Job crafting and well-being : the mediating role of person-job fit

(1)

1

Job crafting and well-being: the mediating role of

person-job fit

Sheila van Teeffelen (10615393) University of Amsterdam

Faculty of Economics and Business

Master Thesis Msc. in Business Administration Leadership and Management track

Academic year: 2014-2015 Supervisor: Dr. C. Boon

Second reader: Dr. C.K. Buengeler Amsterdam, January 2015

(2)

2

Abstract

This study aims at examining the influence of person-job fit (P-J fit) in the relationship between job crafting and well-being. Job crafting was assessed with four dimensions, namely increasing structural job resources, increasing social job resources, increasing challenging job demands and decreasing hindering job demands. Well-being was divided into job satisfaction, work engagement and stress. Data of 201 employees were collected from various organizations in the Netherlands. This study proposes that P-J fit mediates the relationship between job crafting and well-being. The results showed that P-J fit partly mediated the relationship between (a) increasing structural job resources, (b) increasing challenging job demands and (c) decreasing hindering job demands and job satisfaction. Additionally, P-J fit partly mediated the relationship between (a) increasing structural job resources and (b) increasing challenging job demands and work engagement. Furthermore, P-J fit fully mediated the relationship between decreasing hindering job demands and work engagement. Finally, P-J fit did not mediate between increasing social job resources and well-being on the one hand and between the job crafting dimensions and stress on the other. The implications for theoretical and practical job crafting research are provided.

Keywords: Increasing structural job resources; increasing social job resources; increasing challenging job demands; decreasing hindering job demands; well-being; person-job fit

(3)

3

List of figures

Figure 1: Research model ... 22

Figure 2: Mediation model (Model 4 of Hayes) ... 26

Figure 3: Mediation analysis results of job crafting on job satisfaction ... 45

Figure 4: Mediation analysis results of job crafting on work engagement ... 45

List of tables

Table 1: Means, SDs, correlations and reliability coefficients ... 31

Table 2: Regression analysis results of job crafting on well-being ... 36

Table 3: Regression analysis results of job crafting on person-job fit ... 38

Table 4: Regression analysis results of person-job fit on well-being ... 39

Table 5: Direct effects of job crafting on well-being ... 43

(5)

1

1. Introduction

Some employees may find that their needs are met in their current work environments or that their needs are met elsewhere in their lives. Other employees are motivated to fulfill needs at work. Employees who aim at fulfilling their needs at work will try to create more challenging work environments in ways that fulfill their needs (Bakker et al., 2012a; Wrzesniewski and Dutton, 2001). Understanding how employees create a work environment that is resourceful for themselves has become an increasingly important topic (Tims et al., 2013). Reasons for creating more challenging work have to do with the current changes in work environment, the nature of work which is changing (e.g. technical area), employees who are seen as ‘free agents’ (Wrzesniewski and Dutton, 2001) and managers who do not always have time to redesign employee jobs because of limited time and insufficient organizational resources (Chen et al., 2013).

Job design is an important issue in the management field and has changed from a top-down approach to a new, important bottom-up approach. This new approach is about the process (not a one-time event) in which employees actively create a more challenging work environment. It means employees play an (pro)active role in changing and redesigning their work environment (Chen et al., 2013). This concept is called ‘job crafting’ and could be defined as “the physical and cognitive changes individuals make in the task or relational boundaries of their work” (Wrzesniewski and Dutton, 2001, p.179). Job crafting behavior may function as a strategic advantage during ongoing changes (Petrou et al., 2012). This makes it an important concept because a work environment is dynamic and constantly changing (Wrzesniewski and Dutton, 2001). The present study focuses on job crafting because the trend is such that more and more employees will craft their own working lives (Wrzesniewski and Dutton, 2001).

(6)

2 Many studies found that the quality of work environment has a major effect on well-being of employees (Tims et al., 2013). So, job crafting can potentially influence the outcomes of employee well-being (Nielsen and Abildgaard, 2012). This leads to job crafting being seen as a promising concept in organizational psychology (Tims and Bakker, 2010). Job crafting is related to positive and negative outcomes (Wrzesniewski and Dutton, 2001). This study focuses on job satisfaction, work engagement and stress. Job satisfaction is the most common operationalization of job related well-being (Tims et al., 2013; Tinsley, 2000) and the more traditional form (Warr and Inceoglu, 2012). On the other hand, work engagement is the more recently studied form of well-being (Warr and Inceoglu, 2012) and enjoys a growing interest because of its predictive value for job performance (Bakker et al., 2012a). Finally, stress is an important concept because its socially costly and has dysfunctional effects on job performance (Motowidlo et al., 1986).

Employees who craft their jobs create work environments that are more in line with the specific characteristics of the employee (Wrzesniewski and Dutton, 2001). In the literature on organizational behavior, fit or match or congruence between the characteristics of employees and their work environments is one of the topics that is most frequently researched (Greguras et al., 2009; Kristof-Brown et al., 2005; Schneider, 2001). This match is referred to as the person-environment (P-E) fit, “the compatibility between an individual and a work environment that occurs when their characteristics are well matched” (Kristof-Brown et al., 2005, p.281). P-E fit is a general construct and the trend is narrowing the focus to study fit with specific aspects of the environment. Researchers distinguish between the various aspects of fit in order to define each as a unique construct (Jansen and Kristof-Brown, 2006). Several types of fit have received attention and emerge as important research domains. These types include the fit between an individual and his/her job (P-J fit), his/her supervisor (P-S fit), his/her organization (P-O fit) and his/her work group (P-G fit) (Kristof-Brown et al., 2005).

(7)

3 This study focuses on person-job fit (P-J fit), “the relationship between a person’s characteristics and those of the job or tasks that are performed at work” (Kristof-Brown et al., 2005, p.284). P-J fit is a fit that is relevant in the recruitment, selection, socialization and long-term tenure stages (Jansen and Kristof-Brown, 2006).

Job crafting is a topic that only appeared recently (Demerouti, 2014). Therefore, job crafting has received little attention so far (Tims and Bakker, 2010). More research is needed to examine the consequences of job crafting for individual employees (Tims et al., 2012). In order to explain why there is a relationship between job crafting and well-being, mediators are needed to explain this relationship. The current study concentrates on P-J fit as a mediator to explain the relationship between job crafting behavior and well-being outcomes. Through the process of job crafting, employees “act upon the job to create a better fit” (Wrzesniewski and Dutton, 2001, p. 118). Thereby, P-J fit is an important antecedent of job-related attitudes and outcomes (Carless, 2005; Kristof-Brown et al., 2005; Lauver and Kristof-Brown, 2001). The current study strives to answer the following research question: ‘What is the mediating role of person-job fit in the relationship between job crafting and well-being?’

