1
Team job crafting in a prison
How can leader behaviours create team job crafting?
University of Amsterdam
Master of Business Administration Track: Leadership and Management Thesis (final version)
Author: Annemarieke Huisman Student number: 11918845
Date: June 22th, 2018 Supervisor: C.T. Boon
2
Statement of originality
This document is written by Student Annemarieke Huisman who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Table of content 1. Acknowledgements P. 4 2. Abstract P. 5 3. Introduction P. 6 4. Theoretical Framework P. 10 – 21
4.1 Conceptualization job crafting P. 10
4.2 Conceptualization team job crafting P. 12
4.3 Leadership and team job crafting P. 14
5. Methods P. 22 – 31 5.1 Case description P. 23 5.2 Context description P. 24 5.3 Sample description P. 25 5.4 Data collection P. 27 5.5 Data analysis P. 30 6. Results P. 32 – 50
6.1 Dimensional approach to team job crafting P. 32 6.2 Conclusions regarding team job crafting P. 35 6.3 Leader behaviours explaining team job crafting P. 36 6.4 Conclusions regarding leader behaviours P. 49
7. Discussion P. 51 – 57
7.1 Discussion and future research P. 51
7.2 Strengths and limitations P. 56
7.3 Practical implications P. 57
8. Conclusion P. 58
9. References P. 59 – 62
10. Appendix P. 63 – 76
10.1 Interview protocol P. 63
10.2 Interview protocol translated to Dutch P. 67
4
1. Acknowledgements
I would like to start my thesis with expressing my gratitude to the people who enabled me to write it.
First of all, I would like to thank my supervisor for guiding me through the process of writing my thesis. Her feedback has been very helpful in shaping and completing my thesis.
Second of all, I would like to thank the researched prison for providing the opportunity to do the research within their organization. It has been interesting to observe this context.
Especially, I would like to thank the department heads for their efforts of arranging the interviews. Furthermore, I would like to thank the respondents for their willingness to participate in the interviews.
Lastly, I would like to thank my boyfriend for his support during the period of writing my thesis. I am grateful for his patience and help during this period.
5
2. Abstract
This research explores the concept of team job crafting in a prison. The researched prison works with cross-functional teams and job crafting. Job crafting within a prison is surprising since highly regulated job design would be expected. Our single embedded case study explores how leader behaviours create team job crafting within these cross-functional teams. It is important to explore how leader behaviours create team job crafting, since there is a gap in the literature regarding leadership for team job crafting. Our findings reveal that problems arise while using the dimensional structure of team job crafting provided by Tims, Bakker, Derks and Van Rhenen (2013) within our context. This indicates that explaining team job crafting by this structure may not be relevant for every context. Furthermore, our findings reveal that leader behaviours of ‘keeping a distance’ and ‘being involved’ along with
mechanisms of ‘giving responsibility’, ‘making the team think’, and ‘discussing’ explain team job crafting. These leader behaviours and mechanisms provide a more specific explanation for how leader behaviours create team job crafting than literature has done so far. Theoretical and practical implications of these findings are discussed in the discussion section.
6
3. Introduction
Nowadays, there is a rise in complexity of professional jobs due to the increasing popularity of self-managing teams, reengineering, and other organizational innovations coupled with the increased flexibility in work arrangements (Demerouti, 2014). We notice this flexibility in work arrangements within several work environments. Examples of this trend are visible in hospitals, the hairdressing industry, the engineering industry, information technology and restaurants (Wrzesniewski & Dutton, 2001). However, it is even happening in work
environments where you would least expect it. The current case study investigates a prison, where employees work in cross-functional teams which requires them to work flexible. Cross-functional teams consist of a small collection of individuals from diverse Cross-functional
specialization within the organizations (Webber, 2002).
Within this prison, a cross-functional team consist of complex security guards, penitentiary employees, one case manager and one back office employee. This team is directed by the department head. The prison encourages employees to work flexible by using these cross-functional teams. Cross-functional teams enable flexible employability of
employees in different departments. This creates the opportunity for employees to respond independently to changes in the environment. Therefore, it creates the possibility for
employees to engage in job crafting. Job crafting can be seen as proactive behaviour in which the employee initiates changes in job demands and job resources to make their own job more meaningful, engaging and satisfying (Demerouti, 2014). Cross-functional teams and job crafting within a prison is surprising, since it is assumed this work environment consist of a highly regulated and hierarchical job design due to framing of media and tv series like ‘prison break’ and ‘orange is the new black’. By studying this case, we learn more about the new developments within job design theory since we explore a relatively new topic in the field of job redesign: team job crafting.
Job design can be described as the way jobs, tasks, and roles are structured, enacted, and modified and the impact that it has on individual, group, and organizational outcomes (Grant & Parker, 2009). Usually, this process of job design involves the organization creating jobs (Tims & Bakker, 2010). On the contrary, job redesign is the process in which the
supervisor decides to change something in the job, tasks or role of the individual (Tims & Bakker, 2010). Although there is something changed in the role or task of the individual, this change still comes from the organization. The structure and content of the job are redesigned by leaders within the organization, therefore job redesign is still seen as a top down approach
7 (Oldham & Hackman, 2010). However, research has shown this ‘one-size-fits-all’ approach is not sufficient for every work environment anymore (Aust, Rugulies, Finken & Jensen, 2010). In line with the trend of flexibility in work designs, there is now attention for a new form of job redesign: job crafting. In contrast to the top down approach of job redesign, job crafting is a bottom-up redesign approach where the individual initiate changes and craft the design of their own job (Demerouti, 2014)
Why is it interesting to study job crafting? The reasoning lies in the promising
outcomes for individuals as well as for the organization. For example, on the individual level, job crafting has an important effect on employee wellbeing in terms of increased work
engagement and job satisfaction, and it decreases burnout (Tims, Bakker & Derks, 2013). Additionally, job crafting also has some beneficial effects for the organization. For example, Tims, Bakker, Derks and Van Rhenen (2013) showed that job crafting has a positive
relationship with job performance. So, engaging in job crafting results in performance. Determining that job crafting can be beneficial for both parties, it should be noted that these outcomes only relate to the individual- and organizational level. However, an organization consist of teams working together, therefore job crafting at the team level should be taken into account as well.
Job crafting at the team level is referred to as ‘team job crafting’. This can be explained as the collective process about what to craft at work and how in order to achieve shared goals (Leana, Appelbaum & Shevchuk, 2009). This changing of job characteristics is a bottom-up process, where employees themselves, not management, decide as a team which features of their job they would like to alter (Mäkikangas, Bakker & Schaufeli, 2017). Specifically, team job crafting is about the team members jointly shaping their job demands and resources (Tims et al., 2013). The difference with individual job crafting is the team now jointly changes their work environment. This means, that there needs to be interaction
between team members by extensive communication and negotiations. The team decides together which changes they make in job characteristics of the team to increase and/or decrease their job demands and job resources.
