What Drives Employees’ Job Crafting Behavior?

What Drives Employees’ Job Crafting Behavior?

Leadership and Individual Factors as Antecedents for Job Crafting

Research Paper, MSc HRM

University of Groningen, Faculty Economics and Business EBM009B05

July 1

, 2018

E.M. Wassenaar Student number: S3273547 Email: e.m.wassenaar@student.rug.nl

Supervisor: Sanne Feenstra, MSc.

2 ABSTRACT

This paper aims to increase the understanding of the antecedents of job crafting.

Literature suggest that leadership factors and individual factors can influence job crafting behavior of employees. Therefore, within the scope of this study, empowering leadership and self-efficacy are considered as possible antecedents of job crafting. In addition, the mediating role of psychological safety was investigated in the relationship between empowering leadership on job crafting and in the relationship between self-efficacy on job crafting. Moreover, employees’ age was considered as moderator in the relation between psychological safety and job crafting. Data is collected from 205 employees working in a Research and Development environment in the Netherlands. A confirmed relationship between empowering leadership and job crafting was found, concluding that employees craft their jobs more often when they are empowered by their leader.

Psychological safety did also mediate the relationship between empowering leadership on job crafting. Therefore, empowering leaders are found to increase employees’ feelings of psychological safety, which can in turn increase employees’ job crafting behavior.

Furthermore, self-efficacy was not found to be related to job crafting. Employees’ age did not moderate the relationship between psychological safety and job crafting.

Keywords: Job Crafting, Empowering Leadership, Self-Efficacy, Psychological Safety,

Age, Organizational Behavior.

3 INTRODUCTION

In today’s society, organizations regularly face challenges in the transition of job design caused by global competition and technological innovation (Ouweneel, Taris, Van Zolingen & Schreurs, 2009). Traditionally, job designs were characterized by a classic top-down approach, which indicated that leaders needed to design jobs for employees.

However, for the last decades, there has been a transition to a more bottom-up approach in which employees participate more proactively in work design than before (Grant, Parker & Collins, 2009; Wrzesniewski & Dutton, 2001).

Job crafting is a particular form of proactive behavior in which employees take the initiative to optimize their own job characteristics (Wrzesniewski & Dutton, 2001).

According to Leana, Appelbaum and Shevchuk (2009), job crafting can be beneficial for both employees and organizations, because employees make their own jobs more

meaningful, engaging, and satisfying through job crafting (Tims, Bakker & Derks, 2012).

Hence, organizations who facilitate job crafting, enhance the adaptability of employees during organizational change (Petrou, Demerouti & Schaufeli, 2015). Therefore, it is important to understand what the predictors are of employees’ job crafting behavior.

One of the predictors that may be relevant for employees’ job crafting behavior, is the influence of leaders (Breevaart, Bakker, Demerouti & Derks, 2016). Nowadays there is a trend within organizations where leaders empower their employees (Breevaart et al., 2016). This is called an empowering leadership style. Empowering leadership is defined as a leadership style in which leaders use series of management practices, including decentralization, participation, information sharing and training (Hakimi, 2010;

McClelland, 1975). Empowering leadership has also gained increased attention in the

field of science, because it has shown that this style of leadership can act as a motivator

for the performance and well-being of employees (Wrzesniewski & Dutton, 2001). For

4 example, empowering leaders encourage participation of employees in decision-making processes and this increases employees’ involvement and effectiveness (Breevaart et al., 2016). However, only some studies have investigated the positive relationship between empowering leadership on job crafting. Therefore, the first aim of the present study is to show that empowering leadership is positively associated with job crafting.

Previous literature has also shown that there is another possible predictor which may be relevant for job crafting, namely self-efficacy (Bandura, 1997). Self-efficacy is defined as the expectations that employees have about their abilities to execute desired behaviors and to impact their environment successfully (Bandura, 1997). Previous research found that the more employees feel self-efficacious, the more they craft their jobs (Tims et al., 2012). However, other research suggests that employees only craft their jobs when they perceive a lack of self-efficacy, since this can give employees an

immediate cause to optimize their job characteristics (Wrzesniewski & Dutton, 2001). To increase the consistency of the current findings, this research has the purpose to create a more detailed insight in the positive relationship between self-efficacy and job crafting. It is expected that the higher the self-efficacy of employees, the more confident they feel in taking initiatives such as crafting their jobs (Bandura, 1997). Therefore, the second aim of the present study is to show that self-efficacy is positively associated with job crafting.

Further goals of this study is to test the nature of psychological safety as a mediator in two separate relationships. First, the mediating role of psychological safety in the

relationship between empowering leadership on job crafting, and second in the relationship between self-efficacy on job crafting. Psychological safety reflects the feelings of employees that taking interpersonal risks will not result in embarrassment and that employees can employ themselves without the fear of negative consequences

(Wanless, 2016; Edmondson & Lei, 2014). First, psychological safety mediates the

5 relation between empowering leadership and job crafting, because empowering leaders cultivate a climate where followers are encouraged to express themselves about their opinions, ideas and concerns (Fuller, Marler, Hester & Otondo, 2015; Parker & Wu, 2014). Second, psychological safety mediates the relation between self-efficacy and job crafting because, employees who are self-efficacious view activities as something they are able to master (Choi, Price & Vinokur, 2003). Therefore, employees may feel safer to take interpersonal risks (Wanless, 2016).

Consequently, psychological safety may play an important role in predicting job crafting, because employees who feel psychological safe are more likely to show typical job crafting behavior, such as experimenting, exploring, and learning (Edmondson, 1999). Therefore, I introduce psychological safety as a mediator in the relationship between empowering leadership and job crafting and in the relationship between self- efficacy and job crafting. The third aim of the present study is to test the positive

relationship between empowering leadership on job crafting, mediated by psychological safety. Furthermore, the fourth aim is to test the positive relationship between self- efficacy on job crafting, mediated by psychological safety.

Finally, the present study aims to test the nature of the relationship between

psychological safety and job crafting, moderated by employees’ age. In the present study

it is expected that job crafting behavior diminishes with age even though psychological

safety is high. Previous research found that older employees strive for more security in

their jobs, prefer maintenance and they seek for more stability in their jobs (Kooij, de

Lange, Jansen, Kanfer & Dikkers, 2011; Rhodes, 1983). Therefore, it may be more likely

that with older employees the relationship between psychological safety and job crafting

weakens. Studying the moderating role of employees’ age will provide new insight into

the relationship between psychological safety and job crafting, because employees’ age,

6 as such, has not yet been tested. The fifth aim of the present study is therefore to show that employees’ age moderates the relationship between psychological safety and job crafting.

