Job crafting and needs-supplies fit as moderators of the relationship of
perceived transformational leadership and perceived passive leadership
with employee job satisfaction
F.A. Langeveld (10684506) Amsterdam Business School
Executive Programme in Management Studies Specialization: Leadership and Management Master thesis MSc Business Studies
Academic year: 2016 – 2017 Supervisor: Dr. S.T. Mol
Second reader: Dr. W van Eerde March, 2017
ii
Statement of Originality
This document is written by student Felicia Langeveld who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
iii
Table of Content
Abstract ... vi
1. Introduction ... 1
2. Literature review ... 4
2.1. Transformational and passive leadership ... 4
2.2. Job satisfaction ... 6
2.3. Job crafting ... 8
2.4. Needs-supplies fit ... 10
2.5. Research model ... 12
3. Data and method ... 13
3.1. Sampling strategy ... 13
3.2. Sample ... 14
3.3. Measurement and reliability of constructs ... 14
3.3.1. Translation, back-translation procedure ... 14
3.3.2. Dependent variable job satisfaction ... 15
3.3.3. Independent variable transformational and passive leadership ... 15
3.3.4. Moderating variable job crafting ... 15
3.3.5. Moderating variable needs-supplies fit ... 15
3.3.6. Control variables ... 16
3.4. Statistical procedure ... 17
4. Results ... 20
4.1. Correlation analysis ... 20
4.2. Direct effects ... 20
4.3. Moderating effects of job crafting and needs-supplies fit ... 22
iv
5. Discussion ... 35
5.1. Theoretical and practical implications ... 35
5.2. Limitations ... 40 5.3. Further research ... 42 6. Conclusion ... 44 References ... 47 Appendices ... 53 Appendix I Scales ... 53
v
List of tables and figures
Tables
Table 1: Means, Standard Deviations, Correlations and Reliabilities ... 19 Table 2: Hierarchical regression model of antecedents of job satisfaction ... 22 Table 3: Results of job crafting and needs-supplies fit as moderators of the relationship between transformational leadership and job satisfaction ... 31 Table 4: Results of job crafting and needs-supplies fit as moderators of the relationship between passive leadership and job satisfaction ... 31 Table 5: Results of job crafting as an isolated moderator of the relationship between
transformational leadership and job satisfaction ... 32 Table 6: Results of job crafting as an isolated moderator of the relationship between passive leadership and job satisfaction ... 32 Table 7: Results of needs-supplies fit as an isolated moderator of the relationship between transformational leadership and job satisfaction ... 33 Table 8: Results of needs-supplies fit as an isolated moderator of the relationship between passive leadership and job satisfaction ... 33 Table 9: Conditional effect of passive leadership and job satisfaction at levels of job crafting ... 34 Table 10: Conditional effect of passive leadership and job satisfaction at levels of needs-supplies fit ... 34
Figures
Figure 1: Conceptual model ... 12 Figure 2: Job crafting as a moderator of the passive leadership - job satisfaction relationship ... 26 Figure 3: Needs-supplies fit as a moderator of the passive leadership - job satisfaction
vi
Abstract
Many studies have focussed on the beneficial effects of high-quality leadership, but recently authors noticed that these studies did not pay much attention to roles of ineffective leadership styles. This cross-sectional study aims to examine both the relationships of perceived transformational and perceived passive leadership with employee job satisfaction and the moderating effects of both job crafting and need-supplies fit on these relationships. For this study single-source data was collected via an online questionnaire that was completed by 168 respondents working in either a full-time or a part-time job.
The results of this study show a significant positive relationship between perceived transformational leadership and employee job satisfaction. No relationship was found between perceived passive leadership and employee job satisfaction. The moderation analyses did not yield support for any of the hypothesized moderated pathways. Because the hypotheses did not mention the interplay of the interaction effects also the isolated moderation effects were analysed. The analysis found isolated moderation effects of both job crafting and needs-supplies fit on the relationship between perceived passive leadership and employee job satisfaction. The direction of the isolated moderation of job crafting was not in line with the hypothesis.
The results of this study raise a number of questions. The high correlation between the leadership styles could indicate that these styles could be unidimensional opposites instead of separate constructs. Job crafting, leadership, and needs-supplies fit could have conceptual similarities which could explain their overlapping variance in employee job satisfaction. In order to get a better understanding of transformational and passive leadership and the interplay of the variables job crafting, leadership, and needs-supplies fit, and their relationship with employee job satisfaction further research is recommended.
Key words: Perceived transformational and passive leadership; employee job satisfaction; job crafting; needs-supplies fit
Introduction | 1
1. Introduction
Employees are one of the most important resources of an organisation. An effective leader can improve the satisfaction of employees and encourage them through leadership (Chang & Lee, 2007). Many studies have focused on the beneficial effects of high-quality leadership, but research focussed on effective leadership has not paid much attention to the roles of ineffective leadership styles (Kellerman, 2004; Kelloway, Sivanathan, Francis, & Barling, 2005).
In contrast to the transformational leadership style, passive leadership is mostly considered to be ineffective (Kelloway et al., 2005). A transformational leader is proactive, concerned with the well-being of employees, and considered to be effective (Kelloway, Mullen, & Francis, 2006). A passive leader avoids making decisions and does not take responsibility, nor does (s)he use his/her authority (Antonakis, Avolio, & Sivasubramaniam, 2003).
Previous research has mainly focused on the positive organisational impact of active forms of leadership (Kelloway et al. 2005). Research has shown that passive management by exception is negatively related to business unit performance (Howell & Avolio, 1993) and that passive leaders are likely to be unavailable when needed (Bono & Judge, 2004). Kelloway et al. (2005) have suggested that passive leadership is a distinct form of leadership and that it has negative effects beyond those attributable to a lack of transformational leadership skills. Kelloway et al. (2006) responded to this suggestion and concurrently examined the effects of good (transformational) and poor (passive) leadership on safety-related outcomes. Their results show that passive leadership explains unique variance in safety-related outcomes above and beyond that accounted for by transformational leadership. Passive and transformational leadership thus have opposite and complimentary effects on safety-related outcomes. They suggested future research to consider both the nature and the
Introduction | 2 role of passive leadership. Furthermore, passive leadership may explain variance in other work related outcomes as well.
To evaluate the style of a leader the Multi Leadership Questionnaire (MLQ) can be used. The MLQ measures the degree to which an individual engages in different leadership styles (Avolio & Bass, 2004) which implies that elements of transformational and passive leadership can be present in the same individual. Recent research has found support for the positive relationship between transformational leadership and employee job satisfaction (Albion & Gagliardi, 2007; Braun, Peus, Weisweiler, & Frey, 2013). Because transformational leadership could have elements of passive leadership (Avolio & Bass, 2004), which is generally considered to be ineffective (Kelloway et al., 2005), this leader could negatively impact employees’ job satisfaction.
