Job crafting has recently received an increased attention from scholars. It represents ‘an exciting area of research’ (Oldham & Hackman, Reference Oldham and Hackman2010: 470) as it calls for a deep rethinking of mainstream job design. Job crafting is a kind of proactive behavior through which workers reshape their own jobs by changing their tasks and their relational and cognitive boundaries in order to satisfy needs of control, positive self-image and human connection (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). Tims and Bakker (Reference Tims and Bakker2010) conceptualized job crafting within the job demands–resources (JD-R) model; they showed that job crafting implies a significant re-definition of job demands and job resources, while allowing for a better alignment between job characteristics and personal skills, needs and abilities.
Job crafting has significant consequences for both individuals and organizations (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). It is not surprising that scholars are increasing their efforts to investigate both the conditions that facilitate job crafting and its effects. Our study aims at contributing to both these goals.
First, we focus on the antecedents of job crafting. In this respect, scholars found that job crafting may be influenced by several personal variables, such as need for positive self-image, work experience, need for human connection (Niessen, Weseler, & Kostova, Reference Niessen, Weseler and Kostova2016), approach temperament, avoidance temperament (Bipp & Demerouti, Reference Bipp and Demerouti2015), conscientiousness, extraversion, agreeableness, openness to experience, neuroticism (Bell & Njoli, Reference Bell and Njoli2016), self-image, perceived control, readiness for change (Lyons, Reference Lyons2008) and proactive personality (Bakker, Tims, & Derks, Reference Bakker, Tims and Derks2012). Other scholars found that job characteristics such as job enlargement (Berdicchia, Nicolli, & Masino, Reference Berdicchia, Nicolli and Masino2016), work pressure and autonomy (Petrou, Demerouti, Peeters, Schaufeli, & Hetland, Reference Petrou, Demerouti, Peeters, Schaufeli and Hetland2012) may facilitate job crafting behaviors. Finally, other studies show the relevance of leadership in influencing job crafting (Wang, Demerouti, & Bakker, Reference Wang, Demerouti and Bakker2017). Originally, several authors argued that rigid supervision may impede job crafting (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001; Berg, Grant, & Johnson, Reference Berg, Grant and Johnson2010a). More recent studies, however, show that certain leadership styles may indeed facilitate job crafting. Studies show that transformational leadership (Wang, Demerouti, & Le Blanc, Reference Wang, Demerouti and Le Blanc2017), leader–member relationship (Berdicchia, Reference Berdicchia2015) and empowering leadership (Tiago & Miguel Pereira, Reference Tiago and Miguel Pereira2016) may increase job crafting behaviors aimed at seeking challenging job demands. Similarly, Bavik, Bavik, and Tang (Reference Bavik, Bavik and Tang2017) show that servant leadership may support job crafting initiatives aimed at increasing job resources. Thus, there seems to be a need to explore in more depth how different leadership styles may influence job crafting. Our study aims at contributing in such direction by investigating the relationship between job crafting and leadership styles such as promotive control and restrictive control.
Second, we focus on job crafting outcomes. Some studies confirm the positive effects of job crafting on employees, with benefits in terms of person-job fit (Lu, Wang, Lu, Du, & Bakker, Reference Lu, Wang, Lu, Du and Bakker2014), individual well-being (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2013b), work engagement (Bakker, Tims, & Derks, Reference Bakker, Tims and Derks2012), work identity (Mattarelli & Tagliaventi, Reference Mattarelli and Tagliaventi2015) and the development of personal resources. The latter seem to be connected with job crafting in two different ways.
On the one hand, consistently with conservation of resources theory (Hobfoll, Reference Hobfoll1989) and Fredrickson’s (Reference Fredrickson2001) broaden-and-build theory, personal resources constitute an antecedent of job crafting, as they stimulate new ideas, creativity, initiative, motivation, change and proactivity. For example, some authors found that self-efficacious employees may feel more confident in being able to change some elements of their jobs, with positive effects on job crafting behaviors aimed at increasing job resources (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2014) and challenging demands (Berdicchia, Reference Berdicchia2015).
On the other hand, personal resources also represent a possible outcome of job crafting, as the extended perimeter of activities and the acquisition of new social and structural resources may facilitate the development of personal resources. In their three-wave study, Vogt, Hakanen, Brauchli, Jenny, and Bauer (Reference Vogt, Hakanen, Brauchli, Jenny and Bauer2015) found that workers with an extended domain of activities and support from colleagues and supervisors, because of their increased ability to control their work environment, feel more effective and have a stronger belief to be able to act proactively in the future, with positive effects on personal resources such as hope, optimism, efficacy and resilience (psychological capital). Similarly, van Wingerden, Bakker, and Derks (Reference van Wingerden, Bakker and Derks2016) found that job crafters feel more self-efficacious because they experience that they can proactively optimize their work situation. In our study we examine whether job crafting leads to increased self-competence. This idea is already present, in theoretical terms, in seminal contributions about job crafting. Indeed, Wrzesniewski and Dutton (Reference Wrzesniewski and Dutton2001) speculated that becoming an ‘architect’ of one’s job implies the acquisition of better awareness, mastery and knowledge about the job itself, and this may increase the perception of self-competence (as well as autonomy and relatedness). Lyons (Reference Lyons2008) also proposed that, by crafting their own job, individuals may feel more responsible, attribute a deeper meaning to their organizational contribution and feel more capable, competent or skilled. However, as we still do not have empirical evidence about this hypothesis, our study aims at contributing to the empirical investigation of it.
