The negative effects of proactive behavior

(1)

‘The Negative Effects of Proactive Behavior’

Name Ing. Sander (A.A.) Puttenstein

Student number 10317325

Supervisor Renske van Geffen MSc.

Due date 31 December 2014 24:00

Thesis title ‘The Negative Effects of Proactive Behavior’

(2)

ACKNOWLEDGEMENT

Foremost, I would like to express my sincere gratitude to my supervisor Renske van Geffen MSc. for the continuous support, motivation, patience, enthusiasm and knowledge during the process of writing this master thesis.

Besides my supervisor, I would like to thank my colleagues and former colleagues from AWL Techniek Harderwijk and VDL ETG Almelo. My sincere thanks also goes to my dad, Jan Puttenstein, who helped me to convince his colleagues to fill out the online survey. I would like to thank all participants of the online survey for their input and time. Last but not least, I would like to thank my family and friends for supporting me during the whole study.

(3)

ABSTRACT

The research project described in this paper is intended to examine the negative effects of proactive behavior, specifically the intention is to investigate Job Crafting and the dependence on Job Satisfaction and Counterproductive Work Behavior. The expected effects are assumed to be

moderated by Learning and Stress. An online survey was used to gather data from 104 respondents. The respondents mainly came from the high-tech and automotive industry, characterized by mainly highly educated men. The respondents were asked to complete the questionnaire with a colleague on the same hierarchical level which they had to ask and convince themselves. Furthermore, they had to judge the Job Crafting behavior of this colleague.

Although no support was found for the supposed effects, the findings definitely contribute to the context of proactive behavior. Especially the results indicate that employees were unable to

distinguish themselves from their colleagues. Furthermore, Stress slightly moderated the relation between Job Crafting and Job Satisfaction, or Job Crafting moderated the relation between Stress and Job Satisfaction. Next to that, Learning seemed to have an effect on both Job Crafting and Job Satisfaction. Evidence was found that different Control Variables (Age, Genre, Education Level, Tenure and Number of Organizations Worked for) were related to Learning, Counterproductive Work Behavior and Job Crafting.

The results indicate that there is definitely something happening in the context of proactive behavior and provide both practical and theoretical implications. However, future research is needed to provide more insight into the negative effects of proactive behavior for both practical and

theoretical development.

Keywords: Negative effects of proactive behavior, Job Crafting, Learning, Stress, Counterproductive Work Behavior.

(4)

INTRODUCTION

In a dynamic world where organizations try to gain competitive advantage, organizations are increasingly using their employees to differentiate themselves from competitors (Wright et al., 2001). Proactive behavior is seen as an essential part for organizations to remain competitive within this dynamic world (Crant, 2000; Frese and Fay, 2001). Proactive behavior is characterized by three key attributes: It is self-starting, change oriented, and future focused (Parker et al., 2010). Interestingly there is not much known about the negative effects of proactive behavior, therefore the research project described in this paper is intended to examine the negative effects of proactive behavior. The main idea of this research is based on arguments in existing literature namely, research is needed to investigate the effects of stress and friction between more and less proactive employees (Belschak et al., 2010; Bolino et al., 2010). To investigate this, Job Crafting is included as a specific form of proactive behavior. This specific form is chosen because employees who Job Craft are, according to Wrzesniewsk and Dutton (2001), employees who take a more active part in shaping their job to fit with their personal needs and desires. So therefore Job Crafting could, hypothetically, result in friction and stress between colleagues. Next to that there are arguments presented within this paper that, under several circumstances, Job Crafting is related to more or less Job Satisfaction and antisocial behavior (e.g. Counterproductive Work Behavior). To examine this, the expected effects are assumed to be moderated by Stress and Learning.

The necessity of this paper is based on the fact that until this time, proactive behaviors have mainly been characterized as positive, organizationally as well as individually desirable actions (Belschak et al. 2010). Berg et al. (2010, P.181) state that “At a time when employees are increasingly expected to find proactive ways to meet organizational objectives and contribute the best of their talents to their organizations, managers are more likely than ever to rely on the initiative employees take to job craft”. It is therefore important to provide new insights into the negative effects of proactivity because they are currently largely unknown (Belschak et al., 2010) and research is needed to investigate the effects of stress and the friction between coworkers on proactive behavior.

Hence, the main focus of this research project is not to provide support for the specific friction or stress between colleagues but is dedicated to provide a more comprehensive view into the context of proactive behavior and its possible negative effects. In order to learn and provide both practical and theoretical insights based on data and guidelines distracted from previous literature.

(6)

THEORY

When looking at proactive behavior and Job Crafting, Tims et al. (2012) argue that Job Crafting is related to proactive personality which refers to a stable tendency to take action to influence the environment (Crant, 2000). Tims et al. (2012) state that Job Crafting can be seen as a specific form of proactive behavior, it is self-starting and focuses on how employees perceive their work environment. They act in accordance with their own preferences, values and skills. According to Wrzesniewski and Dutton (2001) employees who are Job Crafting take a more active part in shaping their job to fit their personal needs and desires. Berg et al. (2010) found that the perceptions employees have of their freedom to adapt to challenges in Job Crafting does not necessarily reflect their level of formally endowed autonomy and power. Rather, their results suggest that employees at lower ranks occupy positions in which they find it relatively easier to create more opportunities to job craft, while higher-rank employees feel more constrained despite being in positions of greater formal autonomy and power. As Berg et al. (2010) already suggested, Job Crafting is more complex than previously suggested by the literature. It is therefore important to have an understanding of the different Job Crafting forms in order to relate them theoretically to other constructs.

Wrzesniewski and Dutton (2001) described three different forms of Job Crafting. First, employees may alter aspects of their jobs that are task-related. Second, employees may change aspects of their jobs that involve the relationships they have at work. These self-initiated changes may lead to a work environment that is more in line with the specific characteristics of the employee. The third form of Job Crafting is that employees may change the cognitions they have about aspects of their jobs with the aim to increase the meaning of their work. The concept of Wrzesniewski and Dutton (2001) is elaborated on and referred to by Tims et al. (2012), who argued that there are three different dimensions of Job Crafting. These dimensions are increasing job resources, increasing challenging job demands and decreasing hindering job demands. Times et al. (2012) found that an increased level of job resources leads to Job Satisfaction. In turn, Job Satisfaction refers to contentment with a job and its aspects (Fox and Spector, 2002). Crant et al. (2010) state that proactivity is associated with Job Satisfaction because proactive employees create conditions more contributory to personal success at work. Crant et al. (2010) also found a relation between proactivity and Job Satisfaction, however this was mediated by leader-member-exchange. Furthermore, Tims et al. (2012) state that, in reference to challenge demands, an under stimulating job may cause boredom that in turn may lead to job dissatisfaction. Bolino et al. (2010) and Belschak et al. (2010) indicated that there might be friction between more and less proactive employees which, as a result, could influence Job Satisfaction. Additional, Bolino et al. (2010) stated that relation between proactivity and Job Satisfaction needs to be further investigated. To sum up, the different arguments described above are combined in order

(7)

to provide new insights. Explanatory, Job Crafting of an employee is positive related to Job Satisfaction (Crant et. al, 2010) and there might be friction between more and less proactive employees (Belschak et al., 2010; Bolino et al., 2010). So it could very well be that Job Crafting of a colleague fosters Job Dissatisfaction of an employee due to the friction between coworkers, therefore the following hypothesis is investigated.