This study will respond to the suggestion of Kelloway et al., (2005) to consider the potential negative role of passive leadership in predicting desired workplace outcomes and concurrently examines the relationship of transformational and passive leadership with employee job satisfaction. This research will investigate how the leadership style, as perceived by the employee, is related to the job satisfaction of that employee. Therefore the focus will be on perceived passive and perceived transformational leadership. This research will examine if it can confirm the observed positive relationship between transformational leadership and employee job satisfaction and if passive leadership has a negative relationship with employee job satisfaction.
Employees who self-initiate changes in their own job demands and job resources to attain and/or optimize their own (work) goals are known as job crafters (Tims, Bakker, & Derks, 2012). Through job crafting employees can optimize their own well-being when they are allowed to (Tims, Bakker, & Derks, 2013). Tims et al. (2012) developed and validated a generic scale to measure job crafting. Their research showed that four dimensions of job
Introduction | 3 crafting may be distinguished, but in this research job crafting will be investigated as a whole. Tims et.al. (2012) suggested further research is needed to test if this scale is equally applicable to other samples and to investigate the possible consequences of job crafting for individual employees and the organization at large. Focussing on the active role of the employee in shaping their jobs will provide more insight in the interplay between the work situation and the employee.
The degree of employee job satisfaction is in part determined by the extent to which employees perceive a high degree of match between their needs and the supplies of a job (Cable & DeRue, 2002; Locke, 1976). When the needs, desires, or preferences of an employee are met by the job they perform then need-supplies fit occurs(Kristof-Brown, Zimmerman, & Johnson, 2005).
This research will investigate the possible moderating effects of both job crafting and needs-supplies fit on the relationships of transformational and passive leadership with employee job satisfaction. This inspired the following research question: “Do job crafting and needs-supplies fit moderate the positive relationship between perceived transformational leadership and employee job satisfaction and the negative relationship between perceived passive leadership and employee job satisfaction?”
Literature review | 4
2. Literature review
In order to answer the research question, the constructs of interest, the proposed relationships, and hypotheses will be introduced in this chapter.
2.1. Transformational and passive leadership
Leadership has been an important topic of investigation and a lot of research has been conducted on leadership styles and their effectiveness. Researchers have different opinions about the meaning of the term leadership. De Jong and Den Hartog (2007, p. 5) defined leadership as “the process of influencing others towards achieving some kind of desired outcome”.
Different trends in leadership theory, for example traits (who leaders are), styles (behaviour of the leader) and contingencies (effectiveness of the leader depends on the situation), have been studied (Deanne & Paul, 2001). From the early 1980s onward, the interest rose in a renewed leadership concept in both scientific and professional fields, namely transformational leadership. Transformational leaders have a vision and inspire loyalty and emotional attachment in their employees (Deanne & Paul, 2001). The concept of transformational leadership was first introduced by Burns (1978) and has since been further developed by a number of researchers (Hartog, Muijen, & Koopman, 1997; Bono & Judge, 2004; Deanne & Paul, 2001; Scandura & Williams, 2004; Sosik, Godshalk, & Yammarino, 2004; Judge & Piccolo, 2004). Studies using the MLQ and similar questionnaires found transformational leadership to be positively related to indicators of leadership effectiveness,
such as subordinate satisfaction, motivation, and performance (Yukl, 1999).
Transformational leadership consists of four different dimensions: charisma or idealized influence, inspirational motivation, intellectual stimulation and individualized consideration (Judge & Piccolo, 2004). A transformational leader has charisma or idealized influence; the leader behaves in admirable ways so that the followers will identify themselves
Literature review | 5 with the leader. Inspirational motivation means this type of leader articulates a vision that is appealing and inspiring to followers. Effective leaders inspire followers to think beyond their own interest and focus on the importance of the collective (Deanne & Paul, 2001; Yukl, 1999). This type of leadership intellectually stimulates followers by challenging assumptions and taking risks, and soliciting followers’ ideas. The leader attends to followers’ needs, acts as a mentor or coach, and listens to their concerns and needs, which is captured in the dimension individualized consideration (Scandura & Williams, 2004).
Charismatic leadership has, apart from its role in transformational leadership, also been the basis of its own distinct literature (Judge & Piccolo, 2004). In his evaluation of conceptual weaknesses in transformational and charismatic leadership theories, Yukl (1999) showed that transformational leadership and charismatic leadership are often treated as equal. Although there are plausible differences that should not be ignored, they are partially overlapping styles.
The active form leadership, transformational leadership, is often contrasted with the passive laissez-faire leadership (Den Hartog et al., 1997). Passive laissez-faire leadership consists of two categories: abusive or unethical leadership and passive leadership (Kelloway et al. 2005). This study will focus on passive leadership to examine the difference between the active (transformational) and inactive (passive) form of leadership.
One could argue that in some situations an active form of leadership is not desired. Kerr and Jermier (1978) proposed several characteristics in their research that could reduce the importance of active leadership. Passive leadership could lead to empowerment of followers. Giving followers more power, by for example involving them in the decision making process, is a component of the transformational leadership style. So characteristics of the passive leadership style are visible in transformational leadership. This corresponds with the Multi Leadership Questionnaire (MLQ) (Avolio & Bass, 2004). The MLQ is a
well-Literature review | 6 established instrument to measure the transformational, transactional and passive-avoidant leadership styles. The MLQ is a 360-degree feedback method, which allows an individual to measure how they perceive themselves with regard to a specific leadership style and it also includes feedback from subordinates. The MLQ implies that the leadership styles are not independent from one another (i.e. they are correlated). In this research passive and transformational leadership will be operationalized as separate constructs.
2.2. Job satisfaction
Employees have different opinions on many aspects of their jobs, careers and organizations (Saari & Judge, 2004). According to Saari and Judge (2004) the central perspective of employee attitude in research and practice is job satisfaction. The most-used research definition of job satisfaction (Weiss, 2002) by Locke (1976) states that “job satisfaction is a pleasurable or positive emotional state resulting from an appraisal of one’s job or job experiences” (p. 1300). This definition is broad as it includes characteristics relating to the job itself and the work environment (Varela González & García Garazo, 2006). The work environment is important for the employees’ job satisfaction (Lambert, Hogan, & Barton, 2001). For an organisation it is important to pay attention to employee job satisfaction because dissatisfaction can instigate employee turnover (Lambert et al. 2001) and absenteeism (Wegge, Schmidt, Parkes, & Dick, 2007).