This study hopes to contribute to the literature on job crafting in several ways. Research that has been carried out so far is mostly based on qualitative research studies (Tims et al., 2012) and on specific work domains (e.g. Chen at al., 2013). This study focuses on quantitative research and different work domains. Furthermore, this research goes beyond a study of Chen et al. (2013). They did an interesting study in this field about the relationship between job crafting and the effects on work engagement and the mediating role of P-J fit. The study mentioned above is used as a starting point, but the current study creates a more complex model to contribute to the job crafting theory. Chen et al. (2013) focused both on individual and collective job crafting. In accordance with their findings that the indirect effect of individual crafting on work engagement through P-J fit is stronger than that of

(8)

4 collaborative job crafting, the current study focuses on individual job crafting. Additionally, individual job crafting may be divided into the four dimensions of job crafting as developed by Tims et al. (2012). These four dimensions are: increasing structural job resources, increasing social job resources, increasing challenging job demands and decreasing hindering job demands. In this study these four job crafting dimensions have been examined because different dimensions could have different effects on employee well-being outcomes (Tims and Bakker, 2010). Tims et al. (2012) mentioned more research is needed to examine possible job crafting outcomes for employees. Therefore, an active positive well-being outcome work engagement and a less active positive well-being outcome job satisfaction have been examined. Oldham and Hackman (2010) recommended focusing on dysfunctional outcomes of job crafting. Therefore, a negative well-being outcome stress is examined. The aim is to gain insight into the nature of job dimensions and their relationship with well-being, through P-J fit. If the differentiated job dimensions may have a different relation with the well-being of employees, through person-job fit, this study aims at providing a better understanding of the functioning of employees.

This study discusses the current literature on the various concepts. Additionally, a description of the research methodology is presented. Furthermore, in the result analysis eight hypotheses are tested. Finally, in the discussion chapter the theoretical and practical implications, the limitations, future directions and the conclusion are presented.

(9)

5

2. Literature review

This chapter discusses the most relevant findings of the current job crafting literature and states the eight hypotheses of this study. The key concepts of this study: job crafting, well-being and P-J fit, are presented. Furthermore, the relationships between these variables are described. The chapter continuous with describing the direct relationships between the concepts, followed by outlining how P-J fit mediates between job crafting and well-being. Comparing the direct and indirect effects, allows me to examine whether P-J fit fully or partly mediates the relationship. This chapter ends with a research model which graphically illustrates the hypotheses stated.

2.1 Job crafting

The job crafting theory is based on the classic job design theory which focuses on managers designing jobs for employees (Hackman and Oldham, 1980). Although the concept of job crafting was coined by Wrzesniewski and Dutton (2001), the idea was already mentioned by Kulik, Oldman and Hackman (Kulik et al., 1987). They argued that employees may redesign their own jobs with or without supervision. This is in line with Wrzesniewski’s and Dutton’s (2001) definition of job crafting: “the physical and cognitive changes individuals make in the task or relational boundaries of their work” (p.179). An important job design theory change is the switch in responsibility. Employees are also seen as individuals who are responsible for work outcomes (Tims and Bakker, 2010). Wrzesniewski and Dutton (2001) argued the motivation for job crafting arises from three basic individual needs. Firstly, to get some control over their jobs in order to avoid estrangement from their work. Secondly, to get motivation to create a positive self-image in their jobs. Finally, to get connected to others, a basic human need. Demerouti (2014) added a final motivation reason to these three basis

(10)

6 individual needs, namely to create conditions in which employees can work healthily and with motivation.

Employees are hired by the organization and then employees start to change the job to create a better fit with their abilities and preferences (Tims and Bakker, 2010). Job crafting is not about redesigning the job as a whole, but changing aspects of the job in the boundaries of specific job tasks (Tims et al., 2012). According to Wrzesniewski and Dutton (2001), job crafting is referred to as changing the task boundaries (either physically or cognitively), the relational boundaries, or both boundaries of the job. Changing physical task boundaries refers to changes in the form, scope or number of the job tasks. Changing cognitive task boundaries indicates changes in how an employee sees the job. Changing relational boundaries refers to with whom an employee interacts while doing the job. All these changes can be incremental or radical, visible or invisible and affect ‘the meaning of the work’ and ‘one’s work identity’ (Wrzesniewski and Dutton, 2001). The meaning of the work refers to a different purpose for the work one is doing and one’s work identity indicates shaping the form and amount of interaction with others at work. These changes are implemented on their own initiative during their working lives and may lead to a work environment that is more in line with the specific characteristics of the employee (Wrzesniewski and Dutton, 2001).

Although the first instrument to measure job crafting behavior was developed by Wrzeniewski, namely: task, relationship and cognitive crafting (Wrzesniewski and Dutton, 2001), there was no real evidence for this instrument (Demerouti, 2014). Therefore, Tims et al. (2012) developed and validated a scale to measure job crafting behavior. Their operationalization of job crafting is based on the types of job characteristics mentioned in the job demands-resources model (Tims et al., 2013). Based on this model, Tims et al. (2012) define job crafting as “the self-initiated changes that employees make in their own job demands and job resources to attain and/or optimize their personal work goals” (p.173). The

(11)

7 model distinguishes two types of job characteristics: job demands and job resources. Job demands are job characteristics that “require sustained effort from employees and are, therefore, associated with certain costs” (Tims et al., 2013, p.231). Job resources are job characteristics that “contribute toward achieving work-related goals, reducing the effect of job demands and associated costs, and stimulating personal development” (Tims et al., 2013, p.231). Job crafting refers to the changes employees make with their personal abilities and needs to balance their job demands and job resources (Tims et al., 2010). Everyone is capable to craft their jobs, because every job has job demands and job resources that could be increased or decreased (Berg et al., 2010; Wrzesniewski and Dutton, 2001).

Based on the job demands-resources model, Tims et al. (2012) argued that there are four independent job crafting dimensions: increasing structural job resources, increasing social job resources, increasing challenging job demands and decreasing hindering job demands. ‘Structural’ and ‘social’ refer to the type of job resource that are crafted (Tims et al., 2013). The difference between the structural and social job resources is based on the type of job resource. Structural resources have more impact on the job design (such as gaining more responsibilities) and/or knowledge about the job (such as opportunities to develop oneself). Social resources have more impact on the social aspects of the job (such as asking for feedback) and/or interaction (such as social support) (Tims et al., 2012). ‘Challenge’ and ‘hindrance’ refer to the level of job demands. The difference between these demands is based on the type of demand. Job demands are associated with costs (Tims et al., 2013), this does not apply for challenge demands. Crawford et al. (2010) mentioned employees may perceive challenge demands as opportunities to learn, achieve and demonstrate the competence that is required. Examples of challenge demands are: a high level of job responsibility, time pressure and a high workload. Hindrance demands mean they hinder the employee in realizing valued goals (Tims and Bakker, 2010). Examples of hindrance job demands are: role conflict,

(12)

8 hassles, red tape, organizational politics and role ambiguity. Often, employees perceive these hindrance demands as roadblocks, barriers and constraints that hinder progress to achieve the goal (Crawford et al., 2010). Finally, when the levels of job demands and job resources are not balanced, there is a misfit and employees will be triggered to reduce this misfit by using the four dimensions of job crafting behavior (Tims et al., 2012).