There are several reasons why it interesting to study team job crafting. First, different researches have shown beneficial outcomes of team job crafting. Tims et al. (2013) and Mäkikangas, Aunola, Seppälä and Hakanen (2016) both showed that team job crafting through work engagement results in team performance. Additionally, McClelland, Leach, Clegg and McGowan (2014) show that collaborative crafting relates positively to team efficacy, team control, and interdependence which in turn relates positively to work
8 engagement and team performance. For the team, these findings indicate that engaging in team job crafting results in increasement of their work engagement and team performance. Although these outcomes of team job crafting are promising, it is important to note the gap in the literature on how organizations can stimulate team job crafting. Therefore,
stimulating team job crafting should be further explored. Especially, since Tims et al. (2013) show that job crafting behaviour shared within the team inspires individuals to perform job crafting at the individual level as well. So, if teams engage in team job crafting, this finding indicates individuals will perform job crafting as well. More individuals performing job crafting means an increase of wellbeing within the organization (Tims, Bakker & Derks, 2013). Therefore, we need to explore how organizations can stimulate team job crafting. We have to look at what is involved in predicting team job crafting to explore which interventions of organizations stimulate team job crafting. As mentioned before, there is not much literature on this topic yet. However, Leana et al. (2009) found that discretion,
interdependence, social ties and supportive supervision relate to collaborative job crafting. Additionally, a recent study of Mäkikangas et al. (2017) reveals engaging leadership as an organizational antecedent for team job crafting. Comparing these two findings, we conclude there is role for leadership in predicting team job crafting. However, are there also other leadership styles helpful in predicting team job crafting?
We know from previous research on individual level that transformational leadership (Wang, Demerouti & Le Blanc, 2017) and servant leadership (Harju, Schaufeli & Hakanen, 2018) relate to job crafting. However, we do not know if these leadership styles also fit team job crafting. For team job crafting, there is only one research indicating a leadership style. This research by Mäkikangas et al. (2017) shows a relationship between connecting leadership and team job crafting. We conclude from these findings that there is almost no research about leadership related to team job crafting. We need more information on this topic of leadership for team job crafting in order to understand how leader behaviours create this phenomenon of team job crafting with such promising outcomes. Which mechanisms are involved in leader behaviours creating team job crafting? This is a gap in the literature that need further exploration.
Overall, we conclude there is much unknown about the topic of team job crafting. There is a gap in the literature on how to lead this way of working. Therefore, we research the question: ‘How can leader behaviours create team job crafting?’. This study contributes to existing literature in several ways. First, our study contributes to literature by exploring the concept of team job crafting. Secondly, our research contributes to literature by exploring
9 which leadership styles are helpful in explaining how leader behaivours create team job crafting and why. We add new information to literature on how to lead team job crafting and which leader behaviours and mechanisms are important in doing so. Third, the research contributes by studying an exceptional context. There no research on job crafting in a prison as far as we know. For theory, this means our research increases the generalization of the theories of team job crafting in a new context.
However, this research also has practical relevance. Our research provides
practitioners information regarding which leadership styles and behaviours are helpful in creating flexible and proactive behaviours at the team level. Managers will learn which specific behaviours and mechanisms are helpful in creating team job crafting and why. Therefore, this research provides the prison industry, but also other practitioners, insight in how to train leadership in such a way that it increases team performance and work
engagement. Using a qualitative approach by interviewing 19 participants provides us in depth information on how employees feel and approach team job crafting. Furthermore, this approach enables us to reveal information on how employees feel that leaders should behave in order to stimulate them in performing team job crafting. Using a qualitative approach is important since the topic is new in the field and needs more understanding.
10
4. Theoretical Framework
The theoretical framework is structured as follows. First, the conceptualization of job crafting is explained. Secondly, team job crafting is explained. Lastly, research regarding leadership styles that may be related to team job crafting are explained.
4.1 Conceptualization job crafting
As mentioned in the introduction, job crafting is described as an individualized, bottom-up job redesign approach that recognizes the role of individual employees as proactive agents who form their jobs and change their own job characteristics (Fried, Grant, Levi, Hadani & Slowik, 2007; Grant & Parker, 2009). Two perspectives in literature determining the perception of job crafting. First, the perspective of Wrzesniewski and Dutton (2001). Secondly, the Job-Demand Resource (JDR) model of Bakker and Demerouti (2007); Demerouti, Bakker, Nachreiner & Schaufeli (2001). We need an understanding of these perceptions to approach job crafting.
Two perspectives on job crafting
Wrzesniewski and Dutton (2001) are the first authors introducing a theoretical model for job crafting. They referred to job crafting as the physical and cognitive changes individuals make in task or relational boundaries of their work. This means crafting a job includes shaping the physical task boundaries of a job, shaping the cognitive task boundaries of a job, shaping the relational boundaries of the job, or shaping all three (Wrzesniewski & Dutton, 2001).
Changing task boundaries means altering the form or number of activities one engages in while doing the job. This means employees choose to do fewer, more or different tasks than prescribed in their formal job description (Le Blanc, Demerouti, Bakker, Fraccaroli & Sverke, 2017). This is referred to as ‘task crafting’. Changing cognitive task boundaries refers to altering how one sees the job. For example, if someone sees a job as a set of discrete parts or as an integrated whole. This is also referred to as ‘cognitive crafting’. Lastly, changing relational boundaries means exercising discretion over with who someone interacts while doing the job (Wrzesniewski & Dutton, 2001). This involves changes in quality and/or the amount of interactions with people at work (Le Blanc et al., 2017). This is also referred to as ‘relational crafting’.
According to Wrzesniewski and Dutton (2001) job crafting occurs on a daily basis. However, research of Lyons (2008) showed that employees did not report job crafting as a daily activity. Their research reveals employees reporting an average of 1,5 crafting episodes within a period of a year. An episode is defined by actions, processes and outcomes related to
11 the behaviour of making unsupervised, spontaneous changes, which is labeled as job crafting (Lyons, 2008). Defining job crafting as spontaneous changes implies it requires a relatively large amount of work within a short time period. Given the average of 1,5 crafting episodes, it is unlikely that after completing the episode, a new one is started right away. Therefore, it cannot be seen as a daily activity, but only as a daily activity within a short time period. Most time of the year, employees are not daily involved with job crafting.