The type of leadership plays an important role in determining employee job satisfaction (Foels, Driskell, Mullen, & Salas, 2000). Lowe, Kroeck and Sivasubramaniam (1996) found, in their meta-analytic review of studies using the MLQ, that key elements of transformational leadership correlated positively with subordinates’ satisfaction and performance. Taylor (2000) showed in his research that when an organisation cares for its employees it gets support and rewards from its employees in return. Investing in the well-being of employees will result in a higher satisfaction of those employees. The importance of
Literature review | 7 teaching leaders to highlight the link between the job tasks and the broader purpose and vision of the organisation to their followers is confirmed in the research of Bono and Judge (2003). Their study, which examined the self-concordance model at work, found that individuals with more self-concordant work goals are more satisfied with their job. The self-concordance model, which itself is derived from the self-determination theory of Deci and Ryan (1985), predicts that individuals are the happiest when stated goals match enduring interests and values. Self-concordant motives are more likely to induce well-being and are more satisfying.
Transformational leaders are proactive leaders who intellectually stimulate followers by challenging assumptions soliciting followers’ ideas (Scandura & Williams, 2004) and have a vision (Deanne & Paul, 2001). Podsakoff, MacKenzie, and Bommer (1996) found in their research that articulating a vision is positively related to employees’ general satisfaction. The transformational leader enables the employee to put effort into their goals which makes it more likely for the employee to attain goals that will make the employee satisfied (Deci & Ryan, 1985). The overall satisfaction of an employee is a response of an employee concerning his/her particular job and the comparison of the actual outcomes with those that are expected, needed, wanted, or perceived to be fair or just (Lambert et al., 2001). The transformational leader who invests in the well-being of their employees could thus have a positive impact on employee job satisfaction.
A passive leader who only takes action after irregularities have occurred (Den Hartog et al., 1997) is less likely to highlight the link between the job and the broader purpose and vision of the organisation. The work environment is an important element for employee job satisfaction (Lambert et al., 2001). Not sharing a vision could result in employees not knowing what is expected from them. When employees do not know what is expected or wanted from them this could negatively impact their job satisfaction. Therefore the following hypotheses will be tested:
Literature review | 8 Hypothesis 1a: There is a positive relationship between perceived transformational leadership and employee job satisfaction.
Hypothesis 1b: There is a negative relationship between perceived passive leadership and employee job satisfaction.
2.3. Job crafting
Within their areas of responsibility employees may have the opportunity to try out changes or innovations independently (Gebert, Boerner, & Lanwehr, 2003). The capacity to self-initiate changes in one’s own job demands and job resources to attain and/or optimize their (work) goals is captured by the job crafting construct (Tims et al. 2012). Job crafters modify aspects of their job to improve the fit between job characterises and their own needs, abilities, and preferences (Berg, Dutton, & Wrzesniewski, 2008).
At the time the term job crafting was named by Wrzesniewski and Dutton (2001), the concept of redesigning one’s job was not new. Hackman and Oldham (1976) already revealed in their research that individuals become more growth oriented and may act independently to change and adjust to the situation. Because an employee puts changes in motion when crafting his/her job, it is operationalized according to the types of job characteristics suggested in the Job Demands Resources (JD-R) model (Bakker & Demerouti, 2007; Crawford, LePine, & Rich, 2010). Job crafters change the level of job demands, the level of job resources, or both, which is in line with the JD-R model.
According Wrzesniewski and Dutton (2001) an employee may initiate three types of job crafting. The employee may craft the task they fulfil at work, (s)he may craft the quality/ amount of interaction with others when performing his work, or (s)he may change the cognitive boundaries by positively reframing the manner in which (s)he thinks about (aspects of) the job. Crafting the job allows the employee to make their work more engaging and meaningful, and therewith more satisfying.
Literature review | 9 Job crafting allows the employee to change the cognitive boundaries and positively reframe the manner in which (s)he thinks about (aspects of) the job (Wrzesniewski & Dutton, 2001). Tims, Bakker, and Derks (2013) investigated the impact of job crafting on job demands, job resources, and well-being. Their research focussed on the way employees change their work environment to benefit their own well-being, and showed that job crafting has a positive impact on employee well-being. Job crafting allows employees to take initiative in aligning their job resources with their own needs and preferences to experience increased being. A high level of job resources significantly and positively impacts employee well-being (Judge, Bono, Erez, & Locke, 2005). Challenging demands offer experience which can lead to satisfaction and higher levels of self-efficacy (Gorgievski & Hobfoll, 2008).
Transformational leaders highlight the link between the job task and the broader purpose and vision of the organisation to their followers (Bono & Judge, 2003). According to the self-determination theory of Deci and Ryan (1985) individuals are happiest when stated goals match enduring interests and values. The transformational leader who inspires their employees to try out innovations independently (Gebert et al., 2003) can allow the job crafting employee to change one’s job demands and resources (Tims et al., 2012) in a manner in which (s)he thinks positively about (aspects of) the job (Wrzesniewski & Dutton, 2001). Employees who craft their jobs can allow themselves to align their job with what they expect or want, resulting in higher job satisfaction.
The passive leader could lead to empowerment of followers (Kerr & Jermier, 1978). Having more power can allow an employee to optimize their own well-being through job crafting. The job crafting employee can align their job with what they want, resulting in a less negative relationship between passive leadership and employee job satisfaction. The employee who perceives a passive leadership style and does not craft their job could have a negative emotional state from the appraisal of his/her job or job experience resulting in an
Literature review | 10 even less satisfied employee. The relationships of transformational and passive leadership with employee job satisfaction is thus likely to be moderated by job crafting. Therefore the following will be tested:
Hypothesis 2a: Job crafting moderates the positive relationship between perceived transformational leadership and employee job satisfaction, so that this relationship is stronger for higher levels of job crafting.
Hypothesis 2b: Job crafting moderates the negative relationship between perceived passive leadership and employee job satisfaction, so that this relationship is stronger for lower levels of job crafting.
2.4. Needs-supplies fit
A distinct conceptualization of person-environment fit is complementary fit, which is defined as “the mutually offsetting pattern of relevant characteristics between the person and the environment” (Muchinsky & Monahan, 1987, p. 272). Although this type of fit was operationalized strictly as individual skills meeting environmental needs (demands-abilities fit), the model was expanded by Kristof (1996) who included individual needs being met by environmental supplies (needs-supplies fit) (Kristof-Brown et al., 2005).
Need-supplies fit occurs when the needs, desires, or preferences of an employee are met by the job they perform (Kristof-Brown et al., 2005). From an employee perspective needs-supplies fit may be the most important type of fit (Cable & DeRue, 2002), because gaining access to the rewards that an organization offers is a part of the basic motivation for people to enter the labour market and accept a job (Simon, 1951). To allow a needs-supplies fit the organisation can provide the financial resources as well as the task related growth opportunities that are demanded by the individuals (Sekiguchi, 2004).