Finally, we also focus on the organizational outcomes of job crafting. Research shows that collaborative crafting is associated with stronger work satisfaction and commitment (Leana, Appelbaum, & Shevchuk, Reference Leana, Appelbaum and Shevchuk2009). Shusha (Reference Shusha2014) found that task and relational job crafting are predictive variables of organizational citizenship behavior. Others found that job crafting may increase job satisfaction (Ghitulescu, Reference Ghitulescu2006). In their qualitative study, Mattarelli and Tagliaventi (Reference Mattarelli and Tagliaventi2015) found that job crafting may generate changes in products and services offered and in the markets served. Also, while a limited number of studies have shown the potential negative effects of job crafting for organizations, such as counterproductive work behaviors (Demerouti, Bakker, & Halbesleben, Reference Demerouti, Bakker and Halbesleben2015), other authors emphasized the positive effects of job crafting on performance (Rudolph, Katz, Lavigne, & Zacher, Reference Rudolph, Katz, Lavigne and Zacher2017). Overall, it seems to emerge that job crafting may actually improve organizational results, but such effect may depend significantly on several individual and organizational variables (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). Hence, it is important to investigate this relationship in heterogeneous contexts (Bakker, Tims, & Derks, Reference Bakker, Tims and Derks2012). Our study contributes to this literature by exploring the relevance of job crafting on work performance in a manufacturing company.
Theoretical background and hypotheses
The job crafting JD-R model
In their seminal article, Wrzesniewski and Dutton (Reference Wrzesniewski and Dutton2001) identify three types of job crafting behaviors: changing task boundaries, changing relational boundaries and changing cognitive task boundaries. According to this model, the reason why individuals engage in job crafting behaviors is explained in terms of three basic needs (need for control, need for a positive self-image and need for human connections). Job crafting initiatives may be encouraged or inhibited by job characteristics (i.e., autonomy, interdependence), general motivational orientations (i.e., intrinsic vs. extrinsic motivations) or work orientations (i.e., seeing work as a job, a career or a calling). Job crafting generates consequences in terms of work design, social environment at work, meaning of work and work identity.
Tims and Bakker (Reference Tims and Bakker2010) integrate job crafting within the JD-R model. Job resources are physical, psychological, social or organizational job elements that allow pursuing work goals, dealing with job demands and promoting personal growth and development. Insufficient resources generate anxiety, stress, lower motivation, disengagement, withdrawal and prevent employees from goal attainment to personal development (Bakker & Demerouti, Reference Bakker and Demerouti2007). On the contrary, sufficient resources decrease depersonalization, emotional exhaustion (Fernet, Austin, Trépanier, & Dussault, Reference Fernet, Austin, Trépanier and Dussault2013) while increasing satisfaction and motivation, work engagement and the ability to develop further resources (Bakker & Demerouti, Reference Bakker and Demerouti2008; Xanthopoulou, Bakker, Demerouti, & Schaufeli, Reference Xanthopoulou, Bakker, Demerouti and Schaufeli2012).
Job demands are work features that require sustained physical, cognitive or emotional effort. A distinction between ‘challenging job demands’ and ‘hindering job demands’ has been proposed (LePine, Podsakoff, & Lepine, Reference LePine, Podsakoff and Lepine2005). Hindering job demands prevent employees from achieving goals by generating stress and anxiety. Challenging job demands, even when experienced as complex and difficult, provide increased mastery experiences, satisfaction, self-efficacy, work engagement excitement, passion and personal development (Bakker & Demerouti, Reference Bakker and Demerouti2007).
Within the JD-R framework, the job crafting behaviors originally described in Wrzesniewski and Dutton’s (Reference Wrzesniewski and Dutton2001) model are conceptualized as actions aimed at changing job resources and job demands. For example, when workers change their ‘relational boundaries,’ with the goal of improving the social environment and the relations with colleagues and supervisors, in the JD-R model these behaviors are interpreted as job crafting behaviors aimed at increasing their social resources (such as support, feedback, help, etc.). Similarly, when workers reshape their ‘task boundaries’ by changing the number and types of tasks they perform, in the JD-R model these behaviors are interpreted as job crafting behaviors aimed at modifying their job demands. In the JD-R model, the behaviors aimed at changing the ‘cognitive boundaries’ described by Wrzesniewski and Dutton (Reference Wrzesniewski and Dutton2001) lose somewhat their relevance, as these behaviors ‘may be more like coping with the circumstances as they are and not about actively shaping the boundaries of one’s work’ (Tims & Bakker, Reference Tims and Bakker2010: 4,5).
Also, since authors distinguish between structural job resources and social job resources (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2012), the JD-R model identify four different job crafting behaviors: increasing structural resources, increasing social resources, increasing challenging demands and decreasing hindering demands.
Structural resources concern increased responsibility and control, ability to decide about processes and problems, knowledge, competencies and skills, professional development, opportunity to learn and to deploy the full range of one’s capabilities.
Social resources, on the other hand, concern increased support from colleagues, help and guidance, positive relationship with one’s supervisor leading to more feedback and improved coaching.
Challenging job demands concern the participation to new projects and activities that provide opportunity to increase personal and professional growth, learning and mastery.
Finally, hindering demands concern those activities that workers may try to avoid because they make work mentally or emotionally too intense, or because they make work too complex, or because they involve problematic relationships with colleagues or other actors.
Promotive control and restrictive control
Scholl (Reference Scholl1999) defined restrictive control as a form of power exertion through which a subject pushes his wishes through against the interests of another, and promotive control as a process by which an actor influences another actor in line with the latter’s interests. Restrictive control is exercised through formal power, as it is legitimized by the hierarchical authority of the supervisor. Behavioral change happens through coercion and pressure to conform. On the contrary, promotive control embraces dialogue, participation to decision making, respect and exchange of information and ideas between the supervisor and his/her collaborators. As a consequence, this form of influence triggers ‘an authentic change in the target’s preferences and behaviors’ (Elias, Reference Elias2009: 370).