H1: Job Crafting of colleagues is negatively related to Job Satisfaction.

One could argue that the risk of friction between more and less proactive employees (Bolino et al., 2010; Belschak et al., 2010) can, as a result, end in antisocial behavior. This type of behavior that employees engage in could, according to Spector and Fox (2002), be defined as Counterproductive Work Behavior. In other words, acts to harm the organization or its members. In his article, Spector (2011) states that there are a number of constructs defined in the literature that overlap

Counterproductive Work Behavior. These constructs are aggression, deviance, revenge, retaliation, emotional abuse, bullying and mobbing. Next to that Penney and Spector (2002) include behaviors such as theft, sabotage, verbal abuse, and work slowdowns. Furthermore, Robinson and Bennett (1995) distinguish two types of workplace deviance. They describe Counterproductive Work Behaviors that are aimed towards other persons in the organization and they refer to Counterproductive Work Behaviors that are aimed towards the organization.

Interestingly, one could argue that there may be an overlap between Counterproductive Work Behaviors and Job Crafting. For example, when browsing the internet at work. It could be interpreted as a Job Crafting activity but can also be associated with Counterproductive Work Behavior. Although some Job Crafting behaviors may be Counterproductive Work Behaviors it is likely that most are not, given the results of Fox and Spector (2002) who found that a low control perceptions of an employee could result in Counterproductive Work Behavior. Given this result, and the risk of friction between more and less proactive employees (Bolino et al., 2010; Belschak et al., 2010), the relation between Job Crafting of a colleague and Counterproductive Work Behavior aimed towards colleagues is investigated. Reason for this is because an employee might engage more in Counter Productive Work Behavior when a colleague is Job Crafting as a result of friction between more and less proactive employees. In order to investigate this the following hypothesis is investigated.

H2: Job Crafting of colleagues is positively related to Counterproductive Work Behavior.

Bolino et al. (2010) argued that when looking at the resource based view of Barney (1991) there is a risk for organizations to become dependent on proactive employees, which could reduce the

(8)

value of other organizational resources like learning capacity (due to reduced investments in training). When looking at learning within an organization, Watkins and Marsick (1993) argued that the

learning organization is not a collection of individuals learning, but rather a process occurring at different levels of the organization. These levels are respectively, the individual level, the group level and the organizational level (Watkins and Marsick, 1996). The five-discipline model (Senge, 1990) refers and integrates these levels namely, mental models and personal mastery (individual level), team working (group level) and a shared vision and systems thinking (organizational level). The distinction of different levels is also found by Garvin (1993) who suggested that learning at operational, policy and strategic levels needs to be conscious, continuous and integrated.

In order to measure the extent to which an organization meets the different learning criteria, Watkins and Marsick (1996, 2003) developed the dimensions of the Learning Organization

Questionnaire (DLOQ). It consists out of seven dimensions, on the individual level these dimensions are creating continuous learning opportunity and promoting inquiry and dialogues. On the team level it consists out of encouraging collaboration and team learning. On the organization level it consists out of establishing systems to capture and share learning, empowering people towards a collective vision, connecting the organization to its environment and providing strategic leadership for organizational learning.

When looking at Learning and the assumed relations described above (H1 and H2), it would be interesting to investigate whether Learning moderates this relations. This is because one could argue that when an organization scores high on Learning, in combination with employees who are seeking for Learning within an organization, it fosters Job Satisfaction and decreases Counterproductive Work Behavior. Further, employees who are focused on Learning might accept Job Crafting activities of their colleague more than others because they are open to new orientations (Senge, 1990). Contrary to the aim of this research a positive effect in relation to proactivity is assumed, which is done because it contains important insights for the long term negative effects, namely the absence of Learning in relation to Job Dissatisfaction and higher levels of Counterproductive Work Behavior. As mentioned, there is a long term risk for organizations to become dependent on proactive employees which could reduce Learning capacity (Bolino et al., 2010). So if evidence can be found for a positive effect in combination with the presence of Learning, it could very well be that when organizations become dependent on proactive employees and invest less in Learning it would, on the long term, foster negative effects (e.g. Job Dissatisfaction and Counterproductive Work Behavior). In order to investigate this the following hypotheses are formulated.

(9)

H3: Learning positively moderates the relation between Job Crafting of colleagues and

Job Satisfaction.

H4: Learning negatively moderates the relation between Job Crafting of colleagues and

Counterproductive Work Behavior.

Tims et al. (2012) state that, based on the JD–R model, all job characteristics can be categorized into two broad classes; job demands and job resources. Job demands refer to aspects of the job that require sustained physical and/or psychological effort or skills. Job resources refer to those aspects of the job that are functional in achieving work goals. Tims et al. (2012) state that when job demands are high in combination with low levels of job resources this may lead to negative health

consequences such as stress and burnout. In their qualitative research Bolino et al. (2010) argue that when an organization expects proactive behavior form their employees this may also contribute to stress. Next to that Frese and Fay (2001) state that when employees show a high degree of personal initiative this may, in the end, lead high levels of stress. Interestingly, Belschak et al. (2010)

mentioned that most of the literature so far focused on stress as an antecedent instead of consequence of it.

When looking at stress, Sonnentag and Frese (2003) argue that stress can occur form an incongruity between an employee and the environment. They argued that there are two types of misfit, the first consist out of a misfit between demands based on the environmental and

competence related abilities. The second misfit is based on the actual needs of an employee and supplies from the environment. Further, the stress-frustration model by Fox and Spector (1999) stated that constraints prevent goal achievement which in the end result in the fact that employees experience stress. Additional, continuous exposure to stressors will accumulate and lead to job dissatisfaction.

Interestingly, Tims et al. (2012) conclude that when employees experience that their job demands and job resources are not balanced, they may use the three complementary strategies of Job Crafting to reduce the misfit. This is also found by Wrzesniewski and Dutton (2001) who argue that when a job does not meet the skills of an employee there will be motivation to change elements of the job. Furthermore, Crawford et al. (2010) found that challenging job demands were positively related to work engagement even though they can also be appraised as stressful. Finally, as

mentioned before, Bolino et al. (2010) stated that there is a risk of stress between more and less proactive employees which could result in job dissatisfaction. So when combining the fact that proactive behaviors may contribute to Stress (Bolino et al., 2010; Frese and Fay, 2001) and that there is a risk of Stress between more and less proactive employees (Bolino et al. 2010), the objective

(10)

perspective of a colleague might overcome the findings of Crawford et al. (2010) who stated that there is an overlap between Job Crafting of an employee and Stress. Next to that Fox and Spector (1999) stated that continuous exposure to stressors will accumulate and lead to job dissatisfaction. To sum up, when including the perspective of the colleague, in combination with the arguments described above, it could very well be that when employees experiences high levels of Stress in combination with Job Crafting of a colleague, it fosters Job Dissatisfaction. Therefore the following hypothesis is investigated.