Needs-supplies fit are perceptions and judgments between the needs and the rewards employees’ receive in return for their service and contributions on their job (Cable & DeRue,
Literature review | 11 2002). The degree of job satisfaction of an employee is determined by the extent to which an employee perceives a high degree of fit between their needs and the supplies of the job (Cable & DeRue, 2002; Locke, 1976). Kristof-Brown et al., (2005) found in their meta-analysis (K= 47; N = 12,960) of person-job, person-organization, person-group and person-supervisor fit a strong correlation between needs-supplies fit and job satisfaction (.56).
Transformational leadership has a positive relationship with employee job satisfaction (Lowe et al., 1996) because transformational leaders listen to employees’ concerns and needs (Deanne & Paul, 2001) and act as a mentor (Scandura & Williams, 2004). When the resources of the organisation meet the needs of the employee (i.e. high needs-supplies fit) the positive relationship between transformational leadership and job satisfaction could be strengthened (Kristof-Brown et al., 2005).
The passive leader who does not take responsibility nor uses his/her authority (Antonakis et al., 2003) and only takes action after irregularities have occurred (Den Hartog et al., 1997) could negatively impact employees’ job satisfaction. But when the resources of the organisation also do not meet the needs of the employee (i.e. low needs-supplies fit) this could result in an even less satisfied employee. The relationships of transformational and passive leadership with employee job satisfaction is thus likely to be moderated by needs-supplies fit. Therefore the following will be tested:
Hypothesis 3a: Needs-supplies fit moderates the positive relationship between perceived transformational leadership and employee job satisfaction, so that this relationship is stronger for higher levels of needs-supplies fit.
Hypothesis 3b: Needs-supplies fit moderates the negative relationship between perceived passive leadership and employee job satisfaction, so that this relationship is stronger for lower levels of needs-supplies fit.
Literature review | 12
2.5. Research model
In the prior section of this chapter a set of three hypotheses were established. A visual representation of the proposed relationships between the dependent variable job satisfaction and the independent variables perceived transformational and perceived passive leadership and the hypothesized moderating effects of both job crafting and needs-supplies fit is captured in figure 1.
Figure 1: Conceptual model
This research investigated the hypotheses on a group of Dutch employees. In the following chapters the analysis and results of this research will be reported.
Perceived
transformational leadership
Perceived passive leadership
Employee job satisfaction
Data and method | 13
3. Data and method
This chapter represents the start of the empirical part of this study. The characteristics of the collected sample will first be described. After the description of the sampling strategy the characteristics of the sample of this study will be described. Subsequently the variables included in the questionnaire and their reliabilities will be discussed. The chapter will end with a brief description of the statistical approach to test the expected relations as discussed in the previous chapter.
3.1. Sampling strategy
For this study primary data was collected via a distributed questionnaire. Non-probability sampling was used, because no sampling frame was available and therefore random sampling was not possible. A combination of convenience, self-selection and snowball sampling was used. The respondents received an invitation via email, with a short explanation of the study, to participate in the study by completing the questionnaire. Both the personal and professional network of the researcher where used as a starting point for contacting respondents. The respondents could have any type of job, but needed to have some kind of a leader whose leadership style they could assess, in any type of organisation. The expectations, behaviour and performance of employees in different national cultures could differ (Redding, 1990). Because of possible confounding effects of value differences at the individual level (Hofstede, 2011) the respondents all needed to be working in the Netherlands. The measurement level of the variables was at the individual level. The respondents were asked to assess their leader’s levels on transformational and passive leadership, and their own job satisfaction, job crafting, and needs-supplies fit. The data was single source (employee-rated only). Ethical issues such as privacy, anonymity, and voluntary participation were included in the introductory letter accompanying the questionnaire. See appendix II for the introductory letter and questionnaire (in Dutch).
Data and method | 14
3.2. Sample
The sample of this study consisted of Dutch employees working on either a part-time (73.8%) or fulltime (26.2%) basis for their organisation. Of the 194 employees that started filling out the questionnaire, 168 respondents completed the questionnaire in full (response rate 86.6%). Of the respondents (Mage = 24.61, SDage = 11.82, age-range: 18-63 years) 58.3%
were female. The sample covered a broad range of education backgrounds. Only 7.1% of the respondents had completed a secondary education programme as their highest education (VMBO = 1.8%, HAVO/VWO = 5.4%). A majority of 63.7% completed an educational programme at a university of applied sciences (MBO = 12.5%, HBO = 51.2%) and 29.2% completed an education programme at a (research) university. Of the respondents 6% reported an organizational tenure of less than a year, 40.5% reported an organizational tenure of 1-5 years, 27.4% reported an organizational tenure of 6-10 years, 11.3% reported an organizational tenure of 11-15 years, 6% reported an organizational tenure of 16-20 years, 1.8% reported an organizational tenure of 21-25 years, 7.1% reported an organizational tenure of more than 25 years. Of the respondents 64.9% worked for an organization with a non-profit background (e.g. government, health care, educational institute, cultural institution).
3.3. Measurement and reliability of constructs 3.3.1. Translation, back-translation procedure
All items used for the questionnaire were derived from English studies. Since all respondents to the survey were to be Dutch the original items were translated to Dutch. In order to assure that the content of the items remained equivalent in the translation process the translated Dutch items were translated back to English by a second person. The small number of discrepancies between the back-translated items and the original items were corrected in the final Dutch version of the questionnaire.
Data and method | 15
3.3.2. Dependent variable job satisfaction
To measure job satisfaction the scale of Brayfield and Rothe (1951) (19 items, Cronbach’s α = .891) was used. One of these items was: “My job is like a hobby to me”. Respondents were asked to answer according a 5 point Likert-scale, ranging from 1 (strongly disagree) to 5 (strongly agree). The scale of job satisfaction consisted of 10 counter-indicative items.
3.3.3. Independent variable transformational and passive leadership
To measure perceived transformational and passive leadership the scale of Kelloway et al. (2006) was used. These scales where adapted from the MLQ-5 (Bass & Avolio, 1990). Ten items were used to measure perceived transformational leadership (Cronbach’s α = .872). One of these items was “Makes sure that we receive appropriate rewards for achieving targets on the job”. Three items were used to measure perceived passive leadership (Cronbach’s α = .822). One of these items was: “Waits for things to go wrong before taking action”. The scale Kelloway et al. (2006) used was safety specific but is for this research transformed to general items. An example of a transformed question is “Avoids making decisions that affect safety on the job” transformed to “Avoids making decisions that affect the job”. A 5 point Likert-scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used.
3.3.4. Moderating variable job crafting
To measure job crafting the scale of Tims et.al. (2013) was used (21 items, Cronbach’s α = .705). One of these items was: “I try to develop my capabilities”. The measurement was conducted by using a 5 point Likert-scale ranging from 1 (strongly disagree) to 5 (strongly agree).