Promotive control and restrictive control may generate different processes and behavioral consequences, as well as different work outcomes. For example, research shows that promotive control and restrictive control have opposite effects on leader–member relationship, effective commitment and normative commitment (Elias & MacDonald, Reference Elias and MacDonald2006), innovation success (Scholl, Reference Scholl1999) and job satisfaction (Elias, Reference Elias2009). The distinction between promotive and restrictive control is different from other constructs related to leadership style. For example, the concepts of employee orientation-consideration (influence by means of trustful and respectful leadership) and production orientation-initiating structure (influence through emphasis on tasks, Northouse, Reference Northouse2007) focus on tasks completion or the quality of the supervisor–collaborator relationship. Instead, concepts of promotive and restrictive control focus on how leadership may be exercised based on either a strict adherence to the organizational power structure or an open, collaborative, agreement-seeking approach. Also, the classic distinction between transactional and transformational leadership (Bass, Reference Bass1997) does not relate closely to either promotive or restrictive control leadership styles. A more exhaustive overview of the usefulness of promotive/restrictive control distinction in relation to other conceptualizations of leadership can be found in Elias (Reference Elias2009). Overall, it seems that such distinction actually adds a unique, useful perspective on alternative leadership styles, by focusing on two broad and yet clearly different ways through which leaders may choose to exercise their influence on collaborators.
The relationship between promotive control, restrictive control and job crafting
We argue that promotive control may trigger job crafting initiatives aimed at increasing both structural and social job resources. On the one hand, participation to decision-making and open discussion may generate opportunities to increase learning and mental abilities, to develop technical skills and to acquire new competencies (Panari, Guglielmi, Simbula, & Depolo, Reference Panari, Guglielmi, Simbula and Depolo2010). Also, promotive control requires workers to pay attention and process information in order to provide their opinions and to contribute to decision-making, and that may enhance their thinking skills and problem-solving ability (Parker & Wu, Reference Parker and Wu2014). At the same time, a constructive interaction with the supervisor may create conditions for increased seeking feedback and other social resources (help, support, advice, guidance, etc.). A climate of trust may improve the personal relationships and the social bounds in the workplace (Elias & MacDonald, Reference Elias and MacDonald2006). Finally, conservation of resources theory (Hobfoll, Reference Hobfoll1989) suggests that the availability of more resources (as provided by promotive control) may create the psychological conditions that facilitate initiatives aimed at developing further resources (Hakanen, Perhoniemi, & Toppinen-Tanner, Reference Hakanen, Perhoniemi and Toppinen-Tanner2008).
On the contrary, we argue that restrictive control, by generating pathologies related to the procurement, transmission and application of information, may decrease the amount of knowledge generated and shared by employees, and their ability to seek and develop more resources. When the supervisor does not communicate anything but what is strictly necessary for the mere execution of predetermined tasks, employees ‘lack the information requisite to the pursuit of organizational goals’ (Elias & MacDonald, Reference Elias and MacDonald2006: 240) with negative consequences on self-efficacy, self-competence, self-image and also on their ability to identify ways to increase their resources (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). At the same time, initiatives aimed at seeking social support and feedback may also decrease, because the skills and knowledge necessary to identify the most useful information sources are lacking (Grant & Ashford, Reference Grant and Ashford2008).
We now examine how promotive and restrictive control may affect job crafting initiatives in relation to job demands. We first focus on increasing challenging job demands. In their seminal paper, Tims and Bakker (Reference Tims and Bakker2010) argued that when supervisors share more detailed information about their collaborators’ expected contributions, the latter may utilize such information to identify job crafting initiatives aimed at changing their demands and finding new resources. Similarly, promotive control may also create conditions allowing workers to improve their critical thinking about their work activities (Scholl, Reference Scholl1999; Reference Scholl2001), thereby increasing their ability to identify new challenging job demands. Wrzesniewski and Dutton (Reference Wrzesniewski and Dutton2001: 195) also observed that when managers exercise influence by involving workers in conversations and share ‘what they are trying to accomplish and why,’ workers ‘can use this knowledge to motivate and legitimate their own job crafting behaviors.’ Moreover, the feeling of interest created by self-determination and autonomy may increase workers’ willingness to explore new experiences and ideas, to expand the self in the process and to generate a sense of achievement which, in turn, allows them to envision even greater achievements in the future through new challenging demands as an opportunity for further personal development (Fredrickson, Reference Fredrickson2001). Restrictive control, on the contrary, may decrease the resources necessary to find new challenging job demands. It may also generate a passive style of coping (LePine, Podsakoff, & Lepine, Reference LePine, Podsakoff and Lepine2005), characterized by a tendency to avoid or decrease challenging activities (Lyubomirsky, King, & Diener, Reference Lyubomirsky, King and Diener2005). Restrictive control decreases the learning opportunities and, by consequence, the skills and knowledge that may be critical to cope with challenging demands (Morrison, Cordery, Girardi, & Payne, Reference Morrison, Cordery, Girardi and Payne2005). Similarly, the sense of engagement may also be decreased (Schaufeli & Salanova, Reference Schaufeli and Salanova2010), which also seems to decrease the ability and the willingness to face new challenging activities (Hakanen, Perhoniemi, & Toppinen-Tanner, Reference Hakanen, Perhoniemi and Toppinen-Tanner2008). Finally, restrictive control may also have negative effects on workers’ motivation to seek new challenging job demands because of the decreased possibility to express preferences and ideas, which in turn may decrease self-efficacy and self-competence (Bandura, Reference Bandura1977).