In order to better capture the ‘everyday’ changes employees may pursue Petrou, Demerouti, Peeters, Schaufeli, and Hetland (2012) and Tims and Bakker (2010) framed the definition of job crafting in the JDR model of Bakker and Demerouti (2007) and Demerouti et al. (2001). The JDR model is a heuristic model that specifies how employee well-being and effectiveness may be produced by two specific sets of working conditions: job demands and job resources (Tims & Bakker, 2010). Job demands refer to those physical, psychological, social, or organizational aspects of the job that require sustained physical and/or
psychological (cognitive and emotional) effort or skills. Therefore, they are associated with certain physiological and/or psychological costs. Examples of job demands are high working pressure, an unfavorable physical environment, and emotionally demanding interactions with clients (Bakker & Demerouti, 2007). Job resources refer to those physical, psychological, social, or organizational aspects of the job functional in achieving work goals, reduce job demands and the associated physiological and psychological costs, or stimulate personal growth, learning, and development (Bakker & Demerouti, 2007). Tims and Bakker (2010) describe examples of job resources visible at different levels. First, examples at the
organizational level are payment, job security and career opportunities. Secondly, examples at the interpersonal level in terms of social interaction are supervisor/coworker support and team climate. The third level describes how work is organized, examples are role clarity and
participation in decision making. Lastly, the level of the task is described. Examples are skill variety, task identity, task significance, autonomy and performance feedback.
In the JDR model perspective, job crafting is seen as proactive behaviour specifically targeted at job characteristics (Le Blanc, et al., 2017). Job crafting is in this perspective described as changes employees may make to balance their job demands and job resources with their personal abilities and needs (Tims & Bakker, 2010). As mentioned, Wrzesniewski and Dutton (2001) describe job crafting as physical and cognitive changes individuals make in the task or relational boundaries of their work. Tims and Bakker (2010) expand the perception of task crafting and relational crafting by adding the dimension of ‘decreasing the level of job demands’. According to Tims and Bakker (2010), job crafting consist of three dimensions:
12 increasing the level of job demands, increasing the level of job resources and decreasing the level of job demands. Increasing the level of job demands refers to task crafting since both aim at creating more challenge at work by increasing the job demands. This can be done by adding tasks to the jobs, volunteering for interesting project groups, or taking over tasks from the supervisor (Tims & Bakker, 2010). This dimension is also referred to as ‘increasing challenging job demands’ or ‘seeking challenges’.
Increasing the level of job resources relates to relational crafting since both aim at the interpersonal level and social interaction with the work environment. Employees mobilize their job resources to be able to deal with job demands without too much difficulty (Tims & Bakker, 2010). This dimension is also named as ‘increasing job resources’. Later research of Tims, Bakker and Derks (2012) distinguishes this dimension in two separate dimensions. The first dimension is ‘increasing structural job resources’. Examples are seeking for resource variety, looking for opportunities to develop and maximize job autonomy. The second dimension is ‘increasing social job resources’. Examples are social support, supervisory coaching and feedback.
The dimension Tims and Bakker (2010) added to the conceptualization of
Wrzesniewski and Dutton (2001) is referred to as ‘decreasing the level of job demands’. This dimension is included since job demands can exceed the capabilities of employees. If this happens, employees can ask colleagues to help them with their task, or reduce the number of interactions with demanding customers or colleagues. In this way, the employee reaches the work goals without too much effort and staying healthy (Tims & Bakker, 2010). This
dimension is also named as ‘decreasing hindering job demands’. The difference between the dimension of increasing job resources is this hindering job demands dimension is about workload influencing the employees’ health, where job resources are about increasing the resources to extend someone’s capabilities to meet job demands (Tims & Bakker, 2010). The main difference between the two conceptualizations is the JDR perspective focuses on the behavioral component and excludes the cognitive dimension of job crafting, which results in different targets and types of crafting distinguished (Le Blanc, et al. 2017). In this paper we follow the perspective of the JDR model and define job crafting in terms of the behaviours provided by Tims et al. (2012), since these behaviours are more easily to capture on a daily basis in the interviews with the employees.
4.2 Conceptualization team job crafting
13 Therefore, we have to define and explore the concept more exhaustively since there is a lot unknown. Leana et al. (2009) were the first authors exploring the concept of job crafting on the team level. They examined the extent to which childcare workers crafted their jobs and how such crafting affected the class. Based on the definition of Wrzesniewski and Dutton (2001) collaborative job crafting is described here as the joint effort among employees in the service of changing work process and the work of a dyad or group of employees who together make “physical and cognitive changes” in the task or relational boundaries of their work (Leana et al, 2009). It is more understandable to describe it as the process by which groups of employees determine together how they can alter their work to meet their shared work goals (Leana et al., 2009).
To make the concept more tangible, Tims et al. (2013) define team job crafting as the extent to which team members combine efforts to increase structural and social job resources. Furthermore, it includes increasing challenging job demands and decreasing hindering job demands. Mäkikangas et al. (2017) reveal some examples which makes team job crafting more clear. Examples of increasing structural job demands are utilizing the capacity and know-how of every team member. Examples of increasing social job resources include requesting and providing feedback from and to other team members. An example of teams making their job demands more challenging is adding responsibilities to the team. And lastly, examples of decreasing hindering job demands include reducing the monotony of tasks or reducing an emotionally burdensome atmosphere. Team job crafting is changing the job characteristics, where employees themselves, not management, decide as a team which
features of their job they would like to alter (Mäkikangas et al., 2017). It is not just the sum of individual team members performing job crafting behaviours (Tims et al., 2013). Team job crafting means it is about the combination of individuals and how they interact together as a team and behave independent and goal-directed (Morgeson & Hofman, 1999). For example, a team may work with agreements which makes them more efficient. In this way they have more time left to add responsibilities to the team. Therefore, they increase their challenging job demands.
Despite the little amount of research regarding team job crafting, there are already some researches indicating positive outcomes resulting from team job crafting. Leana et al. (2009) showed collaborative crafting results in job satisfaction, work commitment and lower turnover intentions. However, there is more recent evidence that team job crafting results in team efficacy and interdependence (McClelland et al., 2014), but also in increased levels of work engagement and team performance (Tims et al., 2013). Even more recently, Mäkikangas
14 et al. (2016) showed shared job crafting among team members increases team performance. These positive outcomes of work engagement and performance show the importance of examining team job crafting more exhaustively and what is involved in this phenomenon. 4.3 Leadership and team job crafting
The impact of leadership behaviours on team job crafting has hitherto been largely neglected in literature (Mäkikangas et al., 2017). Before 2017, only one research connected leadership and team job crafting. Leana et al. (2009) identified the role of supportive supervision as a predictor of team job crafting. This shows leader behaviours seem to be important in
predicting team job crafting. The involvement of leader behaviours is also described in a more recent study of Berdicchia and Masino (2017). They show the Leader-Member-Exchange (LMX) reveals to be an antecedent for job crafting. The LMX explains the relationship between the leader and the follower defined by mutual trust and frequent exchanges that go beyond the content of the formal job description (Berdicchia & Masino, 2017). The leader shows behaviours of social and emotional support, providing more autonomy for the member, giving responsibilities, feedback and information, and they make the member participate in decision making. In response to these leader behaviours, the member reciprocates with increased effort (Graen & Scandura, 1987). This reciprocity effect shows the important effect leader behaviours have on employees and that it is beneficial for the organization.