H5: Stress negatively moderates the relation between Job Crafting of colleagues and Job

Satisfaction.

Miles et al. (2002) stated that, in reference to stress, employees will respond by high levels of Counterproductive Work Behavior. According to them constraints have been positively correlated with both frustration and Counterproductive Work Behavior. They argue that stressors lead to negative emotions, which result in Counterproductive Work Behavior. Finally, according to aggression-frustration model mentioned by Fox and Spector (1999), frustration in response to stressors is an important predictor of Counterproductive Work Behavior. So when combining this with the arguments described above in relation to H5 (Stress between more and less proactive employees and proactive behaviors that foster Stress), one could state that when an employee experiences high levels of Stress in combination with a colleague that is Job Crafting, it could foster Counterproductive Work Behavior more than compared to low levels of Stress. In order to

investigate this the following hypothesis is formulated.

H6: Stress positively moderates the relation between Job Crafting of colleagues and

Counterproductive Work Behavior.

(11)

Conceptual Model

Figure 1 Conceptual model

Hypothesis

H1: Job Crafting of a colleague is negatively related to Job Satisfaction.

H2: Job Crafting of a colleague is positively related to Counterproductive Work Behavior.

H3: Learning positively moderates the relation between Job Crafting of a colleague and Job

Satisfaction.

H4: Learning negatively moderates the relation between Job Crafting of a colleague and

Counterproductive Work Behavior.

H5: Stress negatively moderates the relation between Job Crafting of a colleague and Job

Satisfaction.

H6: Stress positively moderates the relation between Job Crafting of a colleague and

Counterproductive Work Behavior.

*Note: All Hypothesis which include Job Crafting will be split up into a)Job Crafting of a colleague (self-report) and b)Job Crafting of a colleague (reported by the colleague).

Job Satisfaction

Stress

Counterproductive Work Behavior Job Crafting of a

Colleague

(12)

METHOD

Sample

The target sample consists out of employees who are working in and for various industries and organizations. They should differ in terms of age, tenure, work experience and education level. They also had to have a colleague on the same hierarchical level. When looking at the sample, it consisted out of 104 sets of questionnaires that were completed in pairs of two colleagues (every respondent can be linked to his colleague and vice versa). To prevent a common method bias (Podsakoff et al., 2003), employees were asked to judge the Job Crafting behavior of their colleague who worked on the same hierarchical level. The survey was published in a wide range of industries, respectively 54 (51.9%) respondents came from the high-tech and automotive industry, 28 (26.9%) respondents came from Oil and Gas industry and 22 (21.2%) from diverse industries. Among the respondents were 81 (77.9%) men and 23 (22.1%) women. On average, they were 40.2 years old with a standard deviation of 11.1 years. The youngest respondent was 21 years old and the oldest was 63 years old. When looking at their highest educational level, 18 respondents (17.3%) had a WO diploma and 48 (46.2%) had a HBO diploma. Respectively 31 (29.8%) of the respondents had a MBO diploma and 7 (6.7%) had a secondary school diploma. They worked on average 10.7 years within the organization, with a standard deviation of 9.8 years, a minimum of 1 year and a maximum of 39 years. The total number of organizations where they had worked for in their career was on average 3.3 with a standard deviation of 1.9. The minimum was 1 and the maximum was 10 organizations.

The main difference between the target sample and the gathered data is that the sample mainly consisted out of men (77.9%) with a high Education Level (63.5% HBO or higher) who are working in the high-tech and automotive industry (78.8%). This could have an impact on the outcome and needs to be taken into account when interpreting the results.

Procedure

A hyperlink to an online questionnaire on www.Qualtrics.com was sent to different organizations and placed on different Social Media sites. The questionnaire was accompanied by a cover letter (Appendix A) and a flyer (Appendix B). Recipients were asked to complete the online questionnaire (Appendix C) with a colleague which they had to ask and convince themselves. Regarding the Job Crafting rating, both colleagues needed to agree on a six-digit code so they could be linked to each other anonymously afterwards. Employees were asked to report their own Job Crafting behavior and the Job Crafting behavior of their colleague. Different follow-up mails and updated posts were sent by different people, it took circa three months to gather all 104

questionnaires. Afterwards the respondents and the organizations are going to receive a factsheet with the main results of the research project.

(13)

Measures

The constructs were measured using different items, which were all derived from validated English scales. Due to the fact that all respondents were Dutch, all items were translated to Dutch using ‘translation-back-translation’. The items were back-translated by an independent group of five persons to ensure that there were no discrepancies. All items were averaged to form a single index of the construct. Next to that zero cases were excluded so all 104 respondents were valid. It is checked for each construct if the Cronbach alpha coefficient would improve if items were deleted, this was not the case (<0.05 improvement). The scales used will be described in the next paragraph.

Job Crafting was assessed using a twenty-one-item scale developed by Tims et al. (2012), with a Cronbach alpha of 0.80. The scale measures five items of increasing structural job resources, six items regarding decreasing hindering job demands, five items on increasing social job resources and five items regarding Increasing challenging job demands. Ratings were completed on a seven-point scale (1: Strongly disagree to 7: Strongly agree). As discussed employees needed to rate their own Job Crafting behavior and the Job Crafting behavior of their colleague. Example questions are; “I try to develop my capabilities”, “when an interesting project comes along”, “I offer myself proactively as project co-worker”, “when there is not much to do at work”, “I see it as a chance to start new projects”.

Job Satisfaction was based on a four-item scale developed by Thompson and Phua (2012), with a Cronbach alpha of 0.85. Ratings were completed on a seven-point scale (1: Strongly disagree to 7: Strongly agree). Example questions are; “I Find real enjoyment in my job”, “I like my job better than the average person”, “Most days I am enthusiastic about my job”, “I feel fairly well satisfied with my job”.

Counterproductive Work Behavior (CWB) was assessed by using the CWB-C scale developed by Fox and Spector (2003). The CWB-C scale measures a total of 33 items which can be divided into five dimensions. As discussed, this research project will only focus on CWBP which is in line with the dimension abuse from the CWB-C scale (Robinson and Bennett, 1995), it targets the individuals within the organization (CWBP). Hence, abuse will be assessed using a fifteen-item scale, with a Cronbach alpha of 0.74. The respondents indicated the frequency in which they engage in specific behaviors on a seven-point scale (1: Never to 7: Every day). Example questions are;

“Insulted or made fun of someone at work”, “Blamed someone at work for error you made”, “Looked at someone at work’s private mail/property without permission”.