We now turn our attention to the ways promotive and restrictive control that may affect job crafting initiatives in relation to hindering job demands. Decreasing job demands may be seen as socially undesirable (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2013b) and leading to sanctions (Grant & Ashford, Reference Grant and Ashford2008), as job demands often appear to be more like ‘givens’ and ‘predetermined’ than job resources (Hakanen, Bakker, & Schaufeli, Reference Hakanen, Bakker and Schaufeli2006). Literature shows that when supervisors encourage participation in decision-making, free expression of opinion, awareness of the organizational goals, the psychological risk associated with individual initiative is decreased, intrinsic motivation, engagement and the propensity to proactivity increases (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001, Parker, Bindl, & Strauss, Reference Parker, Bindl and Strauss2010). Thus, it seems reasonable to imply that promotive control may encourage behaviors aimed at decreasing those job demands that are considered ‘hindering’ by the individuals. On the contrary, it seems safe to assume that restrictive control, by emphasizing conformity to predetermined procedures, may induce workers not to engage in decreasing hindering job demands.
The relationship between job crafting, self-competence and work performance
Self-competence concerns the self-evaluation of skills and abilities possessed by the individual in relation to a certain activity or domain, and the perception of oneself as capable, effective and in control (Ford, Reference Ford1985). More recently, Tafarodi and Swann (Reference Tafarodi and Swann2001: 654) defined self-competence as ‘the valuative experience of oneself as a causal agent, an intentional being that can bring about desired outcomes through exercising its will.’ Literature shows that organizations may increase workers’ self-competence in different ways, including practices such as job enrichment, goal setting and participative decision-making (Williams & Lillibridge, Reference Williams and Lillibridge1992). Similarly, job crafting may also help to develop self-competence in a variety of ways. In general, job crafting enables employees to adjust their jobs to their personal skills, abilities, knowledge and preferences. People experiencing a congruence between their characteristics and key aspects of their job will likely feel more positive work-related affective states, as well as a higher perception of being in control and competent (Maslach & Leiter, Reference Maslach and Leiter1997). More specifically, we need to consider separately job crafting in terms of changing job resources and job demands.
According to the JD-R model, job resources can play not only an extrinsic motivational role (as they are instrumental for goal achievement) but also an intrinsic motivational role, because they foster employees’ learning, growth and development. By increasing their structural resources workers increase their opportunities for learning as well as their ability to utilize their capabilities and to intervene autonomously on their work activities (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2012). In turn, this should increase their belief to possess the cognitive and behavioral repertoire needed to control and cope with problems and, ultimately, this should also increase self-efficacy and self-competence (Bandura, Reference Bandura1977; Kelloway & Barling, Reference Kelloway and Barling1991; Williams & Lillibridge, Reference Williams and Lillibridge1992). Behaviors aimed at increasing social resources may also increase self-competence, because help, support, advice and feedback clearly constitute opportunities for learning, which may consolidate the belief to master problems at work with more proficiency (Bakker & Demerouti, Reference Bakker and Demerouti2007).
We also argue that job crafting behaviors such as changing job demands may positively affect self-competence. Indeed, proactivity literature shows that efforts aimed at increasing challenges at work improve self-competence (Parker, Bindl, & Strauss, Reference Parker, Bindl and Strauss2010). Even though new challenges may be perceived as difficult and stressful, they may promote mastery, personal growth, future gains and develop confidence about one’s abilities (Clegg & Spencer, Reference Clegg and Spencer2007). Also, according to job crafting literature, ‘it is expected that employees only increase their level of job demands when they have enough job resources available to deal with them’ (Tims & Bakker, Reference Tims and Bakker2010: 4). Overall, it seems that the combination of increasing resources and increasing challenging job demands may represent the attempt to create a sort of ‘active job,’ characterized by personal development, learning, mastery and competence feeling (Karasek & Theorell, Reference Karasek and Theorell1990; Petrou et al., Reference Petrou, Demerouti, Peeters, Schaufeli and Hetland2012).
The effect of behaviors aimed at decreasing hindering demands on self-competence appears to be two-fold.
Some studies show that while hindrances tend to be appraised as stressful demands that may thwart personal growth, learning and goal attainment (Tims & Bakker, Reference Tims and Bakker2010), they are nonetheless of relevance for workers’ engagement, so decreasing them may contribute to make the work environment less challenging and stimulating (Petrou et al., Reference Petrou, Demerouti, Peeters, Schaufeli and Hetland2012) and to decrease the opportunities to acquire new abilities and to strengthen the existing ones. However, when individuals find it impossible to tackle such demands, these become threatening to the self and a hindrance to achieve a sense of fulfillment. Indeed, literature shows that hindering demands may generate emotional exhaustion and a decline in one’s feelings of competence and successful achievement in one’s job (Maslach & Leiter, Reference Maslach and Leiter1997; Fernet et al., Reference Fernet, Austin, Trépanier and Dussault2013). Accordingly, when such demands are alleviated the employees’ confidence about their skills and abilities may be increased. For these reasons, consistently with the original job crafting JD-R model, we expect that self-competence may benefit from proactive behaviors aimed at decreasing hindering job demands.
We now focus on work performance. Several scholars hypothesized (Clegg & Spencer, Reference Clegg and Spencer2007; Tims & Bakker, Reference Tims and Bakker2010) and found empirically a positive effect of job crafting on performance (Bakker, Tims, & Derks, Reference Bakker, Tims and Derks2012; Petrou et al., Reference Petrou, Demerouti, Peeters, Schaufeli and Hetland2012; Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2012; Tims, Bakker, Derks, & Van Rhenen, Reference Tims, Bakker, Derks and Van Rhenen2013a). Specifically, increasing job resources may improve performance, both through increased extrinsic motivation (as a result of more adequate means for the achievement of goals) and through increased intrinsic motivation (through personal development and mastery), and by decreasing the negative effects (i.e., stress, burnout) generated by chronic job demands (Tims & Bakker, Reference Tims and Bakker2010). At the same time, by crafting job demands, workers may improve their performance as a consequence of an enrichment of their activity through more stimulating and complex tasks, which may allow them to engage in a more active style of coping and to fully utilize their skills and abilities.