3.3.5. Moderating variable needs-supplies fit
Cable & DeRue (2002) created a scale for needs-supplies fit based on the conceptualisations offered by Kristof (1996) and Edwards (1991). To assess needs-supplies fit
Data and method | 16 this scale (3 items, Cronbach’s α = .876) was used. One of these items was “There is a good fit between what my job offers me and what I am looking for in a job”. Respondents were asked to provide their answers on a 5 point Likert-scale, ranging from 1 (strongly disagree) to 5 (strongly agree).
3.3.6. Control variables
Next to items representing the constructs of interest, items where added as control variables. Besides the control variables gender and age, also items for ethical culture constructs where included as control variables.
Earlier research showed that it is paradoxical that the level of job satisfaction is higher for woman than for men (Sousa-Poza & Sousa-Poza, 2000). Research of Kaiser (2007) showed that a gender-job satisfaction paradox will more likely emerge when the labour market access is more restrictive. Gender also has a small direct effect on leadership behaviours; superiors rated female managers more transformational than male managers (Barbuto Jr, Fritz, Matkin, & Marx, 2007; Carless, 1998). Because of the possible gender-job satisfaction paradox this research controlled for gender.
Kalleberg and Loscocco (1983) found age to be positively related to job satisfaction. Kacmar and Ferris (1989) found in their research besides a linear relationship also a curvilinear association between age and job satisfaction. The curvilinear relationship between age and job satisfaction can be due to one’s career stage, extrinsic job satisfaction, and the linear relationship could be more concerned with intrinsic job satisfaction. Because of the relationship between age and job satisfaction this research controlled for age.
The research of Chye Koh & Boo (2004) indicated significant and positive links between ethical culture constructs and job satisfaction. The items Chye Koh & Boo (2004) used to test the ethical culture constructs where developed by Hunt, Chonko, and Wilcox (1984). They added one additional item to the scale of top management support for ethical
Data and method | 17 behaviour “Top management in my organisation should have higher ethical standards than they do now”. Ethical culture constructs consist of the construct top management supporting the ethical behaviour, which will be named ‘support for ethical behaviour’ and the construct association between ethical behaviour and career success within the organisation, which will be named ‘association between ethical behaviour and career success’. The association between ethical behaviour and career success is an organisational indicator but because it is used in this research as a control variable the difference of the level of analysis will be ignored.
To measure support for ethical behaviour and association between ethical behaviour and career success, the scales of Chye Koh & Boo (2004) were used. Four items were used to measure support for ethical behaviour (Cronbach’s α = .748). One of these items was: “Top management in my organisation has clearly conveyed that unethical behaviour will not be tolerated”. Six items were used to measure association between ethical behaviour and career success (Cronbach’s α = .837). One of these items was: “Successful managers in my organisation are more ethical than unsuccessful managers”. The measurement was conducted with the use of a 5 point Likert-scale ranging from 1 (strongly disagree) to 5 (strongly agree).
The scale of support for ethical behaviour consisted of two counter-indicative items and the scale of association between ethical behaviour and career success consisted of five counter-indicative items.
3.4. Statistical procedure
The data for this study was collected by means of an online survey using Qualtrics software. The survey administration started on November 13th, 2016 and was closed four weeks later on December 10th, 2016. The analyses were performed by making use of the Statistical software Package of Social Sciences (SPSS).
Data and method | 18 First a check of frequencies was performed to check if the data had the format it was expected to have. All respondents who had completed the questionnaire were used in the analyses so there was no missing data. After the recoding of the counter-indicative items scale reliabilities, descriptive statistics, skewness, kurtosis and normality tests were computed. The reliability checks showed us that all scales, except job crafting and support for ethical behavior had a Cronbach’s alpha of >.80, which is good according to the rules of thumb of George & Mallery (2003). Job crafting and support for ethical behavior had Cronbach’s alpha’s of >.70 which is acceptable. Not all the variables were normally distributed. The variables where a bit skewed but there were no kurtosis issues. The absence of a normal distribution in these variables can be explained by the fact that the perceived level of the variables in the sample was not normally distributed. This could carry some problems but regression analyses are quite robust to violations of this kind, so the data was be transformed.
Hierarchical regression analysis was used to examine the direct relationship between the independent variables and the dependent variable. A hierarchical multiple regression was performed to investigate the ability of transformational leadership and passive leadership to predict levels of job satisfaction, after controlling for gender, age, support for ethical behaviour, and association between ethical behaviour and career success.
To test the hypothesized moderation effects between the variables regression analyses were undertaken. In order to test the moderating roles of job crafting and needs-supplies fit the SPSS macro of Preacher & Hayes (2008) was used.
SPSS provided us with a correlation coefficients table, called a correlation matrix, for all of the combinations of variables which is presented in table 1.
Data and method | 19 Table 1: Means, Standard Deviations, Correlations and Reliabilities
Variables Number
of items M SD 1 2 3 4 5 6 7 8 9
1. Age 1 .42 .49 -
2. Gender (0=female, 1=male) 1 24.61 11.82 .109 -
3. Support for ethical behaviour 4 3.30 .66 .066 .032 (.748)
4. Association between ethical behaviour and career success 6 3.40 .73 .016 .097 .572** (.837)
5. Job satisfaction 15 3.63 .39 .100 .051 .364** .399* (.891)
6. Transformational leadership 10 3.54 .66 -.087 .037 .300** .324** .420** (.872)
7. Passive leadership 3 2.38 .95 .100 -.055 -.260** -.232** -.333** -.645** (.822)
8. Job crafting 21 3.28 .32 -.210** -.202** .079 -.095 .156* .248** -.128* (.705)
9. Needs-supplies fit 3 3.60 .83 .162* .074 .402** .427** .739** .403** -.323** .069 (.876)
Note: N = 168. Reliabilities are reported along the diagonal.
* Correlation is significant at the 0.05 level (one-tailed). ** Correlation is significant at the 0.01 level (one-tailed).
Results | 20
4. Results
This chapter will start with a discussion of the correlation matrix (see table 1). After outlining the results of the regression analysis the direct relationships between the variables transformational leadership and passive leadership and employee job satisfaction are discussed. The chapter will end with the discussion of the moderation effects.
4.1. Correlation analysis
The correlation coefficients table 1 presents an overview of the descriptive statistics, correlations and scale reliabilities. A first observation derived from the table is that the reliabilities of all scales have a Cronbach’s α above .7 which is acceptable (Kline, 2013). Both transformational and passive leadership significantly correlated with job satisfaction, job crafting and needs-supplies fit, but the size of the effects was small.
A second observation from the correlation coefficients table is that the independent variables transformational leadership and passive leadership correlated, significantly and strongly, negatively with each other (r = -.645, p < .01).