2.2 Well-being

Employee well-being has become an increasingly important concept in management research, because the concept is related to many important work outcomes such as job performance, profitability and employee retention (Wright and Huang, 2012). Employee well-being can be described as the overall quality of the functioning and experience of an employee at work (Warr, 1987). According to Grant et al. (2007), well-being could be divided into three dimensions: happiness, health and relationship. Satisfaction and engagement are examples of happiness-related well-being and are targeted at the job. They are based on subjective functioning and experience at work. Stress is an example of a health-related type of well-being. This is based on functioning and bodily health at work (Grant et al., 2007). Stress could be measured objectively or subjectively (Hobfull, 1989). In the current study all well-being outcomes are measured subjectively.

Job satisfaction is the most common operationalization of work-related well-being (Tims et al., 2013). Satisfaction comes from the Latin ‘satis’ and implies sufficiency or adequacy (Macey and Schneider, 2008). Locke (1969) mentioned, “it is the degree to which the job fulfills or allows the fulfillment of the individual’s needs that determines his or her degree of job satisfaction” (p.1303). Likewise, the theory of work adjustment states that “job satisfaction represents an individual worker’s subjective evaluation of the degree to which his or her requirements are met by the job” (Cable and DeRue, 2002, p.877). Someone who is

(13)

9 satisfied with his or her job believes that it will provide or has provided an acceptable level of what is wanted (Warr and Inceoglu, 2012).

Kahn (1990) was the first who addressed the concept of work engagement. This concept has been defined as the opposite of burnout. Engagement is represented by: energy, involvement and efficacy, which are considered as the direct opposites of the three burnout dimensions: exhaustion, cynicism and lack of professional efficacy (Schaufeli et al., 2002). Furthermore, work engagement is defined as “an active, positive, work-related state characterized by vigor, dedication, and absorption” (Lu et al., 2014, p.142). Vigor implies willingness to put effort into the job, high levels of energy and persistence in the face of difficulties. Dedication is characterized as enthusiasm and the feelings of being challenged by the job. Absorption implies the quality of being fully focused and concentrated on the job (Tims et al., 2013). Vigor and dedication are the core dimensions of engagement. Absorption appears to be related to the concept of flow. It plays a different role compared to the other two dimensions (Schaufeli et al., 2009). However, in almost all previous studies these three dimensions are compared to an overall concept of work engagement (e.g. Chen et al., 2013; Warr and Inceoglu, 2012).

Both job satisfaction and work engagement have a positive meaning, but work engagement differs from job satisfaction in being more strongly activated (Tims et al., 2013). Macey and Schneider (2008) mentioned “it is the sense of energy and enthusiasm in engagement that makes the construct different” (p. 24) and “engagement connotes activation, whereas satisfaction connotes satiation” (p. 8). Finally, whereas engage workers are satisfied, satisfied workers are not necessarily be engaged (Warr and Inceoglu, 2012).

Stress is, in contrast to the previous outcomes, a negative employee well-being outcome because of its dysfunctional and socially costly effects on job performance (Motowidlo et al., 1986). According to Hobfoll (1989), the term stress is ‘borrowed’ from the

(14)

10 field of the physics. Walter Cannon (1932) applied the concept of stress to humans. He was worried by the effects of environmental stressors on persons (e.g. cold). He mentioned low level stressors could be survived, high levels of stressors lead to a breakdown of the biological system (Hofboll, 1989). The lack of consensus among authors ensures there is not one definition of stress (Motowidlo et al., 1986). The concept has become an over-general concept and consists of different forms of stress (Pines and Keinan, 2005). According to Parker and DeCotiis (1983), job stress is “a particular individual's awareness or feeling of personal dysfunction as a result of perceived conditions or happenings in the work setting” (p.161). Motowidlo et al. (1986) referred to stress as “an unpleasant emotional experience associated with elements of fear, dread, anxiety, irritation, annoyance, anger, sadness, grief, and depression” (p. 618). Hobfoll (1989) presented a new stress model, based on the conservation of resources theory. This well-established theory suggests that “people strive to retain, protect and build resources and that what is threatening to them is the potential or actual loss of these valued resources” (Hobfoll, 1989, p.516). Based on this theory Hobfoll (1989) defined stress as “a reaction to the environment in which there is (a) the threat of a net loss of resources, (b) the net loss of resources, or (c) a lack of resource gain following the investment of resources” (p.516).

2.3 Job crafting and well-being

Job crafters view their work as a calling (Bakker et al., 2012a). The dominant view in previous research was that callings are psychologically pleasant experiences (Berg et al, 2010a). Berg et al. (2010a) mentioned employees could have unanswered callings. These unanswered callings can bring a pleasant experience, greater meaning and enjoyment to the job experience of a person, but also unpleasant experience, greater regret and stress due to difficulties in pursuing these unanswered callings.

(15)

11 The job demands-resources model explains the relationship between job crafting and well-being. This well-established model includes many possible working circumstances and focuses on positive and negative employee well-being outcomes (Bakker and Demerouti, 2007). The model underlined two psychological processes that determine the well-being of employees. First, the health impairment process assumes that high demands override employees’ mental and physical resources and reduce employees’ well-being. Second, the motivational process suggests that resources have a motivational function so that employees activate more resources and increase employees’ well-being (Bakker and Demerouti, 2007). Thus, someone who has a high level of job resources, is expected to be associated with an increase in well-being. Additionally, someone who has a high level of job demands, is expected to be associated with a decrease in well-being. However, a new ‘differentiated job demands-resources model’ was developed by Crawford et al. (2010). It was developed because later research, based on this perspective, came up with conflicting findings in the relationship between demands and engagement. They refined the model by making a distinction between challenge and hindrance demands to better explain this relationship. They updated this model by using the transactional theory of stress (Crawford et al., 2010). The transactional theory of stress explains that employees asses stressful situation such as job demands in terms of potentially challenging or threatening demands (Lazarus and Folkman, 1984). However, challenge demands may also have the potential to increase their well-being by for example personal growth (Crawford et al., 2010). Thus, based on this theory, the expectation is that increasing structural and social job resources will lead to higher well-being. Increasing challenging demands could lead to both positive and negative well-well-being. Decreasing hindering job demands may result in higher well-being. Finally, the expectation is that demands are related to stress.