Again, if we focus on hindering demands we could hypothesize two opposite effects. Although it seems reasonable that workers may improve performance by removing hindering job demands, as these generate anxiety, stress, dissatisfaction and, ultimately, obstacles to the achievement of goals (Tims & Bakker, Reference Tims and Bakker2010), some studies show that this is not always the case. The reduction of workload may protect employee well-being in stressful situations but it may also decrease the sense of challenge and work engagement (Petrou, Demerouti, & Schaufeli, Reference Petrou, Demerouti and Schaufeli2016), with negative effects on performance (Demerouti, Bakker, & Halbesleben, Reference Demerouti, Bakker and Halbesleben2015). However, consistently with what originally stipulated within the job crafting JD-R model, we expect a positive relationship between behaviors aimed at decreasing hindering job demands and work performance.
The indirect role of job crafting in mediating supervisory control, self-competence and performance
Given the theoretical premises that we described in earlier paragraphs, we expect that restrictive control and promotive control have an indirect effect on performance and self-competence through job crafting. We also hypothesize that restrictive control and promotive control, by facilitating or inhibiting participation, discussion, dialogue and feedback, may influence job crafting initiatives aimed at changing job resources and job demands, with effects that are respectively positive and negative on performance and self-competence. Here is the complete set of hypotheses:
Hypothesis 1: There is a positive effect of promotive control on self-competence via (H1a) increasing structural resources, (H1b) increasing social resources, (H1c) increasing challenging demands and (H1d) decreasing hindering demands.
Hypothesis 2: There is a positive effect of promotive control on work performance via (H2a) increasing structural resources, (H2b) increasing social resources, (H2c) increasing challenging demands and (H2d) decreasing hindering demands.
Hypothesis 3: There is a negative effect of restrictive control on self-competence via (H3a) increasing structural resources, (H3b) increasing social resources, (H3c) increasing challenging demands and (H3d) decreasing hindering demands.
Hypothesis 4: There is a negative effect of restrictive control on work performance via (H4a) increasing structural resources, (H4b) increasing social resources, (H4c) increasing challenging demands and (H4d) decreasing hindering demands.
Method
Participants and procedure
The study was conducted in an Italian manufacturing company with the voluntary participation of several employees at different levels. The human resources director was informed about the purpose and the methodology of the study, which he reported to all employees. Then, 230 questionnaires were distributed to those who chose to participate. Every employee was clearly informed about the anonymity of the entire process. All participants were asked to describe their job crafting actions, to evaluate their own perceived competence, to describe the characteristics of their supervisors’ style of control, to indicate their supervisors’ names and to put the completed form inside a closed box. After about 3 weeks, the filled forms were collected and 25 (among supervisors and managers) were asked to provide the performance data about their collaborators. All of them were also informed about the anonymity of the process. A total of 162 questionnaires were completely filled. Overall, 162 employees evaluated their supervisors (25 in total) on promotive and restrictive control, while 25 supervisors evaluated the performance of 162 employees. About 72% of respondents are workers with production duties, 15% are supervisors with supervising and administrative duties and 13% are managers. The average age is 38.9 (SD=12.72), the minimum 18 and maximum 67; 56% were male; the average job tenure is 8.6 years (SD=9.44). About 17.9% have a bachelor degree, 70.4% a high school diploma, while 11.7% have a middle school diploma.
Measures
The translation of the questionnaire from English to Italian was validated through the back translation method (Brislin, Lonner, & Thorndike, Reference Brislin, Lonner and Thorndike1973).
Control
We measured promotive control and restrictive control through the respective four items and two items of the scale developed by Scholl (Reference Scholl1999). The scale was anchored on a six-point format ranging from (1=‘not at all’ to 6=‘very often’). Promotive control was measured through four items. An example is the following: ‘when differences of opinion occurred, the process of discussion and decision-making was characterized by mutual convergence.’ Since the four-item scale featured a 0.62α, we improved the reliability by eliminating one item. Removing item number 2 improved α to 0.82. Restrictive control was measured through two items. An example is: ‘when differences of opinion occurred, the process of discussion and decision-making was characterized by pressure from “above”.’ α=0.92.
Job crafting
We measured job crafting behaviors aimed at increasing structural resources, increasing social resources, increasing challenging demands and decreasing hindering demands through the subdimensions of the job crafting scale developed by Tims, Bakker, and Derks (Reference Tims, Bakker and Derks2012). The first three dimensions (increasing structural resources, increasing social resources and increasing challenging demands) include five items each, while decreasing hindering demands utilizes six items. Examples for each variable are: ‘I try to develop my capabilities’ for increasing structural resources; ‘I ask colleagues for advice’ for increasing social resources; ‘When an interesting project comes along, I offer myself proactively as project co-worker’ for increasing challenging demands; ‘I make sure that my work is mentally less intense’ for decreasing hindering demands. The range of responses varied from 1 (never) to 5 (very often). Cronbach’s α estimates for these scales were 0.87, 0.90, 0.87 and 0.92, respectively.
Self-competence
We utilized the subscale of the self-liking/self-competence revisited scale developed by Tafarodi and Swann (Reference Tafarodi and Swann2001). The subscale includes eight items. A typical item is ‘I am highly effective at the things I do.’ At the beginning of each item, we added ‘at work.’ Items were scored on a scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s α was 0.89.
Overall performance
We measured overall performance through a scale developed by Motowidlo and Van Scotter (Reference Motowidlo and Van Scotter1994). It includes three items on a seven-point scales. The supervisor of each respondent assessed how the latter performed in relation to the standard required for the job, in relation to both others of the same rank, and others in the same work unit. α=0.90.