Learning will be assessed using dimensions of the Learning Organization Questionnaire (DLOQ) developed by Watkins and Marsick (1996), 17 items were used with a Cronbach alpha of 0.91. The scale measures five items on creating continuous learning opportunities, six items on promoting

(14)

inquiry and dialogue and six items on encouraging collaboration and team learning. A seven-point Likert scale was used to allow respondents to rate the degree of presence of the learning

organization environment, which is ranging from strongly agree as the highest and strongly disagree as the lowest degree. Example questions are; “In my organization, people openly discuss mistakes in order to learn from them”, “In my organization, teams/groups have the freedom to adapt their goals as needed”, “In my organization, people are rewarded for learning”.

Thirteen items in the questionnaire were on Stress. The scale used is referred to as the Job Stress Questionnaire (JSQ) developed by Caplan et al. (1975), with a Cronbach alpha of 0.71. A perceptual self-report measure of job stress is operationalized by four dimensions; work overload (five items), role ambiguity (three items), under-utilization of skills (three items) and role conflict (two items). The JSQ is presented in a seven-point Likert (1: Never to 7: Always). Example questions are; “How often does your job require you to work very fast or hard?”, “How often are you given a change to do the things you do best?”, “How often are you unclear on what your responsibilities are?”.

Last, there are six control variables included in the last section of the questionnaire to control for spurious relationships between the variables; Age, Gender, Education Level, Tenure, Number of organizations worked for and Liking.

Employees’ age has proven to have an influence on work behaviors, as younger employees are more likely to behave antisocial (Berkowitz, 1993). When looking at gender, men are thought to express more counterproductive behaviors (Martinko et al., 2002). Furthermore, employees with a higher educational background are believed to engage less in counterproductive work behaviors (Ng & Feldman, 2009). Tenure and ‘Number of Organizations Worked for’ is controlled for because one could argue that proactive people seek more for new opportunities. Tims et al. (2012) state that proactive people initiate either by acting in advance of a future situation and/or by taking control and causing change. Last, because employees had to complete the questionnaire together with a colleague, Liking was added to make sure that a positive rating of Job Crafting was not due to the liking of the colleague.

Analysis

Table 1 represents descriptive statistics: the means, standard deviations, correlations and Cronbach Alpha’s. Hierarchical multiple regression was used to assess the ability of the control measures Job Crafting of a colleague (self-report) and Job Crafting of a colleague (reported by the colleague) to predict the level of Job Satisfaction and Counterproductive Work behavior. Hence, the following relations were investigated;

(15)

 (1b) The relation between Job Crafting a colleague (reported by the colleague) and Job Satisfaction;

 (2a) The relation between Job Crafting of a colleague (self-report) and Counterproductive Work Behavior;

 (2b) The relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior.

Furthermore, hierarchical multiple regression analysis and the procedure advocated by Baron and Kenny (1986) was used to investigate the moderation effect of learning and stress in the relation of Job Crafting of a colleague (self-report and reported by the colleague) on Job Satisfaction and Counterproductive Work Behavior. Hence, the following relations were investigated;

 (3a) The moderating effect of Learning on the relation between Job Crafting of a colleague (self-report) and Job Satisfaction;

 (3b) The moderating effect of Learning on the relation Job Crafting of a colleague (reported by the colleague) and Job Satisfaction;

 (4a) The moderating effect of Learning on the relation between Job Crafting of a colleague (self-report) and Counterproductive Work Behavior;

 (4b) The moderating effect of Learning on the relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior.

 (5a) The moderating effect of Stress on the relation between Job Crafting of a colleague (self-report) and Job Satisfaction;

 (5b) The moderating effect of Stress on the relation Job Crafting of a colleague (reported by the colleague) and Job Satisfaction;

 (6a) The moderating effect of Stress on the relation between Job Crafting of a colleague (self-report) and Counterproductive Work Behavior;

 (6b) The moderating effect of Stress on the relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior.

Interaction was investigated after verifying the influence of age, education, gender, liking, number of organizations worked for and Tenure. Furthermore, preliminary analyses were conducted to ensure no violation of the assumptions of normality, linearity, multicollinearity and

homoscedasticity. Prior to the creation of the interaction terms, all variables were centred (Aiken and West, 1991). Statistical analyses were conducted with SPSS 22.0 (IBM, Armonk, NY). The level of significance was set to α=0.05.

(16)

RESULTS

H1: The relation between Job Crafting and Job Satisfaction.

When looking at table 2, after the entry of Job Crafting of a colleague (self-report) at step 2 (model 2), the total variance explained by the model as a whole was 16% , (P <0.01). In the final model, only the measure of Job Crafting of a colleague (self-report) was statistically significant (β=0.37; P<0.01). This indicates that employees who perceive their colleague to score high on Job Crafting are more satisfied with their jobs. However, this indicates that H1a is rejected because a negative relation was assumed. Further, table 3 indicates that, after the entry of Job Crafting of a colleague (reported by the colleague), the total variance explained by the model as a whole was 4.4%, (P >0.05). In the final model, Job Crafting of a colleague (reported by the colleague) was not

statistically significant (β=0.08; P>0.05). This indicates that H1b is rejected because the results indicate that there is no significant relation.

H2: The relation between Job Crafting and Counterproductive Work Behavior.

When looking at table 4, after the entry of Job Crafting of a colleague (self-report) at step 2 (model 2) the total variance explained by the model as a whole was 21.4%, (P >0.05). Job Crafting of a colleague (self-report) was not significant (β=-0.10; P>0.05) which indicates that H2a is rejected because a positive significant relation between Job Crafting and Counterproductive Work Behavior was assumed.When looking at table 5, after the entry of Job Crafting of a colleague (reported by the colleague) at step 2 (model 2), the total variance explained by the model as a whole was 20.6%, (P>0.05). Job Crafting of a colleague (reported by the colleague) was not statistically significant (β=0.04; P>0.05). This indicates that H2b is also rejected because, as discussed above, a significant positive relation was assumed.

H3: The effect of Learning on the relation between Job Crafting and Job Satisfaction.

Table 6 indicates that after the entry of Job Crafting of a colleague (self-report), learning and the product of both at step 2 (model 2), the total variance explained by the model was 29%, (P <0.01). There were two signification effects, Job Crafting of a colleague (self-report) (β=0.24; P<0.05) and learning (β=0.39; P<0.01). This indicates that H3a is rejected because the moderator effect was not significant (β=-0.09; P>0.05). Additionally, Table 7 indicates that, after the entry of Job Crafting of a colleague (reported by the colleague), learning and the moderator effect the total variance explained by the model was 23.9%, (P<0.01). As the results indicate, there were no significant effects which indicates that H3b is rejected.

(17)

H4: The effect of Learning on the relation between Job Crafting and Counterproductive

Work Behavior.

Table 8 indicates that after the entry of Job Crafting of a colleague (self-report), learning and the moderator effect at step 2 (model 2), the total variance explained by the model was 22.1%, (P >0.05). Furthermore, the results in Table 9 (H4b) indicate the same. 23%, (P>0.05) of the variance is explained by this model after the entry of Job Crafting towards a Colleague, learning and the moderator effect. To sum up, hypothesis H4a and H4b are rejected because Learning, the Job Crafting measures and the moderator effect were not statistically significant.

H5: The effect of Stress on the relation between Job Crafting and Job Satisfaction.

