Accountants – Facebook). Finally, several schools and accountancy companies were contacted and asked to distribute our research among incumbent teachers and accountants.
Table 1. Descriptive statistics of study participants
N= 253
Accountant Teacher Total
M SD Range M SD Range M SD Range
Tenure (y) 10.6 9.8 1 - 38 16.7 13 1 -59 12.5 12.3 1 - 59
Age (y) 34.2 11.9 21 -67 43.5 13.3 21 -63 40.2 13.6 21 - 67
Hours worked
per week 39.3 5.5 24-60 30 10.3 5 - 50 33.1 10.0 5 - 60
Percentage Percentage Percentage
Gender
Male 66.7 35.6 44.7
Female 33.3 64.4 51.8
Profession 35.2 64.8 100.0
The total sample existed of 44.7% men and 51.8% women (see Table 1), who worked on average 33 hours a week and who have an average age of 40. However, this study found some differences between the two professions of accountants and teachers. Only 33.3% of the accountants was female, where females were in majority in the teacher sample with a percentage of 64.4. However, according to a report of CBS, the results for teaching are representative to the population teachers on primary schools (Beiro & Ramaekers, 2016). Results in the report of CBS show an even greater majority than this study found, with an increase of women working at primary schools from 82% in 2005 to 87% in 2015. Also for
that their members have a percentage of 79 men in 2016. Furthermore, accountants work on average 40 hours per week, where the teachers on average only work 30 hours per week. The results also show a difference in age. Accountants had an average age of 34 years old, where teachers scored much higher with an average age of 40 years old.
Because of the differences in the descriptive statistics, I have decided to run an independent t-test (see Table 2). The results of Levene’s test for equality of variances not assumed, showed that the variances for the two professions are different for calling F(1,241) = 0.67, p < .001. The results of the independent t-test show that teachers (M = 4.02, SD = 0.52) scored higher on calling than accountants (M =3.58, SD = 0.59), t(241) = 5.77, p < .001. Moreover, the results of Levene’s test for equality of variances for work engagement, and stress, showed also that the variances for the two professions are different. On work engagement, accountants score significantly higher on average, and for stress the teachers score significantly higher on average. For LMX, no significant difference was found. However, on three of the four variables differences were found between the two groups, which indicates that there could also be differences in the analyses for testing the direct effects of calling on well-being, and the moderation effect of change on this relationship. Therefore, the analyses are not only run for
Table 2. t-test results for teachers and accountants
Profession Accountant (N = 86) Teacher (N = 157) Variable M SD M SD t-value Calling 3.58 0.59 4.02 0.52 5.77*** Work engagement 3.46 0.68 3.11 0.86 -3.32** Stress 2.48 0.68 2.70 0.89 1.98* LMX 3.53 0.62 3.41 0.79 -1.28
the whole sample but are also run split per profession. Furthermore, I will run an extra regression with profession as a moderator.
3.3 Design
After conducting a pilot study (N=80), it was decided to use four instead of the originally designed six, because too little variation was observed between the no change and non-threatening change conditions. The four vignettes (described in Appendix A) were manipulated by the way change is framed, as non-threatening or as a threat. Furthermore, the vignettes were manipulated by two types of change, non-identity related change or identity related change, which are not used in this study. In all situations, there is a change in the job tasks due to technological developments (i.e., big data or implementation of new software). In two scenarios the change is framed as threatening. In these scenario’s the person is forced to change, because otherwise he or she would be redundant. In the other two scenario’s, the technological development is framed as an addition to their current job tasks and therefore not threatening, so these changes can be seen as an opportunity to the employee. A between-subjects design is used, which means that every respondent randomly received one of the four vignettes and where comparisons are made across respondents (Aguinis & Bradley, 2014). Without other vignettes as referents points, responses may not reflect the true actions of a respondent. Therefore, we provided all participants with baseline information to create a similar adequate contextual background. An additional matching survey with a number of statements and questions was added after the vignettes. Of the 253 respondents, 129 got the threatening change scenarios, and 124 the non-threatening change scenarios. The randomization of the four scenarios is controlled through testing for orthogonality of factors by cross tabs. This test shows a non-significant Chi-square of 0.76, which indicates that the factors are orthogonal. In other words, the scenarios are indeed randomly divided among the respondents.