Control variables
We controlled for several potentially relevant variables.
First, we controlled for job tenure (number of years in the same job) because job experience may influence job crafting behaviors in a variety of ways (Berg, Wrzesniewski, & Dutton, Reference Berg, Wrzesniewski and Dutton2010b). On the one hand, long-tenured employees may be less inclined to craft their job because they are more accustomed to their tasks. On the other hand, long-tenured members may have a better understanding of available opportunities for job crafting as a result of their deeper knowledge about their job and their organizational context. We also controlled for age and gender since both may influence employees’ preference for certain work characteristics (Bipp, Reference Bipp2010) and, by consequence, their motivation for job crafting. Bipp (Reference Bipp2010) found that older workers turn more toward tasks that protect their self-concept and provide opportunities for positive events. Also, she found that both older workers and women show more interest for autonomy and feedback. Hence, it is possible that gender and age may be associated to specific job crafting behaviors.
Moreover, we controlled for education because higher education may facilitate job crafting initiatives in several ways (Tims, Bakker, & Derks, Reference Tims, Bakker and Derks2012). First, education may provide more knowledge and other resources, and also highly educated employees usually occupy roles at higher levels, which may allow for more job crafting opportunities (Bakker, Tims, & Derks, Reference Bakker, Tims and Derks2012). For the same reason we also controlled for formal position. We distinguished between three formal levels: a lower level, which includes workers with production duties; an intermediate level, including direct supervisors with administrative duties, and a managerial level. At different ranks, employees may enjoy different degrees of autonomy or freedom to act as job crafters (Berg, Wrzesniewski, & Dutton, Reference Berg, Wrzesniewski and Dutton2010b).
Data analysis and results
In Table 1 means, SDs and intercorrelations of all variables are reported.
Descriptive statistics and inter-correlations of the variables
Note. n=162. *p<.05, **p<.01, ***p<.001, Cronbach’s α is listed in parentheses on the diagonal. Gender: male=1; female=0. Education: 1=middle school diploma or less; 2=high school diploma; 3=bachelor degree or more. Position: 1=production staff/ production duties; 2=direct supervisors/administrative duties; 3=managers/managerial duties.
Before testing our hypotheses, we conducted confirmatory factor analyses (we utilized AMOS 21.0) in order to verify the structural validity of the individual scales. We compared the eight-factor model with a six-factor model (in which promotive control, restrictive control, self-competence and overall performance load on their respective factors, increasing structural resources and increasing social resources load on one factor, increasing challenging demands and decreasing hindering demands load on another factor), with a five-factor model (in which all items related to job crafting loaded on a single latent variable) and with a single-factor model (where we constrained all observed items to load on a single latent variable). The eight-factor model exhibits an acceptable fit [χ2=830.37 (df=602), comparative fit index (CFI)=0.94, incremental fit index (IFI)=0.94, Tucker Lewis index (TLI)=0.93, root mean square error of approximation (RMSEA)=0.05], whereas the six-factor model [χ2=1802.83 (df=615), CFI=0.67, IFI=0.67, TLI=0.64, RMSEA=0.11], the five-factor model [χ2=1985.31 (df=620), CFI=0.62, IFI=0.62, TLI=0.59, RMSEA=0.12] and the single-factor model [χ2=3116.22 (df=631), CFI=0.30, IFI=0.31, TLI=0.27, RMSEA=0.16] show a poorer fit of the data.
Owing to the nested composition of our sample (employees are nested within supervisors), we tested for possible statistical dependence in our data. If the nonindependence phenomenon is detected, a multilevel analysis is required (Hox, Reference Hox2002). To assess this possibility, we calculated ICC(1)s for all variables included in the study (Bliese, Reference Bliese2000). According to Hox (Reference Hox2002), in general, coefficients of 0.05–0.09 indicate a low effect, coefficients of 0.10–0.14 represent a moderate effect and coefficients of 0.15 above indicate a large effect. Since the ICC(1)s of all variables were <0.05, we decided to apply ordinary least squares regression analysis in our research.
Finally, Harman’s one-factor test was used to empirically address the common method variance issue. The results of the factor analysis did not indicate a single-factor structure, so common method variance does not seem a serious concern.
The total, direct and indirect effects of promotive control and restrictive control on overall performance were estimated by the SPSS version of the PROCESS macro (Hayes, Reference Hayes2013) and are reported in Tables 2 and 3.
Model A: Regression results of promotive control as independent variable (PROCESS)
Note. n=162. Standard errors in parentheses. Entries are unstandardized coefficient estimates. DHID=decreasing hindering demands; ICHD=increasing challenging demands; ISOR=increasing social resources; ISTR=increasing structural resources; OP=overall performance; PC=promotive control; SC=self-competence. *p<.05, **p<.01, ***p<.001.
Model B: Regression results of PROCESS (restrictive control as independent variable)
Note. n=162. Standard errors in parentheses. Entries are unstandardized coefficient estimates. DHID=decreasing hindering demands; ICHD=increasing challenging demands; ISOR=increasing social resources; ISTR=increasing structural resources; OP=overall performance; RC=restrictive control; SC=self-competence. *p<.05, **p<.01, ***p<.001.
We separately conducted four multiple mediation analyses. In the first two analysis, promotive control was entered as the predictor variable, self-competence as dependent variable in one analysis and overall performance in the other, and increasing structural resources, increasing social resources, increasing challenging demands, decreasing hindering demands as proposed multiple mediators. Afterwards, we repeated the analysis with restrictive control as the predictor variable. We controlled for age, gender, job tenure, education and position of participants in each analysis.