Table 10 indicates that after the entry of Job Crafting of a colleague (self-report), Stress and the moderator effect at step 2 (model 2) the total variance explained by the model was 22.1%, (P <0.01). There were two signification effects; Job Crafting towards a colleague (β=0.33; P<0.01) and, although one sided, the moderator effect (β=0.17; P=0.05). The results indicate that the moderator has a positive effect on the relation between Job Crafting and Job Satisfaction. Which means that when employees experience more Stress, the relation between Job Crafting and Job Satisfaction will be slightly more positive. Therefore the hypothesis is rejected because a negative moderator effect was assumed on the relation between Job Crafting and Job Satisfaction. Furthermore, the results in Table 11 (H5b) illustrate that H5b is rejected. 7.7%,(P>0.05) of the variance is explained by this model after the entry of Job Crafting of a colleague (reported by the colleague)s, Stress and the moderator effect. None of the effects were statistically significant.

H6: The effect of Stress on the relation between Job Crafting and Counterproductive Work

Behavior.

Table 12 display that after the entry of Job Crafting of a colleague (self-report), Stress and the moderator effect at step 2 (model 2) the total variance explained by the model was 21.4%, (P >0.05). Further, when looking at table 13 (H6b) the results indicate the same, 21%, (P>0.05) was explained by the model after the entry of Job Crafting received from a colleague, Stress and the moderator. This means that, as the results indicate, hypothesis H6a and H6b are rejected because a significant positive moderating effect was assumed.

Control Variables

The Control Variables used within this research project were respectively Age, Education, Gender, Linking, Number of organizations worked for and Tenure. First, when looking at the sample, employees with a higher Education Level showed more Counterproductive Work Behavior (β=-0.29 P<0.01), next to that there was a negative correlation between Age and Education Level (β=-0.33; P<0.01), which indicates that younger employees have a higher Education Level compared to older employees. Besides that there is a negative relation between Tenure and Education Level (β=-0.45;

(18)

P<0.01). Interestingly, there was a negative correlation between the Number of Organizations Worked for and Counterproductive Work Behavior (β=-0.24; P<0.05) and a negative correlation between Counterproductive Work Behavior and Age (β=-0.21; P<0.05), so younger employees showed more Counterproductive Work Behavior and the ones that have worked for more organizations showed less Counterproductive Work Behavior.

In each analysis the control variables were entered at step 1 (model 1). When the control variables were entered by Job Satisfaction they explained 3.8% of the variance in Job Satisfaction (P>0.05). None of the control variables were statistically significant related to Job Satisfaction. When looking at Counterproductive Work Behavior the control variables were also entered at step 1 (model 1), explaining 21% of the variance in Counterproductive Work behavior (P<0.01). The following measures were significant; Education (β=0.34; P<0.01), Gender (β=-0.21; P<0.05) and the Number of organizations worked for (β=-0.24; P<0.05). So as the results indicate, the control variables

mentioned explain some of the variance of the model. This would mean that, as discussed above, there is a relation between Counterproductive Work Behavior and these Control Variables.

(19)

Table 1 Means. Standard Deviations. Correlations. Cronbach Alpha

Variables Mean Std. Dev. 1 2 3 4 5 6 7 8 9 10 11 12

1. Age 40.18 11.08

2. Education Level 2.74 0.83 -0.33**

3. Tenure 10.72 9.84 0.64** -0.45**

4. Number of Organization 3.26 1.87 0.33** 0.10 -0.11

5. Counterproductive Work Behavior 1.34 0.22 -0.21* 0.29** -0.08 -0.24* (0.74)

6. Job Crafting of a colleague (self-report) 4.78 0.66 -0.11 0.01 -0.21* 0.07 -0.08 (0.80)

7. Job Crafting of a colleague (by colleague) 4.78 0.66 -0.10 0.05 -0.14 0.07 0.01 0.04 (0.80)

8. Job Crafting (self-report) 4.81 0.56 -0.07 0.05 -0.21* -0.02 -0.06 0.50** 0.07 (0.76)

9. Job Satisfaction 5.57 0.93 -0.06 -0.01 -0.13 0.01 -0.06 0.38** 0.12 0.33** (0.85)

10. Learning 4.66 0.86 -0.09 0.27** -0.13 0.12 0.02 0.28** -0.04 0.29** 0.41** (0.91)

11. Liking 5.73 0.86 -0.08 -0.07 0.07 -0.10 0.08 0.11 0.17 0.05 0.13 -0.08 (0.77)

12. Stress 2.95 0.45 -0.04 0.09 0.03 0.10 0.08 -0.01 -0.13 0.19 0.14 0.15 0.21* (0.71)

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

(20)

R² R² Change B SE β t Model 1 _0.04 Age 0.00 0.01 -0.02 -0.15 Education -0.03 0.13 -0.03 -0.23 Gender -0.07 0.22 -0.03 -0.32 Liking 0.16 0.12 0.15 1.35

Number of organization worked for 0.00 0.06 0.02 0.12 Tenure -0.01 0.02 -0.14 -0.85 Model 2 0.16 _0.12** Age 0.00 0.01 -0.04 -0.28 Education 0.01 0.12 0.01 0.08 Gender 0.02 0.20 0.01 0.08 Liking 0.10 0.11 0.09 0.89

Number of organization worked for 0.00 0.06 0.00 0.02 Tenure 0.00 0.01 -0.03 -0.19 Job Crafting of colleague (self-report) 0.50 0.14 0.37** 3.61 **. Correlation is significant at the 0.01 level (2-tailed).

*. Correlation is significant at the 0.05 level (2-tailed).

Table 3 H1b Job Crafting of a colleague (reported by the colleague) is negatively related to Job Satisfaction. R² R² Change B SE β t Model 1 _0.04 Age 0.00 0.01 -0.02 -0.15 Education -0.03 0.13 -0.03 -0.23 Gender -0.07 0.22 -0.03 -0.32 Liking 0.16 0.12 0.15 1.35

Number of organization worked for 0.01 0.06 0.02 0.12 Tenure -0.01 0.02 -0.14 -0.85 Model 2 _0.04 _0.00 Age _0.00 _0.01 _-0.02 _-0.09 Education _-0.03 _0.13 _-0.03 _-0.22 Gender _-0.08 _0.22 _-0.04 _-0.35 Liking _0.15 _0.12 _0.13 _1.23

Number of organization worked for _0.00 _0.06 _0.01 _0.04 Tenure _-0.01 _0.02 _-0.13 _-0.81 Job Crafting of colleague (reported by colleague) _0.11 _0.15 _0.08 _0.75 **. Correlation is significant at the 0.01 level (2-tailed).