The present study contributes to the literature of job crafting in different ways. The first theoretical contribution is that the present research contributes to the current knowledge about job crafting by testing the construct of job crafting in relation to a leadership antecedent, which is a relatively underexposed topic in current studies about job crafting. Second, psychological safety is often viewed as a variable that can affect factors such as empowering leadership and self-efficacy. This research aims to test to what extent psychological safety can be influenced by empowering leadership and self- efficacy. Research on the relationships between empowering leadership on psychological safety, and self-efficacy on psychological safety is relatively scarce. Hence, little is known about how individual perceptions of employees can affect the social interaction domain in the workplace. Third, the present study will expand the available theory of job crafting by testing this relationship in a Research and Development environment.

According to my knowledge, there are only limited studies which tested the relationship

in a Research and Development environment before.

7 THEORETICAL BACKGROUND

The Concept of Job Crafting

Wrzesniewski and Dutton (2001) define job crafting as the physical, cognitive and relational changes employees make in the task and relational boundaries of their work.

The literature on job crafting is relatively new, but the subject has received a fair amount of attention since Wrzesniewski and Dutton published their 2001 article, in fact launching the field of job crafting. Also, the job crafting definition and framework of Wrzesniewski and Dutton (2001) is most frequently used in studies about job crafting. The present study will use their definition as theoretical starting point as well.

Wrzesniewski and Dutton (2010) make a distinction between three types of job crafting; task crafting, relational crafting and cognitive crafting. Task crafting is a form of job crafting in which employees add tasks that are meaningful for them. Employees can change the boundaries of tasks by redesigning them. For example, when there are too many constraints and adding tasks and emphasizing tasks becomes difficult, employees can redesign existing tasks that lead to a better person-job fit (Wrzesniewski & Dutton, 2010).

Relational crafting is a form of job crafting in which employees change the

boundaries of work, including interactions with others. Employees can craft the relational boundaries by building relationships, reframing relationships and adapting relationships and interactions with others (Berg, Grant & Johnson, 2010). For example, employees from one departments can form a collaborative relationship with employees from other department. Instead of changing the purpose of relationships, crafters can also cultivate relationships by proving valuable help and support in carrying out their jobs

(Wrzesniewski & Dutton, 2010).

8 Cognitive crafting is a form of job crafting in which think about the tasks, how

employees think about relations and the job as a whole (Wrzesniewski & Dutton, 2010).

This mental form of job crafting explains how employees subjectively experience their work, without changing anything physical or objective about the job itself (Crum &

Langer, 2007). For example, cognitive crafting involves employees reframing how they see their job by expanding perceptions, focusing on perceptions or linking perceptions (Crum & Langer, 2007).

Employees who use a combination of these three types of job crafting techniques, become job crafters in which they can change the boundaries of their work in ways to make the job better fit their own personal needs. Job crafting is a continuous process, instead of a one-time event (Fried, Grant, Levi, Hadani & Slowik, 2007). Job crafting can be applied across a variety of roles with different management levels and degrees of autonomy (Berg, Wrzesniewski, & Dutton, 2010; Wrzesniewski & Dutton, 2001). Hence, studies argue that job crafting is plausible and that it is possible even in the most

restricted and routine jobs for employees to initiate changes to influence their work experience (Slemp, Kerm & Vella-Brodrick, 2015).

The Relationship Between Empowering Leadership and Job Crafting The first predictor of job crafting, suggested in this research is empowering

leadership. Empowering leaders can influence attitudes, work environments and behavior of employees (Breevaart et al., 2016). Empowering leaders may therefore play an

important role in the development of employee behavior, such as job crafting (Fuller et

al., 2015; Leana et al., 2009). Empowering leaders are characterized as leaders who are

willing to share their power, they encourage participation of employees and provide

employees with a voice (Conger & Kanungo, 1988; Kirkman & Rosen, 1997, Spreitzer,

9 1995; Thomas & Velthouse, 1990). Empowering leaders can therefore increase the

personal control of employees, which is important for employees to craft their jobs (Wrzesniewski & Dutton, 2001).

Empowering leaders also delegate their tasks to employees, which may enhance employees’ perceptions of importance Delegation suggests independent thinking and it can help employees to develop skills and knowledge (Wrzesniewski & Dutton, 2001). In turn, employees create a more positive self-image, which can increase employees’ job crafting behavior (Cohen, Chang & Ledford, 1997; Manz & Sims, 1987).

Evidence from survey research amongst leaders working in small and medium-sized manufacturing companies, also imply that empowering leadership may influence job crafting behaviors (Tims et al., 2013). In this research, empowering leaders were significantly more able to affect, so called, innovative behavior of subordinates. Job crafting can be seen as innovative behavior, because it emphasizes the changes that one can make relating to cognitive processes, tasks and relations in order to create a better job-fit (Tims et al., 2013). Findings emphasize the critical role of empowering leaders in encouraging employees’ job crafting. Similar to the research of Tims et al., (2013), I expect that empowering leadership is positively related to job crafting behaviors.

Consequently, I hypothesize that:

Hypothesis 1: Empowering leadership is positively associated with job crafting.

The Relationship Between Self-Efficacy and Job Crafting

The second predictor of job crafting that is suggested in this research is self-efficacy.

Self-efficacy is defined as the expectations that employees have about their abilities to execute desired behaviors and to impact their environment successfully (Bandura, 1997).

Self-efficacious employees believe in their own abilities, feel personal control and are

10 more likely to regulate the activities and efforts than people who are less self-efficacious (Bandura, 1997). Employees with a high self-efficacy have a more positive self-image, which makes it more likely that they have a positive stance towards learning and growth (Petrou et al., 2015). In addition, employees with a high self-efficacy, set more

challenging goals for themselves, are more critical and put more effort in exploring the environment for opportunities (Mohammed & Billings, 2002). Furthermore, self-

efficacious employees are likely to engage in innovation and to change and align the job to their own characteristics (Niessen, Weseler & Kostova, 2016).