In model A, the indirect positive effect of promotive control on self-competence is significant via increasing structural resources (B=0.04 [0.01, 0.10]) and via increasing challenging demands (B=0.05 [0.02, 0.10]), but it is not significant via increasing social resources or decreasing hindering demands. The indirect positive effect of promotive control on performance is significant via increasing structural resources (B=0.10 [0.03, 0.24]), via increasing challenging demands (B=0.10 [0.03, 0.20]) and via decreasing hindering demands (B=0.04 [0.01, 0.11]). However, this latter effect results from a combination of relationships (a negative effect of promotive control on decreasing hindering demands and another negative effect of decreasing hindering demands on performance), which is the opposite, for both effects, of what we expected. The same effect is not significant via increasing social resources. Thus, H1a, H1c, H2a, H2c and H2d are supported, while H1b, H1d and H2b are not supported.
In model B, the indirect negative effect of restrictive control on self-competence is significant via increasing structural resources (B=–0.02 [–0.06, –0.01]) and via increasing challenging demands (B=–0.03 [–0.07, –0.01]), but it is not significant via increasing social resources or decreasing hindering demands. The indirect negative effect of restrictive control on performance is significant via increasing structural resources (B=–0.05 [–0.13, –0.01]) and via increasing challenging demands (B=–0.05 [–0.12, –0.01]), but it is not significant via increasing social resources and decreasing hindering demands. Thus, H3a, H3c, H4a and H4c are supported, whereas H3b, H3d, H4b and H4d are not supported.
Lastly, our results show that job crafting fully mediates the relationship between promotive control and self-competence (total effect B=0.08, p<.05; direct effect B=–0.01, p=.97) and between promotive control and performance (total effect B=0.17, p<.05; direct effect B=–0.05, p=.51) (Figure 1).
Schematic representation of the model
Also, job crafting fully mediates the relationship between restrictive control and self-competence (total effect B=–0.09, p<.01; direct effect B=–0.03, p=.33) and partially mediates the relationship between restrictive control and performance (B=–0.28, p<.001; B=–0.17, p<.01).
Discussion
In our study we found support for the idea that promotive and restrictive control create contextual conditions (resource availability and psychological environment) which may influence significantly job crafting, with relevant consequences on self-competence and performance.
Literature argues that supervisory control may inhibit job crafting (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). Our study helps to clarify such argument. Although we found that restrictive control is negatively associated to two out of four job crafting behaviors (we did not find significant results for increasing social resources and decreasing hindering demands), we found that promotive control may create a resourceful environment that increases workers’ ability to proactively craft their jobs. Also, we found that promotive control is negatively related to decreasing hindering demands, whereas restrictive control is not related to decreasing hindering demands. Overall, these results seem to be consistent with the premises of the JD-R model (Demerouti, Bakker, Nachreiner, & Schaufeli, Reference Demerouti, Bakker, Nachreiner and Schaufeli2001) and with the gain spiral argument proposed by conservation of resources theory (Hobfoll, Reference Hobfoll1989). Although it is probably true that all workers may be inclined to decrease their hindering demands (LePine, Podsakoff, & Lepine, Reference LePine, Podsakoff and Lepine2005), the availability of more resources, provided by a promotive style of control, may decrease the need to reduce effort, while at the same time persistence and resilience are increased. Similar reasons may also explain why we found no significant relation between restrictive control and decreasing hindering demands. In this case, two opposite phenomena seem to compensate each other. On the one hand, restrictive control may increase conformity to existing orders and procedures, so that workers’ motivation to avoid their hindering demands is decreased. On the other hand, restrictive control may also decrease the available resources to tackle such demands, and that may induce workers to protect themselves from stress and exhaustion sources by trying to decrease such demands. Therefore, it seems possible that the two phenomena compensate each other and lead to a neutral effect.
We also found no significant relationship between restrictive control and increasing social resources. Again, two opposite phenomena may be at play. Restrictive control may constrain and limit the concrete opportunities for workers to seek work-related social resources (such as feedback, advice, help, etc.) because such leadership style emphasizes conformity to procedures and authority. Also, in case of restrictive control, the workers’ need for social resources in terms of technical information and feedback may be less frequent, because conformity to procedures does not require employees to be involved in problem-solving. At the same time, however, workers may feel the need to compensate for such insufficient participation by looking for nontechnical or nonwork-related social interactions (moral support, etc.). The two phenomena may compensate each other and generate a neutral effect.
We expected positive relationships between decreasing hindering demands and both self-competence and performance, but we found no significant effect in the first case, and a negative one in the second case. We believe that these findings may be better understood if we consider what other authors pointed out. Decreasing hindering demands may reduce the workload, but at the same time it may also reduce the triggers or the necessity for action (Csikszentmihalyi, Reference Csikszentmihalyi1990) and make work less interesting and motivating (Petrou et al., Reference Petrou, Demerouti, Peeters, Schaufeli and Hetland2012), with negative effects on work engagement (Petrou, Demerouti, & Schaufeli, Reference Petrou, Demerouti and Schaufeli2016) and, ultimately, on performance (Demerouti, Bakker, & Halbesleben, Reference Demerouti, Bakker and Halbesleben2015) and on perceived competence (Macey, Schneider, Barbera, & Young, Reference Macey, Schneider, Barbera and Young2009).
We also found that job crafting behaviors aimed at increasing social resources does not generate significant effects on neither performance nor self-competence. In the first case, this is a result that other studies already found (Weseler & Niessen, Reference Weseler and Niessen2016). In our study, an explanation could be that performance was measured through the reports of supervisors. Literature on help seeking clarifies that seeking help and support from colleagues and supervisors may signal a lack of sufficient competence and abilities (Lee, Reference Lee1997). Hence, a supervisor could interpret behaviors aimed at increasing social resources as typical of bad performers, and that may compensate the positive effect on performance that we expected.