(21)

Work Behavior. R² R² Change B SE β t Model 1 _0.21** Age _0.00 _0.00 _-0.09 _-0.60 Education _0.09 _0.03 _0.34** _3.26 Gender _-0.10 _0.05 _-0.21* _-2.17 Liking _0.03 _0.03 _0.10 _1.07 Number of organization worked for _-0.03 _0.01 _-0.24* _-2.10 Tenure _0.00 _0.00 _0.11 _0.72 Model 2 _0.21 _0.00 Age _0.00 _0.00 _-0.08 _-0.57 Education _0.09 _0.03 _0.33** _3.16 Gender _-0.11 _0.05 _-0.22* _-2.27 Liking _0.03 _0.03 _0.12 _1.21 Number of organization worked for _-0.03 _0.01 _-0.24* _-2.07 Tenure 0.00 0.00 0.08 0.52 Job Crafting of colleague (self-report) _-0.03 _0.03 _-0.10 _-0.98 **. Correlation is significant at the 0.01 level (2-tailed).

Table 5 H2b The relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior.

R² R² Change B SE β t Model 1 _0.21** Age _0.00 _0.00 _-0.09 _-0.60 Education _0.09 _0.03 _0.34** _3.26 Gender _-0.10 _0.05 _-0.21* _-2.17 Liking _0.03 _0.03 _0.10 _1.07

Number of organization worked for _-0.03 _0.01 _-0.24* _-2.10

Tenure _0.00 _0.00 _0.11 _0.72 Model 2 _0.21 _0.00 Age _0.00 _0.00 _-0.09 _-0.57 Education _0.09 _0.03 _0.34** _3.25 Gender _-0.10 _0.05 _-0.21* _-2.17 Liking _0.03 _0.03 _0.10 _1.00

Tenure _0.00 _0.00 _0.11 _0.74

Job Crafting of colleague (reported by colleague) _0.01 _0.03 _0.04 _0.37 **. Correlation is significant at the 0.01 level (2-tailed).

(22)

(self-report) and Job Satisfaction. R² R² Change B SE β t Model 1 _0.04 Age 0.00 0.01 -0.02 -0.15 Education -0.03 0.13 -0.03 -0.23 Gender -0.07 0.22 -0.03 -0.32 Liking 0.16 0.12 0.15 1.35

Number of organization worked for 0.01 0.06 0.02 0.12 Tenure -0.01 0.02 -0.17 -0.85 Model 2 _0.29 _0.25** Age 0.00 0.01 -0.01 -0.08 Education -0.15 0.12 -0.13 -1.25 Gender 0.18 0.19 0.09 0.94 Liking 0.11 0.11 0.10 1.01

Number of organization worked for -0.01 0.06 -0.02 -0.18 Tenure -0.01 0.01 -0.08 -0.55 Learning 0.43 0.11 0.39** 3.80 Job Crafting of colleague (self-report) 0.32 0.14 0.24* 2.35 Learning X Job Crafting of colleague (self-report) -0.13 0.14 -0.09 -0.91 **. Correlation is significant at the 0.01 level (2-tailed).

Table 7 H3b The moderating effect of Learning on the relation Job Crafting of a colleague (reported by the colleague) and Job Satisfaction.

R² R² Change B SE β t Model 1 _0.04 Age 0.00 0.01 -0.02 -0.15 Education -0.03 0.13 -0.03 -0.23 Gender -0.07 0.22 -0.03 -0.32 Liking 0.16 0.12 0.15 1.35

Number of organization worked for 0.01 0.06 0.02 0.12 Tenure -0.01 0.02 -0.17 -0.85 Model 2 _0.24 _0.20** Age _0.01 _0.01 _0.07 _0.44 Education _-0.20 _0.13 _-0.18 _-1.62 Gender _0.13 _0.20 _0.07 _0.66 Liking _0.15 _0.11 _0.14 _1.39

Number of organization worked for _-0.03 _0.06 _-0.06 _-0.48 Tenure _-0.02 _0.01 _-0.20 _-1.32 Learning _0.87 _0.65 _0.79 _1.33 Job Crafting of colleague (reported by colleague) 0.10 0.14 0.07 0.68 Learning X Job Crafting of colleague (reported by

colleague) -0.07 0.14 -0.32 -0.54 **. Correlation is significant at the 0.01 level (2-tailed).

(23)

report) and Counterproductive Work Behavior; R² R² Change B SE β t Model 1 _0.21** Age _0.00 _0.00 _-0.09 _-0.60 Education _0.09 _0.03 _0.34 _3.26 Gender _-0.10 _0.05 _-0.21 _-2.17 Liking _0.03 _0.03 _0.10 _1.07

Number of organization worked for _-0.03 _0.01 _-0.24 _-2.10

Tenure _0.00 _0.00 _0.11 _0.72 Model 2 _0.22 _0.01 Age _0.00 _0.00 _-0.11 _-0.69 Education _0.10 _0.03 _0.36** _3.23 Gender _-0.12 _0.05 _-.24* _-2.37 Liking _0.03 _0.03 _0.11 _1.11

Tenure 0.00 0.00 0.10 0.66

Learning _-0.03 _0.03 _-0.10 _-0.94 Job Crafting of colleague (self-report) _-0.02 _0.04 _-0.07 _-0.68 Learning X Job Crafting of colleague (self-report) -0.01 0.04 -0.02 -0.20 **. Correlation is significant at the 0.01 level (2-tailed).

Table 9 H4b The moderating effect of Learning on the relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior

R² R² Change B SE β t Model 1 _0.21** Age _0.00 _0.00 _-0.09 _-0.60 Education _0.09 _0.03 _0.34 _3.26 Gender _-0.10 _0.05 _-0.21 _-2.17 Liking _0.03 _0.03 _0.10 _1.07

Number of organization worked for _-0.03 _0.01 _-0.24 _-2.10

Tenure _0.00 _0.00 _0.09 _0.58

Learning _0.15 _0.16 _0.57 _0.94 Job Crafting of colleague (reported by colleague) _0.00 _0.04 _-0.01 _-0.07 Learning X Job Crafting of colleague (reported by

colleague) -0.04 0.03 -0.69 -1.16 **. Correlation is significant at the 0.01 level (2-tailed).

(24)

(self-report) and Job Satisfaction. R² R² Change B SE β t Model 1 _0.04 Age _0.00 _0.01 _-0.02 _-0.15 Education _-0.03 _0.13 _-0.03 _-0.23 Gender _-0.07 _0.22 _-0.03 _-0.32 Liking _0.16 _0.12 _0.15 _1.35

Number of organization worked for _0.01 _0.06 _0.02 _0.12 Tenure _-0.01 _0.02 _-0.14 _-0.85 Model 2 _0.20 _0.16** Age _0.00 _0.01 _0.03 _0.17 Education _-0.02 _0.12 _-0.01 _-0.13 Gender _0.00 _0.20 _0.00 _-0.02 Liking _0.09 _0.11 _0.09 _0.82

Number of organization worked for _-0.03 _0.06 _-0.07 _-0.56 Tenure -0.01 0.02 -0.08 -0.50

Stress _0.30 _0.21 _0.15 _1.43

Job Crafting of colleague (self-report) _0.46 _0.14 _0.33** _3.29 Stress X Job Crafting of colleague (self-report)1

0.47 0.28 0.17 1.66 **. Correlation is significant at the 0.01 level (2-tailed).