Empirical studies indeed showed that self-efficacy beliefs affect proactive job crafting behavior. For example, Tims, Bakker and Derks (2015) conducted a diary study using data from various organizations. Results of the study suggested that employees who felt more self-efficacious turned out to search more proactively for new opportunities and variety in their jobs than employees who had a lower score on self-efficacy (Tims,

Bakker and Derks, 2015). Employees were also more likely to redefine and change things in their jobs, when they believed that they were able to successfully master the tasks (i.e.

self-efficacy). Similar, to the study of Tims et al., (2015), I expect that self-efficacy is positively related to employees’ job crafting behavior. Consequently, I hypothesize that:

Hypothesis 2: Self-efficacy is positively associated with job crafting.

The Mediating Role of Psychological Safety

Psychological safety is important for employees in order to feel comfortable in

expressing similarities, differences or other views that employees have (Fuller et all.,

2015; Parker & Wu, 2014). Psychological safety is a cognitive construct that refers to the

feeling that taking interpersonal risks will not result in embarrassment, ridicule, or shame,

enables employees to engage, connect, change, and learn (Edmondson & Lei, 2014). One

11 of the main mechanisms that may explain the relationship between empowering

leadership on job crafting and between self-efficacy on job crafting, is psychological safety.

First, psychological safety can mediate the relationship between empowering leadership on job crafting. Empowering leaders are willing to share their power and delegate their responsibilities to employees lower in the formal hierarchy (Mills &

Ungson, 2003). This can give employees the feeling that they are trusted by their leaders in performing tasks, which in turn may increase the psychological safety of employees (Gao, Janssen & Shi, 2011). Empowering leaders also remove barriers, such as a lack of autonomy, which can be viewed by employees as a threat to their feelings of

psychological safety (Kahn, 1990). Empowering also help employees to believe that taking interpersonal risks will not pose an intolerable level of threat to their identity or sense of self (Edmondson & Lei, 2014). Instead, empowering leaders create optimal development, which stimulates employees to feel comfortable by taking interpersonal risks (Wanless, 2016).

Empirical studies amongst leaders and subordinates, also found that empowering leadership can influence the perceptions of safety of employees (Edmondson, 1999).

Results showed that employees who were empowered by their leaders, are likely to feel safer to develop themselves and feel safer to try out ways of doing things (Edmondson, 1999). Furthermore, they showed that employees who are empowered by their leader discuss mistakes and learn from these behaviors and were more likely to put in effort without feeling feelings or fear or exclusion (Edmondson, 1999). In summary, the present study expects that empowering leadership is positively related to employees’

psychological safety.

12 Second, psychological safety can also mediate the relationship between self-efficacy on job crafting. Employee with a high self-efficacy believe that their actions will lead to successful outcomes (Choi et al., 2003). Therefore, employees may feel confident and safe, because they are less scared for the negative consequences when they perform tasks.

Instead, self-efficacious employees believe that the things that they do will lead to positive outcomes which decrease feelings of fear and anxiety (Maslow, Hirsh, Stein &

Honigmann, 1945).

Empirical evidence from a research amongst 549 students examined the relationships between self-efficacy and psychological safety (i.e. level of anxiety, depression, social avoidance and worry) (Greco & Morris, 2002). This study showed that low levels of self- efficacy are accompanied by high levels of anxiety. Anxiety can influence employees’

psychological safety in a negative way. Furthermore, they showed that the stronger employees’ self-efficacy, the more stress was reduced and the less employees perceive personal threats in performing their activities (Greco & Morris, 2002). In summary, the present study expects that self-efficacy is positively related to employees’ psychological safety.

Lastly, psychological safety may enhance employees’ job crafting behavior, because employees who feel psychological safe perceive decreasing barriers to freely exercise (Kahn, 1990; Wrezesniewski & Dutton, 2001). In addition, employees who feel

psychological safe, enact more often in self-regulated strategies, such taking initiatives and bringing up ideas (Edmondson & Lei, 2014; Hirak, Peng, Carmeli, & Schaubroeck, 2012; Holley & Steiner, 2005). This behavior forms the basis for job crafting as well.

Similar insights and empirical findings were provided by the study of Gilson and

Shalley (2004), who found that teams engaged in more creative processes when they felt

psychological safe in their job. Employees who felt psychological safe, were taking more

13 risks and tried out new approaches while they were performing their tasks (Gilson &

Shalley, 2004). Taken all together, I expect that psychological safety can mediate the relationship between empowering leadership on job crafting and between self-efficacy on job crafting. Consequently, I hypothesize that:

Hypothesis 3: Psychological safety will mediate the relationship between empowering leadership and job crafting such that empowering leadership is associated with increased psychological safety (3a), which in turn predicts enhanced job crafting behavior of employees (3b).

Hypothesis 4: Psychological safety will mediate the relationship between self-efficacy and job crafting such that self-efficacy is associated with increased psychological safety (4a), which in turn predicts enhanced job crafting behavior of employees (4b).

The Moderating Role of Employees’ Age

In the present study is expected that the relationship between psychological safety on job crafting is moderated by employees’ age, such that job crafting behavior diminishes with age even though psychological safety is high. Employees’ age is refers to the chronological age of employees (Peeters, De Jonge & Taris, 2014).

A study who related age to work outcomes, suggest that employees go through five

phases during their work career (Inkson, Dries & Arnold, 2014). The first phase is called

the grow phase. This is the phase prior to one’s working life (0-14 years). The second

phase, called the exploitation phase, is characterized as the phase in which employees

start their careers and employees learn, experiment and look for opportunities that are

relevant for their personal development (15-24 years). The third phase is called progress

phase. In this phase, employees are more focused on achievement, grow and getting

higher responsibilities (25-44 years). Furthermore, the fourth phase is called the

14 maintenance phase. In this phase employees look for more continuity and employees more often stay in their current position (45-64 years). Finally, in the fifth phase, employees generally experience reduced energy, delegate task more often and they are more focused on their retirement (45-67 years) (Inkson et al., 2014).

From this study can be concluded that employees develop different needs and preferences when they become older and when they get more experienced in their jobs (Inkson et al., 2014). Previous research also showed that older employees are less likely to search for new opportunities in their jobs and prefer more stability (Lang &

Carstensen, 2002; Kanfer & Ackerman, 2004). In creating stability, employees develop routines (Midtsundstad, 2011). Preferences for stability and routines may conflict with job crafting, because job crafting is a way in which employees proactively reshape the boundaries of jobs. This asks for a proactive attitude in which employees are open for new ways of doing things (Wrzesniewski & Dutton, 2001). Therefore, it is expected that job crafting behavior diminishes with age even though psychological safety is high.