(16)

12 In addition, there is another theory that explains the relationship between job crafting and well-being. The conservation of resources theory argues that the main human motivation is focused towards the maintenance and accumulation of resources (Hobfoll, 2001). Maximizing social job resources and task resources will have a positive effect on well-being. The reason is that larger pools of resources make it easier for the individuals to protect themselves from resource reduction and to deal with the demands of the environment (Hobfoll, 1989). Stress occurs when resources are lost or threatened (Hobfoll and Freedy, 1993). Furthermore, the expectancy value theory states that “when employees are confronted with job demands, they are likely to assess the effort required to deal with the demands, the probability of success and the instrumentality of dealing with the demands to obtain highly valued outcomes”(Van den Broeck et al., 2010, p.5). Employees may be deal with challenge demands because they believe they have the effort to deal with the demand. They believe they will make it to a success and they believe they will achieve valued outcomes (e.g. work goal attainment). On the other hand, employees may be avoid hindrance demands because they believe these demands are more difficult to overcome. They do not believe in success and achieving their goals (Van den Broeck et al., 2010). Finally, when employees believe they lack resources, they may feel stressed (Lazarus and Folkman, 1984). Thus, based on these theories the expectation is that increasing structural and social resources are related to positive well-being, and also increasing challenging job demands and decreasing hindering job demands. Finally, the expectation is that resources are related to stress.

As mentioned earlier, job crafting is a topic in the literature concerned that appeared recently (Demerouti, 2014). This means there is little empirical evidence for the relationship between job crafting and well-being. Based on the studies that were carried out, Podsakoff et al. (2007) mentioned challenge stressors were positively associated with job satisfaction and hindrance stressors were negatively associated with job satisfaction. Berg et al (2010) argued

(17)

13 challenge demands were positively associated with job satisfaction. Crawford et al. (2010) mentioned resources were positively related to engagement and negatively related to burnout. Furthermore, challenge demands were positively related to engagement and positively related to burnout. Finally, hindrance demands were negatively related to engagement and positively to burnout. That the dimensions are related to burnout suggest it may be appraised as stressful. Burnout is often conceptualized within the concept of stress. It may be a sub-category of stress (Pines and Keinan, 2005).

In summary, based on the previous literature, the expectation is that the dimensions of job crafting are positively related to job satisfaction and work engagement. Furthermore, the expectation is that there is a relationship between the dimensions and stress.

Hypothesis 1a: Increasing structural job resources is positively related to job satisfaction Hypothesis 1b: Increasing social job resources is positively related to job satisfaction Hypothesis 1c: Increasing challenging job demands is positively related to job satisfaction Hypothesis 1d: Decreasing hindering job demands is positively related to job satisfaction

Hypothesis 2a: Increasing structural job resources is positively related to work engagement Hypothesis 2b: Increasing social job resources is positively related to work engagement Hypothesis 2c: Increasing challenging job demands is positively related to work engagement Hypothesis 2d: Decreasing hindering job demands is positively related to work engagement

Hypothesis 3a: Increasing structural job resources is related to stress Hypothesis 3b: Increasing social job resources is related to stress Hypothesis 3c: Increasing challenging job demands is related to stress Hypothesis 3d: Decreasing hindering job demands is related to stress

(18)

14

2.4 Person-job fit

Person-job fit (P-J fit) is defined as “the match between individual knowledge, skills, and abilities (KSA) and demands of the job or the needs or desires of an individual and what is provided by the job (Carless, 2005, p.412). The interaction between the person and the job is grounded in the interactionist theory of behavior. This theory describes the interaction of personal and environmental variables which account for the greatest variance (Muchinsky and Monahan, 1987). The match between the job and employee characteristics is based on the job characteristics theory (Hackman and Oldman, 1976). This theory emphasizes the importance of the match. In addition, when there is a match this theory predicts outcomes which are desirable for the employee and the organization (Kulik et al., 1987). The match could be described as a subjective fit, because the fit between the individual and the job is perceived and reported by the individual (Kristof-Brown et al., 2005). P-J fit should be judged according to the tasks performed and not according to the organization in which the job exists. Job is referred to as the tasks an employee is expected to achieve in exchange for employment and the characteristics of those tasks (Kristof, 1996).

P-J fit is one of the types of P-E fit which has received attention in the available literature (Kristof-Brown et al., 2005). Although the current study focuses on P-J fit, first two higher conceptualization of fit will be discussed. There are two conceptualizations which make it clear what exactly constitutes a fit in the P-E literature (Kristof, 1996; Muchinsky and Monahan, 1987). The first conceptualization, supplementary fit, implies “a person fits into some environmental context because he or she supplements, embellishes, or possesses characteristics which are similar to other individuals in this environment” (Muchinsky and Monahan, 1987, p.269). The second conceptualization, complementary fit, refers to “the match between an individual’s talents and the corresponding needs of the environment” (Muchinsky and Monahan, 1987, p.268). Whereas complementary fit has dominated the P-J

(19)

15 ﬁt literature (Edwards 1991; Kristof-Brown et al., 2005; Lauver and Kristof-Brown, 2001), supplementary ﬁt has been the focus of other fits such as person-organization (P-O) fit, person-supervisor (P-S) fit and person-group (P-G) fit (Kristof-Brown et al., 2005).

In the complementary fit model, the environment is described separately from the people who live in it. It is defined according to its demands and the abilities of a person to match the needs of the environment (Muchinsky and Monahan, 1987). Thus, P-J fit contains both concepts: demands–abilities (D-A) fit and needs–supplies (N-S) fit (Edwards, 1991). The difference between these two fits is the focus. Firstly, D-A fit focuses on the organization. There are organizational demands and employees fulfill these tasks, because they have the right abilities (Cable and DeRue, 2002). Abilities include knowledge, skills, energy and time of the person to meet the demands. Demands refer to qualitative and quantitative requirements placed on the persons and could be socially constructed (e.g. role expectations) or objectively constructed (e.g. length of a workday) (Edwards, 1996). Secondly, N-S Fit focuses on the employee. The needs of the employee are fulfilled through the supplies of the organization. Needs may be financial (e.g pay level), psychological (e.g. power over others) or social (e.g. good peers) (Cable and DeRue, 2002). Supplies refer to the amount, frequency and quality of attributes that may fulfill the values of a person (Edwards, 1996). From an employee perspective, N-S fit may be the most important type of fit (Cable and DeRue, 2002).

2.5 Job crafting and person-job fit

Job crafting is different from earlier studied proactive constructs because the changes that job crafters make are mainly aimed at improving their P-J fit and work motivation (Tims et al., 2012). Job crafters actively shape and utilize their jobs to fit their preferences, needs and values (Lu et al., 2014). Job crafting means that employees will experience meaning in their jobs and this will improve their P-J fits (Bakker et al., 2012a).