Table 11 H5b The moderating effect of Stress on the relation Job Crafting of a colleague (reported by the colleague) and Job Satisfaction.

R² R² Change B SE β t Model 1 _0.04 Age _0.00 _0.01 _-0.02 _-0.15 Education _-0.03 _0.13 _-0.03 _-0.23 Gender _-0.07 _0.22 _-0.03 _-0.32 Liking _0.16 _0.12 _0.15 _1.35

Number of organization worked for _0.01 _0.06 _0.02 _0.12

Tenure _-0.01 _0.02 _-0.14 _-0.85 Model 2 0.08 0.04 Age _0.00 _0.01 _-0.02 _-0.11 Education _-0.08 _0.14 _-0.07 _-0.59 Gender _-0.12 _0.22 _-0.06 _-0.54 Liking _0.11 _0.12 _0.10 _0.91

Number of organization worked for _0.00 _0.06 _-0.01 _-0.06

Tenure _-0.01 _0.02 _-0.13 _-0.81

Stress _-1.77 _1.67 _-0.85 _-1.06

Job Crafting of colleague (reported by colleague) _0.08 _0.16 _0.06 _0.51 Stress X Job Crafting of colleague (reported by colleague) _0.43 _0.35 _0.99 _1.24 **. Correlation is significant at the 0.01 level (2-tailed).

(25)

(self-report) and Counterproductive Work Behavior. R² R² Change B SE β t Model 1 _0.21** Age _0.00 _0.00 _-0.09 _-0.60 Education _0.09 _0.03 _0.34** _3.26 Gender _-0.10 _0.05 _-0.21* _-2.17 Liking _0.03 _0.03 _0.10 _1.07

Number of organization worked for2

-0.03 0.01 -0.23 -1.94

Tenure 0.00 0.00 0.08 0.51

Stress _0.00 _0.05 _0.01 _0.06

Job Crafting of a colleague (self-report) _-0.03 _0.03 _-0.09 _-0.90 Stress X Job Crafting of colleague (self-report) -0.01 0.07 -0.02 -0.21 **. Correlation is significant at the 0.01 level (2-tailed).

Table 13 H6b The moderating effect of Stress on the relation between Job Crafting of a colleague (reported by the colleague) and Counterproductive Work Behavior.

R² R² Change B SE β t Model 1 _0.21** Age 0.00 0.00 -0.09 -0.60 Education _0.09 _0.03 _0.34** _3.26 Gender _-0.10 _0.05 _-0.21* _-2.17 Liking 0.03 0.03 0.10 1.07

Tenure _0.00 _0.00 _0.12 _0.77

Stress _-0.21 _0.37 _-0.41 _-0.55 Job Crafting of colleague (reported by colleague) _0.01 _0.04 _0.02 _0.18 Stress X Job Crafting of a colleague (reported by

colleague) 0.05 0.08 0.43 0.58 **. Correlation is significant at the 0.01 level (2-tailed).

(26)

CONCLUSION AND DISCUSSION

This research project had the aim to examine the negative effects of proactive behavior, especially the intention was to investigate Job Crafting and the dependence on Job Satisfaction and Counterproductive Work Behavior. Furthermore, the expected effects were assumed to be

moderated by Learning and by Stress. Although, even with a multi-level analysis, no direct evidence was found for the supposed negative effects of proactive behavior, there are some very interesting indirect results that provide both practical and theoretical insights for future research.

As discussed, employees with a higher Education Level showed more Counterproductive Work Behavior and young employees had a higher Education Level compared to older employees. Interestingly, employees who had worked for more organizations showed less Counterproductive Work Behavior and younger employees showed more Counterproductive Work Behavior. These results indicate that there is no support found for the findings mentioned by Martinko et al. (2002) who argued that men are thought to express more Counterproductive Work Behavior. Next to that there was no support found for the results of Ng and Feldman (2009) who argued that employees with a higher Education Level engage less in Counterproductive Work Behavior. These contradictions could be due to the fact that, within this sample, 77.9% were male. So future research should look at the influences of including more women. Next to that, the results also emphasize one of the main characteristics of the high-tech and automotive industry (78.8% of the sample). This is of course a limitation within this research because this industry is characterized by fluctuations in terms of demand. The so called bullwhip effect (Aelker et al., 2013) describes these lager swings in inventory in response to changing demand. These enormous amplitude of the bullwhip indicates uncertainty and affects the turnover of employees. So one could argue that these younger employees in this specific industry need to be innovative and creative in order to survive within an organization. This could lead to Counterproductive Work Behavior (e.g. discussions). Bolino et al. (2010) also describes an overlap between Counterproductive Work Behavior and proactivity. So one could argue that, in order to be innovative for an organization and to maintain knowledge, the question rises; where is the line of demarcation between Counterproductive Work Behavior and Proactivity (e.g. Job Crafting)? This is definitely something that managers need to be aware of when looking at their Human

Resource Management. Furthermore, next to the industry limitation the sample mainly consisted out of employees with a high Educated Level (63.5% HBO or higher). Therefore it would be interesting for future research to look at other sectors and industries (e.g. non-profit) with more divers Education Levels. These different Education Levels can very well be related to different task types, which in turn could result in the statement of Berg et al. (2010), who argued that lower ranks occupy positions in which they find it relatively easier to adapt their work environments to create more opportunities to

(27)

job craft, while higher-rank employees feel more constrained despite being in positions of greater formal autonomy and power. Berg et al. (2010) conclude that the perception that employees have regarding the freedom they have to adapt to challenges in Job Crafting does not necessarily reflect their level of formally endowed autonomy and power, which is valuable information for

organizations.

When looking at this Job Crafting, within this sample, employees were unable to distinguish themselves from others, based on the fact that there was a correlation between the self-report of Job Crafting and the self-report of Job Crafting of a colleague (β=0.50; P<0.01). Interestingly, there was no relation found between Job Crafting of colleagues towards each other. So colleagues did not agree on their Job Crafting rating. A possible clarifications for the fact that employees could not distinguish themselves from others and did not agree on the Job Crafting rating can be found in theory about the self-concept (Cooper and Thatcher, 2010), it is based on relation identification. Sluss and Ashforth (2007) describe this relation identification as Coworker Identification, the extent in which individuals define themselves in terms of specific role relationships with other individuals in the workplace. Cooper and Thatcher (2010) argue that self-concepts of employees could influence their motivation because motives represent psychological needs that drive intentions and behaviors (Ryan and Deci, 2000). Cooper and Thatcher (2010) argue that self-concept orientations are driven by gender and cultural considerations. Controlling them would mean that one can predict associated identification motives and targets. Therefore, it is important for organizations to understand the self-concept orientations and identification motives, because understanding these motives provide an opportunity for organizations to communicate with employees and to encourage identification with proactive behavior. Gender was already discussed within this chapter, however it would be very interesting in this context to look at Job Crafting rating between gender types. Next to that, as mentioned by Cooper and Thatcher (2010), culture could be an important force when looking at motivation (e.g. proactivity, Job Crafting). Therefore, future research needs to look at the role of culture when looking at proactive behavior.