Consequently, I hypothesize that:

Hypothesis 5: The relationship between psychological safety is moderated by employees’ age, such that psychological safety is less strongly related to job crafting when employees are older, rather than when employees are younger.

The hypotheses are visualized in a conceptual framework (see figure 1).

15 FIGURE 1

Based on the theoretical framework, the research and the hypotheses that will be tested are presented below:

Empowering Leadership

Self-Efficacy

Psychological Safety Job Crafting Task Crafting Relational Crafting

Cognitive Crafting

Employees’ age

16 METHODOLOGY

Procedure

This study was conducted in a large provider of electronics and medical equipment, located in the north of the Netherlands. Data for this research was collected using an online-survey. In order to test the hypotheses, questionnaires were distributed among participants who work at the R&D department. All employees of the R&D department (N= 493) received an email along with an URL survey link. The email was hierarchically sent by the highest plant manager of the company, located in the North of the

Netherlands. The plant manager supported the study and encouraged participation among employees. After one week, all participants (N= 493) received an email with a reminder to fill in the survey. Finally, a total of 205 digital questionnaires were completed

(response rate of 41.6%).

Participants

The procedure explained above yields for a sample of 205 responses, all working at in the same company (100%). Of the total number of respondents (N=205), 20% were female (N=42), 79% were male (N=162) and 1% did not want to answer the question (N=1). In the present study, 2.9% of the employees were between the ages 26-34, 26.8%

of the employees were between the ages 25-34, 33.7% of them were between the ages 35- 44 and 25.4% of the employees the ages 45-54, whereas 11.2% were 55 or older. The education level of 41% was ‘HBO’, 42.4% reported ‘WO’ and 12.2% indicated ‘MBO’.

Other respondents had primary school (5%), high school (5%) or PHD (3.4%). The

majority of the surveyed employees had a permanent contract (96.17%) and 3.83% had a

flexible contract. Participants with a flexible contract work 38.07 hour per week, while

employees with a permanent contract work 37.78 hours per week.

17 Measurements

The specific measures are described below, along with the results of calculation of Cronbach alpha coefficients for the various measures. All measures were completed by employees. Appendix A includes all survey measures used in this research.

Job crafting was measured with the Job Crafting Scale from Slemp and Vella- Brodrick, (2013). The scale is distributed according to the three dimensions (task,

relational and cognitive crafting) of job crafting. Task crafting items included: ‘Introduce new approaches to improve your work’. Relational crafting items include: ‘Engage in networking activities to establish more relationships’. Lastly, cognitive items included:

‘Think about how your job gives your life purpose’. The complete measure consisted of 14 items and participants indicated the frequency with which they have engaged in each job-crafting activity, ranging from 1 (Hardly ever) to 6 (Very often). Cronbach’s alpha for the scale was .798.

Empowering leadership was measured with the Empowerment scale from Spreitzer (1995). This scale is distributed according to the four dimensions (meaning, competence, self-determination, impact) that capture the essence of empowerment (Spreitzer, 1995).

Example items include: ‘My supervisor helps me see that the work I do is very important’

and; ‘My supervisor helps me be confident about my ability to do my job’. Lastly, ‘My supervisors allow me to decide on my own how to go about doing my work’. The

complete measure consists of 12 items and participants indicated to what extent they felt empowered by their leader, ranging from 1 (Strongly disagree) to 5 (Strongly agree).

Cronbach’s alpha for the scale was .895.

Self-efficacy was measured with the Self-efficacy scale from Sherer, Maddux, Mercandante, Prentice-Dunn, Jacobs & Rogers (1982). Items for self-efficacy include:

‘When I make plans, I am certain I can make them work’; and ‘One of my problems is

18 that I cannot get down to work when I should’. Lastly, ‘If I cannot do a job the first time.

I keep trying until I can’. The complete measure consisted of 17 items and participants indicated to what extent they agreed to the statements from 1 (Strongly disagree) to 5 (Strongly agree). Cronbach’s alpha for the scale was .693.

Psychological safety was measured with the Psychological Safety scale from

Edmondson (1999). These seven items were transformed to be used on the organizational level by Baer and Frese (2003). Items for psychological safety include: ‘When someone in our company makes a mistake, it is often held against them’; and ‘No one in this organization would deliberately act in a way that undermines others’ efforts’. Lastly,

‘The people in our company value others’ unique skills and talents’. The complete measure consisted of 7 items and participants indicated to what extent they felt

psychologically safe at work, ranging from 1 (Hardly ever) to 5 (Very often). Cronbach’s alpha for the scale was .640.

Participants’ chronological age was measured by using a single item question.

Participants’ age was categorized as 18-24 years, 25-34 years, 35-44 years, 45-54 years, 55-64 years.

Control variables that were considered in the present study are gender, the level of

education and the number of years they have been working in their current position. The

variables were considered to be related to job crafting, because previous research has

shown that these variables are related to job crafting. For example, previous research

argues that crafting can vary according to participants’ level of education. Education is

expected to be related to job crafting because education can influence the skills, abilities

and knowledge of employees (Wrzesniewski & Dutton, 2001). Gender is included in the

present study because previous research showed that females scored slightly higher than

males on job performance measures (Roth, Purvis & Bobko, 2012). Years in current

19 experience is included in this research because tenure is found to influence job crafting (Berg, Wrzesniewski, & Dutton, 2010; Tims et al., 2013).

All control variables were measured by single item questions and age and education

were categorized. For example, the control variables educational level was categorized

into primary school, high school, vocational training, bachelor's degree, master's degree,

advanced graduate work or Ph.D. The variable gender was categorized into male, female

or no answer. Years in current position was not categorized.

20 RESULTS

Descriptive Statistics and Correlation

Table 1 provides descriptive statistics and inter-correlations for all variables that are included in this study. In support of hypothesis 1, a positive and significant relationship was found between empowering leadership and job crafting (R= .179, p= .010).

Furthermore, in contrast with hypothesis 2, self-efficacy and job crafting were negatively associated and this relationship did not reach significance (R= -.097, p= .166).

Hypothesis 3 stated that psychological safety mediates the relationship between empowering leadership and job crafting. In accordance with this reasoning, empowering leadership and psychological safety were positively associated (R= .321, p= .000).

Moreover, in line with expectations, psychological safety and job crafting were positively and significantly related (R= .183, p= .009).