Furthermore, the reciprocity effect indicates that leaders could play an important role in stimulating team job crafting.
However, it is more interesting which mechanisms are involved in explaining the relationship between leader behaviours and job crafting, since mechanisms provide a better understanding of how and why leader behaviours relate to job crafting. For example, Ashford, Blatt and VandeWalle (2003) explain supervisors influence individual feedback seeking through reducing fear of potential cost. This is important, since feedback seeking is part of job crafting. Furthermore, mechanisms are practically applicable for leaders which makes them valuable to research. However, leader behaviours are also important since research of Berdicchia and Masino (2017) reveal that a high quality LMX relationship based on mutual exchange, trust and support between a leader and a follower results in individual job crafting. This results indicates LMX could also be important for team job crafting. For instance, the leader provides the team emotional support by guiding their group sessions, or helping them with the group process of collaboration and fostering trust within the team. Additionally, the leader provides the team responsibilities, feedback and information, involves the team in
15 decision making, or encourages the team make decisions independently. This would result in the team reciprocating by performing team job crafting.
So, the presence of a leader role is clear, but what entails this role? Wang, Demerouti and Bakker (2016) propose general leader behaviours which can be helpful in creating job crafting. They propose a leader can promote job crafting by developing personal resources of employees (i.e. self- efficacy or positive emotions towards work) and designing resourceful jobs with urgent to craft (i.e. share resources that helps employees to better enact their job and create active jobs with high job demands and resources so that there is urgency to craft). Additionally, leaders can stimulate job crafting by promoting organizational identification of employees and act as a role model (Wang et al., 2016). However, these proposed leader behaviours only suggest to result in job crafting. Furthermore, they are suggested on the individual level. Nevertheless, it could be possible that these leader behaviours also apply to the team level. For example, the leader can develop personal resources of the team, like self-efficacy of the group or an innovative and cohesive team climate. This is important since Mäkikangas et al. (2017) reveal a cohesive team climate as an antecedent of team job crafting. However, these leader behaviours of Wang et al. (2016) are just suggested and not proven yet. There is enough research indicating a role for leader behaviours in creating team job crafting. However, we do not know how and why it works.
How leadership exactly relates to job crafting has since 2016 only been speculative (Wang et al., 2016). However, the topic increased in attention and evidence is showing up. First, there is evidence for a relationship between transformational leadership and job crafting behaviour on the individual level (Wang et al., 2017). Furthermore, there is a relationship assumed between empowering leadership and individual job crafting by Wang et al. (2016). However, this relationship is assumed and not tested. A proven leadership style for job crafting is servant leadership (Harju et al., 2018). However, it should be noted that these results apply to the individual level of job crafting. The only research identifying a leadership style for team job crafting is provided by Mäkikangas et al. (2017). They indicate connecting leadership as an antecedent of team job crafting. There is no further research regarding leadership for team job crafting yet.
It is shocking there is so little research regarding leadership for team job crafting. There are several researches indicating leadership as important for job crafting. However, we know little about leadership for team job crafting and which mechanisms are helpful in explaining how the two relate. Therefore, we need to use the insights of research regarding leadership for individual job crafting to explore the possible leadership styles and mechanisms
16 helpful in explaining how leader behaviours create team job crafting. That is why we discuss transformational, empowering and servant leadership in our theoretical framework. These leadership styles relate to individual job crafting in a different context. Therefore, they might be useful in explaining how leaders behaviours create team job crafting within our context. Furthermore, we discuss connecting leadership since this leadership style is proven to be related to team job crafting. Therefore, it could be helpful in explaining which mechanisms explain team job crafting.
Transformational leadership
Research on leadership related to job crafting is scarce. However, there are reasonable
arguments for predicting the relationship between job crafting and transformational leadership exists. Wang et al. (2016) speculate a relationship between transformational leadership and job crafting based on the meta-analysis of Chiaburu, Smith, Wang & Zimmerman (2014) which reveals a positive relationship between transformational leadership and employee proactive behaviour. Wang et al. (2016) indicate transformational leadership as a meaningful leadership style for job crafting since it is change-oriented. This change oriented construct is in line with Bass (1985) who defines transformational leadership as a leadership style where the leader transforms the norms and values of the subordinates and motivates them to perform beyond their own expectations.
The relationship between transformational leadership and individual job crafting is proven by Wang et al. (2017). They use six transformational leadership behaviours indicated by Podsakoff, MacKenzie, Moorman, and Fetter (1990). First, transformational leaders articulate a vision. This means inspiring followers with a vision of the future. Secondly, transformational leaders provide an appropriate model. This means leaders setting an example for followers consistent with the values the leaders espouse. Third, transformational leaders foster acceptance of group goals by promoting cooperation among employees and getting them to work together toward a common goal. Fourth, transformational leaders communicate high performance expectations. This includes getting the best out of followers. Fifth,
transformational leaders provide individualized support by respecting followers and being concerned about their personal feelings and needs. Lastly, transformational leaders provide intellectual stimulation which means challenging followers about the ways they see their work and how they perform their tasks..
Wang et al. (2017) assumed these transformational leadership behaviours would lead to employees seeking resources, seeking challenges and reducing resources. However, their
17 results only showed a significant relationship for transformational leadership and seeking resources. When they included adaptability as mediator, an indirect effect showed up between transformational leadership and seeking challenges. There was no relationship between transformational leadership and decreasing hindering demands. These findings show that we need a better understanding of how transformational leadership relates to job crafting and which mechanisms are involved in explaining this relationship. However, it is even more important to discover how transformational leadership reveals on the team level. There is no research yet explaining how transformational leadership relates to team job crafting.
Therefore, it important to explore how this leadership style could explain team job crafting. For suggesting how transformational leadership influences team job crafting, we use the leader behaviours that belong to this leadership style. Since transformational leadership is change oriented, leaders can encourage employees to change their job resources, challenging job demands and hindering job demands. Furthermore, they can inspire followers by
articulating a vision of an organization defined by completely independent teams. This could be helpful, since such a vision invites team members to engage in team job crafting. It becomes a goal the team wants to accomplish. The leader behaviour of fostering acceptance of group goals by promoting cooperation among employees can be helpful in driving the team to work together towards the common goals of team job crafting. Furthermore,
transformational leaders can provide an appropriate role model by showing job crafting behaviours themselves. This shows the team it possible to achieve the goal of team job crafting which could enhance the motivation to engage in team job crafting.