3.4 Measurements
The survey was administered in Dutch. A translation-back-translation procedure was employed to make sure the meaning of the items did not change (Brislin, 1980). The items were also adjusted to the scenario to which they relate. For all items a 5-point Likert scale was used, ranging from ‘totally disagree (1)’ to ‘totally agree (5)’.
Independent variable: Calling. Calling was measured using the four items of
Wrzesniewski et al. (1997). An example item is “I would choose my current work life again if I had the opportunity” (see Appendix B for all items). All items were indicative. This study found an unreliable Cronbach’s alpha of .59. Even if we looked at the Cronbach alpha when deleted one item, the reliability did not increase. Although this Cronbach’s alpha is not reliable (< .70), the scale is still used in this research because it is a well-developed scale and tested a lot in previous research where they have found Cronbach alpha’s of .73 (Leana, Appelbaum & Shevchuk, 2009). Moreover, the low Cronbach alpha means that any significant effect that I will find, must be a quite strong effect.
Dependent variable: Happiness. Happiness-related well-being is measured by the five
items of the Utrecht Work Engagement Scale (UWES) (Schaufeli, et al., 2002). An example item is “I would feel strong and vigorous in my work”. All items were indicative. The items originally needed to be answered on a 7-point Likert scale ranging from ‘does not apply at all’(1) to ‘fully applies’(7), however in this study a 5-point Likert scale was used for all items to make it less complex for the respondent to fill in the survey. Originally, 15 items were administrated for the UWES, with a coefficient alpha of .92. However, the scale is shortened into five items and even then, they found a reliable coefficient alpha (α = .92). The combination of the high internal consistency and the high correlation of the 15 items with the mean level of work engagement participants (.61), shows that the short scale of work engagement is
appropriately assessed (Bledow, Schmitt, Frese & Kühnel, 2011). Current study, using the subscale, found a reliable Cronbach’s alpha of .90.
Dependent variable: Health. Health-related well-being is measured by the job stress
scale of Motowidlo et al., (1986). An example item is “My job would be extremely stressful”. The last item “I would almost never feel stressed because of my job” is counter-indicative, so this item has been recoded. Based on the four items of this scale, previous research found a Cronbach’s alpha of .87 (Bolino, & Turnley, 2005). The current study only used three items (see Appendix B) to maximize the amount of respondents by limiting the length of the survey, and still found a reliable Cronbach’s alpha (α = .80).
Dependent variable: Relations. Relational-related well-being is measured by the Liking
for subordinate scale of Wayne and Ferris (1990). This scale is adjusted to Liking for supervisor. An example item is: “I would get along well with my supervisor”. All items were indicative. Previous research found a Cronbach’s alpha of .94. The current study used only three items (see Appendix B) to maximize the amount of respondents by limiting the length of the survey, and still found a reliable Cronbach’s alpha (α = .82).
A principal axis factoring analysis (PAF) is conducted to confirm the underlying structure of the three types of well-being. Results of this test showed a good fit for a three factor model. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for the analysis, KMO = .846. Bartlett’s test of sphericity χ² (55) =1418.568, p < .001, indicated that correlations between items were sufficiently large for PAF. An initial analysis was run to obtain eigenvalues for each component in the data. Three components had eigenvalues over Kaiser’s criterion of 1 and in combination explained 62.00% of the variance (Kaiser, 1960). In agreement with Kaiser's criterion, examination of the scree plot revealed a levelling off after the third factor. Thus, three factors were retained and rotated with an Oblimin with Kaiser normalization rotation. Table 3 shows the factor loadings after rotation. The items that cluster on the same factors suggest that
factor 1 represents work engagement, factor 2 stress and factor 3 leader member relation. As the results suggest, all items correspond with their own variable. This test shows us that all three types of well-being are correlated, however, all different constructs of well-being.