Hypothesis 4 stated that psychological safety mediates the relationship between self- efficacy and job crafting. In contrast to expectations, self-efficacy and psychological safety were negatively and non-significantly associated (R= -.010, p= .886).

For the possible control variables, only the variable years in current position was

negatively and significantly associated with psychological safety (R= -.228, p=.001). All

other control variables were not found to be related to job crafting. Therefore, I will only

take the variable years in current position into account when performing the regression

analysis, as this variable influences the predictor and/ or explanatory variables.

TABLE 1

Descriptive Statistics and Inter-correlations for all study variables

Notes. N=205;

p < .10,

p < .05,

p < .01. For gender 0 = male and 1 = female. For education: (1= Primary school, 2= High School,

3=Vocational training (MBO), 4=Bachelor's Degree, 5= Master's Degree, 6= Advanced Graduate work or Ph.D. For age: (1= 18-24 years, 2= 25- 34 years, 3= 35-44 years, 4= 45-54 years, 5= 55-64 years)

Variables M SD 1 2 3 4 5 6 7 8 9

1 Gender .010 .210 1

2 Educational level 4.350 .793 -.119

1

3 Years in current position 5.881 6.048 -.033 -.173

1

4 Empowering leadership 4.024 .360 -.070 .150* .015 1

5 Self-efficacy 4.349 .269 -.030 .144

.084 .106 1

6 Psychological safety 4.048 .359 .019 .106 -.228

.321** -.010 1

7 Employees’ Age 3.15 1.034 .125

-.172

.401

.068 .041 -.093 1

8 Job crafting 4.100 .516 .102 .054 -.084 .179* -.097 .183** -.011 1

22 Regression Analysis

Table 2 depicts the results of the stepwise linear regression analysis of the relationship between empowering leadership and job crafting. The control variable years in current

position was taken into account when performing the regression analysis. Hence, self-efficacy was added as control variable to test whether the control variables (years in current position and self-efficacy) influence the independent variable empowering leadership in such a way that they indirectly influence job crafting. Results show that the variable empowering

leadership had a positive and significant effect on job crafting (B= .148, SE= .055, p= .007).

Hence, the model explained 3.5% of the variance in job crafting. The results show support for

hypothesis 1, which stated that empowering leadership was positively associated with job

crafting.

23 TABLE 2

Linear regression analysis of empowering leadership on job crafting, including the control variable

Variable Coefficient (SE) t p

Model 1

Years in current position -.024 (.020) -1.207 .229

R² change= .007 Model 2

Years in current position -.024 (.020) -1.185 .237

Self-efficacy -.102 (.111) -.920 .359

R² change= .004 Model 3

Years in current position -.240 (.020) -1.064 .217

Self-efficacy .127 (.110) -1.157 .248

Empowering leadership .148 (.055) 2.709 .007

R² change= .035

Table 3 depicts the results of the stepwise linear regression analysis of the relationship

between self-efficacy and job crafting. The control variable years in current position was

taken into account when performing the regression analysis. Hence, empowering leadership

was added as control variable to test whether the control variables (years in current position

and empowering leadership) influence the independent variable self-efficacy in such a way

that they indirectly influence job crafting. Results show that the variable self-efficacy was not

significantly related to job crafting (B= .127, SE= .110, p= .248). Therefore, no support was

found for hypothesis 2. The model explained 0.6% of the variance in job crafting.

24 TABLE 3

Linear regression analysis of self-efficacy on job crafting, including control the variable

Variable Coefficient (SE) t p

Model 1

Years in current position -.024 (.020) -1.207 .229 R² change=.007

Model 2

Years in current position -.025 (.020) -1.262 .208

Empowering leadership .143 (.054) 2.620 .009

R² change=0.33 Model 3

Years in current position -.024 (.020) -1.238 .217

Empowering leadership .148 (.055) 2.709 .007

Self-efficacy .127 (.110) -1.157 .248

R² change=.006

In order to test hypothesis 3, the Process program of Hayes (2013) was used.

Psychological safety was first included as a mediator in the relationship between empowering leadership and job crafting (table 4). Results indicated that empowering leadership was a significant predictor of psychological safety (B= .240, SE=.070, p= .000) and that

psychological safety was a significant predictor of job crafting (B= .050, SE= .020, p= .017).

Therefore, the mediation results do support the mediational hypothesis (B=.012, 95% CI

[.002, 026]).

25 TABLE 4

Psychological safety as mediator in the relationship between empowering leadership and job crafting

Variable Coefficient (SE) t p

A- path .240 (.070) 3.523 .000

B-path .049 (.021) .2408 .017

Direct effect (c’) 0.007 (.020) -6.760 .003

Indirect effect .012 (.002, .026)

Years in current position -.014 (020) -.676 .500

Note. N = 205

In order to test hypothesis 4, psychological safety was included as a mediator in the relationship between self-efficacy and job crafting (table 5). Results indicated that self- efficacy was not a significant predictor of psychological safety (B= - .008, SE= .069, p=

.910), but psychological safety was a significant predictor of job crafting (B= .051, SE= .020, p= .011). Consequently, the mediation results do not support the mediational hypothesis (B=

-.000, 95% CI [-.009, .007]).

26 TABLE 5

Psychological safety as mediator in the relationship between self-efficacy and job crafting

Variable Coefficient (SE) t p

A- path -.008 (.069) -.113 .910

B-path .051 (.020) 2.564 .011

Direct effect (c’) -.030 (.020) -1.504 .134

Indirect effect -.000 (-.009, .007)

Years in current position -.011 (.020) -.540 .590

Note. N = 205

Lastly, in order to test hypothesis 5, employees’ age was included as a moderator in the relationship between psychological safety and job crafting (table 6). The regression analysis revealed a positive and significant relationship between psychological safety and job crafting (B= .055 SE= .020, p= .006). In addition, a non-significant relationship between employees’

age and job crafting was found (B= .001, SE= .025, p= .998. The interaction effect did not

reach significance (B= -.889, SE= .018, p= .635). Consequently, the moderation results do

not support the moderation hypothesis.

27 TABLE 6

Employees’ age as moderator in the relationship between psychological safety and job crafting

Coefficient (SE) t p

Psychological safety .055 (.020) 2.775 .006

Employees’ age .001(.020) .025 .998

Interaction -.889 (.018) -.476 .635

Years in current position -.017 (.022) -.787 .432

Note. N = 205

28 CONCLUSION AND DISCUSSION

The main purpose of this study was to investigate whether empowering leadership and self-efficacy influence job crafting. The research also examined the role of psychological safety that could possibly influence job crafting behaviors of employees. Finally, employees’

age was examined as moderator in the relation between psychological safety and job crafting.