(20)

16 The theory of adjustment is one of the most widely applied P-E fit theories (Tinsley, 2000). This theory could explain the relationship between job crafting and P-J fit (Dawis and Lofquist, 1984). Work adjustment means “the continuous and dynamic process by which the individual seeks to achieve and maintain correspondence with the work environment” (Rounds et al., 1987, p. 298). Employees show adjust behavior to complete work that is needed by the organization, in exchange for receiving rewards to satisfy their individual needs (Gregory et al., 2010). Therefore, an employee attempts to change his/her work environment, to achieve correspondence with his/her work environment. Thus, someone who craft their job, is expected to be associated with an increase in P-J fit. Additionally, the theory of proactive personality states that proactive individuals create favorable work conditions and work opportunities for themselves (Bakker et al., 2012). Proactive employees strive for a fit with their environment in terms of needs and abilities. These employees create a work environment in such a way that their abilities and the job demands fit with their needs and the supplies of the organization (Tims and Bakker, 2010). Job crafting behavior is related to proactive behavior, because the dimensions refer to self-initiated changes in their work environment (Tims et al., 2012). Thus, someone who craft their job, is expected to be associated with an increase in P-J fit. Finally, the conservation of resources theory states that the main human motivation is focused towards the maintenance and accumulation of resources (Hobfoll, 2001). Job crafters are engaged with having or creating the right pool of resources (Tims and Bakker, 2010). Thus, someone who craft their job, is expected to be associated with an increase in P-J fit. Based on these theories, the expectation is that the job crafting dimensions are positively related to P-J fit.

As mentioned earlier, job crafting is a topic in the literature that only appeared recently (Demerouti, 2014). This means that there is little empirical evidence for the relationship between job crafting and P-J fit. To my knowledge, there is one study of Chen et al. (2013)

(21)

17 that carried out a research into the relationship between job crafting and P-J fit. However, this study makes a distinction between individual and collective job crafting. To my knowledge, there are no studies that make a distinction between the dimensions of job crafting on P-J fit.

In summary, based on the previous literature, the expectation is that the different independent dimensions have a positive relationship with P-J fit. When employees are active in changing their work environment (increasing structural job resources, increasing social job resources, increasing challenging job demands and decreasing hindering job demands), they are more likely to experience meaning in their jobs and thus they will more likely match their jobs with their personal abilities and preferences.

Hypothesis 4a: Increasing structural job resources is positively related to person-job fit Hypothesis 4b: Increasing social job resources is positively related to person-job fit Hypothesis 4c: Increasing challenging job demands is positively related to person-job fit Hypothesis 4d: Decreasing hindering job demands is positively related to person-job fit

2.6 Person-job fit and well-being

P-J fit is important for employee well-being (e.g. Bakker et al, 2012; Kristof-Brown et al., 2005; Lascbinger et al., 2006; Lauver and Kristof-Brown, 2001; Warr and Inceoglu, 2012). A high level of fit typically results in positive outcomes for the employee (Edwards, 1991; Kristof, 1996).

The psychological need fulfillment theory explains the relationship between P-J fit and job satisfaction. This theory is the most common theory researches have used to conceptualize and operationalize complementary fit (Edwards, 1991), and thus P-J fit. This theory supports “how people’s attitudes are affected by the fit between their desires and the supplies in the work environment available to meet those desires” (Cable and Edwards, 2004, p.822). The

(22)

18 theory of psychological need fulfillment indicates that people become more satisfied when the supplies (psychological needs) provided by the environment, increase to what the person desires (Cable and Edwards, 2004). Thus, someone who has a high level of P-J fit, is expected to be associated with an increase in job satisfaction. In addition, the theory of adjustment argues that job satisfaction represents an employee’s subjective evaluation of the degree to which the requirements of that individual are met by the work environment (Cable and DeRue, 2002), which in itself corresponds with P-J fit. It is the relationship between the job and a person instead of the person and a job separately, that determines the degree of job satisfaction (Locke, 1969). Job satisfaction is a function of the fit between the reinforcer pattern of the work environment and the individual’s needs. The theory relates this ‘need-reinforcer fit’ to job satisfaction. Specifically, an increase in ‘need-‘need-reinforcer fit’ is expected to be associated with an increase in job satisfaction and vice-versa (Rounds et al., 1987). Thus, someone who has a high level of P-J fit, increases his/her ‘need-reinforcer fit’ and is expected to be associated with an increase in job satisfaction. Based on these theories, the expectation is that P-J fit and job satisfaction are positively related.

Furthermore, the self-concept based theory explains the relationship between P-J fit and work engagement. This theory links self-engagement with work engagement. Employees seeing their work as consistent with their personal values will be more engaged (Macey and Schneider, 2008). Thus, someone who has a high level of P-J fit, is expected to be associated with an increase in work engagement. In addition, the conservation of resources theory states that the main human motivation is focused towards the maintenance and accumulation of the right pool of resources (Hobfoll, 2001). A better fit between the work environment and the person, making him/her more engaged with his/her job. Based on these theories, the expectation is that P-J fit and work environment are positively related.

(23)

19 Next, the theory of reasoned action explains the relationship between P-J fit and stress. This theory states that “attitudes toward a given object will generally result in behaviors that are consistent with those attitudes” (Kristof-Brown et al., 2005, p.289). Thus, if the needs of employees are being met at work, their attitudes will be positive and may reduce strain (Kristof-Brown et al., 2005). Furthermore, the P-E fit approach to stress, a widely accepted theory among stress researchers, regards stress as a lack of a match between characteristics of the person and the environment (Edwards and Cooper, 1990; Hobfoll, 2001; Kristof-Brown et al., 2005). Stress is a misfit between an employee’s abilities and demands of the job and a misfit of an employee’s needs supplied by the job environment (Furnham and Schaeffer, 1984; Schuler,1980). Therefore, being in a misfit situation is stressful (Schuler, 1980). Based on these theories, the expectation is that P-J fit and stress are related.

Several studies have already examined the relationship between P-J fit and well-being outcomes. Previous studies found a positive relationship between P-J fit and job satisfaction (e.g. Cable and Judge, 1996; Lauver and Kristof-Brown, 2001; Warr and Inceoglu, 2012; Yoo and Hyun, 2003). Whereas, the study of Warr and Inceoglu (2012) argued that P-J fit and job engagement are negatively related, most previous studies have shown a positive relationship between P-J fit and work engagement (e.g. Bakker et al., 2012; Chen et al., 2013; Lascbinger et al., 2006; Maslach and Leiter, 2008; Schuck et al., 2011). Finally, poorer P-J fit is associated with greater strain (Kristof-Brown et al., 2005).

In summary, based on the previous literature, the expectation is that when employees characteristics match with their job characteristics, the level of employees’ job satisfaction and work engagement are likely to increase. Additionally, based on the theories about misfit, the expectation is that P-J fit is related to stress.