The relation between Job Crafting and Job Satisfaction was investigated. Support was found for the argument of Crant et al. (2010) who found a positive relation between proactivity and Job Satisfaction. The main difference with this research is that a direct positive effect was found instead of a mediating effect. However this research assumed a negative effect, based on Bolino et al. (2010) who stated that a negative relation was assumed due to friction between colleagues. This could also be explained by the fact that, as mentioned above, employees could not distinguish themselves from others and clarification therefore can be found in the self-concept theory.

(28)

In their qualitative research Bolino et al. (2010) argue that when an organization expects proactive behavior form their employees this may also contribute to Stress. Within this research there was no direct relation found between Stress and Job Crafting or Job Satisfaction. After entering Stress as a moderator on the relation between Job Crafting and Job Satisfaction a slightly positive effect was found. Which confirms the findings of Crawford et al. (2010) who argued that challenging job demands were positively related to work engagement even though they can also be appraised as stressful. However, one could argue that Stress needs Job Crafting in order to have a significant impact on Job Satisfaction so it is also possible that Job Crafting moderates the relation between Stress and Job Satisfaction.

When looking at the relation between Job Crafting and Stress, the JD-R model discussed by Tims et al. (2012), stated that when job demands are high in combination with low levels of job resources this may lead to stressors. The reason that, within this research, no relation was found could be, next to the limitations mentioned before, caused by other limitations; inadequate attention for the combinations of job resources and demands within the survey which would mean that future research should investigate the different forms of Job Crafting in relation to other constructs (e.g. Stress, Learning and Counterproductive Work Behavior). Furthermore, there was no longitudinal data collection to measure the impact Job Crafting over time but only cross-sectional data collection. Therefore, it is important for future research and organizations to pay attention to the long-term effects of Job Crafting in relation to Stress because Lee and Ashforth (1996) indicated that stressful events are positively related to emotional exhaustion and burnout. Next to that Fox and Spector (1999) found that frustration in response to stressors is an important predictor of Counterproductive Work Behavior which in turn was negatively related to Job Satisfaction.

Finally, when looking at the aim of this research project and the last construct; Learning, the assumptions was that organizations become, based on the resource based view (Barney,1991), dependent on proactive employees. Which could reduce the learning capacity within an organization (Bolino et al., 2010). Next to that, there is a risk of creating friction between more and less proactive employees (Bolino et al., 2010; Belschak et al., 2010) in combination with low learning resources. Looking at the results, a relation was found between Learning and Education Level (β=0.27; P<0.01) which makes sense and could be, as discussed, due to the sample (e.g. high innovative and high Learning). Learning also correlates with Job Crafting of a colleague (self-report) (β=0.28; P<0.01) and, because employees could not distinguish themselves from others, also with Job Crafting self-report (β=0.29; P<0.01). Interestingly no moderating effect of Learning was found, while Learning correlated with Job Satisfaction (β=0.41; P<0.01). The correlation between Job Crafting and Learning could also be due to the similarity of the items (e.g. opportunity to learn). Although no moderating effect was found, the results indicate that it is important for organizations and future research to include

(29)

Learning when proactive behavior is needed. Within this research only positive effects were found because, unfortunately, the absence of a high Learning was not included in the sample, all

organizations scored relative high on Learning (mean 4.66 and standard deviation of 0.86). However this can also be due to a self-selection bias (Wright, 2006) which would mean that some individuals are more likely to complete an online survey (e.g. eager to learn) than others. Next to the other constructs it would be very interesting for future research to include different Education Levels in combination different task levels. Next to that, it would be interesting to look at different hierarchies within the organization to examine if image management is applicable. Stobbeleir et al. (2010) argue that image management could be a costly downside of proactive behavior. For this specific research this could be done by letting the supervisor, next to the colleagues, judge the Job Crafting behavior of the employees.

To sum up, a lot of future research is needed to provide more insight in the dark side of proactive behavior. However this research project managed to provide evidence regarding the context of Job Crafting in terms of Counterproductive Work Behavior, Job Satisfaction, Stress and Learning. The main highlights are the fact that employees were unable to distinguish themselves from their colleagues and the slightly positive moderating effect of Stress on the relation between Job Crafting and Job Satisfaction, or the moderating effect of Job Crafting on the relation between Stress and Job Satisfaction. Next to that, Learning seems to have an effect on both Job Crafting and Job Satisfaction. Last, the evidence found for different Control Variables (Age, Genre, Education Level, Tenure and Number of Organizations Worked for) in relation to Learning, Counterproductive Work Behavior and Job Crafting. These results added an interestingly depth into the research and confirmed some of the statements in existing literature. All these highlighted results indicate that the aim of this research, to learn and provide both practical and theoretical insights based on data and guidelines distracted from previous literature, is achieved. There is definitely something happening in the context of proactive behavior and that future research is needed to provide more insight for both practical and theoretical development.

(30)

REFERENCES

Aelker J., Bauernhansl T., Ehm H. (2013). Managing Complexity in Supply Chains: A Discussion of Current Approaches on the Example of the Semiconductor Industry. Procedia CIRP, Volume 7,

Pages 79-84.

Aiken, L. S., West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Ne

wbury Park: Sage.

Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of

Management, 17, 99–120.

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

Belschak, F. D., Hartog, D. N. D., Fay, D. (2010). Exploring positive, negative and context-dependent aspects of proactive behaviors at work. Journal of Occupational and Organizational

Psychology, 83(2), 267–273.

Berg, J.M., Wrzesniewski, A., Dutton, J.E. (2010). Perceiving and responding to challenges in Job Crafting at different ranks: When proactivity requires adaptively. Journal of Organizational

Behavior, 31, 158-186

Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. Philadelphia, PA:

Temple University Press.

Bolino, M., Valcea, S., Harvey, J. (2010). Employee, manage thyself: The potentially negative implications of expecting employees to behave proactively. Journal of Occupational and

Organizational Psychology, 83(2), 325–345.

Caplan, R., Cobb, S,, French, J., Van Harrison, R., Pinneau, S. (1975). Demands and worker health: Main effects and organizational differences. Washington, DC: U.S Government Printing Office.

Cooper, D., Thatcher, S.M.B. (2010). Identification in organizations: The role of Self-Concept orientations and identification motives. Academy of Management Review 2010, Vol. 35, No. 4, 516–

538.

Crant, J. M. (2000). Proactive behavior in organizations. Journal of Management, 26(3), 435– 462. Grant, A. M., and Ashford, S. J. (2008). The dynamics of proactivity at work. Research in

Organizational Behavior, 28(1), 3–34.

Crant, J.M., Li, N., Liang, J. (2010). The Role of Proactive Personality in Job Satisfaction and Organizational Citizenship Behavior: A Relational Perspective Journal Of Applied

Psychology, 2010 Mar, Vol.95(2), pp.395-404.

Crawford, E. R., LePine, J. A., Rich, B. L. (2010). Linking job demands and resources to employee engagement and burnout: A theoretical extension and meta-analytic test. Journal of

The negative effects of proactive behavior