More specifically for the dimension of seeking resources, the behaviour of
communicating high performance expectations can be important. This can result in the team asking for feedback and advise from the leader or team members to improve their
performance and meet the expectations of the leader (Wang et al., 2017). Furthermore, research has shown transformational leaders are open and willing to share resources that help the team in performing their tasks (Wang, Law, Hackett, Wang & Chen, 2005). This makes it easier for a team to approach the leader and asking him for resources, like feedback or
training opportunities.
For the dimension of seeking challenges, the behaviours of communicating high performance expectations and intellectual stimulation can be important. Leaders can encourage their teams to get the best out of their team by performing team job crafting. Furthermore, they can challenge the team on their working process. They can motivate the team to go beyond their own expectations (Bass, 1985). For team job crafting, this would
18 mean the leader encourages the team to independently seek for challenges that go beyond their assigned team goals.
For the dimension of reducing hindering demands, the leader behaviours of providing individualized support toteam members and being concerned about their personal feelings and needs can be helpful. When the leader is concerned about personal feelings from team
members, it is likely that he is involved with the team and checks up on them several times. The involvement of the leader enables the possibility to notice similarities in stories from different team members regarding stress or workload. If the team is not comfortable discussing these problem with each other, the leader can help in making the first move to discuss these issues within the team. Afterwards, the team can take action to reduce their hindering demands. However, these explanations are all speculative and we should explore how transformational leadership really explains team job crafting.
Empowering leadership
The relationship between empowering leadership and job crafting has only been speculative. In their review, Wang et al. (2016) base their arguments on research regarding proactive behaviour. Research of Martin, Liao and Campbell (2013) shows empowering leadership increases proactive behaviour. This is important, since Bakker, Tims and Derks (2012) show a proactive personality results in job crafting. Furthermore, other research on job crafting reveals a difference in power between high-rank employees and low-rank employees affects the way employees craft their jobs (Berg, Wrzesniewski & Dutton, 2010). This means we have to look at empowering leadership, since it influences power of employees. Zhang and Bartol (2010) indicate leader behaviours of empowering leadership that could be involved predicting job crafting. First, empowering leaders provide job autonomy and delegate power to their followers which could stimulate job crafting. Secondly, empowering leaders enhance significance of work, foster participation in decision making, express confidence in high performance, and provide autonomy from bureaucratic constraints (Ahearne, Mathieu, & Rapp, 2005; Zhang & Bartol, 2010).
However, this research is only speculative. There is no proven relationship between empowering leadership and job crafting yet and we do not know the explanation for how and why the two relate. Therefore, this research contributes to literature by exploring how
empowering leadership could explain team job crafting. For example, we can explore how mechanisms of job autonomy and delegating power to the team increases team job crafting. Is the explanation that increasing the power of employees results in the team having more access
19 to increase their resources? Or could it be that job autonomy provides the team freedom to do the task in their own way, which makes them more efficient, therefore resulting in the team seeking for more challenges? We can also explore what kind of empowering leader
behaviours are helpful in reducing hindering job demands. Empowering leadership is about empowering the team so which leader behaviours can help the team decrease their hindering demands, but still make them feel empowered? This is something we can explore.
Engaged leadership / connecting leadership
A leadership style recently proven to be an antecedent for team job crafting is engaged leadership (Mäkikangas et al., 2017). Engaged leadership consist of three dimensions. First, engaged leaders inspire their followers. Examples involve leaders enthusing followers for their vision and plans, and making followers feel that they contribute to an important mission. Secondly, engaged leaders strengthen their followers. Examples involve leaders granting the followers freedom and responsibility, and delegating tasks to them. Third, engaged leaders connect their followers. Examples involve leaders encouraging collaboration and
interpersonal bonding of the team by promoting a high team spirit (Schaufeli, 2015). These leader behaviours promote the fulfillment of the follower’s basic psychological needs of autonomy, competence and relatedness which may result in engagement (Schaufeli, 2015). Mäkikangas et al. (2017) only focused on the connecting dimension of engaged leadership. It could be helpful to explore the other dimensions of engaged leadership as well. However, in this paper, inspiring behaviours of a leader (the first dimension of engaged leadership) is defined by transformational leadership when using the leader behaviours of Podsakoff et al. (1990). Secondly, strengthen behaviour (the second dimension of engaged leadership) is similar to providing job autonomy and delegating power as leader behaviours of empowerment leadership. Therefore, we focus in this research on the connecting dimension of engaging leadership.
According to Mäkikangas et al. (2017) connecting leadership behaviour that encourages team job crafting is the ability of the leader to foster social relationship among team members. The leaders ability to establish a good and trusting atmosphere within the team facilitates trust within the team, communication about ideas and information sharing, openness for new ways of working and it makes the team reduce job demands. So creating an open, trusting and supportive work clime is important. However, how do leaders exactly do that in a way that it results in team job crafting? The scale Mäkikangas et al. (2017) used to measure connecting leadership only consisted of 3 measures. An example item includes ‘my
20 supervisor encourages team members to cooperate’ (Schaufeli, 2015). However, answering this question with yes or no does not explain how and why it relates to team job crafting. Therefore, we need to explore how this leader behaviour explains team job crafting. Servant Leadership
The recent study of Harju et al. (2018) showed a relationship between servant leadership and job crafting. The study aimed at finding a cross-level effect of team-level servant leadership on job boredom while mediated by job crafting. However, they did not find that result. Their main finding revealed job crafting mediated the cross-level effect of team-level servant
leadership on job boredom three years later (Harju et al., 2018). The research shows a positive relationship between servant leadership and individual job crafting, which requires us to understand how servant leadership and team job crafting relate. Servant leadership
distinguishes itself from the other leadership styles by focusing on the development of the potential of followers (Van Dierendonck, Stam, Boersma, De Windt & Alkema, 2014). Servant leadership behaviours include empower and develop employees, expressing humility, authenticity, and interpersonal acceptance (Dierendonck, 2011).
Van Dierendonck and Nuijten (2011) explain these behaviours more extensively. They describe servant leadership characterized by six dimensions. The first dimension includes empowerment, which describes development of employees. The second dimension includes accountability which means leaders holding employees accountable for their performance. The third dimension includes stewardship which means leaders having a vision. The fourth dimension includes standing back which means leaders placing themselves in the background and giving credits to others. The fifth dimension includes authenticity which means leaders being open about their limitations. The last dimension includes humility which means leaders learning from other views and opinions. According to Harju et al. (2018), servant leadership is best suited for job crafting since it focuses primarily on the good of the individual instead of the organization. We have seen the existence of a significant relationship between servant leadership and job crafting. However, this is on the individual level. Is servant leadership also helpful in explaining how leadership and team job crafting relate? Based on Harju et al. (2018) one could argue that it is not, since team job crafting is on the team level and servant leadership focuses on individual development. Therefore, we explore if servant leadership could be helpful in explaining how leadership and team job crafting relate.