(24)

20 Hypothesis 5a: Person-job fit is positively related to job satisfaction

Hypothesis 5b: Person-job fit is positively related to work engagement Hypothesis 5c: Person-job fit is related to stress

2.7 The mediating role of person-job fit

P-J fit may play an important role in the relation between the job crafting dimensions and well-being outcomes. Through the process of job crafting, employees “act upon the job to create a better fit” (Wrzesniewski and Dutton, 2001, p. 118). It is proposed that a better fit is positively related to well-being and related to stress.

The conservation of resources theory explains the mediating role of P-J fit. This theory argues that that the main human motivation is focused towards the maintenance and accumulation of resources (Hobfoll, 2001). Maximizing social job resources and task resources will have a positive effect on well-being. The reason is that larger pools of resources make it easier for the individual to protect himself/herself from resource reduction (Hofboll, 1989). Preventing resource reduction also means less stress. Stress occurs when resources are lost or threatened (Hobfoll and Freedy, 1993). Additionally, it makes it easier for the individual to deal with the demands of the environment (Hofboll, 1989). Job resources could buffer the negative effects of job demands (Tims et al., 2012). Well-being is in this case the result of having or creating the right pool of resources (Tims and Bakker, 2010).

Based on this theory, the expectation is that P-J fit mediates the relationship between the job crafting dimensions and (a) job satisfaction, (b) work engagement and (c) stress.

(25)

21 Hypothesis 6a: P-J fit mediates the relationship between increasing structural job resources and job satisfaction

Hypothesis 6b: P-J fit mediates the relationship between increasing social job resources and job satisfaction

Hypothesis 6c: P-J fit mediates the relationship between increasing challenging job demands and job satisfaction

Hypothesis 6d: P-J fit mediates the relationship between decreasing hindering job demands and job satisfaction

Hypothesis 7a: P-J fit mediates the relationship between increasing structural job resources and work engagement

Hypothesis 7b: P-J fit mediates the relationship between increasing social job resources and work engagement

Hypothesis 7c: P-J fit mediates the relationship between increasing challenging job demands and work engagement

Hypothesis 7d: P-J fit mediates the relationship between decreasing hindering job demands and work engagement

Hypothesis 8a: P-J fit mediates the relationship between increasing structural job resources and stress

Hypothesis 8b: P-J fit mediates the relationship between increasing social job resources and stress

Hypothesis 8c: P-J fit mediates the relationship between increasing challenging job demands and stress

Hypothesis 8d: P-J fit mediates the relationship between decreasing hindering job demands and stress

(26)

22

2.8 Research model

The previous paragraphs stated the eight hypotheses. The hypothesized relationships are illustrated in the following research model (figure 1). The first set op hypotheses presents the direct relationships. The final three hypotheses present the indirect relationships.

Figure 1: Research model

Increasing challenging job demands Increasing structural job

resources

Increasing social job resources

Decreasing hindering job demands Person-job fit Job satisfaction Work engagement Stress H1 H2 H3 H4 H6,7,8 H5

(27)

23

3. Method

This chapter starts with the characteristics of the collected sample and the procedure. The chapter continues with an analytical strategy. This paragraph explains how the hypotheses are tested. Finally, a description of the measures will be given and corresponding reliabilities are presented.

3.1 Sample and procedure

This study was conducted by a research team consisting of two students of the Master in Business Administration. The study was supervised by Dr. C. Boon (coordinator Leadership and Management track at the UVA). This study built on the data that were obtained by a research team consisting of four bachelor students at the UVA in 2013-2014, under supervision of Dr. C. Boon. Their research design was a survey designed for supervisors and employees. This team already translated the items of the survey into Dutch. Survey administration of the current research team started on September 1st 2014. The survey was closed eight weeks later on November 1st 2014.

This study used non-probability sampling methods, convenience sampling and snowball sampling. Convenience sampling was used to gain access to supervisors and employees through personal contacts. Snowball sampling was used with the help of already participating supervisors and employees, who made an effort to have more supervisors and employees participating. Employees and supervisors were invited by e-mail to complete the online survey. The survey was distributed to respondents to complete the survey on a voluntary basis. The data were collected via qualtrics.com.

For the data set the current research team was, just like the previous research team, looking for couples consisting of a supervisor and one or two employees. Additionally, in the introduction of the survey it was emphasized that for this study the average instead of a

(28)

24 specific response was important. It was done this way in order to ensure the respondents would answer as honestly as possible. Data from completed surveys were returned directly to us. Reminders were sent after three weeks to the supervisors and employees who had not yet finished the survey. The supervisor and employee surveys were linked using a code that matched these respondents. This explains why participation was not fully anonymous. On the other hand the research team promised to delete these codes after the data were imported.

The previous research team already collected data from 90 supervisors and 144 employees. The current research team expanded this data collection to a final sample size of 127 supervisors and 201 employees. This study focuses only on the employee because the concepts for this study are related to them. Data obtained for the concepts of job crafting, P-J fit, job satisfaction, work engagement and stress were used (appendix). Therefore, the employees needed to fill in the survey of questions about their work environments and their jobs. It took 20-25 minutes to complete this survey. The response rate of the data collection was between 80%-90%. This rate was very high because the respondents were personal contacts of the research team. In general, direct relations will complete the survey more often than indirect relations.

For the current study the sample contains data from employees from different Dutch organizations in the Netherlands. They are working in a wide range of job types and levels. The final sample consists of 42,9% male and 57,1% female respondents. The average age of the respondents was 33.5 years. The data consist of respondents of different educational backgrounds. In general, they were highly educated, namely 1 respondent PhD (0,5%), 30 respondents WO master’s degree (14,9%), 11 respondents WO bachelor’s degree (5,5%), 91 respondents HBO (45,3%), 35 respondents MBO (17,4%) and 27 respondents enjoyed a secondary education program (13,4%). The respondents worked on an average 6,8 years for

(29)

25 their current companies (SD =7,87) and on an average 2,7 years with the same supervisors (SD=3,12). Finally, the respondents worked on an average 33 hours per week (SD= 10,96).

3.2 Analytical strategy

To perform the statistical analyses, the data from qualtrics.com were imported into the Statistical Package for Social Sciences (SPSS) 22.0. This software was used to analyze the collected data. To clean the dataset, only the employees who skipped almost all questions were deleted from the dataset. Missing values were marked with 999. There was one counterbalanced item for job satisfaction that had to be reversed coded, namely “In general I don’t like my job”. Additionally, there were two counterbalanced items for stress that had to be reversed coded, namely “Very few stressful things happen to me at work” and “I almost never feel stressed at work”. All other variables were positively formulated, excepting the variable decreasing hinder job demands. The items of this variable were all formulated negatively. All variables relevant for this study were compared to a total by computing scale means. Furthermore, a correlation was created showing the mean, standard deviation (SD) and correlations between the variables. The internal consistency was tested through reliability tests. Scales are reliable when Cronbach’s alpha is higher than 0.7 (Nunnally and Bernstein, 1994). Finally, the hypotheses were tested through different analyses. In order to test hypotheses 1-5, a hierarchical regression analysis was used to test direct relationships between the variables. In the hierarchical regression analysis three control variables were included: gender, age and education. Furthermore, to test hypotheses 6-8 a simple mediation analysis (model 4 of Hayes) was conducted (figure 2). This means this study examines the indirect relation between job crafting and well-being, through P-J fit. Finally, when P-J fit mediates, the direct relation of the regression analysis will be compared to the direct relation of the mediation analysis to decide whether it is a full or partial mediation. Full mediation occurs

(30)

26 when the direct effect loses its significance when controlling for the mediator. Partial mediation occurs when the direct effect becomes smaller, but still significant (Baron and Kenny, 1986).