Results mainly showed support for the relationship between empowering leadership and job crafting behavior. Moreover, the present study found that psychological safety mediates the relationship between empowering leadership and job crafting. Therefore, empowering leadership was associated with increased psychological safety, which in turn predicted employees’ job crafting behavior. No support was found for the mediating role of

psychological safety in the relationship between self-efficacy and job crafting. Hence, no support was found for the direct effect of self-efficacy on job crafting. Lastly, the moderating role of employees’ age in the relationship between psychological safety and job crafting was not found.

Theoretical Implications

The findings of the present study make several contributions to the insights about the

antecedents of job crafting. First, the present study links the literature of job crafting and

empowering leadership (hypothesis 1), responding to call to identify leadership antecedents of

job crafting (Wang, Demerouti & Le Blanc, 2017). I tested both leadership (empowering

leadership) and individual (self-efficacy) antecedents on job crafting and the present study

demonstrated that the leadership antecedent is the only influential predictor of employees’ job

crafting behavior. This view balances that of previous studies on job crafting, which attributed

it largely to stable individual characteristics (Bateman & Crant, 1993).

29 Second, the present research shows that psychological safety mediates the relationship between empowering leadership and job crafting (hypothesis 3). Therefore, this study shows that when employees feel empowered by their leaders, employees are also more likely to feel psychologically safe. In turn, psychological safety can enhance employees’ job crafting behavior.

Third, the present study shows that the role of self-efficacy may be more nuanced than previously believed, because no support was found for the relationship between self-efficacy and job crafting (hypothesis 2). One possible explanation for the lack of support for the relationship between self-efficacy and job crafting may be, because of the study design. In the study design of the present study, self-efficacy was only measured on one day. However, there is a possibility that self-efficacy may vary on a daily basis, because self-efficacy considers the feelings of employees with regard to their abilities to execute desired behavior and to affect their environment successfully (Bandura, 1997). Employees’ feelings are likely to vary from day to day and job crafting can vary accordingly.

Fourth, this study did not found support that a relationship exists between self-efficacy

and job crafting, mediated by psychological safety (hypothesis 4). Although the present study

shows that psychological safety and job crafting were positively related, self-efficacy was not

related to psychological safety. Therefore, no relationship was found between self-efficacy

and job crafting, mediated by psychological safety. One possible explanation for this may be

that employees who are highly self-efficacious do not perceive psychological safety as

necessary in order to craft their jobs (Bandura, 1997). Self-efficacy employees see potential

difficulties as something that can be mastered and they can feel less dependent on perceptions

of psychological safety (Bandura, 1997). Therefore, self-efficacious employees may not

perceive psychological safety as an important condition in order to craft their jobs.

30 Fifth, the present study found a lack of support for the moderating role of employees’ age in the relationship between psychological safety and job crafting (hypothesis 5). It was expected that the relationship between psychological safety and job crafting would be less strongly related, when employees were older, rather than younger. The finding of this research do not show support for these predictions. One possible explanation for the lack of support for this finding may be because the impairments with regard to older employees may be not consistent among all employees (Wong & Tetrick, 2017). Hence, there may be a general misconception about older employees, because employees’ vary considerably from each other. Therefore, there may be no reason to suggest that older employees craft their jobs differently than younger employees (Wong & Tetrick, 2017). Besides, there is a trend going on within organizations that encourage employees towards a lifelong learning. This can foster employees’ employability and may increase job crafting behavior of all employees, without making a distractions in age (Lyons, 2008; Kooij, Tims & Kanfer, 2015).

Practical Implications

The results of the current study include several practical implications. This research

showed that empowering leadership has a positive influence on employees’ psychological

safety and their job crafting behavior. Therefore, it is important for organizations to invest in

job crafting activities by stimulating leaders to empower their employees. This may help

employees to shape the boundaries of their jobs, since empowering leadership can have a

positive impact on employees’ feelings of psychological safety. For instance, leaders should

stimulate employees to take personal initiatives by stimulating participation and autonomy to

make their own choices (Leana et al., 2009). Leaders can also play an important role in

implementing human resource policies and practices to facilitate the opportunity of

employees to craft their jobs.

31 Additionally, this research confirmed the importance of psychological safety for

employees’ job crafting. The present research found that when the environment is

psychological safe for employees, they will be more likely to craft their jobs. Leaders can be encouraged to monitor the individual level of psychological safety in order to stimulate job crafting behavior of employees. Research showed that positive relationships are important for employee’s to perceive psychological safety (Gong, Cheung, Wang, & Huang, 2012; Seligson

& MacPhee, 2004). Positive relationships can provide employees with trust, belonging and empathy. Organizations can therefore develop relationships in the organization, by providing training on role-playing and mirroring of experiences (Gong et al., 2012; Seligson &

MacPhee, 2004). Furthermore, organizations and leaders can increase the psychological safety of employees by limiting the perceived threats of employees or by increasing

employees’ tolerance level when they perceive low levels of psychological safety (Wanless, 2016). Taken all together, there is much that organizations and leaders can do to increase the psychological safety and job crafting behavior of employees.

Limitations and Future Research

The present research is not without limitations. This section will reflect upon the

limitations of this study and describes how future research can address these limitations. The

first limitation is that this research used cross-sectional and single-source data, meaning that

data was collected from a group of employees at one point in time. According to Podsakoff

(2003), this enhances potential concerns for common method bias. Consequently, I cannot be

totally confident that the answers of the respondents would be similar when the research was

examined at a different point in time. Future research should therefore conduct longitudinal

research by using multiple moments of testing in order to address the insights in the effects of

job crafting and the predictors of job crafting over time.

32 The second limitation is that the sample consisted of employees all working in the same company. The sampling strategy was therefore not broad, as employees were recruited from only one company. This maximizes the likelihood that the results are a function of a specific industry or occupation (Podsakoff, 2003). Therefore, the sample used in this study offers a relatively low degree of generalizability. Future studies should take different companies into account and focus on different industries in order to ensure generalizability.