A final observation is that job satisfaction significantly correlated with both needs-supplies fit and job crafting. The correlation between job satisfaction and need-needs-supplies fit was particularly strong (r = .739, p <.01) compared to the small effect size of the correlation between job satisfaction and job crafting (r = .156, p < .05) (Cohen, 1992). Testing for multicollinearity showed that there were no multicollinearity issues. The tolerance (.683) and VIF (1.465) values of needs-supplies fit did not indicate a problem and were within the boundaries given by Menard (2002) and Myers (2000).
4.2. Direct effects
To test the direct effects of hypothesis H1a and H1b a hierarchical multiple regression was performed. As a first step of the hierarchical multiple regression, four predictors were entered: gender, age, support for ethical behaviour and association between ethical behaviour
Results | 21 and career success. This model was statistically significant F (4, 163) = 9,746; p < .000 and explained 19.3% of variance in job satisfaction. As presented in table 2, transformational leadership was positively related to job satisfaction (β = .268, p < .01). Furthermore, the control variable association between ethical behaviour and career success significantly predicted job satisfaction (β = .207, p < .05).
After entry of transformational leadership and passive leadership at step 2 the total variance explained by the model as a whole was 28.3% F (2, 161) = 10,096, p < .000. The introduction of transformational leadership and passive leadership explained an additional 9% of the variance in satisfaction, after controlling for gender, age, support for ethical behaviour and association between ethical behaviour and career success (R2 Change = .090; F (2 ,161) = 10,096, p < .000). In the final model two out of six predictor variables were statistically significant, with transformational leadership recording a higher Beta value (β = .268, p < .01) than the association between ethical behaviour and career success (β = .207, p < .05). In other words, if transformational leadership increases by one unit employee job satisfaction will increase with .268 units. On the other hand, if association between ethical behaviour and career success increases by one unit, job satisfaction will increase by .207 units. The model could explain 28.3% of the total variance of job satisfaction whereof 9% was explained by transformational leadership and passive leadership.
Hypothesis H1a predicted a positive relationship between perceived transformational leadership and employee job satisfaction. The hypothesis testing provides evidence that job satisfaction is positively related to transformational leadership, supporting hypothesis H1a.
Hypothesis H1b predicted a negative relationship between perceived passive leadership and employee job satisfaction. Hypothesis testing showed that passive leadership had no significant effect on job satisfaction. Thus hypothesis H1b was not supported; the null hypothesis could not be rejected.
Results | 22 The results of the hierarchical multiple regression without the control variables showed no differences. The model was significant F (2,165) = 18.473, p <.000 and explained 18.3% of variance in job satisfaction. Only the predictor variable transformational leadership was statistically significant (β = .352, p < .01).
Table 2: Hierarchical regression model of antecedents of job satisfaction
R R2 R2 Change B SE β t Step 1 .439 .193*** .193*** Gender .008 .071 .008 .118 Age .003 .003 .082 1.150
Support for ethical behaviour .131 .058 .195* 2.266
Association between ethical behaviour and
career success .211 .064 .285** 3.304
Step 2 .532 .283*** .090***
Gender .000 .067 .000 .003
Age .005 .003 .118 1.739
Support for ethical behaviour .093 .056 .137 1.659
Association between ethical behaviour and
career success .154 .062 .207* 2.462
Transformational leadership .213 .066 .286** 3.217
Passive leadership -.029 .046 -.055 -.624
Note: N = 168
Statistical significance: * p<.05; ** p<.01; *** p<.001
Results of this analysis without the control variables showed no differences
4.3. Moderating effects of job crafting and needs-supplies fit
The set of hypotheses H2a, H2b and H3a, H3b proposed that the relationship between the independent variables transformational and passive leadership and the dependent variable employee job satisfaction would be moderated by both job crafting and needs-supplies fit. To test the moderating effects the SPSS macro of Preacher & Hayes (2008) was used. The leadership style that was not included in the model was included as a covariate.
Even though the relationship between passive leadership and employee job satisfaction was not significant this research aimed to test the moderating effect of job crafting and needs-supplies fit on the relationships as suggested in hypothesis H2b and H3b.
The analyses for the moderation effect was started with a model to test both the possible moderation effects of job crafting and needs-supplies on the relationship between the leadership style and job satisfaction at the same time.
Results | 23 To test hypothesis H2a and H3a both the moderating effect of job crafting and needs-supplies fit on the relationship between transformational leadership and job satisfaction was tested after controlling for passive leadership, gender, age, support for ethical behaviour, and association between ethical behaviour and career success. The result of this analysis is presented in table 3. According to the model summary of the moderation analysis the model was significant with F (10,157) = 4.5310; p < .001, R2 = .5871. Looking at the significance of the interaction effects it was observed that neither of the interaction effects were significant according to the model. Job crafting did not significantly moderate the relationship between transformational leadership and job satisfaction (B = .0299; t(157) = .1704; p = .8649) and
needs-supplies fit did not significantly moderate the relationship between transformational leadership and job satisfaction (B = -.0792; t(157) = -1.0792; p = .2821). Thus, hypothesis H2a
and H3a were not supported; the null hypothesis could not be rejected. The results of the examination of the moderating effect of job crafting and needs-supplies fit on the relationship between transformational leadership and job satisfaction without all the control variables and without the control variables but still including passive leadership as a control variable both showed no differences.
To test hypothesis H2b and H3b both the moderating effect of job crafting and needs-supplies fit on the relationship between passive leadership and job satisfaction were tested after controlling for transformational leadership, gender, age, support for ethical behaviour, and association between ethical behaviour and career success. The result of this analysis is presented in table 4. According to the model summary of the moderation analysis the model was significant with F (10,157) = 14.0351; p <.001, R2 = .6084. Looking at the significance of the interaction effects it was observed that both interaction effects were not significant according to the model. Job crafting did not significantly moderate the relationship between passive leadership and job satisfaction (B = -.1123; t(157) = -1.1882; p = .2366) and
needs-Results | 24 supplies fit did not significantly moderate the relationship between passive leadership and job satisfaction (B = .0752; t(157) = 1.9024; p = .0590). Although formally not supported it could
be referred to as being marginally significant. Thus, hypothesis H2b and H3b were not supported; the null hypothesis could not be rejected. The results of the examination of the moderating effects of job crafting and needs-supplies fit on the relationship between passive leadership and job satisfaction without the control variables (B = .0795; t(162) = 2.8028; p =
.0057) and without the control variables but still including transformational leadership as a control variable (B = .0793; t(161) = 2.8058; p = .0056) both showed a significant moderation
effect of needs-supplies fit on the relationship between passive leadership and job satisfaction. In sum, none of the moderation hypotheses (H2a, H2b, H3a, and H3b) were supported. An explanation for this could be that the interplay of the variables and the interaction terms outshone the possible moderation effects. The variance explained by the models was high, so it could be that the predictors explained an overlapping variance in the dependent variable job satisfaction, which could lead to not having moderation effects because this variance was already explained. Because the hypotheses were not mentioning the interplay of the interaction effects different analyses were conducted to see if the moderation effects were present when the effects were isolated. Simple moderation analyses were conducted to test the isolated two-way interaction effects and a three-way interaction effect was tested.