Figure 2: Mediation model (Model 4 of Hayes)

3.3 Description of measures

Existing and well-established scales were used to measure job crafting, well-being outcomes: job satisfaction, work engagement and stress, and P-J fit. Results of current study are controlled for three control variables: gender, age and education.

Job crafting. Job crafting was measured using the 21-items scale developed by Tims et al. (2012). The concept job crafting is divided into four dimensions: increasing structural job resources (Cronbach’s α=.781), increasing social job resources (Cronbach’s α=.775), increasing challenging job demands (Cronbach’s α=.753) and decreasing hindering job demands (Cronbach’s α=.821). Each dimension included five items, except the dimension decreasing hindering job demands (six items). Ratings were completed by a 5-point scale ranging from 1 (never) to 5 (very often). A high score reflected a greater degree of employees’ job crafting. A sample item for increasing structural job resources was, “I try to

(31)

27 develop my capabilities”, for decreasing hindering job demands, “I make sure that my work is emotionally less intense”, for increasing social job resources, “I ask my supervisor to coach me” and for increasing challenging job demands, “When an interesting project comes along, I offer myself proactively as project co-worker”.

Well-being. Well-being was measured using three employee well-being outcomes: job satisfaction, work engagement and stress.

Job satisfaction. Job satisfaction was measured using the three-item scale of the Michigan Organizational Assessment Questionnaire Job Satisfaction Subscale (MOAQ-JSS) developed by Cammann et al. (1979, 1983). The three-items of job satisfaction (Cronbach’s α=.844) contains: “All in all I am satisfied with my job”, “In general, I don’t like my job” and “In general, I like working here.” The second item is reversed-scored. The original version of the MOAQ-JSS, a 7-point scale was used ranging from 1 (strongly disagree) to 7 (strongly agree). A high score means a greater degree of employees’ satisfaction related to their jobs.

Work engagement. Work engagement was measured using the short nine-item version of the Utrecht Work Engagement Scale (UWES-9) developed by Schaufeli et al. (2006). This work engagement scale (Cronbach’s α=.944) reflects the three underlying dimensions: vigor, dedication and absorption. Every underlying dimension contains three items. The seven-point response scale ranged from 1 (never) to 7 (always). A high score reflected a greater degree of employees’ engagement to their jobs. A sample item for vigor was, “At my job, I feel strong and vigorous”, for dedication, “My job inspires me” and for absorption, “I feel happy when I am working intensely”.

(32)

28 Stress. Stress was measured using the four-item scale developed by Motowidlo et al. (1986). The four-item scale of stress (Cronbach’s α=.703) contains: “I feel a great deal of stress because of my job”, “Very few stressful things happen to me at work”, “My job is extremely stressful”' and “I almost never feel stressed at work.” Responses were on a 7-point scale from 1 (strongly disagree) to 7 (strongly agree). A high score means a greater degree of stress.

Person-job fit. Person-job fit was measured using the Cable and DeRue (2002) six-item scale. Both types of P-J fit (Cronbach’s α=.927), need-supplies fit and demand-abilities fit were measured with a three-item scale each. The seven-point response scale ranged from 1 (strongly disagree) to 7 (strongly agree). A high score means a greater degree of employees’ perceived P-J fit. A sample item for need-supplies fit was, “There is a good fit between what my job offers me and what I am looking for in a job” and for demand-abilities fit, “The match is very good between the demands of my job and my personal skills”.

(33)

29

4. Results

This chapter outlines the correlation matrix, including the reliabilities of the variables. In order to test the direct relationships the results of the hierarchical regression analyses will be discussed. Finally, the indirect relationships will be analyzed.

4.1 Correlation analysis

An overview of the means, standard deviations (SDs), correlations and Cronbach’s alpha coefficients are shown in table 1. The internal consistency was tested through reliability tests. Scales are reliable when Cronbach’s alpha is higher than 0.7 (Nunnally and Bernstein,1994). As can be seen from table 1, the Cronbach’s alpha of all variables meet the criterion of .70.

A number of significant correlations between job crafting and well-being implied direct relationships. Namely, all job crafting dimensions correlate significantly with job satisfaction (increasing structural job resources: r=.269, p<.01, increasing social job resources: r=.147, p<.05, increasing challenging job demands: r=.263, p<.01, decreasing hindering job demands r=-.364, p<.01). Additionally, they correlate significantly with work engagement (increasing structural job resources: r=.331, p<.01, increasing social job resources: r=.291, p<.01, increasing challenging job demands: r=.385, p<.01, decreasing hindering job demands: r=-.188, p<.05). Correlations with work engagement are in all cases more strongly correlated. The job crafting dimensions do not correlate significantly with stress. Furthermore, P-J fit correlates significantly with job satisfaction (r=.573, p<0.01) and work engagement (r=.641, p<0.01). Finally, P-J fit does not correlate significantly with stress. Correlations with both job satisfaction and work engagement appeared to be stronger for P-J fit than for the job crafting dimensions, which might indicate that the effect of these dimensions on these well-being outcomes is mediated through P-J fit (paragraph 4.3 for a formal test of mediation effect). Finally, with respect to the controls, gender and education

(34)

30 correlate significantly. Furthermore age correlates significantly to increasing social job resources, P-J fit and stress. Education correlates significantly with increasing structural job resources and increasing challenging job demands.

(35)

31

Table 1: Means, SDs, correlations and reliability coefficients

Variables M SD 1 2 3 4 5 6 7 8 9 10 11 Controls: 1. Gender 1.57 .496 2. Age 33.49 12.025 -.109 3. Education 3.03 1.350 .166* .026 Variables:

4. Increasing structural job resources 3.5750 .65259 -.040 -.012 .151* (.781) 5. Increasing social job resources 2.7668 .76315 .010 -.219** .068 .590** (.775) 6. Increasing challenging job demands 2.9539 .72821 .029 -.060 .180* .668** .569** (.753) 7. Decreasing hindering job demands 1.9119 .64291 .030 -.130 -.098 -.053 .232** -.021 (.821) 8. Person-job fit 5.0418 1.23349 .009 .236** .018 .221** .165* .233** -.211** (.927) 9. Job satisfaction 5.8209 1.01372 .086 .058 -.065 .269** .147* .263** -.364** .573** (.844) 10. Work engagement 5.2023 1.22086 .058 .027 .019 .331** .291** .385** -.188* .641** .672** (.944)

11. Stress 3.9318 1.12449 .036 .199** -.051 .093 .094 .090 .075 .064 -.086 -.014 (.703)

(36)

32

4.2 Direct effects

To test the direct relationships (hypotheses 1-5), a hierarchical regression analysis was used. The direct effects of the job crafting variables on the well-being outcomes are summarized in table 2. Followed by the direct effects of the job crafting dimensions on P-J fit that are summarized in table 3. Finally, the direct effects of P-J fit on the well-being outcomes are summarized in table 4.