A third limitation is the presence of mainly highly educated people in this study. By analyzing the data, results showed that 87% of all participants completed a Bachelor Degree (UBO), Master Degree (WO) or PhD. Wrzesniewski and Dutton (2001) suggested that job crafting can differ depending on the educational levels of employees. Specifically, they suggest that higher educated employees are more likely to craft their job compared to lower educated employees. Using a sample of mostly high-educated employees in this study may have influenced the study results in a more positive way. Therefore, future studies should use participants with a variety in the levels of education in order to enhance the generalizability of the findings.

A fourth limitation is that, in this study, a percentage of 23% employees (n=48) had only

worked in their current position for one year. According to Frese and Fay (2010), it takes

about 1-10 years to make a right fit between employees’ proactive behavior with jobs that

require significant amounts of personal initiative. Considering the results of this study, there

is a relatively large group of employees included in this research who have only worked in

one specific position for a year. Future studies should examine longitudinal research on

whether job crafting behavior of employees change when employees work longer in the same

positions, since the match between proactive behavior of employees and their jobs may

become stronger over time (Frese & Fay, 2001).

33 The fifth limitation is that in this study, empowering leadership and self-efficacy were the only antecedents used to test employees’ motivation to craft their jobs. Considering other antecedents, it could be interesting to look at other predictors that will motivate employees to craft their jobs, especially other leadership antecedents. Wang et al., 2017), underscores the need for future research on the role of leaders with regard to job crafting, because this role is so far mostly speculative. In practice, leaders may exhibit different leadership styles,

depending upon the situation (Sagie, 1997). Therefore, it may be interesting for future research to test the effectiveness of hybrid leadership on job crafting.

According to Bolden (2011), hybrid leadership is characterized as a leadership style in which leaders combine instrumental, expressive, reactive and proactive traits. In other words, hybrid leaders use both hierarchical and non-hierarchical leadership styles. In addition to empowering leadership, hybrid leadership may be also effective for job crafting because of the mix of hierarchical and non-hierarchical traits. Hierarchical traits may be effective for job crafting because leaders can, for example, reward employees who take initiatives or ask employees to be more proactive (Martin, Liao & Campbell, 2013). A hierarchical leadership style in combination with a non-hierarchical leadership style, such as empowering leadership, may be even more effective for job crafting than when leaders have only traits of non-

hierarchical leaders. Therefore, future research should conduct research on the relationship between hybrid leadership and job crafting.

Furthermore, it may be interesting for future research to use psychological safety as a

mediator in the relationship between hybrid leadership and job crafting, because perceptions

of psychological safety may become even more necessary for employees when their leaders

have more traits of a hierarchical leader (i.e. stimulates less participation of employees)

(Wanless, 2016).

34

Another possibility for future research is to include a separate analysis of how leaders

craft their jobs differently in comparison with employees lower in formal hierarchy. Instead,

the present research has analyzed how employees in general craft their jobs, without making

the distinction between leaders and employees. However, only little is known about the ways

leaders in particular craft their jobs (Wang et al., 2017). It is known that leaders have, in

general, high autonomy and power, which gives them considerable room to craft. Future

studies should gain more specific insights in how leaders craft their jobs in terms of task

crafting, relational crafting and cognitive crafting. It is expected that when leaders actively

craft their jobs, employees are more likely to follow this example-behavior because leaders

can act as role-models (Gardner, Avolio, Luthans, May & Walumbwa, 2005).

35 APPENDIX A

MEASURES USED IN STUDY

General Measures

1. What is your age?

-18-24 years old -25-34 years old -35-44 years old -45-54 years old -55-64 years old -65-74 years old 2. What is your gender?

-Male -Female -No answer

3. What is the highest level of education you have obtained?

-Primary school -High School

-Vocational training (MBO) -Bachelor's Degree

-Master's Degree

-Advanced Graduate work or Ph.D.

4. How many years have you been working in your current position?

Job Crafting (adapted from Slemp and Vella-Brodrick, 2013)

The following questions are about how you behave at your place of work. Please indicate how often you engage in the behaviors described below using the following scale:

1= Hardly ever, to 5 = Very often.

Note: 'Very Often' means as often as possible in your workplace.

1. I introduce new approaches to improve my work

2. I change the scope or types of tasks that I complete at work 3. I introduce new work tasks that better suit to my skills or interests 4. I choose to take on additional tasks at work

5. I think about how my job gives my life purpose

6. I remind myself about the significance my work has for the success of the organization

7. I remind myself of the importance of my work for the broader community 8. I think about the ways in which my work positively impacts my life 9. I reflect on the role my job has for my overall well-being

10. I make an effort to get to know people well at work

11. I organize or attend work related social functions

36 12. I organize special events in the workplace (e.g., celebrating a co-worker's birthday 13. I choose to mentor new employees (officially or unofficially)

14. I make friends with people at work who have similar skills or interest

Empowering Leadership (adapted from Spreitzer, 1995)

The following statements are about your supervisor, the person who oversees your work or makes decisions regarding your work. Please indicate to what extent you disagree or agree with the following statements, using the following scale:

1= Strongly disagree to 5= Strongly agree.

My supervisor...

1. Helps me see that the work I do is very important

2. Supports that my job activities are personally meaningful to me 3. Wants the work I do to be meaningful to me

4. Helps me be confident about my ability to do my job

5. Makes me self-assured about my capabilities to perform my work activities 6. Helps me master the skills necessary for my job

7. Provides significant autonomy in determining how I do my job 8. Allows me to decide on my own how I want to do my work

9. Provides considerable opportunity for independence and freedom in how I do my job

10. Shows me that my impact on what happens in my department is large 11. Provides me a great deal of control over what happens in my department 12. Let me have significant influence over what happens in my department

Psychological Safety (adapted from Edmondson, 1999)

The following questions are about you as a person. Please indicate to what extent the following statements apply to you, using the following scale:

1= Strongly disagree, to 5= Strongly agree.

1. In my company some employees are rejected for being different

2. When someone in my company makes a mistake, it is often held against them.

3. No one in my company would deliberately act in a way that undermines others’

efforts.

4. It is difficult to ask others for help in my company 5. In my company one is free to take risks.

6. The people in my company value others’ unique skills and talents.

7. As an employee in my company one is able to bring up problems and tough issues.

37 Self-efficacy (adapted from Sherer et all., 1982)

The following questions are about you as a person. Please indicate to what extent the following statements apply to you, using the following scale:

1= Strongly disagree, to 5= Strongly agree.