4.4. Isolated moderating effects of job crafting and needs-supplies fit
For hypothesis H2a the isolated moderating effect of job crafting on the relationship between transformational leadership and job satisfaction was examined after controlling for passive leadership, gender, age, support for ethical behaviour, and association between ethical behaviour and career success. The result of this analysis is presented in table 5. According to the model summary of the moderation analysis the model was significant with F (8,159) = 4.5310; p = <.01, R2 = .2295. Looking at the significance of the interaction effect it was
Results | 25 observed that the interaction effect was not significant according to the model, p = 0.7065. Thus, hypothesis H2a, stating that job crafting moderates the relationship between transformational leadership and job satisfaction is also not supported looking at the isolated moderation effect, the null hypothesis could not be rejected. Looking into the isolated moderation effect of job crafting on the relationship between transformational leadership and job satisfaction with the control variables and also controlling for needs-supplies fit showed no differences. The results of the examination of the isolated moderating effect of job crafting on the relationship between transformational leadership and job satisfaction without the control variables also showed no differences.
To test the isolated moderating effects of hypothesis H2b the moderating effect of job crafting on the relationship between passive leadership and job satisfaction were examined after controlling for transformational leadership, gender, age, support for ethical behaviour and association between ethical behaviour and career success. The results of this analysis are presented in table 6. According to the model summary of the moderation analysis the model was significant with F (8, 159) = 7.0654; p <.001, R2 = .3498. If the interaction effect is significant and the direction of the interaction is in line with the hypothesis H2b the hypothesis can be confirmed regardless of the significance of the main effect. Looking at the significance of the interaction effect it was found to be significant (B = -.3098; p= <05). The total variance explained by the model as a whole was 34.98% F (8, 159) = 7.0654; p <.001. If the interaction term significantly increases the amount of variance explained in the criterion variable (i.e. R2 change is significant), then, as proposed in hypothesis H2b, job crafting serves as a moderator of the relationship between passive leadership and job satisfaction. Step 2 in table 4 shows that the interaction term accounted for an additional 5.16% of the variance in job satisfaction, which was significant (R2 Change = .0516; F (1, 159) = 5.7698; p < .001). The method of Johnson-Neyman gives an approach to simple slopes (Field, 2013) (see table
Results | 26 9). With the use of this method the data showed that for employees with low levels of job crafting (up to -.4934) the relationship between passive leadership and job satisfaction is positive and significant (job crafting -.5964, B = .1553, 95% CI [.0115; .2991], t(159) = p <.05).
For employees with high levels of job crafting (higher than .4807 as at .5226 the CI excludes 0) the relationship between passive leadership and job satisfaction becomes negative and significant (job crafting .5226, B = -1914, 95% CI [-.3792; -.0035], t(159) = p <.05). The
direction of the moderation effect is presented in figure 2. As figure 2 shows, employees with high levels of job crafting where more satisfied when they perceived lower levels of passive leadership than employees who perceived higher levels of passive leadership. Employees with low levels of job crafting where less satisfied when they perceived lower levels of passive leadership than employees who perceived higher levels of passive leadership.
Figure 2: Job crafting as a moderator of the passive leadership - job satisfaction relationship Although there is a significant isolated moderating effect, the direction of the interaction is not in line with hypothesis H2b. Results of the isolated moderation analysis do not support hypothesis H2b. Thus, hypothesis H2b, stating that job crafting moderates the negative relationship between passive leadership and job satisfaction so that this relationship
Results | 27 is stronger for lower levels of job crafting is not supported; the null hypothesis could not be rejected. Looking into the isolated moderation effect of job crafting on the relationship between passive leadership and job satisfaction with the control variables and also controlling for needs-supplies fit showed no differences. The results of the examination of the isolated moderating effect of job crafting on the relationship between passive leadership and job satisfaction without the control variables also showed no differences.
For hypothesis H3a, the isolated moderating effect of needs-supplies fit on the relationship between transformational leadership and job satisfaction was examined after controlling for passive leadership gender, age, support for ethical behaviour and association between ethical behaviour and career success. The result of this analysis is presented in table 7. According to the model summary of the moderation analysis the model was significant with F (8,159) = 16.9029; p <.001, R2 = .5769. Looking at the significance of the interaction effect it was found not to be significant according to the model, p = 0.3269. Thus, hypothesis H3a, pertaining needs-supplies fit moderating the relationship between transformational leadership and job satisfaction is not supported looking at the isolated moderation effect, as the null hypothesis could not be rejected. Looking into the isolated moderation effect of needs-supplies fit on the relationship between transformational leadership and job satisfaction with the control variables and also controlling for job crafting showed no differences. The results of the examination of the isolated moderating effect of needs-supplies fit on the relationship between transformational leadership and job satisfaction without the control variables also showed no differences.
To test the isolated moderating effects of hypothesis H3b the moderating effect of needs-supplies fit on the relationship between passive leadership and job satisfaction were examined after controlling for transformational leadership, gender, age, support for ethical behaviour and association between ethical behaviour and career success. The result of this
Results | 28 analysis is presented in table 8. According to the model summary of the moderation analysis the model was significant with F (8, 159) = 13.7150; p <.001, R2 = .5872. If the interaction effect is significant and the direction of the interaction is in line with the hypothesis H3b the hypothesis can be confirmed regardless of the significance of the main effect. Looking at the significance of the interaction effect it was observed that the interaction effect was significant (B = .0731; p <.05). The total variance explained by the model as a whole was 58.72% F (8, 159) = 13.7150; p <.05. If the interaction term significantly increases the amount of variance explained in the criterion variable (i.e. R2 change is significant), then, as proposed in hypothesis H3b, needs-supplies fit serves as a moderator of the relationship between passive leadership and job satisfaction. Step 2 in table 4 shows that the interaction term accounted for an additional 1.81% of the variance in job satisfaction, which was significant (R2 Change = .0181; F (1, 159) = 4.7933; p <.05). Using the Johnson-Neyman method (Field, 2013) (see table 10) to look into the data it showed that for employees with low levels of needs-supplies fit (up to -.4934) the relationship between passive leadership and job satisfaction is negative and significant (needs-supplies fit -2.4032, B = -.1850 95% CI [-.3676; -.0024], t(159) = p
<.05). In other words, if an employee has a needs-supplies fit of -2.4032 and passive leadership increases by one unit job satisfaction will decrease by .200 units. The direction of the moderation effect is presented in figure 3. As displayed in figure 3, employees who perceive a low needs-supplies fit will become less satisfied when they perceive higher levels of passive leadership.