21 4.5 Short summary
The length of our theoretical framework makes it look like there is a lot known about leadership regarding team job crafting. However, it is important to emphasize that the research about transformational and servant leadership are the only two studies proving leadership styles for individual job crafting. Furthermore, the research about connecting leadership is the only one proving a leadership style for team job crafting. There is much speculation about leadership regarding team job crafting instead of proven relationships. Noticing this little amount of research regarding leadership for team job crafting, shows the urgency to explore how leadership explains team job crafting. All leadership styles described in the previous section are possibilities which could be helpful in explaining how leader behaviours create team job crafting and which mechanisms are helpful in doing so. Therefore, we use theories of transformational, empowering, connecting and servant leadership to
explore the leader behaviours and mechanisms explaining team job crafting. We compare leader behaviours and mechanisms arising from our research with the ones described in our theoretical framework. First of all, to explore if we can add new information to literature by explaining why; the leader behaviours from these leadership styles explain team job crafting. Secondly, to see if we can indicate a leadership style for team job crafting.
22
5. Methods
Since there is a lot unknown about the topic of team job crafting, the purpose of this study is explorative. According to Saunders, Lewis and Thornhill (2009) this is the appropriate approach, since it helps understanding what is happening and it assesses new insights about a phenomena. The study is cross sectional, since we conducted interviews about a particular phenomenon at a particular time (Saunders et al., 2009). The research is a ‘snapshot’ of how team job crafting is experienced among employees of the prison. In this research, we assess new insights about the role of leader behaviours in creating the phenomena of team job crafting. We try to answer the research question of how leader behaviours can create team job crafting. A disadvantage of this explorative research is the impossibility to establish causality since the research is not about specific relationships between variables, like explanatory research (Saunders et al., 2009). However, explorative research explains why something happens. It is an advantage that explorative research is flexible. This provides the opportunity to change the direction of the research if new insights appear (Saunders et al., 2009).
To a great extent, this research takes an inductive approach. The purpose of an inductive approach is making sense of the data collected by analyzing those data. The result of this inductive analysis is formulation of theory (Saunders et al., 2009). It is important to take an inductive approach, since it helps us understand the way human ascribe meaning to their social world (Saunders et al., 2009). This is important for our research since
understanding how team members ascribe meaning to leader behaviours may be helpful in explaining how leader behaviours stimulate them to perform team job crafting. In this way, we add new information to literature regarding leadership for team job crafting. However, the research is also partly deductive, since we use the conceptualization of team job crafting by Tims et al. (2013) which resulted in a structure for the interview questions. Furthermore, we use theories from our theoretical framework regarding leadership for individual job crafting as a conceptual lens to identify a leadership style for team job crafting. Using our theoretical framework based on scientific principles to analyze the data describes our research as partly deductive (Saunders et al,, 2009). We start our findings inductive by exploring which behaviours and mechanisms are helpful in explaining the relationship between leader behaviours and team job crafting. Afterwards, we compare the leader behaviours and mechanism arising from our findings to known literature from our theoretical framework to see if we can add new information. Although critics say that this approach is not able to explain causal-effect relationships, an important advantage of induction is that it provides a good understanding of the research context (Saunders et al., 2009). This is needed since
23 leadership on team job crafting is not researched within a prison yet. An inducive approach allows alternative explanations of what is going on (Saunders et al., 2009). Therefore, it is helpful in understanding the phenomenon of team job crafting within this context.
5.1 Case description
A case study is defined by Yin (2003) as an empirical inquiry that investigates a
contemporary phenomenon within its real-life context. In this context, the boundaries between a phenomenon and the context are not clearly evident. However, in our research we define boundaries by researching the phenomenon and dynamics of team job crafting within the case of a prison. (Eisenhardt, 1989). Within this single case, we intended to examine a number of logical sub-units: three cross-functional teams within the prison. Having more than one unit of analysis, defines our study as an embedded case study (Saunders et al., 2009). Overall, we define our study as a single embedded case study, since we only research the case of the prison and no other cases. A limitation of using a single case study is the impossibility to check the findings in another context. Therefore, we cannot generalize our findings (Yin, 2003). However, having multiple units of analysis provides us the opportunity to explore existing theory in different teams. We gain different approaches from different teams. This makes it possible to generalize our findings to the whole organization. Further research could explore our findings in another context. In that way, it is still worthwhile since our research can provide a source of new research questions (Saunders et al., 2009).
We used purposive sampling to select the prison case, since it provides particularly information about team job crafting. The prison encourages their employees to perform job crafting by using cross-functional teams. This flexible way of working is unusual for a prison, since the context consists of a highly regulated environment because of the dangerous target group. Therefore, it is important to explore how this team job crafting phenomenon works within this special context. By studying the prison case, we show the prison industry how team job crafting is possible in such a highly regulated environment. Using our case to make this point defines our case as a unique case (Saunders et al., 2009). We need to understand what happens in this unique case to make analytical generalizations to other cases. Analytical generalizations includes a reasoned judgment about the extent to which findings in one study can be used as a guide to what might occur in another situation (Kvale, 1996). This can be done by the researcher pointing out similarities and differences between situations and using a relevant theoretical framework for an interpretive understanding to support their
24 applicability of the generalizations (Daly, Willis, Small, Green, Welch, Kealy & Hughes, 2007). Taking this point of external validity into account, we took two steps. First, by using the conceptualization of team job crafting by Tims et al. (2013) as a framework for our interview questions. Secondly, by using relevant known literature of leadership regarding job crafting to analyze the answers of the participant and describe our findings. Although we cannot make statistical generalizations, a strength of this research is the opportunity to get a good understanding of what is happening in this unique case. This makes our findings useful as a guide for further research on leadership for team job crafting. Therefore, analytical generalizations are possible.
5.2 Context description
As we mentioned, the context in this case is special. It distinguishes itself from business contexts by the degree of danger and regulation needed to prevent employees and society from crimes of detainees. We use this paragraph to explain more about the researched organization and how it differs from other prison facilities within the Netherlands. The researched prison is the most modern prison facility in the Netherlands. It is opened since September 2016 and still in development. The organization works according to the vision of ‘independency’ which is defined by own responsibility. First, this vision appeals to the detainees. Using modern technology enables detainees to become independent. Within the organization, there is a digital environment where detainees fill in forms, make requests, order their groceries, plan their visits, search for information or follow an education by themselves. Some of these proceedings can be done in their cells, some in workrooms. This modern approach provides detainees the opportunity to keep up with the developments within the society and help them become independent.