1. When I make plans, I am certain I can make them work.

2. One of my problems is that I cannot get down to work when I should.

3. If I cannot do a job the first time. I keep trying until I can.

4. When I set important goals for myself, I rarely achieve them.

5. I give up on things before completing them.

6. I avoid facing difficulties.

7. If something looks too complicated, I will not even bother to try it.

8. When I have something unpleasant to do, I stick to it until I finish it.

9. When I decide to do something, I go right to work on it.

10. When trying to learn something new, I soon give up if I am not initially successful.

11. When unexpected problems occur, I do not handle them well.

12. I avoid trying to learn new things when they look too difficult for me.

13. Failure just makes me try harder.

14. I feel insecure about my ability to do things.

15. I am a self-reliant person.

16. I give up easily.

17. I do not seem capable of dealing with most problems that come up in life.

38 APPENDIX B

INVITATION FOR PARTICIPATION FIELD STUDY ON JOB CRAFTING

Dear all,

The University of Groningen is looking into the subject of Leadership and antecedents of Proactive behavior. As part of this research, Evelien Wassenaar is currently assigned as an Intern within our HR Department in Drachten. By means of a digital survey, I would like to kindly ask you to fill out the questionnaire as soon as possible, but ultimately before April 7

. We realize how precious your time is, that is why we made sure this survey will only take around 10 minutes to complete. Would you be so kind in contributing to this research?

Please be assured that your input will be 100% anonymous, meaning that there will be no trace or link in any way regarding your input and you as a person. Thus, nobody can tell who submitted which information. Philips will only receive some high level feedback on general correlations regarding leadership and proactive behavior, which Evelien will convert in to some advices on how we could improve. If you would be interested to receive a summary of the results, please send an email to Evelien Wassenaar (Evelien.wassenaar@philips.com). If you have any further questions with regard to the survey or the research, please address your questions to Evelien only.

Please find below the link to the survey.

https: https://ph08.eyeqsoft.com/eqphi/r1/czg2m

A sincere thanks in advance, your input is highly appreciated!

Best wishes,

Pieter van Groos

39 APPENDIX C

CHECKING FOR ASSUMPTIONS

• Assumptions for regression

• Before a regression analysis can be examined, five assumptions must be met. First, the assumption of independent observations is checked. It is assumed that the assumption has been violated, because the respondents all working at the same company and some are working at the same departments. Therefore, independency in the responses of participants cannot be prevented. In contrast employees have participated independent from each other and were able to fill in the questionnaire anonymous.

• Second, the assumption of a linear relationship is checked. For this purpose, a

scattergram of the residuals of the dependent variable is made. This shows that the Loess line is not completely on the red line, so the assumption of the linear relationship is slightly violated. But since we work with real data, it is

virtually impossible to get the residuals exactly on the average line. With this in mind, it can be said that the assumption of a linear relationship is almost not violated and in this research a linear regression analysis can be used.

• Third, the assumption of homoscedasticity is checked. There is analyzes whether the

conditional standard deviation has remained constant. This can be read by checking in the

scatter diagram whether the distribution of the points are in balance for both above and below

the average line. Most outliers are both below and above the red line. From this it can be

concluded that the assumption of homoscedasticity has not been violated because the

conditional standard deviation is constant. Thus, the data can still be used for a hierarchical

linear regression analysis.

40

Variable VIF

Model 1

Education 1.031

Years in current position 1.031

Model 2

Education 1.034

Years in current position 1.031

Empowering leadership 1.003

Model 3

Education 1.060

Years in current position 1.044

Empowering leadership 1.007

Self-efficacy 1.038

Model 4

Education 1.064

Years in current position 1.095

Empowering leadership 1.067

Self-efficacy 1.039

Psychological safety 1.123

Model 5

Education 1.179

Years in current position 1.117

Empowering leadership 1.159

Self-efficacy 1.045

Psychological safety 1.158

Employees’ age 1.303

41 Fourth, the normality assumption is checked. The distribution of the dependent variable job crafting can be seen in figure demonstrated above, in which the histogram and the P-P plot of this distribution are shown. From the histogram of the residuals it appears that there is a reasonable normal distribution. In order to be able to make a better statements or the

assumptions of a normal distribution in the conditional distribution of job crafting, we look at the P-P plot. Here we see that the points fall reasonably close to the line, but there is one place where the points do not touch the line. On the basis of this reasoning it can be concluded that the assumption of normality has not been violated. Therefore, a linear regression can be performed for this study.

Finally, in in the final model is checked for outliers. Not all outliers influence the

regression line, but it is therefore important to find those outliers that do have an influence on

the regression line. For this, the studentized residuals are viewed. The points must lie between

a z-score of -3 and 3. If they are above or below this, we speak of outliers. To determine

whether these points are indeed an influential outlier, the leverage method is used. The

formula is as follows: leverage = 3 * p / n. The p stands for the number of parameters in the

model and n stands for the number of respondents in the sample. In this sample, this limit is

3 * (7/502) = 0.10. In this sample there are 0 influential outliers. For this the regression

42 analysis has been carried out again, without the influential points. From these data it can be concluded that the removal of the influential points makes no difference in the violations of the assumptions and in whether or not they make significant. For this reason, no outliers need to be removed from the sample.

REFERENCES

• Baer, M., & Frese, M. (2003). Innovation is not enough: Climates for initiative and psychological safety, process innovations, and firm performance. Journal of organizational behavior, 24(1), 45-68.

• Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change.

Psychological review, 84(2), 191.

• Bateman, T. S., & Crant, J. M. (1993). The proactive component of organizational behavior: A measure and correlates. Journal of organizational behavior, 14(2), 103-118.

• Berg, J. M., Grant, A. M., & Johnson, V. (2010). When callings are callings: Crafting work and leisure in pursuit of unanswered occupational callings. Organization Science, 21, 973-994.

• Berg, J. M., Wrzesniewski, A., & Dutton, J. E. (2010). Perceiving and responding to challenges in job crafting at different ranks: When proactivity requires

adaptivity. Journal of Organizational Behavior, 31(2‐3), 158-186.

• Bolden, R. (2011). Distributed leadership in organizations: A review of theory and research. International Journal of Management Reviews, 13(3), 251-269.

• Breevaart, K., Bakker, A. B., Demerouti, E., & Derks, D. (2016). Who takes the lead? A multi-source diary study on leadership, work engagement, and job

performance. Journal of Organizational Behavior, 37(3), 309-325.