Results | 29 Figure 3: Needs-supplies fit as a moderator of the passive leadership - job satisfaction
relationship
This isolated moderating analyses showed that hypothesis H3b, pertaining to needs-supplies fit moderating the negative relationship between passive leadership and job satisfaction so that this relationship is stronger for lower levels of needs-supplies fit is supported. Looking into the isolated moderation effect of needs-supplies fit on the relationship between passive leadership and job satisfaction with the control variables and also controlling for job crafting showed no differences. The results of the examination of the isolated moderating effect of needs-supplies fit on the relationship between passive leadership and job satisfaction without the control variables also showed no differences.
Based on the significant isolated two-way interaction effect of needs-supplies fit on the relationship between passive leadership and job satisfaction the model was extended with a second moderator job crafting. The SPSS macro of Preacher & Hayes (2008) was used to perform hierarchical regression analyses and subsequently compute the test of individual slopes differences as explained by Dawson and Richter (2006).
Results | 30 The potential three-way interaction effect of needs-supplies fit and job crafting on the relationship between passive leadership and job satisfaction was tested after controlling for transformational leadership, gender, age, support for ethical behaviour, and association between ethical behaviour and career success. None of the interactions were significant according to the model so there were no three-way interaction effects of needs-supplies fit and job crafting on the relationship between passive leadership and job satisfaction. The results of the examination of the three-way interaction effects without the control variables showed no differences.
Results | 31 Table 3: Results of job crafting and needs-supplies fit as moderators of the relationship
between transformational leadership and job satisfaction Model Summary R R-sq MSE F df1 df2 p .7665 .5871 .1060 13.6473 10.0000 157.0000 .0000 Model B SE B t p LLCI ULCI Interaction variables Constant 3.4191 .2045 16.7160 .0000*** 3.0151 3.8231 Job crafting .1794 .1089 1.6474 .1015 -.0357 .3945 Transformational leadership .0657 .0704 .9341 .3517 -.0733 .2048 Job crafting x transformational leadership .0299 .1757 .1704 .8649 -.3171 .3770
Needs-supplies fit .3572 .0465 7.6789 .0000*** .2653 .4490
Needs-supplies fit x transformational leadership -.7092 .0734 -.1.0792 .2821 -.2242 .0658 Control variables
Passive leadership -.0124 .0423 -.2923 .7705 -.0960 .0712
Gender -.0013 .0505 -.0258 .9795 -.1011 .0985
Age .0010 .0026 .3809 .7038 -.0042 .0062
Team manager support -.0065 .0517 .1264 .8996 -.1085 .0955
Association between ethical behaviour and
career success .0768 .0553 1.3896 .1666 -.0324 .1861
Note: N = 168
Statistical significance: * p<.05; ** p<.01; *** p<.001
Results of this analysis without the control variables showed no differences
Table 4: Results of job crafting and needs-supplies fit as moderators of the relationship between passive leadership and job satisfaction
Model Summary R R-sq MSE F df1 df2 p .7800 .6084 .1006 14.0351 10.0000 157.0000 .0000 Model B SE B t p LLCI ULCI Interaction variables 3.1534 .2397 13.1553 .0000*** 2.6799 3.6268 Constant .2189 .0915 2.3924 .0175* .0382 .3997 Job crafting -.0114 .0339 -.3358 .7375 -.0783 .0555 Passive leadership -.1123 .0945 -1.1882 .2366 -.2989 .0744
Job crafting x passive leadership .3391 .0472 7.1882 .0000*** .2459 .4322
Needs-supplies fit .0752 .0396 1.9024 .0590 -.0029 .1534
Needs-supplies fit x passive leadership Control variables
Transformational leadership .0739 .0537 1.3753 .1710 -.0355 .1800
Gender -.0067 .0520 -.1283 .8981 -.1093 .0960
Age .0015 .0025 .5716 .5684 -.0036 .0065
Team manager support .0011 .0503 .0217 .9827 -.0982 .1004
Association between ethical behaviour and
career success .0582 .0527 1.1062 .2703 -.0458 .1622
Note: N = 168
Results | 32 Table 5: Results of job crafting as an isolated moderator of the relationship between
transformational leadership and job satisfaction Model Summary R R-sq MSE F df1 df2 p .5472 .2295 .1776 4.5310 8.000 159.000 .0001 Model B SE B t p LLCI ULCI Interaction variables Constant 2.6954 .2914 9.2502*** .0000 2.1199 3.2709 Job crafting .2335 .1357 1.7201 .0874 -.0346 .5015 Transformational leadership .1849 .0881 2.0986* .0374 .0109 .3588 Job crafting x transformational leadership .0751 .1990 .3773 .7065 -.3179 .4680 Control variables
Passive leadership -.0300 .0594 -.5045 .6146 -.1473 .0874
Gender .0226 .0660 .3427 .7323 -.1078 .1530
Age .0062 .0033 1.8948 .0599 -.0003 .0126
Team manager support .0762 .0677 1.1255 .2621 -.0575 .2100
Association between ethical behaviour and
career success .1757 .0682 2.5748 .0109 .0409 .3104
Note: N = 168
Statistical significance: * p<.05; ** p<.01; *** p<.001
Results of this analysis without the control variables showed no differences
Table 6: Results of job crafting as an isolated moderator of the relationship between passive leadership and job satisfaction
Model Summary R R-sq MSE F df1 df2 p .5914 .3498 .1648 7.0654 8.0000 159.0000 .0000 Model B SE B t p LLCI ULCI Interaction variables Constant 1.9996 .3642 5.4904*** .0000 .1.2803 2.7189 Job crafting .2442 .1141 2.1405* .0338 .0189 .4696 Passive leadership -.0295 .0460 -.6408 .5226 -1.1202 .0613
Job crafting x passive leadership -.3098 .1290 -2.4020* .0175 .5645 -.0551 Control variables
Transformational leadership .1813 .0682 2.6581** .0087 .0466 .3160
Gender -.0072 .0666 -.1074 .9146 -.1387 .1244
Age .0058 .0033 1.7828 .0765 -.0006 .0122
Team manager support .0908 .0615 1.4756 .1420 -.0307 .2124
Association between ethical behaviour
and career success .1595 .0660 2.4160 .0168 .0291 .2899
Step 2
R-square increase due to interaction(s) R2 Change F df1 df2 p
Job crafting x Passive leadership .0516 5.7698 1.000 159.000 .0175
Note: N = 168
Statistical significance: * p<.05; ** p<.01; *** p<.001