However, this vision also appeals to the employees. Their work changed from closing doors to motivating and guiding detainees towards independency. However, employees should also be independent themselves. The organization encourages this by the use of cross-functional teams. The modern technology enables flexible employability of employees in different departments. This makes employees able to do their work efficiently, goal oriented and respond independent to changes in the environment. Five core values define the prison industry: safety awareness, integrity, collaboration, humanity, and responsibility. Within this industry, the researched prison adds the value of flexibility to these 5 cores values. This value explains the goal of the organization to have independent employees who take their own responsibility within their work and solve problems by themselves.
25 5.3 Sample description
In order to reduce the amount of data, we sampled a sub-group to study the population of employees within our case. We conducted 19 interviews with employees of different cross-functional teams within the prison. We intended to research three teams, in which we would interview 5 members of each team and the 3 accompanying department heads. The teams are distinguished by their performance and selected by the management of the organization. Performance within the prison is defined as to what extend employees are engaged with cross-functional working. The sample consist of a good, middle and bad performing teams.
However, we do not know which team belongs to which category. The organization did not provide us this information. Nevertheless, the distinction between these teams is purposively sampled, since having members of different performing teams in our sample gives an
representative sample of the organization. Furthermore, we used heterogeneous sampling to select the participants. Heterogeneous sampling focusses on obtaining the maximum variation in the cases selected (Saunders et al., 2009). This sampling strategy helps us compare the variation in findings within different teams. Therefore, it enables us to describe and explain the key theme of leadership on team job crafting within different teams. Although the sample is small, patterns that do emerge across the different teams are likely to be of particular
interest and represent the key theme of leadership on team job crafting (Saunders et al., 2009) It is important to emphasize the department heads choose the participants for us. This makes our sample sensitive for bias. The department heads may for example choose
participants who are likely to give answers in line with the organization or supervisor, which is defined as a participant bias (Saunders et al., 2009). This is likely, since the organization is involved in a change process. Therefore, it is understandable to assume the organization does not want to create a negative image to the outside world. Therefore, this selection method can be seen as a limitation to our research. However, we tried to overcome this participant bias by ensuring anonymity of respondents (Saunders et al., 2008). This gives the respondents the opportunity to talk freely. Nevertheless, during the interviews, respondents sometimes reacted suspicious when we asked about recording the interview. Others also referred to the recording during interview by saying ‘I cannot say this because of the recording’. The participants being aware of the recording could result in a participant bias. Furthermore, since the respondents have busy work schedules, the time-consuming requirements of the interview process sometimes resulted in an reduction of willingness to take part. This can bias how participants answer questions. For example, we noticed some of the participants lost their concentration at the end of the interview or they shortened their answers to get it over with.
26 Having the organization choose the participant for us resulted in not having the sample we intended to have. Participants were selected base on availability, therefore participants were not always part of the teams we intended to research. In a few cases, we started the interview with the information that the respondent participated in one of the three intended teams. However, during the interview we discovered this was not the case when we asked about their leader. We ended up having 6 team members of cross-functional team A, 4 team members of cross-functional team B, and 2 team members of cross-functional team C. In addition, we ended up having 4 team members from various cross-functional teams within the organization. We merge these participants to one team, where we refer to as ‘Diverse’. So, ‘Diverse’ consist of team members from different departments. The team members do not work together in one cross-functional team.
This distribution of participants has some disadvantages. First, the incompleteness of Team C and the combination of different team members in the Diverse team makes that we cannot verify stories of respondents by their team members. Therefore, for these two teams, we could not check if team job crafting is experienced by the whole team or just by one person. However, an advantage of this distribution is the possibility to verify leader
behaviours that arise from the three intended teams also revealed in other teams within the organization with different leaders by using team ‘Diverse’. So, having the distribution of these four teams provides the opportunity to discover patterns within the organization and it gives a more representative image of what leadership within the organization contains. For an overview of the participating teams and the amount of respondents per team, see table 1. For an explanation of the role distribution within the teams and what these roles entail, see table 2. For a description of the teams and in which departments they work, see table 3.
Team A Team B Team C Diverse
Department head 1 1 1 -
Case manager 1 1 1 -
Back office - - - 1
Penitentiary employee 5 2 1 1
Complex Guard - 1 - -
Complex Guard / IBT (internal assistance team)
- - - 2
27
Function Description
Department head Runs the department. Leads different teams. Functions as middle manager, these leaders are the intermediate between the board and the employees.
Case manager Makes assessments of detainees and helps the detainees with issues that play outside the prisons. For example, freezing debts and property perseveration.
Back office Responsible for the administrative file of the detainee. Collects different documents from different systems and unites them into one file for the case manager. Also described as ‘the right hand of the case manager’.
Complex guard Responsible for monitoring the security of the department , for example by checking who runs in and out of the department on video cameras. However, they can also alternate penitentiary employees on the work floor.
Penitentiary employee Running the department with detainees. This means motivating the detainee to show proactive behaviour towards reintegration. However, work also consist of checking cells, opening cell doors, guiding activities, providing medicine, etc.
IBT (internal assistance team) When a calamity happens in the prison, this team comes into play. For example, when there is fire, when there is a fight, if someone needs to go to the isolation cells, etc.
Table 2: Overview Functions Team Description
Team A This team is responsible for a special group of detainees: the top600. This is a target group with low intellectual ability that needs special treatment and guiding. They are convicted for crimes that have a big impact on the safety of citizens. This team work in the department of the house of preservation. Prisoners need to stay in this department till they receive a verdict of the judge. Team B This team functions in the Extra Care Department. This department contains of detainees with
mental illness. Therefore, they need extra care. In contrast to a regular departments with two-person cells, this department contains of one-two-person cells. This means that a prisoner has the cell for himself instead of sharing it with another prisoner.
Team C This team functions in the prison system department. This department includes detainees who have already been convicted for a longer duration. They have to serve their sentence.