Hypothesis 1 suggests that the dimensions of increasing structural job resources (H1a), increasing social job resources (H1b), increasing challenging job demands (H1c) and decreasing hindering job demands (H1d) are positively related to job satisfaction. Using job satisfaction as a dependent variable, in the first step, the control variables gender, age and education were entered. In the second step, the dimension increasing structural job resources was entered. In the third step, the dimension increasing social job resources was entered. In the fourth step, the dimension increasing challenging job demands was entered. Finally, in the fifth step, the dimension decreasing hindering job demands was entered. The results are shown in table 2. Increasing structural job resources was found to have a significant positive relationship with job satisfaction (β=.296, p<.001). Increasing social job resources was significantly positively related to job satisfaction (β=.164, p<.05). Increasing challenging job demands was found to have a significant positive relationship with job satisfaction (β =.288, p<.001). Decreasing hindering job demands was significantly negatively related to job satisfaction (β=-.376, p<.001). Therefore, hypotheses 1a-d are supported.

Hypothesis 2 claims that the dimensions of increasing structural job resources (H2a), increasing social job resources (H2b), increasing challenging job demands (H2c) and decreasing hindering job demands (H2d) are positively related to work engagement. Using work engagement as a dependent variable, the control variables gender, age and education were entered in the first model. In the second step, the dimension increasing structural job

(37)

33 resources was entered. In the third step, the dimension increasing social job resources was entered. In the fourth step, the dimension increasing challenging job demands was entered. Finally, in the fifth step, the dimension decreasing hindering job demands was entered. The results are shown in table 2. Increasing structural job resources significantly positively related to work engagement (β=.339, p < .001). Increasing social job resources was found to have a significant positive relationship with work engagement (β=.307, p<.001). Increasing challenging job demands was significantly positively related to work engagement (β=.392, p<.001). Decreasing hindering job demands was found to have a significant negative relationship with work engagement (β=-.187, p<.05). Therefore, hypotheses 2a-d are supported.

Hypothesis 3 claims that the dimensions of increasing structural job resources (H3a), increasing social job resources (H3b), increasing challenging job demands (H3c) and decreasing hindering job demands (H3d) are related to stress. Using stress as dependent variable, in the first step, the control variables gender, age and education were entered. In the second step, the dimension increasing structural job resources was entered. In the third step, the dimension increasing social job resources was entered. In the fourth step, the dimension increasing challenging job demands was entered. Finally, in the fifth step, the dimension decreasing hindering job demands was entered. The results are shown in table 2. Increasing structural job resources was found to have no significant relationship with stress (β=.109, p>.05). Increasing social job resources was not significantly related to stress (β=.146, p>.05). Increasing challenging job demands was found to have no significant relationship with stress (β=.110, p>.05). Decreasing hindering job demands was not significantly related to stress (β=.109, p>.05). Therefore, hypotheses 3a-d are not supported.

Hypothesis 4 suggests that the dimensions of increasing structural job resources (H4a), increasing social job resources (H4b), increasing challenging job demands (H4c) and

(38)

34 decreasing hindering job demands (H4d) have a positive relationship with P-J fit. Using P-J fit as dependent variable, in the first step, the control variables gender, age and education were entered. In the second step, the dimension increasing structural job resources was entered. In the third step, the dimension increasing social job resources was entered. In the fourth step, the dimension increasing challenging job demands was entered. Finally, in the fifth step, the dimension decreasing hindering job demands was entered. The results are shown in table 3. Increasing structural job resources was found to have a significant positive relationship with P-J fit (β=.247, p<.01). Increasing social job resources was significantly positively related to P-J fit (β=.245, p<.01). Increasing challenging job demands was found to be positive significant related to P-J fit (β=.262, p<.001). Decreasing hindering job demands was significantly negatively related to P-J fit (β=-.208, p<.01). Therefore, hypotheses 4a-d are supported.

Hypothesis 5 suggests that P-J fit is positively related to job satisfaction (H5a) and work engagement (H5b). In addition, P-J fit is related to stress (H5c). Using job satisfaction as dependent variable, in the first step the control variables gender, age and education were entered. In the second step, P-J fit was entered. The results are shown in table 4. P-J fit was found to have a significant positive relationship with job satisfaction (β=.599, p<.001). Using work engagement as a dependent variable, in the first step the control variables gender, age and education were entered. In the second step, P-J fit was entered. The results are shown in table 4. P-J fit was significantly positively related to work engagement (β=.669, p<.001). Using stress as a dependent variable, in the first step the control variables gender, age and education were entered. In the second step, P-J fit was entered. The results are shown in table 4. P-J fit was found to have no significant relationship with stress (β=.018, p>.05). Therefore, hypothesis 5a and 5b are supported and 5c is not supported.

(39)

35 In summary, all the dimensions of job crafting were found significantly positively related to (a) job satisfaction and (b) work engagement, and not significantly related to stress. Furthermore, all dimensions of job crafting were found to be positively significant with P-J fit. Finally, P-J fit was found to be positively significantly related to (a) job satisfaction and (b) work engagement, and not significantly related to stress.

(40)

36

Table 2: Regression analysis results of job crafting on well-being

Job satisfaction Work engagement Stress

Variable B SE Β B SE Β B SE β Step1 Gender .225 .155 .110 .159 .189 .064 .152 .170 .067 Age .007 .006 .080 .004 .008 .042 .020 .007 .213** Education -.059 .058 -.077 .015 .070 .016 -.043 .063 -.051 Step2 Gender .264 .149 .129 .213 .179 .086 .168 .170 .074 Age .007 .006 .080 .004 .007 .042 .020 .007 .213** Education -.095 .056 -.124 -.035 .067 -.038 -.058 .064 -.068

Increasing structural job resources .463 .113 .296*** .642 .136 .339*** .189 .129 .109

Step3

Gender .229 .153 .112 .168 .181 .068 .156 .169 .069

Age .009 .006 .113 .010 .007 .104 .022 .007 .242**

Education -.067 .057 -.087 -.004 .067 -.004 -.051 .063 -.060

Job crafting and well-being : the mediating role of person-job fit