Literature on help seeking may also illuminate the interpretation of our result about seeking social resources on self-competence. Although obtaining social support and feedback may allow to develop new competencies related to specific tasks, it may also slow down or even inhibit the capacity to develop autonomous problem-solving skills or to master the task in question. Thus, it is possible that those workers seeking social resources may perceive themselves as not fully able to do their job by themselves, counting solely or mostly on their own abilities and skills. Accordingly, the net effect of seeking social resources on self-competence may be neutral.
Overall, an articulated framework seems to emerge. Promotive control shows positive effects on performance and self-competence because it creates, for workers, material and psychological conditions helping them to proactively increase their resources, to seek new, challenging job demands and to avoid hindering job demands. On the contrary, restrictive control has negative effects on performance and self-competence mostly because it creates conditions that do not allow workers to increase their resources and to extend their job boundaries in terms of more challenging job demands.
Practical implications
Traditional top-down job design approaches (including those proposing enlarged or enriched jobs) implicitly suggest that the value of workers’ initiative is limited at best (Lawler, Reference Lawler1969; Hackman & Oldham, Reference Hackman and Oldham1980). In other words, these approaches assume either that organizations do not face significant uncertainty or that tight control is more important than creativity and problem-solving. However, it seems obvious that today these assumptions are no longer valid for most organizations (Oldham & Hackman, Reference Oldham and Hackman2010). A viable option for organizations is to try to establish organizational and work conditions that allow employees to utilize their abilities to react more autonomously to the unexpected, to take initiatives and engage in problem-solving. Facilitating workers’ proactivity represents, for organizations, a promising way to create those conditions. However, proactivity does not necessarily lead to outcomes that are consistent with organizational goals. With higher employees’ autonomy, the problem of aligning individual and organizational goals becomes even more salient. Our main finding is that supervisory control style may represent a key element to achieve a good balance between the need for proactivity and need for consistency between individual behavior and organizational goals. Restrictive control, while increasing employees’ compliance to rules and authority, inhibits proactive job crafting behaviors aimed at increasing resources or challenging demands, with negative consequences on individual performance and self-competence. Instead, promotive control facilitates job crafting, a prominent reason being that it increases employees’ awareness and knowledge about the relation between their actions and the organizational goals. In other words, it seems that promotive control helps to align the behavior of job crafters to the organizational interests. Indeed, the fact that promotive control is also associated with better work performance may be seen as indirect evidence of the ‘aligning function’ played by such supervisory style. Hence, promotive control seems to have a double, positive effect. On the one hand, it generates the kind of organizational ‘atmosphere’ that encourages employees to be more proactive. On the other hand, it seems to increase the probability that employee’s proactivity will be exercised toward positive directions for the organization. Overall, there are significant implications for both employees and human resource managers. On the one hand, since promotive control seems to foster self-competence of employees by encouraging a proactive attitude at work, a number of positive psychological consequences that are usually associated with increased self-competence, such as intrinsic motivation, engagement and personal development, should be observed in work environments characterized by widespread promotive leadership. On the other hand, our study suggests that human resource managers may significantly benefit their organizations by selecting, training and encouraging supervisors to use a promotive style of control, especially in those work situations where proactivity may generate significant value because of the intrinsic uncertainty of activities.
Limits and future research
Our research has several limitations.
First, not all the linkages in our model were immune to the effects of common method (source) variance. However, ensuring the anonymity of all respondents and utilizing multisource data (from supervisors and employees) helps to reduce the problem. The Harman’s one-factor test was used to empirically address the common method variance issue, and the result did not indicate a single-factor structure, so common method variance is not a serious concern.
Second, we used data collected at one point in time, with the exception of performance data. The cross-sectional nature of our research makes it difficult to establish with certainty the direction of causal relations. Further studies adopting a longitudinal methodology are therefore necessary.
Third, both our small sample size and the fact that our study focused on one company imply caution in the generalizability of our results. Further studies are necessary to confirm our results in other organizations and other industries.
Also, our research design did not include moderating variables. Elias (Reference Elias2009) showed the relevance of locus of control when power dynamics are investigated, as employees with internal or external loci may react differently to social influence. Some authors argued that locus of control may influence the propensity to act proactively (Tornau & Frese, Reference Tornau and Frese2013). More generally, there are many variables, both personal (i.e., personality traits) and organizational, that may play a significant role in shaping the way supervisory control is related to job crafting. Further research about such relations is certainly needed.
Finally, in this study we utilized the JD-R model of job crafting, which is the most utilized job crafting conceptualization in current empirical studies (Rudolph, et al., Reference Rudolph, Katz, Lavigne and Zacher2017). Although this model provides a very specific articulation of different job crafting behaviors, it does not consider cognitive forms of job crafting. From a theoretical point of view, the way workers ‘see’ their job may have very significant effects, at both the individual and at the organizational level (Wrzesniewski & Dutton, Reference Wrzesniewski and Dutton2001). For example, a study by Slemp and Vella-Brodrick (Reference Slemp and Vella-Brodrick2014) shows that cognitive crafting is connected with intrinsic needs satisfaction. Hopefully, future empirical studies will extend our knowledge about cognitive and other job crafting forms.
Acknowledgments
The authors would like to thank the editor and anonymous reviewers for their constructive comments.
About the Authors
Domenico Berdicchia is Researcher (Tenure Track) in the Department of Economics and Management, University of Ferrara, Italy. He received his PhD at the University of Ferrara, Italy. His current research interests include two major areas: proactivity at work and organizational change.
Giovanni Masino is Full Professor of Organizational Behavior at the Department of Economics and Management, University of Ferrara, Italy. He is Pro-Chancellor for students’ orientation and post-lauream. His research interests concern organizational change, work organization, proactivity, job crafting, leadership.