Diverse This team consist of employees from different departments. Two participants function as complex guards / IBT employees. One participant functions as back office employee in the house of preservation department for the top400 (hese detainees are similar to the top600, but they are younger). One penitentiary employee works on the prison system department for the top600. Table 3: Overview teams
5.4 Data collection
We took several steps to enhance the construct validity, reliability, the external validity, and the internal validity to build rigor case study. We use a mono method qualitative data collection technique, since we only use semi-structured interviews for our data collection. However, it is important to note this influences the construct validity of the study. Construct validity is the extent to which a study investigates what it claims to investigate. This means the extent to which a procedure leads to an accurate observation of reality (Denzin & Lincoln,
28 1994). To increase the construct validity, we observed two hours on the work floor while waiting for our interview and we observed a two hour meeting. This enabled us to verify answers of participants in a real working situation. According to Saunders et al. (2009) using only semi-structured interviews is appropriate for exploratory research. Semi-structured interviews include a list of themes and questions, although researchers may omit or add some questions in particular interviews (Saunders et al., 2009). This gives the opportunity to explore team job crafting and the meaning respondents ascribe to this phenomena (Saunders et al., 2009). Therefore, it helps us gain new insights to existing literature.
Reliability in qualitative research is concerned whether alternative researchers would reveal similar information (Silverman, 2007). To make the research reliable, we took several actions. First, we tried to minimize interviewer bias. An interviewer bias shows up if
comments, tone or non-verbal behaviour of the interviewer creates bias in the way
interviewees respond to questions being asked. To reduce this bias, we practiced the interview two times to be aware of our non-verbal behaviours and comments. Practicing the interview also provided us the insight of distinguishing our questions into logical steps to gain answers we looked for. This will be explained in the next paragraph. Secondly, we tape-recorded the interviews and transcribed carefully. An averaged interview took approximately 64 minutes in length. Using this strategy enabled us a clear overview of answers from the respondents to answer our research question accurate. Third, we made an interview protocol which can be found in section 10.1 and 10.2 in the appendix. Using this protocol enhances the external validity, as we explained in section 5.1. Using a protocol helps reaching an understanding of how leader behaviours specifically explain every dimension of team job crafting.
Furthermore, the distinction in dimensions allows us to do a cross-case comparison after collecting the data since it enables the possibility to look for within-group similarities or differences coupled with intergroup similarities or differences. Therefore, it increases the likelihood of accurate and reliable theory building (Eisenhardt, 1989). The interview protocol of Berg et al. (2010) inspired us to distinguish the interview questions into the dimensions of Tims et al. (2013). Berg et al. (2010) distinguish their interview questions into different dimensions of job crafting using the conceptualization of Wrzesniewski and Dutton (2001). We divided the interview protocol in seven subjects. First, we started with some introduction questions to get to know the respondent. Afterwards, we asked about the subject of individual job crafting to start easy and make the respondent comfortable to speak. We used some example questions of Berg et al. (2010) to do this. For example, we asked ‘has your job changed since you first started? If so, how has it changed?’. Thereafter, we asked
29 about the subject of team job crafting in general. We used the same question structure of Berg et al. (2010) for job crafting, only now for the team level. For example, we asked ‘has the collaboration within your team changed since the beginning?’. However, sometimes
respondent answered within this general subject of team job crafting something in line with a specific dimensions of team job crafting. We responded to this by asked the questions related to that subject of the specific dimension. For example, if the respondent answered something in line with workload, we asked the questions from subject 5 of decreasing hindering
demands. However, we followed the interview structure in order when respondents did not provide a specific answer in line with one of the dimensions of team job crafting. Subject four, five, six and seven contain these questions about the dimensions of team job crafting. While practicing the interview, we noticed the difficulty for respondents to directly talk about team level changes. Therefore, we created logical steps in the interview questions to make it easy for respondents to answer. First, we asked about the meaning respondents ascribed to a dimension. For example, ‘what do you see as a challenging job demands within your work?’ Secondly, we asked if team members experienced the same or if there was an common perception within the team about a dimension. Third, we asked if the team made changes to increase or reduce the dimension to discover team job crafting. These steps helped the respondent to answer the questions without too much difficulty because of the logical steps guiding them to the key question: the changes the team makes together to increase or reduce the dimension. Furthermore, it provided a more accurate image of what respondents meant by the dimensions of Tims et al. (2013) and we could see the differences between individual and team perceptions. After establishing the occurrence of team job crafting, we asked about which role the leader played in the team making those changes (or not).
Additionally, we asked which specific leader behaviours influenced the team in making those changes and how this occurred. To avoid participant bias, we formulated open questions. This helped us avoid sending the participants in a direction. Furthermore, we avoid theoretical concepts since the understanding of such terms can vary from interviewees or it could be that they do not understand them et al (Saunders et al., 2009).
Using an interview protocol based on the conceptualization of Tims et al. (2013) also enhances the internal validity of the interviews. Internal validity for qualitative research is concerned whether the researcher manages to construct a plausible causal argument that is cogent enough to defend the research conclusions (Gibbert & Ruighok, 2010). This protocol brings structure in the argumentation and enables us to answer the research question
30 research if leader behaviours from our findings care comparable to leadership styles from previous studies in our theoretical framework. We enhance the internal validity by comparing empirically observed patterns with either predicted or patterns established in previous studies in different contexts (Denzin & Lincoln, 1994).
5.5 Data analysis
For analyzing the data, we used a computer-assisted qualitative data software, Nvivo 2012. This allowed us to order and structure the data. Furthermore, we used within case analysis to search for unique patterns within each team to emerge before generalizing patterns across the teams (Eisenhardt, 1989). Within each team, we started with establishing the phenomenon of team job crafting. Therefore, we categorized the data by codes derived from our theoretical framework including the dimensions of team job crafting. For these codes derived from our theoretical frame work, see table 1 in the appendix. We used open coding to name the changes related to team job crafting. Open coding is defined as providing a label to units within the data. The same label or name will be given to similar units of data (Saunders et al., 2009). To prevent researcher bias, we started with ‘in vivo’ codes. These codes are terms emerging from the data. They are based on actual terms used by participants (Saunders et al., 2009). We continued using the first in vivo code we found for a phenomenon. Especially, when we found other people describing the same phenomenon, but using a different term for it. We placed the related unit of text under the same code. During this process of open coding for specific changes of team job crafting and accompanying leader behaviours, we merged some codes. For example, we merged the codes of ‘own decisions and solutions’ into one code, since respondents use these codes together to describe their independency. Furthermore, we merged the codes of ‘discussing and asking’ into one code since, both describe similar ways to utilize the capacity of every team member. For the same reason we merged the code of ‘inviting and asking specialist’. Lastly, we merged ‘delay or not doing tasks’ into one code, since this code describes the same phenomenon of procrastinate tasks to reduce hindering demands.
However, we also split some codes. For example, we dissembled the leader behaviour codes of ‘being involved’ and ‘discussing’ since the being involved code explains why the
discussing code happens within teams.
After coding, we categorized the open codes under the dimensional codes derived from literature. After having lots of codes, we reduced the data of team job crafting to changes confirmed by at least two team members to know for sure the phenomenon happens in the team. This reduces the possibility that it is not just a perception from someone. For an