Moderator: Framing of change. The vignettes were manipulated by change as
non-threatening and change as non-threatening. When organizations need to change, the framing of the change is important because it determines from which perspective employees are perceiving the change (Chong & Druckman, 2007). When the change is framed as a threat, the implementation of a new technology requires employees to adjust themselves to the new way of working, as otherwise they would become superfluous. In the condition where change is
Table 3. Factor analysis
Rotated Factor Loadings
Item
Work
Engagement LMX Stress
Me sterk en krachtig voelen in mijn werk 0.66 -0.04 -0.12
Me vol energie voelen op het werk 0.94 -0.06 0.09
Enthousiast zijn over mijn werk 0.86 -0.01 -0.04
Ik zou vrolijk helemaal op kunnen gaan in
mijn werk 0.77 0.04 -0.09
Geïnspireerd 0.57 0.19 0.04
Me bijna nooit gestrest voelen door mijn
baan 0.02 -0.05 0.52
Mijn baan als extreem stressvol ervaren -0.04 0.02 0.83
Gestrest raken door mijn baan -0.04 0.03 0.91
Goed overweg kunnen met mijn
leidinggevende -0.02 0.84 -0.05
Het een plezier vinden om onder mijn
leidinggevende te werken 0.03 0.89 0.07
Ik denk dat de leidinggevende in dit scenario
een goede vriend zou kunnen zijn 0.02 0.53 -0.05
Eigenvalues 5.03 1.57 1.23 % of variance 42.7 11.4 7.9 Correlations Work Engagement 1.00 0.52 -0.54 Stress 0.52 1.00 -0.33 LMX -0.54 -0.33 1.00
new technology. The combination of implementing the new technology and their current job tasks will lead to better results. The manipulation check inquired whether the implementation of the new technology in the vignette makes the employee superfluous in doing their job or that they play an important role in the implementation.
Control variables. Control variables were used to rule out possible spurious relations.
The respondents were asked for their age and gender. For both age and gender, evidence is found of influencing the chance of getting a burnout (Russell, Altmaier & Van Velzen, 1987). Dobrow and Tosti-Kharas (2011) also found that age is related to calling. These control variables were included in the first section of the questionnaire.
3.5 Analyses
Hierarchical multiple regression analyses were undertaken to test the direct effects of callings on well-being in hypotheses 1, 2, and 3. In step 1 of the regression, the control variables gender and age were entered. In step 2, calling was added to the model.
To test the moderation effect of change on callings and the three types of well-being after controlling for age and gender in hypotheses 4, 5, and 6, the regression analyses through SPSS process of Hayes (2013) is used. Furthermore, Mplus is used to perform structural equation modelling (SEM) with robust maximum likelihood estimation. This test is used to find out if the effects of calling are the same whether the regressions are ran separately or at the same time. By doing these tests, there is taken care of interactions among the dependent variables. The fit indices and cut-off values based on Hu and Bentler (1999) are used to assess model fit (i.e., χ² non-significant, RSMEA < .05, CFI > .90, TLI > .90, SRMR < .08).
To be able to use all these type of analyses, it is important that the variables are normally distributed. Therefore, all numerical variables were checked on normal distribution, measured by Skewness and Kurtosis. Both stress and LMX are normally distributed, with for stress found
a skewness of 0.53 (SE = 0.15) and kurtosis of 0.01 (SE = 0.31); and for LMX a skewness of -0.72 (SE = 0.16) and kurtosis of 0.81 (SE = 0.31). Whereas the sharpness of the peak of distribution for stress is normal, for LMX it has slightly more scores in the tails and therefore the form is more pointy (leptokurtic). However, for both variables zero is found between the upper and lower boundary CI, and therefore we can conclude that the two variables are normal distributed. Furthermore, for both variables a normally distributed histogram and Q-Q plot were observed (see Appendix C).
Work engagement is non-normally distributed, with a skewness of -0.27 (SE = 0.15) and kurtosis of -0.39 (SE = 0.31), the sharpness of the peak of the distribution is flatter than the normal (platykurtic). However, the Q-Q plots shows us a normal distribution. Moreover, because this study has a large sample, and the skewness and kurtosis are relatively close to zero and therefore do not deviate much of the normal distribution, it is likely that work engagement is normally distributed.
For calling there were some abnormalities found in the normal distribution, namely a moderate negative skewness of -0.68 (SE = 0.15) and a slightly positive kurtosis of 0.27 (SE = 0.31). Therefore, the data of calling is transformed by the formula Calling*=√(K-Calling), where K is the highest value of the variable calling plus one. After computing this transformed variable of calling, a normal distribution with a skewness of 0.32 (SE =0.15) and kurtosis of -0.10 (SE = 0.31) was found. The analyses were run but both on the transformed variable and the original one and no meaningful differences except the mean were observed. Therefore, it was decided to keep the old values of the variable calling to keep the original data. Indeed, Tabachnick, Fidell and Osterlind (2001), have also asserted that for large samples, skewness will not make a substantive difference to the results. In the last step in preparing the data, the standardized variables were created and checked on outliers. For LMX, ten outliers were identified which are excluded from the data with an exclusion criteria of < -3 and > 3.
